From 54b2f09d6e1c130794ebb85a3db7cba6de43bbdf Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Mon, 3 Oct 2005 12:03:13 -0700 Subject: [PATCH] --- yaml --- r: 9612 b: refs/heads/master c: c2681dd8c71c8fb4ca530f94536550fcd843aae4 h: refs/heads/master v: v3 --- [refs] | 2 +- trunk/Documentation/DocBook/libata.tmpl | 1072 ------- trunk/Documentation/SubmittingPatches | 24 +- trunk/Documentation/networking/ip-sysctl.txt | 10 +- trunk/arch/x86_64/kernel/setup.c | 5 +- trunk/drivers/net/bonding/bond_main.c | 2 +- trunk/drivers/net/cassini.c | 502 ++- trunk/drivers/net/skge.c | 24 +- trunk/drivers/net/tg3.c | 27 +- trunk/drivers/net/wan/sdlamain.c | 23 +- trunk/drivers/net/wan/syncppp.c | 2 +- trunk/drivers/net/wireless/strip.c | 4 +- trunk/drivers/parisc/led.c | 5 +- trunk/drivers/s390/net/qeth_main.c | 4 +- trunk/drivers/scsi/3w-9xxx.c | 55 +- trunk/drivers/scsi/3w-9xxx.h | 17 +- trunk/drivers/scsi/Makefile | 1 + trunk/drivers/scsi/aacraid/aachba.c | 283 +- trunk/drivers/scsi/aacraid/aacraid.h | 17 +- trunk/drivers/scsi/aacraid/comminit.c | 17 +- trunk/drivers/scsi/aacraid/commsup.c | 581 +++- trunk/drivers/scsi/aacraid/linit.c | 12 +- trunk/drivers/scsi/ahci.c | 31 +- trunk/drivers/scsi/aic7xxx/aic7770_osm.c | 3 + trunk/drivers/scsi/aic7xxx/aic79xx_osm.c | 8 +- trunk/drivers/scsi/aic7xxx/aic79xx_osm_pci.c | 3 + trunk/drivers/scsi/aic7xxx/aic7xxx_osm.c | 8 +- trunk/drivers/scsi/aic7xxx/aic7xxx_osm_pci.c | 3 + trunk/drivers/scsi/hosts.c | 2 +- trunk/drivers/scsi/libata-core.c | 433 +-- trunk/drivers/scsi/libata-scsi.c | 491 +-- trunk/drivers/scsi/libata.h | 4 - trunk/drivers/scsi/lpfc/lpfc_attr.c | 8 +- trunk/drivers/scsi/lpfc/lpfc_hbadisc.c | 4 +- trunk/drivers/scsi/lpfc/lpfc_hw.h | 4 +- trunk/drivers/scsi/lpfc/lpfc_init.c | 6 +- trunk/drivers/scsi/megaraid.c | 70 +- trunk/drivers/scsi/megaraid/Kconfig.megaraid | 9 + trunk/drivers/scsi/megaraid/Makefile | 1 + trunk/drivers/scsi/megaraid/megaraid_sas.c | 2805 +++++++++++++++++ trunk/drivers/scsi/megaraid/megaraid_sas.h | 1142 +++++++ trunk/drivers/scsi/qla2xxx/qla_rscn.c | 2 + trunk/drivers/scsi/sata_mv.c | 1142 ++----- trunk/drivers/scsi/sata_nv.c | 2 +- trunk/drivers/scsi/sata_promise.c | 6 +- trunk/drivers/scsi/sata_sis.c | 2 +- trunk/drivers/scsi/sata_uli.c | 2 +- trunk/drivers/scsi/sata_via.c | 2 +- trunk/drivers/scsi/scsi_scan.c | 96 +- trunk/drivers/scsi/scsi_transport_sas.c | 9 +- trunk/include/linux/ata.h | 20 +- trunk/include/linux/inetdevice.h | 12 +- trunk/include/linux/ipv6.h | 5 +- trunk/include/linux/libata.h | 34 +- trunk/include/linux/pci_ids.h | 2 + trunk/include/linux/skbuff.h | 12 +- trunk/include/linux/tc_ematch/tc_em_meta.h | 2 +- trunk/include/net/inet6_hashtables.h | 21 +- trunk/include/net/inet_hashtables.h | 64 +- trunk/include/net/inet_timewait_sock.h | 2 +- trunk/include/net/sock.h | 5 +- trunk/include/scsi/scsi_device.h | 1 + trunk/net/atm/clip.c | 2 +- trunk/net/atm/common.c | 2 +- trunk/net/core/netpoll.c | 2 +- trunk/net/core/pktgen.c | 3 +- trunk/net/core/skbuff.c | 5 - trunk/net/dccp/ipv4.c | 12 +- trunk/net/econet/af_econet.c | 2 +- trunk/net/ipv4/arp.c | 21 +- trunk/net/ipv4/devinet.c | 22 +- trunk/net/ipv4/fib_frontend.c | 4 +- trunk/net/ipv4/fib_semantics.c | 4 +- trunk/net/ipv4/icmp.c | 2 +- trunk/net/ipv4/igmp.c | 2 +- trunk/net/ipv4/inet_timewait_sock.c | 6 +- trunk/net/ipv4/ip_gre.c | 4 +- trunk/net/ipv4/ipmr.c | 6 +- .../ipv4/netfilter/ip_conntrack_netbios_ns.c | 2 +- trunk/net/ipv4/netfilter/ip_queue.c | 4 +- trunk/net/ipv4/netfilter/ipt_REDIRECT.c | 2 +- trunk/net/ipv4/netfilter/ipt_ULOG.c | 4 +- trunk/net/ipv4/route.c | 6 +- trunk/net/ipv4/tcp_ipv4.c | 11 +- trunk/net/ipv6/addrconf.c | 2 +- trunk/net/ipv6/ip6_output.c | 2 +- trunk/net/ipv6/netfilter/ip6_queue.c | 4 +- trunk/net/ipv6/tcp_ipv6.c | 18 +- trunk/net/ipv6/udp.c | 14 +- trunk/net/irda/irlan/irlan_eth.c | 2 +- trunk/net/netfilter/nfnetlink_log.c | 4 +- trunk/net/netfilter/nfnetlink_queue.c | 4 +- trunk/net/packet/af_packet.c | 4 +- trunk/net/sched/em_meta.c | 6 +- trunk/net/sctp/protocol.c | 2 +- trunk/net/sysctl_net.c | 2 + 96 files changed, 5846 insertions(+), 3513 deletions(-) create mode 100644 trunk/drivers/scsi/megaraid/megaraid_sas.c create mode 100644 trunk/drivers/scsi/megaraid/megaraid_sas.h diff --git a/[refs] b/[refs] index c9de928835f2..29b182e64ecf 100644 --- a/[refs] +++ b/[refs] @@ -1,2 +1,2 @@ --- -refs/heads/master: e12669e774be403c84baa651306d305752a35cd8 +refs/heads/master: c2681dd8c71c8fb4ca530f94536550fcd843aae4 diff --git a/trunk/Documentation/DocBook/libata.tmpl b/trunk/Documentation/DocBook/libata.tmpl index d260d92089ad..375ae760dc1e 100644 --- a/trunk/Documentation/DocBook/libata.tmpl +++ b/trunk/Documentation/DocBook/libata.tmpl @@ -415,362 +415,6 @@ and other resources, etc. - - Error handling - - - This chapter describes how errors are handled under libata. - Readers are advised to read SCSI EH - (Documentation/scsi/scsi_eh.txt) and ATA exceptions doc first. - - - Origins of commands - - In libata, a command is represented with struct ata_queued_cmd - or qc. qc's are preallocated during port initialization and - repetitively used for command executions. Currently only one - qc is allocated per port but yet-to-be-merged NCQ branch - allocates one for each tag and maps each qc to NCQ tag 1-to-1. - - - libata commands can originate from two sources - libata itself - and SCSI midlayer. libata internal commands are used for - initialization and error handling. All normal blk requests - and commands for SCSI emulation are passed as SCSI commands - through queuecommand callback of SCSI host template. - - - - How commands are issued - - - - Internal commands - - - First, qc is allocated and initialized using - ata_qc_new_init(). Although ata_qc_new_init() doesn't - implement any wait or retry mechanism when qc is not - available, internal commands are currently issued only during - initialization and error recovery, so no other command is - active and allocation is guaranteed to succeed. - - - Once allocated qc's taskfile is initialized for the command to - be executed. qc currently has two mechanisms to notify - completion. One is via qc->complete_fn() callback and the - other is completion qc->waiting. qc->complete_fn() callback - is the asynchronous path used by normal SCSI translated - commands and qc->waiting is the synchronous (issuer sleeps in - process context) path used by internal commands. - - - Once initialization is complete, host_set lock is acquired - and the qc is issued. - - - - - SCSI commands - - - All libata drivers use ata_scsi_queuecmd() as - hostt->queuecommand callback. scmds can either be simulated - or translated. No qc is involved in processing a simulated - scmd. The result is computed right away and the scmd is - completed. - - - For a translated scmd, ata_qc_new_init() is invoked to - allocate a qc and the scmd is translated into the qc. SCSI - midlayer's completion notification function pointer is stored - into qc->scsidone. - - - qc->complete_fn() callback is used for completion - notification. ATA commands use ata_scsi_qc_complete() while - ATAPI commands use atapi_qc_complete(). Both functions end up - calling qc->scsidone to notify upper layer when the qc is - finished. After translation is completed, the qc is issued - with ata_qc_issue(). - - - Note that SCSI midlayer invokes hostt->queuecommand while - holding host_set lock, so all above occur while holding - host_set lock. - - - - - - - - How commands are processed - - Depending on which protocol and which controller are used, - commands are processed differently. For the purpose of - discussion, a controller which uses taskfile interface and all - standard callbacks is assumed. - - - Currently 6 ATA command protocols are used. They can be - sorted into the following four categories according to how - they are processed. - - - - ATA NO DATA or DMA - - - ATA_PROT_NODATA and ATA_PROT_DMA fall into this category. - These types of commands don't require any software - intervention once issued. Device will raise interrupt on - completion. - - - - - ATA PIO - - - ATA_PROT_PIO is in this category. libata currently - implements PIO with polling. ATA_NIEN bit is set to turn - off interrupt and pio_task on ata_wq performs polling and - IO. - - - - - ATAPI NODATA or DMA - - - ATA_PROT_ATAPI_NODATA and ATA_PROT_ATAPI_DMA are in this - category. packet_task is used to poll BSY bit after - issuing PACKET command. Once BSY is turned off by the - device, packet_task transfers CDB and hands off processing - to interrupt handler. - - - - - ATAPI PIO - - - ATA_PROT_ATAPI is in this category. ATA_NIEN bit is set - and, as in ATAPI NODATA or DMA, packet_task submits cdb. - However, after submitting cdb, further processing (data - transfer) is handed off to pio_task. - - - - - - - How commands are completed - - Once issued, all qc's are either completed with - ata_qc_complete() or time out. For commands which are handled - by interrupts, ata_host_intr() invokes ata_qc_complete(), and, - for PIO tasks, pio_task invokes ata_qc_complete(). In error - cases, packet_task may also complete commands. - - - ata_qc_complete() does the following. - - - - - - - DMA memory is unmapped. - - - - - - ATA_QCFLAG_ACTIVE is clared from qc->flags. - - - - - - qc->complete_fn() callback is invoked. If the return value of - the callback is not zero. Completion is short circuited and - ata_qc_complete() returns. - - - - - - __ata_qc_complete() is called, which does - - - - - qc->flags is cleared to zero. - - - - - - ap->active_tag and qc->tag are poisoned. - - - - - - qc->waiting is claread & completed (in that order). - - - - - - qc is deallocated by clearing appropriate bit in ap->qactive. - - - - - - - - - - - So, it basically notifies upper layer and deallocates qc. One - exception is short-circuit path in #3 which is used by - atapi_qc_complete(). - - - For all non-ATAPI commands, whether it fails or not, almost - the same code path is taken and very little error handling - takes place. A qc is completed with success status if it - succeeded, with failed status otherwise. - - - However, failed ATAPI commands require more handling as - REQUEST SENSE is needed to acquire sense data. If an ATAPI - command fails, ata_qc_complete() is invoked with error status, - which in turn invokes atapi_qc_complete() via - qc->complete_fn() callback. - - - This makes atapi_qc_complete() set scmd->result to - SAM_STAT_CHECK_CONDITION, complete the scmd and return 1. As - the sense data is empty but scmd->result is CHECK CONDITION, - SCSI midlayer will invoke EH for the scmd, and returning 1 - makes ata_qc_complete() to return without deallocating the qc. - This leads us to ata_scsi_error() with partially completed qc. - - - - - ata_scsi_error() - - ata_scsi_error() is the current hostt->eh_strategy_handler() - for libata. As discussed above, this will be entered in two - cases - timeout and ATAPI error completion. This function - calls low level libata driver's eng_timeout() callback, the - standard callback for which is ata_eng_timeout(). It checks - if a qc is active and calls ata_qc_timeout() on the qc if so. - Actual error handling occurs in ata_qc_timeout(). - - - If EH is invoked for timeout, ata_qc_timeout() stops BMDMA and - completes the qc. Note that as we're currently in EH, we - cannot call scsi_done. As described in SCSI EH doc, a - recovered scmd should be either retried with - scsi_queue_insert() or finished with scsi_finish_command(). - Here, we override qc->scsidone with scsi_finish_command() and - calls ata_qc_complete(). - - - If EH is invoked due to a failed ATAPI qc, the qc here is - completed but not deallocated. The purpose of this - half-completion is to use the qc as place holder to make EH - code reach this place. This is a bit hackish, but it works. - - - Once control reaches here, the qc is deallocated by invoking - __ata_qc_complete() explicitly. Then, internal qc for REQUEST - SENSE is issued. Once sense data is acquired, scmd is - finished by directly invoking scsi_finish_command() on the - scmd. Note that as we already have completed and deallocated - the qc which was associated with the scmd, we don't need - to/cannot call ata_qc_complete() again. - - - - - Problems with the current EH - - - - - - Error representation is too crude. Currently any and all - error conditions are represented with ATA STATUS and ERROR - registers. Errors which aren't ATA device errors are treated - as ATA device errors by setting ATA_ERR bit. Better error - descriptor which can properly represent ATA and other - errors/exceptions is needed. - - - - - - When handling timeouts, no action is taken to make device - forget about the timed out command and ready for new commands. - - - - - - EH handling via ata_scsi_error() is not properly protected - from usual command processing. On EH entrance, the device is - not in quiescent state. Timed out commands may succeed or - fail any time. pio_task and atapi_task may still be running. - - - - - - Too weak error recovery. Devices / controllers causing HSM - mismatch errors and other errors quite often require reset to - return to known state. Also, advanced error handling is - necessary to support features like NCQ and hotplug. - - - - - - ATA errors are directly handled in the interrupt handler and - PIO errors in pio_task. This is problematic for advanced - error handling for the following reasons. - - - First, advanced error handling often requires context and - internal qc execution. - - - Second, even a simple failure (say, CRC error) needs - information gathering and could trigger complex error handling - (say, resetting & reconfiguring). Having multiple code - paths to gather information, enter EH and trigger actions - makes life painful. - - - Third, scattered EH code makes implementing low level drivers - difficult. Low level drivers override libata callbacks. If - EH is scattered over several places, each affected callbacks - should perform its part of error handling. This can be error - prone and painful. - - - - - - - libata Library !Edrivers/scsi/libata-core.c @@ -787,722 +431,6 @@ and other resources, etc. !Idrivers/scsi/libata-scsi.c - - ATA errors & exceptions - - - This chapter tries to identify what error/exception conditions exist - for ATA/ATAPI devices and describe how they should be handled in - implementation-neutral way. - - - - The term 'error' is used to describe conditions where either an - explicit error condition is reported from device or a command has - timed out. - - - - The term 'exception' is either used to describe exceptional - conditions which are not errors (say, power or hotplug events), or - to describe both errors and non-error exceptional conditions. Where - explicit distinction between error and exception is necessary, the - term 'non-error exception' is used. - - - - Exception categories - - Exceptions are described primarily with respect to legacy - taskfile + bus master IDE interface. If a controller provides - other better mechanism for error reporting, mapping those into - categories described below shouldn't be difficult. - - - - In the following sections, two recovery actions - reset and - reconfiguring transport - are mentioned. These are described - further in . - - - - HSM violation - - This error is indicated when STATUS value doesn't match HSM - requirement during issuing or excution any ATA/ATAPI command. - - - - Examples - - - - ATA_STATUS doesn't contain !BSY && DRDY && !DRQ while trying - to issue a command. - - - - - - !BSY && !DRQ during PIO data transfer. - - - - - - DRQ on command completion. - - - - - - !BSY && ERR after CDB tranfer starts but before the - last byte of CDB is transferred. ATA/ATAPI standard states - that "The device shall not terminate the PACKET command - with an error before the last byte of the command packet has - been written" in the error outputs description of PACKET - command and the state diagram doesn't include such - transitions. - - - - - - - In these cases, HSM is violated and not much information - regarding the error can be acquired from STATUS or ERROR - register. IOW, this error can be anything - driver bug, - faulty device, controller and/or cable. - - - - As HSM is violated, reset is necessary to restore known state. - Reconfiguring transport for lower speed might be helpful too - as transmission errors sometimes cause this kind of errors. - - - - - ATA/ATAPI device error (non-NCQ / non-CHECK CONDITION) - - - These are errors detected and reported by ATA/ATAPI devices - indicating device problems. For this type of errors, STATUS - and ERROR register values are valid and describe error - condition. Note that some of ATA bus errors are detected by - ATA/ATAPI devices and reported using the same mechanism as - device errors. Those cases are described later in this - section. - - - - For ATA commands, this type of errors are indicated by !BSY - && ERR during command execution and on completion. - - - For ATAPI commands, - - - - - - !BSY && ERR && ABRT right after issuing PACKET - indicates that PACKET command is not supported and falls in - this category. - - - - - - !BSY && ERR(==CHK) && !ABRT after the last - byte of CDB is transferred indicates CHECK CONDITION and - doesn't fall in this category. - - - - - - !BSY && ERR(==CHK) && ABRT after the last byte - of CDB is transferred *probably* indicates CHECK CONDITION and - doesn't fall in this category. - - - - - - - Of errors detected as above, the followings are not ATA/ATAPI - device errors but ATA bus errors and should be handled - according to . - - - - - - CRC error during data transfer - - - This is indicated by ICRC bit in the ERROR register and - means that corruption occurred during data transfer. Upto - ATA/ATAPI-7, the standard specifies that this bit is only - applicable to UDMA transfers but ATA/ATAPI-8 draft revision - 1f says that the bit may be applicable to multiword DMA and - PIO. - - - - - - ABRT error during data transfer or on completion - - - Upto ATA/ATAPI-7, the standard specifies that ABRT could be - set on ICRC errors and on cases where a device is not able - to complete a command. Combined with the fact that MWDMA - and PIO transfer errors aren't allowed to use ICRC bit upto - ATA/ATAPI-7, it seems to imply that ABRT bit alone could - indicate tranfer errors. - - - However, ATA/ATAPI-8 draft revision 1f removes the part - that ICRC errors can turn on ABRT. So, this is kind of - gray area. Some heuristics are needed here. - - - - - - - - ATA/ATAPI device errors can be further categorized as follows. - - - - - - Media errors - - - This is indicated by UNC bit in the ERROR register. ATA - devices reports UNC error only after certain number of - retries cannot recover the data, so there's nothing much - else to do other than notifying upper layer. - - - READ and WRITE commands report CHS or LBA of the first - failed sector but ATA/ATAPI standard specifies that the - amount of transferred data on error completion is - indeterminate, so we cannot assume that sectors preceding - the failed sector have been transferred and thus cannot - complete those sectors successfully as SCSI does. - - - - - - Media changed / media change requested error - - - <<TODO: fill here>> - - - - - Address error - - - This is indicated by IDNF bit in the ERROR register. - Report to upper layer. - - - - - Other errors - - - This can be invalid command or parameter indicated by ABRT - ERROR bit or some other error condition. Note that ABRT - bit can indicate a lot of things including ICRC and Address - errors. Heuristics needed. - - - - - - - - Depending on commands, not all STATUS/ERROR bits are - applicable. These non-applicable bits are marked with - "na" in the output descriptions but upto ATA/ATAPI-7 - no definition of "na" can be found. However, - ATA/ATAPI-8 draft revision 1f describes "N/A" as - follows. - - -
- - 3.2.3.3a N/A - - - A keyword the indicates a field has no defined value in - this standard and should not be checked by the host or - device. N/A fields should be cleared to zero. - - - - -
- - - So, it seems reasonable to assume that "na" bits are - cleared to zero by devices and thus need no explicit masking. - - - - - - ATAPI device CHECK CONDITION - - - ATAPI device CHECK CONDITION error is indicated by set CHK bit - (ERR bit) in the STATUS register after the last byte of CDB is - transferred for a PACKET command. For this kind of errors, - sense data should be acquired to gather information regarding - the errors. REQUEST SENSE packet command should be used to - acquire sense data. - - - - Once sense data is acquired, this type of errors can be - handled similary to other SCSI errors. Note that sense data - may indicate ATA bus error (e.g. Sense Key 04h HARDWARE ERROR - && ASC/ASCQ 47h/00h SCSI PARITY ERROR). In such - cases, the error should be considered as an ATA bus error and - handled according to . - - - - - - ATA device error (NCQ) - - - NCQ command error is indicated by cleared BSY and set ERR bit - during NCQ command phase (one or more NCQ commands - outstanding). Although STATUS and ERROR registers will - contain valid values describing the error, READ LOG EXT is - required to clear the error condition, determine which command - has failed and acquire more information. - - - - READ LOG EXT Log Page 10h reports which tag has failed and - taskfile register values describing the error. With this - information the failed command can be handled as a normal ATA - command error as in and all - other in-flight commands must be retried. Note that this - retry should not be counted - it's likely that commands - retried this way would have completed normally if it were not - for the failed command. - - - - Note that ATA bus errors can be reported as ATA device NCQ - errors. This should be handled as described in . - - - - If READ LOG EXT Log Page 10h fails or reports NQ, we're - thoroughly screwed. This condition should be treated - according to . - - - - - - ATA bus error - - - ATA bus error means that data corruption occurred during - transmission over ATA bus (SATA or PATA). This type of errors - can be indicated by - - - - - - - ICRC or ABRT error as described in . - - - - - - Controller-specific error completion with error information - indicating transmission error. - - - - - - On some controllers, command timeout. In this case, there may - be a mechanism to determine that the timeout is due to - transmission error. - - - - - - Unknown/random errors, timeouts and all sorts of weirdities. - - - - - - - As described above, transmission errors can cause wide variety - of symptoms ranging from device ICRC error to random device - lockup, and, for many cases, there is no way to tell if an - error condition is due to transmission error or not; - therefore, it's necessary to employ some kind of heuristic - when dealing with errors and timeouts. For example, - encountering repetitive ABRT errors for known supported - command is likely to indicate ATA bus error. - - - - Once it's determined that ATA bus errors have possibly - occurred, lowering ATA bus transmission speed is one of - actions which may alleviate the problem. See for more information. - - - - - - PCI bus error - - - Data corruption or other failures during transmission over PCI - (or other system bus). For standard BMDMA, this is indicated - by Error bit in the BMDMA Status register. This type of - errors must be logged as it indicates something is very wrong - with the system. Resetting host controller is recommended. - - - - - - Late completion - - - This occurs when timeout occurs and the timeout handler finds - out that the timed out command has completed successfully or - with error. This is usually caused by lost interrupts. This - type of errors must be logged. Resetting host controller is - recommended. - - - - - - Unknown error (timeout) - - - This is when timeout occurs and the command is still - processing or the host and device are in unknown state. When - this occurs, HSM could be in any valid or invalid state. To - bring the device to known state and make it forget about the - timed out command, resetting is necessary. The timed out - command may be retried. - - - - Timeouts can also be caused by transmission errors. Refer to - for more details. - - - - - - Hotplug and power management exceptions - - - <<TODO: fill here>> - - - - - - - - EH recovery actions - - - This section discusses several important recovery actions. - - - - Clearing error condition - - - Many controllers require its error registers to be cleared by - error handler. Different controllers may have different - requirements. - - - - For SATA, it's strongly recommended to clear at least SError - register during error handling. - - - - - Reset - - - During EH, resetting is necessary in the following cases. - - - - - - - HSM is in unknown or invalid state - - - - - - HBA is in unknown or invalid state - - - - - - EH needs to make HBA/device forget about in-flight commands - - - - - - HBA/device behaves weirdly - - - - - - - Resetting during EH might be a good idea regardless of error - condition to improve EH robustness. Whether to reset both or - either one of HBA and device depends on situation but the - following scheme is recommended. - - - - - - - When it's known that HBA is in ready state but ATA/ATAPI - device in in unknown state, reset only device. - - - - - - If HBA is in unknown state, reset both HBA and device. - - - - - - - HBA resetting is implementation specific. For a controller - complying to taskfile/BMDMA PCI IDE, stopping active DMA - transaction may be sufficient iff BMDMA state is the only HBA - context. But even mostly taskfile/BMDMA PCI IDE complying - controllers may have implementation specific requirements and - mechanism to reset themselves. This must be addressed by - specific drivers. - - - - OTOH, ATA/ATAPI standard describes in detail ways to reset - ATA/ATAPI devices. - - - - - PATA hardware reset - - - This is hardware initiated device reset signalled with - asserted PATA RESET- signal. There is no standard way to - initiate hardware reset from software although some - hardware provides registers that allow driver to directly - tweak the RESET- signal. - - - - - Software reset - - - This is achieved by turning CONTROL SRST bit on for at - least 5us. Both PATA and SATA support it but, in case of - SATA, this may require controller-specific support as the - second Register FIS to clear SRST should be transmitted - while BSY bit is still set. Note that on PATA, this resets - both master and slave devices on a channel. - - - - - EXECUTE DEVICE DIAGNOSTIC command - - - Although ATA/ATAPI standard doesn't describe exactly, EDD - implies some level of resetting, possibly similar level - with software reset. Host-side EDD protocol can be handled - with normal command processing and most SATA controllers - should be able to handle EDD's just like other commands. - As in software reset, EDD affects both devices on a PATA - bus. - - - Although EDD does reset devices, this doesn't suit error - handling as EDD cannot be issued while BSY is set and it's - unclear how it will act when device is in unknown/weird - state. - - - - - ATAPI DEVICE RESET command - - - This is very similar to software reset except that reset - can be restricted to the selected device without affecting - the other device sharing the cable. - - - - - SATA phy reset - - - This is the preferred way of resetting a SATA device. In - effect, it's identical to PATA hardware reset. Note that - this can be done with the standard SCR Control register. - As such, it's usually easier to implement than software - reset. - - - - - - - - One more thing to consider when resetting devices is that - resetting clears certain configuration parameters and they - need to be set to their previous or newly adjusted values - after reset. - - - - Parameters affected are. - - - - - - - CHS set up with INITIALIZE DEVICE PARAMETERS (seldomly used) - - - - - - Parameters set with SET FEATURES including transfer mode setting - - - - - - Block count set with SET MULTIPLE MODE - - - - - - Other parameters (SET MAX, MEDIA LOCK...) - - - - - - - ATA/ATAPI standard specifies that some parameters must be - maintained across hardware or software reset, but doesn't - strictly specify all of them. Always reconfiguring needed - parameters after reset is required for robustness. Note that - this also applies when resuming from deep sleep (power-off). - - - - Also, ATA/ATAPI standard requires that IDENTIFY DEVICE / - IDENTIFY PACKET DEVICE is issued after any configuration - parameter is updated or a hardware reset and the result used - for further operation. OS driver is required to implement - revalidation mechanism to support this. - - - - - - Reconfigure transport - - - For both PATA and SATA, a lot of corners are cut for cheap - connectors, cables or controllers and it's quite common to see - high transmission error rate. This can be mitigated by - lowering transmission speed. - - - - The following is a possible scheme Jeff Garzik suggested. - - -
- - If more than $N (3?) transmission errors happen in 15 minutes, - - - - - if SATA, decrease SATA PHY speed. if speed cannot be decreased, - - - - - decrease UDMA xfer speed. if at UDMA0, switch to PIO4, - - - - - decrease PIO xfer speed. if at PIO3, complain, but continue - - - -
- - - - - - - ata_piix Internals !Idrivers/scsi/ata_piix.c diff --git a/trunk/Documentation/SubmittingPatches b/trunk/Documentation/SubmittingPatches index 1d96efec5e8f..237d54c44bc5 100644 --- a/trunk/Documentation/SubmittingPatches +++ b/trunk/Documentation/SubmittingPatches @@ -305,7 +305,7 @@ point out some special detail about the sign-off. The canonical patch subject line is: - Subject: [PATCH 001/123] [:] + Subject: [PATCH 001/123] subsystem: summary phrase The canonical patch message body contains the following: @@ -330,9 +330,25 @@ alphabetically by subject line - pretty much any email reader will support that - since because the sequence number is zero-padded, the numerical and alphabetic sort is the same. -See further details on how to phrase the "" in the -"Subject:" line in Andrew Morton's "The perfect patch", referenced -below. +The "subsystem" in the email's Subject should identify which +area or subsystem of the kernel is being patched. + +The "summary phrase" in the email's Subject should concisely +describe the patch which that email contains. The "summary +phrase" should not be a filename. Do not use the same "summary +phrase" for every patch in a whole patch series. + +Bear in mind that the "summary phrase" of your email becomes +a globally-unique identifier for that patch. It propagates +all the way into the git changelog. The "summary phrase" may +later be used in developer discussions which refer to the patch. +People will want to google for the "summary phrase" to read +discussion regarding that patch. + +A couple of example Subjects: + + Subject: [patch 2/5] ext2: improve scalability of bitmap searching + Subject: [PATCHv2 001/207] x86: fix eflags tracking The "from" line must be the very first line in the message body, and has the form: diff --git a/trunk/Documentation/networking/ip-sysctl.txt b/trunk/Documentation/networking/ip-sysctl.txt index ab65714d95fc..b433c8a27e2d 100644 --- a/trunk/Documentation/networking/ip-sysctl.txt +++ b/trunk/Documentation/networking/ip-sysctl.txt @@ -355,10 +355,14 @@ ip_dynaddr - BOOLEAN Default: 0 icmp_echo_ignore_all - BOOLEAN + If set non-zero, then the kernel will ignore all ICMP ECHO + requests sent to it. + Default: 0 + icmp_echo_ignore_broadcasts - BOOLEAN - If either is set to true, then the kernel will ignore either all - ICMP ECHO requests sent to it or just those to broadcast/multicast - addresses, respectively. + If set non-zero, then the kernel will ignore all ICMP ECHO and + TIMESTAMP requests sent to it via broadcast/multicast. + Default: 1 icmp_ratelimit - INTEGER Limit the maximal rates for sending ICMP packets whose type matches diff --git a/trunk/arch/x86_64/kernel/setup.c b/trunk/arch/x86_64/kernel/setup.c index 257f5ba17902..cb28df14ff6f 100644 --- a/trunk/arch/x86_64/kernel/setup.c +++ b/trunk/arch/x86_64/kernel/setup.c @@ -967,13 +967,12 @@ static int __cpuinit intel_num_cpu_cores(struct cpuinfo_x86 *c) static void srat_detect_node(void) { #ifdef CONFIG_NUMA - unsigned apicid, node; + unsigned node; int cpu = smp_processor_id(); /* Don't do the funky fallback heuristics the AMD version employs for now. */ - apicid = phys_proc_id[cpu]; - node = apicid_to_node[apicid]; + node = apicid_to_node[hard_smp_processor_id()]; if (node == NUMA_NO_NODE) node = 0; cpu_to_node[cpu] = node; diff --git a/trunk/drivers/net/bonding/bond_main.c b/trunk/drivers/net/bonding/bond_main.c index 6d00c3de1a83..bf81cd45e4d4 100644 --- a/trunk/drivers/net/bonding/bond_main.c +++ b/trunk/drivers/net/bonding/bond_main.c @@ -2776,7 +2776,7 @@ static u32 bond_glean_dev_ip(struct net_device *dev) return 0; rcu_read_lock(); - idev = __in_dev_get(dev); + idev = __in_dev_get_rcu(dev); if (!idev) goto out; diff --git a/trunk/drivers/net/cassini.c b/trunk/drivers/net/cassini.c index 45831fb377a0..2e617424d3fb 100644 --- a/trunk/drivers/net/cassini.c +++ b/trunk/drivers/net/cassini.c @@ -4423,18 +4423,14 @@ static struct { #define CAS_REG_LEN (sizeof(ethtool_register_table)/sizeof(int)) #define CAS_MAX_REGS (sizeof (u32)*CAS_REG_LEN) -static u8 *cas_get_regs(struct cas *cp) +static void cas_read_regs(struct cas *cp, u8 *ptr, int len) { - u8 *ptr = kmalloc(CAS_MAX_REGS, GFP_KERNEL); u8 *p; int i; unsigned long flags; - if (!ptr) - return NULL; - spin_lock_irqsave(&cp->lock, flags); - for (i = 0, p = ptr; i < CAS_REG_LEN ; i ++, p += sizeof(u32)) { + for (i = 0, p = ptr; i < len ; i ++, p += sizeof(u32)) { u16 hval; u32 val; if (ethtool_register_table[i].offsets < 0) { @@ -4447,8 +4443,6 @@ static u8 *cas_get_regs(struct cas *cp) memcpy(p, (u8 *)&val, sizeof(u32)); } spin_unlock_irqrestore(&cp->lock, flags); - - return ptr; } static struct net_device_stats *cas_get_stats(struct net_device *dev) @@ -4561,316 +4555,251 @@ static void cas_set_multicast(struct net_device *dev) spin_unlock_irqrestore(&cp->lock, flags); } -/* Eventually add support for changing the advertisement - * on autoneg. - */ -static int cas_ethtool_ioctl(struct net_device *dev, void __user *ep_user) +static void cas_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) +{ + struct cas *cp = netdev_priv(dev); + strncpy(info->driver, DRV_MODULE_NAME, ETHTOOL_BUSINFO_LEN); + strncpy(info->version, DRV_MODULE_VERSION, ETHTOOL_BUSINFO_LEN); + info->fw_version[0] = '\0'; + strncpy(info->bus_info, pci_name(cp->pdev), ETHTOOL_BUSINFO_LEN); + info->regdump_len = cp->casreg_len < CAS_MAX_REGS ? + cp->casreg_len : CAS_MAX_REGS; + info->n_stats = CAS_NUM_STAT_KEYS; +} + +static int cas_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) { struct cas *cp = netdev_priv(dev); u16 bmcr; int full_duplex, speed, pause; - struct ethtool_cmd ecmd; unsigned long flags; enum link_state linkstate = link_up; - if (copy_from_user(&ecmd, ep_user, sizeof(ecmd))) - return -EFAULT; - - switch(ecmd.cmd) { - case ETHTOOL_GDRVINFO: { - struct ethtool_drvinfo info = { .cmd = ETHTOOL_GDRVINFO }; - - strncpy(info.driver, DRV_MODULE_NAME, - ETHTOOL_BUSINFO_LEN); - strncpy(info.version, DRV_MODULE_VERSION, - ETHTOOL_BUSINFO_LEN); - info.fw_version[0] = '\0'; - strncpy(info.bus_info, pci_name(cp->pdev), - ETHTOOL_BUSINFO_LEN); - info.regdump_len = cp->casreg_len < CAS_MAX_REGS ? - cp->casreg_len : CAS_MAX_REGS; - info.n_stats = CAS_NUM_STAT_KEYS; - if (copy_to_user(ep_user, &info, sizeof(info))) - return -EFAULT; - - return 0; + cmd->advertising = 0; + cmd->supported = SUPPORTED_Autoneg; + if (cp->cas_flags & CAS_FLAG_1000MB_CAP) { + cmd->supported |= SUPPORTED_1000baseT_Full; + cmd->advertising |= ADVERTISED_1000baseT_Full; } - case ETHTOOL_GSET: - ecmd.advertising = 0; - ecmd.supported = SUPPORTED_Autoneg; - if (cp->cas_flags & CAS_FLAG_1000MB_CAP) { - ecmd.supported |= SUPPORTED_1000baseT_Full; - ecmd.advertising |= ADVERTISED_1000baseT_Full; + /* Record PHY settings if HW is on. */ + spin_lock_irqsave(&cp->lock, flags); + bmcr = 0; + linkstate = cp->lstate; + if (CAS_PHY_MII(cp->phy_type)) { + cmd->port = PORT_MII; + cmd->transceiver = (cp->cas_flags & CAS_FLAG_SATURN) ? + XCVR_INTERNAL : XCVR_EXTERNAL; + cmd->phy_address = cp->phy_addr; + cmd->advertising |= ADVERTISED_TP | ADVERTISED_MII | + ADVERTISED_10baseT_Half | + ADVERTISED_10baseT_Full | + ADVERTISED_100baseT_Half | + ADVERTISED_100baseT_Full; + + cmd->supported |= + (SUPPORTED_10baseT_Half | + SUPPORTED_10baseT_Full | + SUPPORTED_100baseT_Half | + SUPPORTED_100baseT_Full | + SUPPORTED_TP | SUPPORTED_MII); + + if (cp->hw_running) { + cas_mif_poll(cp, 0); + bmcr = cas_phy_read(cp, MII_BMCR); + cas_read_mii_link_mode(cp, &full_duplex, + &speed, &pause); + cas_mif_poll(cp, 1); } - /* Record PHY settings if HW is on. */ - spin_lock_irqsave(&cp->lock, flags); - bmcr = 0; - linkstate = cp->lstate; - if (CAS_PHY_MII(cp->phy_type)) { - ecmd.port = PORT_MII; - ecmd.transceiver = (cp->cas_flags & CAS_FLAG_SATURN) ? - XCVR_INTERNAL : XCVR_EXTERNAL; - ecmd.phy_address = cp->phy_addr; - ecmd.advertising |= ADVERTISED_TP | ADVERTISED_MII | - ADVERTISED_10baseT_Half | - ADVERTISED_10baseT_Full | - ADVERTISED_100baseT_Half | - ADVERTISED_100baseT_Full; - - ecmd.supported |= - (SUPPORTED_10baseT_Half | - SUPPORTED_10baseT_Full | - SUPPORTED_100baseT_Half | - SUPPORTED_100baseT_Full | - SUPPORTED_TP | SUPPORTED_MII); - - if (cp->hw_running) { - cas_mif_poll(cp, 0); - bmcr = cas_phy_read(cp, MII_BMCR); - cas_read_mii_link_mode(cp, &full_duplex, - &speed, &pause); - cas_mif_poll(cp, 1); - } - - } else { - ecmd.port = PORT_FIBRE; - ecmd.transceiver = XCVR_INTERNAL; - ecmd.phy_address = 0; - ecmd.supported |= SUPPORTED_FIBRE; - ecmd.advertising |= ADVERTISED_FIBRE; - - if (cp->hw_running) { - /* pcs uses the same bits as mii */ - bmcr = readl(cp->regs + REG_PCS_MII_CTRL); - cas_read_pcs_link_mode(cp, &full_duplex, - &speed, &pause); - } + } else { + cmd->port = PORT_FIBRE; + cmd->transceiver = XCVR_INTERNAL; + cmd->phy_address = 0; + cmd->supported |= SUPPORTED_FIBRE; + cmd->advertising |= ADVERTISED_FIBRE; + + if (cp->hw_running) { + /* pcs uses the same bits as mii */ + bmcr = readl(cp->regs + REG_PCS_MII_CTRL); + cas_read_pcs_link_mode(cp, &full_duplex, + &speed, &pause); } - spin_unlock_irqrestore(&cp->lock, flags); + } + spin_unlock_irqrestore(&cp->lock, flags); - if (bmcr & BMCR_ANENABLE) { - ecmd.advertising |= ADVERTISED_Autoneg; - ecmd.autoneg = AUTONEG_ENABLE; - ecmd.speed = ((speed == 10) ? - SPEED_10 : - ((speed == 1000) ? - SPEED_1000 : SPEED_100)); - ecmd.duplex = full_duplex ? DUPLEX_FULL : DUPLEX_HALF; + if (bmcr & BMCR_ANENABLE) { + cmd->advertising |= ADVERTISED_Autoneg; + cmd->autoneg = AUTONEG_ENABLE; + cmd->speed = ((speed == 10) ? + SPEED_10 : + ((speed == 1000) ? + SPEED_1000 : SPEED_100)); + cmd->duplex = full_duplex ? DUPLEX_FULL : DUPLEX_HALF; + } else { + cmd->autoneg = AUTONEG_DISABLE; + cmd->speed = + (bmcr & CAS_BMCR_SPEED1000) ? + SPEED_1000 : + ((bmcr & BMCR_SPEED100) ? SPEED_100: + SPEED_10); + cmd->duplex = + (bmcr & BMCR_FULLDPLX) ? + DUPLEX_FULL : DUPLEX_HALF; + } + if (linkstate != link_up) { + /* Force these to "unknown" if the link is not up and + * autonogotiation in enabled. We can set the link + * speed to 0, but not cmd->duplex, + * because its legal values are 0 and 1. Ethtool will + * print the value reported in parentheses after the + * word "Unknown" for unrecognized values. + * + * If in forced mode, we report the speed and duplex + * settings that we configured. + */ + if (cp->link_cntl & BMCR_ANENABLE) { + cmd->speed = 0; + cmd->duplex = 0xff; } else { - ecmd.autoneg = AUTONEG_DISABLE; - ecmd.speed = - (bmcr & CAS_BMCR_SPEED1000) ? - SPEED_1000 : - ((bmcr & BMCR_SPEED100) ? SPEED_100: - SPEED_10); - ecmd.duplex = - (bmcr & BMCR_FULLDPLX) ? - DUPLEX_FULL : DUPLEX_HALF; - } - if (linkstate != link_up) { - /* Force these to "unknown" if the link is not up and - * autonogotiation in enabled. We can set the link - * speed to 0, but not ecmd.duplex, - * because its legal values are 0 and 1. Ethtool will - * print the value reported in parentheses after the - * word "Unknown" for unrecognized values. - * - * If in forced mode, we report the speed and duplex - * settings that we configured. - */ - if (cp->link_cntl & BMCR_ANENABLE) { - ecmd.speed = 0; - ecmd.duplex = 0xff; - } else { - ecmd.speed = SPEED_10; - if (cp->link_cntl & BMCR_SPEED100) { - ecmd.speed = SPEED_100; - } else if (cp->link_cntl & CAS_BMCR_SPEED1000) { - ecmd.speed = SPEED_1000; - } - ecmd.duplex = (cp->link_cntl & BMCR_FULLDPLX)? - DUPLEX_FULL : DUPLEX_HALF; + cmd->speed = SPEED_10; + if (cp->link_cntl & BMCR_SPEED100) { + cmd->speed = SPEED_100; + } else if (cp->link_cntl & CAS_BMCR_SPEED1000) { + cmd->speed = SPEED_1000; } + cmd->duplex = (cp->link_cntl & BMCR_FULLDPLX)? + DUPLEX_FULL : DUPLEX_HALF; } - if (copy_to_user(ep_user, &ecmd, sizeof(ecmd))) - return -EFAULT; - return 0; + } + return 0; +} - case ETHTOOL_SSET: - if (!capable(CAP_NET_ADMIN)) - return -EPERM; +static int cas_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) +{ + struct cas *cp = netdev_priv(dev); + unsigned long flags; - /* Verify the settings we care about. */ - if (ecmd.autoneg != AUTONEG_ENABLE && - ecmd.autoneg != AUTONEG_DISABLE) - return -EINVAL; + /* Verify the settings we care about. */ + if (cmd->autoneg != AUTONEG_ENABLE && + cmd->autoneg != AUTONEG_DISABLE) + return -EINVAL; - if (ecmd.autoneg == AUTONEG_DISABLE && - ((ecmd.speed != SPEED_1000 && - ecmd.speed != SPEED_100 && - ecmd.speed != SPEED_10) || - (ecmd.duplex != DUPLEX_HALF && - ecmd.duplex != DUPLEX_FULL))) - return -EINVAL; + if (cmd->autoneg == AUTONEG_DISABLE && + ((cmd->speed != SPEED_1000 && + cmd->speed != SPEED_100 && + cmd->speed != SPEED_10) || + (cmd->duplex != DUPLEX_HALF && + cmd->duplex != DUPLEX_FULL))) + return -EINVAL; - /* Apply settings and restart link process. */ - spin_lock_irqsave(&cp->lock, flags); - cas_begin_auto_negotiation(cp, &ecmd); - spin_unlock_irqrestore(&cp->lock, flags); - return 0; + /* Apply settings and restart link process. */ + spin_lock_irqsave(&cp->lock, flags); + cas_begin_auto_negotiation(cp, cmd); + spin_unlock_irqrestore(&cp->lock, flags); + return 0; +} - case ETHTOOL_NWAY_RST: - if ((cp->link_cntl & BMCR_ANENABLE) == 0) - return -EINVAL; +static int cas_nway_reset(struct net_device *dev) +{ + struct cas *cp = netdev_priv(dev); + unsigned long flags; - /* Restart link process. */ - spin_lock_irqsave(&cp->lock, flags); - cas_begin_auto_negotiation(cp, NULL); - spin_unlock_irqrestore(&cp->lock, flags); + if ((cp->link_cntl & BMCR_ANENABLE) == 0) + return -EINVAL; - return 0; + /* Restart link process. */ + spin_lock_irqsave(&cp->lock, flags); + cas_begin_auto_negotiation(cp, NULL); + spin_unlock_irqrestore(&cp->lock, flags); - case ETHTOOL_GWOL: - case ETHTOOL_SWOL: - break; /* doesn't exist */ + return 0; +} - /* get link status */ - case ETHTOOL_GLINK: { - struct ethtool_value edata = { .cmd = ETHTOOL_GLINK }; +static u32 cas_get_link(struct net_device *dev) +{ + struct cas *cp = netdev_priv(dev); + return cp->lstate == link_up; +} - edata.data = (cp->lstate == link_up); - if (copy_to_user(ep_user, &edata, sizeof(edata))) - return -EFAULT; - return 0; - } +static u32 cas_get_msglevel(struct net_device *dev) +{ + struct cas *cp = netdev_priv(dev); + return cp->msg_enable; +} - /* get message-level */ - case ETHTOOL_GMSGLVL: { - struct ethtool_value edata = { .cmd = ETHTOOL_GMSGLVL }; +static void cas_set_msglevel(struct net_device *dev, u32 value) +{ + struct cas *cp = netdev_priv(dev); + cp->msg_enable = value; +} - edata.data = cp->msg_enable; - if (copy_to_user(ep_user, &edata, sizeof(edata))) - return -EFAULT; - return 0; - } +static int cas_get_regs_len(struct net_device *dev) +{ + struct cas *cp = netdev_priv(dev); + return cp->casreg_len < CAS_MAX_REGS ? cp->casreg_len: CAS_MAX_REGS; +} - /* set message-level */ - case ETHTOOL_SMSGLVL: { - struct ethtool_value edata; +static void cas_get_regs(struct net_device *dev, struct ethtool_regs *regs, + void *p) +{ + struct cas *cp = netdev_priv(dev); + regs->version = 0; + /* cas_read_regs handles locks (cp->lock). */ + cas_read_regs(cp, p, regs->len / sizeof(u32)); +} - if (!capable(CAP_NET_ADMIN)) { - return (-EPERM); - } - if (copy_from_user(&edata, ep_user, sizeof(edata))) - return -EFAULT; - cp->msg_enable = edata.data; - return 0; - } +static int cas_get_stats_count(struct net_device *dev) +{ + return CAS_NUM_STAT_KEYS; +} - case ETHTOOL_GREGS: { - struct ethtool_regs edata; - u8 *ptr; - int len = cp->casreg_len < CAS_MAX_REGS ? - cp->casreg_len: CAS_MAX_REGS; - - if (copy_from_user(&edata, ep_user, sizeof (edata))) - return -EFAULT; - - if (edata.len > len) - edata.len = len; - edata.version = 0; - if (copy_to_user (ep_user, &edata, sizeof(edata))) - return -EFAULT; - - /* cas_get_regs handles locks (cp->lock). */ - ptr = cas_get_regs(cp); - if (ptr == NULL) - return -ENOMEM; - if (copy_to_user(ep_user + sizeof (edata), ptr, edata.len)) - return -EFAULT; - - kfree(ptr); - return (0); - } - case ETHTOOL_GSTRINGS: { - struct ethtool_gstrings edata; - int len; - - if (copy_from_user(&edata, ep_user, sizeof(edata))) - return -EFAULT; - - len = edata.len; - switch(edata.string_set) { - case ETH_SS_STATS: - edata.len = (len < CAS_NUM_STAT_KEYS) ? - len : CAS_NUM_STAT_KEYS; - if (copy_to_user(ep_user, &edata, sizeof(edata))) - return -EFAULT; - - if (copy_to_user(ep_user + sizeof(edata), - ðtool_cassini_statnames, - (edata.len * ETH_GSTRING_LEN))) - return -EFAULT; - return 0; - default: - return -EINVAL; - } - } - case ETHTOOL_GSTATS: { - int i = 0; - u64 *tmp; - struct ethtool_stats edata; - struct net_device_stats *stats; - int len; - - if (copy_from_user(&edata, ep_user, sizeof(edata))) - return -EFAULT; - - len = edata.n_stats; - stats = cas_get_stats(cp->dev); - edata.cmd = ETHTOOL_GSTATS; - edata.n_stats = (len < CAS_NUM_STAT_KEYS) ? - len : CAS_NUM_STAT_KEYS; - if (copy_to_user(ep_user, &edata, sizeof (edata))) - return -EFAULT; - - tmp = kmalloc(sizeof(u64)*CAS_NUM_STAT_KEYS, GFP_KERNEL); - if (tmp) { - tmp[i++] = stats->collisions; - tmp[i++] = stats->rx_bytes; - tmp[i++] = stats->rx_crc_errors; - tmp[i++] = stats->rx_dropped; - tmp[i++] = stats->rx_errors; - tmp[i++] = stats->rx_fifo_errors; - tmp[i++] = stats->rx_frame_errors; - tmp[i++] = stats->rx_length_errors; - tmp[i++] = stats->rx_over_errors; - tmp[i++] = stats->rx_packets; - tmp[i++] = stats->tx_aborted_errors; - tmp[i++] = stats->tx_bytes; - tmp[i++] = stats->tx_dropped; - tmp[i++] = stats->tx_errors; - tmp[i++] = stats->tx_fifo_errors; - tmp[i++] = stats->tx_packets; - BUG_ON(i != CAS_NUM_STAT_KEYS); - - i = copy_to_user(ep_user + sizeof(edata), - tmp, sizeof(u64)*edata.n_stats); - kfree(tmp); - } else { - return -ENOMEM; - } - if (i) - return -EFAULT; - return 0; - } - } +static void cas_get_strings(struct net_device *dev, u32 stringset, u8 *data) +{ + memcpy(data, ðtool_cassini_statnames, + CAS_NUM_STAT_KEYS * ETH_GSTRING_LEN); +} - return -EOPNOTSUPP; +static void cas_get_ethtool_stats(struct net_device *dev, + struct ethtool_stats *estats, u64 *data) +{ + struct cas *cp = netdev_priv(dev); + struct net_device_stats *stats = cas_get_stats(cp->dev); + int i = 0; + data[i++] = stats->collisions; + data[i++] = stats->rx_bytes; + data[i++] = stats->rx_crc_errors; + data[i++] = stats->rx_dropped; + data[i++] = stats->rx_errors; + data[i++] = stats->rx_fifo_errors; + data[i++] = stats->rx_frame_errors; + data[i++] = stats->rx_length_errors; + data[i++] = stats->rx_over_errors; + data[i++] = stats->rx_packets; + data[i++] = stats->tx_aborted_errors; + data[i++] = stats->tx_bytes; + data[i++] = stats->tx_dropped; + data[i++] = stats->tx_errors; + data[i++] = stats->tx_fifo_errors; + data[i++] = stats->tx_packets; + BUG_ON(i != CAS_NUM_STAT_KEYS); } +static struct ethtool_ops cas_ethtool_ops = { + .get_drvinfo = cas_get_drvinfo, + .get_settings = cas_get_settings, + .set_settings = cas_set_settings, + .nway_reset = cas_nway_reset, + .get_link = cas_get_link, + .get_msglevel = cas_get_msglevel, + .set_msglevel = cas_set_msglevel, + .get_regs_len = cas_get_regs_len, + .get_regs = cas_get_regs, + .get_stats_count = cas_get_stats_count, + .get_strings = cas_get_strings, + .get_ethtool_stats = cas_get_ethtool_stats, +}; + static int cas_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) { struct cas *cp = netdev_priv(dev); @@ -4883,10 +4812,6 @@ static int cas_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) */ down(&cp->pm_sem); switch (cmd) { - case SIOCETHTOOL: - rc = cas_ethtool_ioctl(dev, ifr->ifr_data); - break; - case SIOCGMIIPHY: /* Get address of MII PHY in use. */ data->phy_id = cp->phy_addr; /* Fallthrough... */ @@ -5112,6 +5037,7 @@ static int __devinit cas_init_one(struct pci_dev *pdev, dev->get_stats = cas_get_stats; dev->set_multicast_list = cas_set_multicast; dev->do_ioctl = cas_ioctl; + dev->ethtool_ops = &cas_ethtool_ops; dev->tx_timeout = cas_tx_timeout; dev->watchdog_timeo = CAS_TX_TIMEOUT; dev->change_mtu = cas_change_mtu; diff --git a/trunk/drivers/net/skge.c b/trunk/drivers/net/skge.c index fd398da4993b..c2e6484ef138 100644 --- a/trunk/drivers/net/skge.c +++ b/trunk/drivers/net/skge.c @@ -2837,21 +2837,29 @@ static void skge_netpoll(struct net_device *dev) static int skge_set_mac_address(struct net_device *dev, void *p) { struct skge_port *skge = netdev_priv(dev); - struct sockaddr *addr = p; - int err = 0; + struct skge_hw *hw = skge->hw; + unsigned port = skge->port; + const struct sockaddr *addr = p; if (!is_valid_ether_addr(addr->sa_data)) return -EADDRNOTAVAIL; - skge_down(dev); + spin_lock_bh(&hw->phy_lock); memcpy(dev->dev_addr, addr->sa_data, ETH_ALEN); - memcpy_toio(skge->hw->regs + B2_MAC_1 + skge->port*8, + memcpy_toio(hw->regs + B2_MAC_1 + port*8, dev->dev_addr, ETH_ALEN); - memcpy_toio(skge->hw->regs + B2_MAC_2 + skge->port*8, + memcpy_toio(hw->regs + B2_MAC_2 + port*8, dev->dev_addr, ETH_ALEN); - if (dev->flags & IFF_UP) - err = skge_up(dev); - return err; + + if (hw->chip_id == CHIP_ID_GENESIS) + xm_outaddr(hw, port, XM_SA, dev->dev_addr); + else { + gma_set_addr(hw, port, GM_SRC_ADDR_1L, dev->dev_addr); + gma_set_addr(hw, port, GM_SRC_ADDR_2L, dev->dev_addr); + } + spin_unlock_bh(&hw->phy_lock); + + return 0; } static const struct { diff --git a/trunk/drivers/net/tg3.c b/trunk/drivers/net/tg3.c index 25f85fb9df46..1802c3b48799 100644 --- a/trunk/drivers/net/tg3.c +++ b/trunk/drivers/net/tg3.c @@ -67,8 +67,8 @@ #define DRV_MODULE_NAME "tg3" #define PFX DRV_MODULE_NAME ": " -#define DRV_MODULE_VERSION "3.41" -#define DRV_MODULE_RELDATE "September 27, 2005" +#define DRV_MODULE_VERSION "3.42" +#define DRV_MODULE_RELDATE "Oct 3, 2005" #define TG3_DEF_MAC_MODE 0 #define TG3_DEF_RX_MODE 0 @@ -9284,8 +9284,8 @@ static int __devinit tg3_get_invariants(struct tg3 *tp) static struct pci_device_id write_reorder_chipsets[] = { { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_FE_GATE_700C) }, - { PCI_DEVICE(PCI_VENDOR_ID_AMD, - PCI_DEVICE_ID_AMD_K8_NB) }, + { PCI_DEVICE(PCI_VENDOR_ID_VIA, + PCI_DEVICE_ID_VIA_8385_0) }, { }, }; u32 misc_ctrl_reg; @@ -9300,15 +9300,6 @@ static int __devinit tg3_get_invariants(struct tg3 *tp) tp->tg3_flags2 |= TG3_FLG2_SUN_570X; #endif - /* If we have an AMD 762 or K8 chipset, write - * reordering to the mailbox registers done by the host - * controller can cause major troubles. We read back from - * every mailbox register write to force the writes to be - * posted to the chip in order. - */ - if (pci_dev_present(write_reorder_chipsets)) - tp->tg3_flags |= TG3_FLAG_MBOX_WRITE_REORDER; - /* Force memory write invalidate off. If we leave it on, * then on 5700_BX chips we have to enable a workaround. * The workaround is to set the TG3PCI_DMA_RW_CTRL boundary @@ -9439,6 +9430,16 @@ static int __devinit tg3_get_invariants(struct tg3 *tp) if (pci_find_capability(tp->pdev, PCI_CAP_ID_EXP) != 0) tp->tg3_flags2 |= TG3_FLG2_PCI_EXPRESS; + /* If we have an AMD 762 or VIA K8T800 chipset, write + * reordering to the mailbox registers done by the host + * controller can cause major troubles. We read back from + * every mailbox register write to force the writes to be + * posted to the chip in order. + */ + if (pci_dev_present(write_reorder_chipsets) && + !(tp->tg3_flags2 & TG3_FLG2_PCI_EXPRESS)) + tp->tg3_flags |= TG3_FLAG_MBOX_WRITE_REORDER; + if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5703 && tp->pci_lat_timer < 64) { tp->pci_lat_timer = 64; diff --git a/trunk/drivers/net/wan/sdlamain.c b/trunk/drivers/net/wan/sdlamain.c index 74e151acef3e..7a8b22a7ea31 100644 --- a/trunk/drivers/net/wan/sdlamain.c +++ b/trunk/drivers/net/wan/sdlamain.c @@ -57,6 +57,7 @@ #include /* request_region(), release_region() */ #include /* WAN router definitions */ #include /* WANPIPE common user API definitions */ +#include #include #include /* phys_to_virt() */ @@ -1268,37 +1269,41 @@ unsigned long get_ip_address(struct net_device *dev, int option) struct in_ifaddr *ifaddr; struct in_device *in_dev; + unsigned long addr = 0; - if ((in_dev = __in_dev_get(dev)) == NULL){ - return 0; + rcu_read_lock(); + if ((in_dev = __in_dev_get_rcu(dev)) == NULL){ + goto out; } if ((ifaddr = in_dev->ifa_list)== NULL ){ - return 0; + goto out; } switch (option){ case WAN_LOCAL_IP: - return ifaddr->ifa_local; + addr = ifaddr->ifa_local; break; case WAN_POINTOPOINT_IP: - return ifaddr->ifa_address; + addr = ifaddr->ifa_address; break; case WAN_NETMASK_IP: - return ifaddr->ifa_mask; + addr = ifaddr->ifa_mask; break; case WAN_BROADCAST_IP: - return ifaddr->ifa_broadcast; + addr = ifaddr->ifa_broadcast; break; default: - return 0; + break; } - return 0; +out: + rcu_read_unlock(); + return addr; } void add_gateway(sdla_t *card, struct net_device *dev) diff --git a/trunk/drivers/net/wan/syncppp.c b/trunk/drivers/net/wan/syncppp.c index b56a7b516d24..a6d3b55013a5 100644 --- a/trunk/drivers/net/wan/syncppp.c +++ b/trunk/drivers/net/wan/syncppp.c @@ -769,7 +769,7 @@ static void sppp_cisco_input (struct sppp *sp, struct sk_buff *skb) u32 addr = 0, mask = ~0; /* FIXME: is the mask correct? */ #ifdef CONFIG_INET rcu_read_lock(); - if ((in_dev = __in_dev_get(dev)) != NULL) + if ((in_dev = __in_dev_get_rcu(dev)) != NULL) { for (ifa=in_dev->ifa_list; ifa != NULL; ifa=ifa->ifa_next) { diff --git a/trunk/drivers/net/wireless/strip.c b/trunk/drivers/net/wireless/strip.c index 4b0acae22b0d..7bc7fc823128 100644 --- a/trunk/drivers/net/wireless/strip.c +++ b/trunk/drivers/net/wireless/strip.c @@ -1352,7 +1352,7 @@ static unsigned char *strip_make_packet(unsigned char *buffer, struct in_device *in_dev; rcu_read_lock(); - in_dev = __in_dev_get(strip_info->dev); + in_dev = __in_dev_get_rcu(strip_info->dev); if (in_dev == NULL) { rcu_read_unlock(); return NULL; @@ -1508,7 +1508,7 @@ static void strip_send(struct strip *strip_info, struct sk_buff *skb) brd = addr = 0; rcu_read_lock(); - in_dev = __in_dev_get(strip_info->dev); + in_dev = __in_dev_get_rcu(strip_info->dev); if (in_dev) { if (in_dev->ifa_list) { brd = in_dev->ifa_list->ifa_broadcast; diff --git a/trunk/drivers/parisc/led.c b/trunk/drivers/parisc/led.c index e90fb72a6962..286902298e33 100644 --- a/trunk/drivers/parisc/led.c +++ b/trunk/drivers/parisc/led.c @@ -37,6 +37,7 @@ #include #include #include +#include #include #include #include @@ -358,9 +359,10 @@ static __inline__ int led_get_net_activity(void) /* we are running as tasklet, so locking dev_base * for reading should be OK */ read_lock(&dev_base_lock); + rcu_read_lock(); for (dev = dev_base; dev; dev = dev->next) { struct net_device_stats *stats; - struct in_device *in_dev = __in_dev_get(dev); + struct in_device *in_dev = __in_dev_get_rcu(dev); if (!in_dev || !in_dev->ifa_list) continue; if (LOOPBACK(in_dev->ifa_list->ifa_local)) @@ -371,6 +373,7 @@ static __inline__ int led_get_net_activity(void) rx_total += stats->rx_packets; tx_total += stats->tx_packets; } + rcu_read_unlock(); read_unlock(&dev_base_lock); retval = 0; diff --git a/trunk/drivers/s390/net/qeth_main.c b/trunk/drivers/s390/net/qeth_main.c index 86582cf1e19e..71de834ece1a 100644 --- a/trunk/drivers/s390/net/qeth_main.c +++ b/trunk/drivers/s390/net/qeth_main.c @@ -5200,7 +5200,7 @@ qeth_free_vlan_addresses4(struct qeth_card *card, unsigned short vid) if (!card->vlangrp) return; rcu_read_lock(); - in_dev = __in_dev_get(card->vlangrp->vlan_devices[vid]); + in_dev = __in_dev_get_rcu(card->vlangrp->vlan_devices[vid]); if (!in_dev) goto out; for (ifa = in_dev->ifa_list; ifa; ifa = ifa->ifa_next) { @@ -7725,7 +7725,7 @@ qeth_arp_constructor(struct neighbour *neigh) goto out; rcu_read_lock(); - in_dev = rcu_dereference(__in_dev_get(dev)); + in_dev = __in_dev_get_rcu(dev); if (in_dev == NULL) { rcu_read_unlock(); return -EINVAL; diff --git a/trunk/drivers/scsi/3w-9xxx.c b/trunk/drivers/scsi/3w-9xxx.c index a6ac61611f35..a748fbfb6692 100644 --- a/trunk/drivers/scsi/3w-9xxx.c +++ b/trunk/drivers/scsi/3w-9xxx.c @@ -60,6 +60,7 @@ Remove un-needed eh_abort handler. Add support for embedded firmware error strings. 2.26.02.003 - Correctly handle single sgl's with use_sg=1. + 2.26.02.004 - Add support for 9550SX controllers. */ #include @@ -82,7 +83,7 @@ #include "3w-9xxx.h" /* Globals */ -#define TW_DRIVER_VERSION "2.26.02.003" +#define TW_DRIVER_VERSION "2.26.02.004" static TW_Device_Extension *twa_device_extension_list[TW_MAX_SLOT]; static unsigned int twa_device_extension_count; static int twa_major = -1; @@ -892,11 +893,6 @@ static int twa_decode_bits(TW_Device_Extension *tw_dev, u32 status_reg_value) writel(TW_CONTROL_CLEAR_QUEUE_ERROR, TW_CONTROL_REG_ADDR(tw_dev)); } - if (status_reg_value & TW_STATUS_SBUF_WRITE_ERROR) { - TW_PRINTK(tw_dev->host, TW_DRIVER, 0xf, "SBUF Write Error: clearing"); - writel(TW_CONTROL_CLEAR_SBUF_WRITE_ERROR, TW_CONTROL_REG_ADDR(tw_dev)); - } - if (status_reg_value & TW_STATUS_MICROCONTROLLER_ERROR) { if (tw_dev->reset_print == 0) { TW_PRINTK(tw_dev->host, TW_DRIVER, 0x10, "Microcontroller Error: clearing"); @@ -930,6 +926,36 @@ static int twa_empty_response_queue(TW_Device_Extension *tw_dev) return retval; } /* End twa_empty_response_queue() */ +/* This function will clear the pchip/response queue on 9550SX */ +static int twa_empty_response_queue_large(TW_Device_Extension *tw_dev) +{ + u32 status_reg_value, response_que_value; + int count = 0, retval = 1; + + if (tw_dev->tw_pci_dev->device == PCI_DEVICE_ID_3WARE_9550SX) { + status_reg_value = readl(TW_STATUS_REG_ADDR(tw_dev)); + + while (((status_reg_value & TW_STATUS_RESPONSE_QUEUE_EMPTY) == 0) && (count < TW_MAX_RESPONSE_DRAIN)) { + response_que_value = readl(TW_RESPONSE_QUEUE_REG_ADDR_LARGE(tw_dev)); + if ((response_que_value & TW_9550SX_DRAIN_COMPLETED) == TW_9550SX_DRAIN_COMPLETED) { + /* P-chip settle time */ + msleep(500); + retval = 0; + goto out; + } + status_reg_value = readl(TW_STATUS_REG_ADDR(tw_dev)); + count++; + } + if (count == TW_MAX_RESPONSE_DRAIN) + goto out; + + retval = 0; + } else + retval = 0; +out: + return retval; +} /* End twa_empty_response_queue_large() */ + /* This function passes sense keys from firmware to scsi layer */ static int twa_fill_sense(TW_Device_Extension *tw_dev, int request_id, int copy_sense, int print_host) { @@ -1613,8 +1639,16 @@ static int twa_reset_sequence(TW_Device_Extension *tw_dev, int soft_reset) int tries = 0, retval = 1, flashed = 0, do_soft_reset = soft_reset; while (tries < TW_MAX_RESET_TRIES) { - if (do_soft_reset) + if (do_soft_reset) { TW_SOFT_RESET(tw_dev); + /* Clear pchip/response queue on 9550SX */ + if (twa_empty_response_queue_large(tw_dev)) { + TW_PRINTK(tw_dev->host, TW_DRIVER, 0x36, "Response queue (large) empty failed during reset sequence"); + do_soft_reset = 1; + tries++; + continue; + } + } /* Make sure controller is in a good state */ if (twa_poll_status(tw_dev, TW_STATUS_MICROCONTROLLER_READY | (do_soft_reset == 1 ? TW_STATUS_ATTENTION_INTERRUPT : 0), 60)) { @@ -2034,7 +2068,10 @@ static int __devinit twa_probe(struct pci_dev *pdev, const struct pci_device_id goto out_free_device_extension; } - mem_addr = pci_resource_start(pdev, 1); + if (pdev->device == PCI_DEVICE_ID_3WARE_9000) + mem_addr = pci_resource_start(pdev, 1); + else + mem_addr = pci_resource_start(pdev, 2); /* Save base address */ tw_dev->base_addr = ioremap(mem_addr, PAGE_SIZE); @@ -2148,6 +2185,8 @@ static void twa_remove(struct pci_dev *pdev) static struct pci_device_id twa_pci_tbl[] __devinitdata = { { PCI_VENDOR_ID_3WARE, PCI_DEVICE_ID_3WARE_9000, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, + { PCI_VENDOR_ID_3WARE, PCI_DEVICE_ID_3WARE_9550SX, + PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, { } }; MODULE_DEVICE_TABLE(pci, twa_pci_tbl); diff --git a/trunk/drivers/scsi/3w-9xxx.h b/trunk/drivers/scsi/3w-9xxx.h index 8c8ecbed3b58..46f22cdc8298 100644 --- a/trunk/drivers/scsi/3w-9xxx.h +++ b/trunk/drivers/scsi/3w-9xxx.h @@ -267,7 +267,6 @@ static twa_message_type twa_error_table[] = { #define TW_CONTROL_CLEAR_PARITY_ERROR 0x00800000 #define TW_CONTROL_CLEAR_QUEUE_ERROR 0x00400000 #define TW_CONTROL_CLEAR_PCI_ABORT 0x00100000 -#define TW_CONTROL_CLEAR_SBUF_WRITE_ERROR 0x00000008 /* Status register bit definitions */ #define TW_STATUS_MAJOR_VERSION_MASK 0xF0000000 @@ -285,9 +284,8 @@ static twa_message_type twa_error_table[] = { #define TW_STATUS_MICROCONTROLLER_READY 0x00002000 #define TW_STATUS_COMMAND_QUEUE_EMPTY 0x00001000 #define TW_STATUS_EXPECTED_BITS 0x00002000 -#define TW_STATUS_UNEXPECTED_BITS 0x00F00008 -#define TW_STATUS_SBUF_WRITE_ERROR 0x00000008 -#define TW_STATUS_VALID_INTERRUPT 0x00DF0008 +#define TW_STATUS_UNEXPECTED_BITS 0x00F00000 +#define TW_STATUS_VALID_INTERRUPT 0x00DF0000 /* RESPONSE QUEUE BIT DEFINITIONS */ #define TW_RESPONSE_ID_MASK 0x00000FF0 @@ -324,9 +322,9 @@ static twa_message_type twa_error_table[] = { /* Compatibility defines */ #define TW_9000_ARCH_ID 0x5 -#define TW_CURRENT_DRIVER_SRL 28 -#define TW_CURRENT_DRIVER_BUILD 9 -#define TW_CURRENT_DRIVER_BRANCH 4 +#define TW_CURRENT_DRIVER_SRL 30 +#define TW_CURRENT_DRIVER_BUILD 80 +#define TW_CURRENT_DRIVER_BRANCH 0 /* Phase defines */ #define TW_PHASE_INITIAL 0 @@ -334,6 +332,7 @@ static twa_message_type twa_error_table[] = { #define TW_PHASE_SGLIST 2 /* Misc defines */ +#define TW_9550SX_DRAIN_COMPLETED 0xFFFF #define TW_SECTOR_SIZE 512 #define TW_ALIGNMENT_9000 4 /* 4 bytes */ #define TW_ALIGNMENT_9000_SGL 0x3 @@ -417,6 +416,9 @@ static twa_message_type twa_error_table[] = { #ifndef PCI_DEVICE_ID_3WARE_9000 #define PCI_DEVICE_ID_3WARE_9000 0x1002 #endif +#ifndef PCI_DEVICE_ID_3WARE_9550SX +#define PCI_DEVICE_ID_3WARE_9550SX 0x1003 +#endif /* Bitmask macros to eliminate bitfields */ @@ -443,6 +445,7 @@ static twa_message_type twa_error_table[] = { #define TW_STATUS_REG_ADDR(x) ((unsigned char __iomem *)x->base_addr + 0x4) #define TW_COMMAND_QUEUE_REG_ADDR(x) (sizeof(dma_addr_t) > 4 ? ((unsigned char __iomem *)x->base_addr + 0x20) : ((unsigned char __iomem *)x->base_addr + 0x8)) #define TW_RESPONSE_QUEUE_REG_ADDR(x) ((unsigned char __iomem *)x->base_addr + 0xC) +#define TW_RESPONSE_QUEUE_REG_ADDR_LARGE(x) ((unsigned char __iomem *)x->base_addr + 0x30) #define TW_CLEAR_ALL_INTERRUPTS(x) (writel(TW_STATUS_VALID_INTERRUPT, TW_CONTROL_REG_ADDR(x))) #define TW_CLEAR_ATTENTION_INTERRUPT(x) (writel(TW_CONTROL_CLEAR_ATTENTION_INTERRUPT, TW_CONTROL_REG_ADDR(x))) #define TW_CLEAR_HOST_INTERRUPT(x) (writel(TW_CONTROL_CLEAR_HOST_INTERRUPT, TW_CONTROL_REG_ADDR(x))) diff --git a/trunk/drivers/scsi/Makefile b/trunk/drivers/scsi/Makefile index 1e4edbdf2730..48529d180ca8 100644 --- a/trunk/drivers/scsi/Makefile +++ b/trunk/drivers/scsi/Makefile @@ -99,6 +99,7 @@ obj-$(CONFIG_SCSI_DC395x) += dc395x.o obj-$(CONFIG_SCSI_DC390T) += tmscsim.o obj-$(CONFIG_MEGARAID_LEGACY) += megaraid.o obj-$(CONFIG_MEGARAID_NEWGEN) += megaraid/ +obj-$(CONFIG_MEGARAID_SAS) += megaraid/ obj-$(CONFIG_SCSI_ACARD) += atp870u.o obj-$(CONFIG_SCSI_SUNESP) += esp.o obj-$(CONFIG_SCSI_GDTH) += gdth.o diff --git a/trunk/drivers/scsi/aacraid/aachba.c b/trunk/drivers/scsi/aacraid/aachba.c index a8e3dfcd0dc7..93416f760e5a 100644 --- a/trunk/drivers/scsi/aacraid/aachba.c +++ b/trunk/drivers/scsi/aacraid/aachba.c @@ -313,18 +313,37 @@ int aac_get_containers(struct aac_dev *dev) } dresp = (struct aac_mount *)fib_data(fibptr); + if ((le32_to_cpu(dresp->status) == ST_OK) && + (le32_to_cpu(dresp->mnt[0].vol) == CT_NONE)) { + dinfo->command = cpu_to_le32(VM_NameServe64); + dinfo->count = cpu_to_le32(index); + dinfo->type = cpu_to_le32(FT_FILESYS); + + if (fib_send(ContainerCommand, + fibptr, + sizeof(struct aac_query_mount), + FsaNormal, + 1, 1, + NULL, NULL) < 0) + continue; + } else + dresp->mnt[0].capacityhigh = 0; + dprintk ((KERN_DEBUG - "VM_NameServe cid=%d status=%d vol=%d state=%d cap=%u\n", + "VM_NameServe cid=%d status=%d vol=%d state=%d cap=%llu\n", (int)index, (int)le32_to_cpu(dresp->status), (int)le32_to_cpu(dresp->mnt[0].vol), (int)le32_to_cpu(dresp->mnt[0].state), - (unsigned)le32_to_cpu(dresp->mnt[0].capacity))); + ((u64)le32_to_cpu(dresp->mnt[0].capacity)) + + (((u64)le32_to_cpu(dresp->mnt[0].capacityhigh)) << 32))); if ((le32_to_cpu(dresp->status) == ST_OK) && (le32_to_cpu(dresp->mnt[0].vol) != CT_NONE) && (le32_to_cpu(dresp->mnt[0].state) != FSCS_HIDDEN)) { fsa_dev_ptr[index].valid = 1; fsa_dev_ptr[index].type = le32_to_cpu(dresp->mnt[0].vol); - fsa_dev_ptr[index].size = le32_to_cpu(dresp->mnt[0].capacity); + fsa_dev_ptr[index].size + = ((u64)le32_to_cpu(dresp->mnt[0].capacity)) + + (((u64)le32_to_cpu(dresp->mnt[0].capacityhigh)) << 32); if (le32_to_cpu(dresp->mnt[0].state) & FSCS_READONLY) fsa_dev_ptr[index].ro = 1; } @@ -460,7 +479,7 @@ static int aac_get_container_name(struct scsi_cmnd * scsicmd, int cid) * is updated in the struct fsa_dev_info structure rather than returned. */ -static int probe_container(struct aac_dev *dev, int cid) +int probe_container(struct aac_dev *dev, int cid) { struct fsa_dev_info *fsa_dev_ptr; int status; @@ -496,12 +515,30 @@ static int probe_container(struct aac_dev *dev, int cid) dresp = (struct aac_mount *) fib_data(fibptr); + if ((le32_to_cpu(dresp->status) == ST_OK) && + (le32_to_cpu(dresp->mnt[0].vol) == CT_NONE)) { + dinfo->command = cpu_to_le32(VM_NameServe64); + dinfo->count = cpu_to_le32(cid); + dinfo->type = cpu_to_le32(FT_FILESYS); + + if (fib_send(ContainerCommand, + fibptr, + sizeof(struct aac_query_mount), + FsaNormal, + 1, 1, + NULL, NULL) < 0) + goto error; + } else + dresp->mnt[0].capacityhigh = 0; + if ((le32_to_cpu(dresp->status) == ST_OK) && (le32_to_cpu(dresp->mnt[0].vol) != CT_NONE) && (le32_to_cpu(dresp->mnt[0].state) != FSCS_HIDDEN)) { fsa_dev_ptr[cid].valid = 1; fsa_dev_ptr[cid].type = le32_to_cpu(dresp->mnt[0].vol); - fsa_dev_ptr[cid].size = le32_to_cpu(dresp->mnt[0].capacity); + fsa_dev_ptr[cid].size + = ((u64)le32_to_cpu(dresp->mnt[0].capacity)) + + (((u64)le32_to_cpu(dresp->mnt[0].capacityhigh)) << 32); if (le32_to_cpu(dresp->mnt[0].state) & FSCS_READONLY) fsa_dev_ptr[cid].ro = 1; } @@ -655,7 +692,7 @@ int aac_get_adapter_info(struct aac_dev* dev) fibptr, sizeof(*info), FsaNormal, - 1, 1, + -1, 1, /* First `interrupt' command uses special wait */ NULL, NULL); @@ -806,8 +843,8 @@ int aac_get_adapter_info(struct aac_dev* dev) if (!(dev->raw_io_interface)) { dev->scsi_host_ptr->sg_tablesize = (dev->max_fib_size - sizeof(struct aac_fibhdr) - - sizeof(struct aac_write) + sizeof(struct sgmap)) / - sizeof(struct sgmap); + sizeof(struct aac_write) + sizeof(struct sgentry)) / + sizeof(struct sgentry); if (dev->dac_support) { /* * 38 scatter gather elements @@ -816,8 +853,8 @@ int aac_get_adapter_info(struct aac_dev* dev) (dev->max_fib_size - sizeof(struct aac_fibhdr) - sizeof(struct aac_write64) + - sizeof(struct sgmap64)) / - sizeof(struct sgmap64); + sizeof(struct sgentry64)) / + sizeof(struct sgentry64); } dev->scsi_host_ptr->max_sectors = AAC_MAX_32BIT_SGBCOUNT; if(!(dev->adapter_info.options & AAC_OPT_NEW_COMM)) { @@ -854,7 +891,40 @@ static void io_callback(void *context, struct fib * fibptr) dev = (struct aac_dev *)scsicmd->device->host->hostdata; cid = ID_LUN_TO_CONTAINER(scsicmd->device->id, scsicmd->device->lun); - dprintk((KERN_DEBUG "io_callback[cpu %d]: lba = %u, t = %ld.\n", smp_processor_id(), ((scsicmd->cmnd[1] & 0x1F) << 16) | (scsicmd->cmnd[2] << 8) | scsicmd->cmnd[3], jiffies)); + if (nblank(dprintk(x))) { + u64 lba; + switch (scsicmd->cmnd[0]) { + case WRITE_6: + case READ_6: + lba = ((scsicmd->cmnd[1] & 0x1F) << 16) | + (scsicmd->cmnd[2] << 8) | scsicmd->cmnd[3]; + break; + case WRITE_16: + case READ_16: + lba = ((u64)scsicmd->cmnd[2] << 56) | + ((u64)scsicmd->cmnd[3] << 48) | + ((u64)scsicmd->cmnd[4] << 40) | + ((u64)scsicmd->cmnd[5] << 32) | + ((u64)scsicmd->cmnd[6] << 24) | + (scsicmd->cmnd[7] << 16) | + (scsicmd->cmnd[8] << 8) | scsicmd->cmnd[9]; + break; + case WRITE_12: + case READ_12: + lba = ((u64)scsicmd->cmnd[2] << 24) | + (scsicmd->cmnd[3] << 16) | + (scsicmd->cmnd[4] << 8) | scsicmd->cmnd[5]; + break; + default: + lba = ((u64)scsicmd->cmnd[2] << 24) | + (scsicmd->cmnd[3] << 16) | + (scsicmd->cmnd[4] << 8) | scsicmd->cmnd[5]; + break; + } + printk(KERN_DEBUG + "io_callback[cpu %d]: lba = %llu, t = %ld.\n", + smp_processor_id(), (unsigned long long)lba, jiffies); + } if (fibptr == NULL) BUG(); @@ -895,7 +965,7 @@ static void io_callback(void *context, struct fib * fibptr) static int aac_read(struct scsi_cmnd * scsicmd, int cid) { - u32 lba; + u64 lba; u32 count; int status; @@ -907,23 +977,69 @@ static int aac_read(struct scsi_cmnd * scsicmd, int cid) /* * Get block address and transfer length */ - if (scsicmd->cmnd[0] == READ_6) /* 6 byte command */ - { + switch (scsicmd->cmnd[0]) { + case READ_6: dprintk((KERN_DEBUG "aachba: received a read(6) command on id %d.\n", cid)); - lba = ((scsicmd->cmnd[1] & 0x1F) << 16) | (scsicmd->cmnd[2] << 8) | scsicmd->cmnd[3]; + lba = ((scsicmd->cmnd[1] & 0x1F) << 16) | + (scsicmd->cmnd[2] << 8) | scsicmd->cmnd[3]; count = scsicmd->cmnd[4]; if (count == 0) count = 256; - } else { + break; + case READ_16: + dprintk((KERN_DEBUG "aachba: received a read(16) command on id %d.\n", cid)); + + lba = ((u64)scsicmd->cmnd[2] << 56) | + ((u64)scsicmd->cmnd[3] << 48) | + ((u64)scsicmd->cmnd[4] << 40) | + ((u64)scsicmd->cmnd[5] << 32) | + ((u64)scsicmd->cmnd[6] << 24) | + (scsicmd->cmnd[7] << 16) | + (scsicmd->cmnd[8] << 8) | scsicmd->cmnd[9]; + count = (scsicmd->cmnd[10] << 24) | + (scsicmd->cmnd[11] << 16) | + (scsicmd->cmnd[12] << 8) | scsicmd->cmnd[13]; + break; + case READ_12: + dprintk((KERN_DEBUG "aachba: received a read(12) command on id %d.\n", cid)); + + lba = ((u64)scsicmd->cmnd[2] << 24) | + (scsicmd->cmnd[3] << 16) | + (scsicmd->cmnd[4] << 8) | scsicmd->cmnd[5]; + count = (scsicmd->cmnd[6] << 24) | + (scsicmd->cmnd[7] << 16) | + (scsicmd->cmnd[8] << 8) | scsicmd->cmnd[9]; + break; + default: dprintk((KERN_DEBUG "aachba: received a read(10) command on id %d.\n", cid)); - lba = (scsicmd->cmnd[2] << 24) | (scsicmd->cmnd[3] << 16) | (scsicmd->cmnd[4] << 8) | scsicmd->cmnd[5]; + lba = ((u64)scsicmd->cmnd[2] << 24) | + (scsicmd->cmnd[3] << 16) | + (scsicmd->cmnd[4] << 8) | scsicmd->cmnd[5]; count = (scsicmd->cmnd[7] << 8) | scsicmd->cmnd[8]; + break; } - dprintk((KERN_DEBUG "aac_read[cpu %d]: lba = %u, t = %ld.\n", + dprintk((KERN_DEBUG "aac_read[cpu %d]: lba = %llu, t = %ld.\n", smp_processor_id(), (unsigned long long)lba, jiffies)); + if ((!(dev->raw_io_interface) || !(dev->raw_io_64)) && + (lba & 0xffffffff00000000LL)) { + dprintk((KERN_DEBUG "aac_read: Illegal lba\n")); + scsicmd->result = DID_OK << 16 | COMMAND_COMPLETE << 8 | + SAM_STAT_CHECK_CONDITION; + set_sense((u8 *) &dev->fsa_dev[cid].sense_data, + HARDWARE_ERROR, + SENCODE_INTERNAL_TARGET_FAILURE, + ASENCODE_INTERNAL_TARGET_FAILURE, 0, 0, + 0, 0); + memcpy(scsicmd->sense_buffer, &dev->fsa_dev[cid].sense_data, + (sizeof(dev->fsa_dev[cid].sense_data) > sizeof(scsicmd->sense_buffer)) + ? sizeof(scsicmd->sense_buffer) + : sizeof(dev->fsa_dev[cid].sense_data)); + scsicmd->scsi_done(scsicmd); + return 0; + } /* * Alocate and initialize a Fib */ @@ -936,8 +1052,8 @@ static int aac_read(struct scsi_cmnd * scsicmd, int cid) if (dev->raw_io_interface) { struct aac_raw_io *readcmd; readcmd = (struct aac_raw_io *) fib_data(cmd_fibcontext); - readcmd->block[0] = cpu_to_le32(lba); - readcmd->block[1] = 0; + readcmd->block[0] = cpu_to_le32((u32)(lba&0xffffffff)); + readcmd->block[1] = cpu_to_le32((u32)((lba&0xffffffff00000000LL)>>32)); readcmd->count = cpu_to_le32(count<<9); readcmd->cid = cpu_to_le16(cid); readcmd->flags = cpu_to_le16(1); @@ -964,7 +1080,7 @@ static int aac_read(struct scsi_cmnd * scsicmd, int cid) readcmd->command = cpu_to_le32(VM_CtHostRead64); readcmd->cid = cpu_to_le16(cid); readcmd->sector_count = cpu_to_le16(count); - readcmd->block = cpu_to_le32(lba); + readcmd->block = cpu_to_le32((u32)(lba&0xffffffff)); readcmd->pad = 0; readcmd->flags = 0; @@ -989,7 +1105,7 @@ static int aac_read(struct scsi_cmnd * scsicmd, int cid) readcmd = (struct aac_read *) fib_data(cmd_fibcontext); readcmd->command = cpu_to_le32(VM_CtBlockRead); readcmd->cid = cpu_to_le32(cid); - readcmd->block = cpu_to_le32(lba); + readcmd->block = cpu_to_le32((u32)(lba&0xffffffff)); readcmd->count = cpu_to_le32(count * 512); aac_build_sg(scsicmd, &readcmd->sg); @@ -1031,7 +1147,7 @@ static int aac_read(struct scsi_cmnd * scsicmd, int cid) static int aac_write(struct scsi_cmnd * scsicmd, int cid) { - u32 lba; + u64 lba; u32 count; int status; u16 fibsize; @@ -1048,13 +1164,48 @@ static int aac_write(struct scsi_cmnd * scsicmd, int cid) count = scsicmd->cmnd[4]; if (count == 0) count = 256; + } else if (scsicmd->cmnd[0] == WRITE_16) { /* 16 byte command */ + dprintk((KERN_DEBUG "aachba: received a write(16) command on id %d.\n", cid)); + + lba = ((u64)scsicmd->cmnd[2] << 56) | + ((u64)scsicmd->cmnd[3] << 48) | + ((u64)scsicmd->cmnd[4] << 40) | + ((u64)scsicmd->cmnd[5] << 32) | + ((u64)scsicmd->cmnd[6] << 24) | + (scsicmd->cmnd[7] << 16) | + (scsicmd->cmnd[8] << 8) | scsicmd->cmnd[9]; + count = (scsicmd->cmnd[10] << 24) | (scsicmd->cmnd[11] << 16) | + (scsicmd->cmnd[12] << 8) | scsicmd->cmnd[13]; + } else if (scsicmd->cmnd[0] == WRITE_12) { /* 12 byte command */ + dprintk((KERN_DEBUG "aachba: received a write(12) command on id %d.\n", cid)); + + lba = ((u64)scsicmd->cmnd[2] << 24) | (scsicmd->cmnd[3] << 16) + | (scsicmd->cmnd[4] << 8) | scsicmd->cmnd[5]; + count = (scsicmd->cmnd[6] << 24) | (scsicmd->cmnd[7] << 16) + | (scsicmd->cmnd[8] << 8) | scsicmd->cmnd[9]; } else { dprintk((KERN_DEBUG "aachba: received a write(10) command on id %d.\n", cid)); - lba = (scsicmd->cmnd[2] << 24) | (scsicmd->cmnd[3] << 16) | (scsicmd->cmnd[4] << 8) | scsicmd->cmnd[5]; + lba = ((u64)scsicmd->cmnd[2] << 24) | (scsicmd->cmnd[3] << 16) | (scsicmd->cmnd[4] << 8) | scsicmd->cmnd[5]; count = (scsicmd->cmnd[7] << 8) | scsicmd->cmnd[8]; } - dprintk((KERN_DEBUG "aac_write[cpu %d]: lba = %u, t = %ld.\n", + dprintk((KERN_DEBUG "aac_write[cpu %d]: lba = %llu, t = %ld.\n", smp_processor_id(), (unsigned long long)lba, jiffies)); + if ((!(dev->raw_io_interface) || !(dev->raw_io_64)) + && (lba & 0xffffffff00000000LL)) { + dprintk((KERN_DEBUG "aac_write: Illegal lba\n")); + scsicmd->result = DID_OK << 16 | COMMAND_COMPLETE << 8 | SAM_STAT_CHECK_CONDITION; + set_sense((u8 *) &dev->fsa_dev[cid].sense_data, + HARDWARE_ERROR, + SENCODE_INTERNAL_TARGET_FAILURE, + ASENCODE_INTERNAL_TARGET_FAILURE, 0, 0, + 0, 0); + memcpy(scsicmd->sense_buffer, &dev->fsa_dev[cid].sense_data, + (sizeof(dev->fsa_dev[cid].sense_data) > sizeof(scsicmd->sense_buffer)) + ? sizeof(scsicmd->sense_buffer) + : sizeof(dev->fsa_dev[cid].sense_data)); + scsicmd->scsi_done(scsicmd); + return 0; + } /* * Allocate and initialize a Fib then setup a BlockWrite command */ @@ -1068,8 +1219,8 @@ static int aac_write(struct scsi_cmnd * scsicmd, int cid) if (dev->raw_io_interface) { struct aac_raw_io *writecmd; writecmd = (struct aac_raw_io *) fib_data(cmd_fibcontext); - writecmd->block[0] = cpu_to_le32(lba); - writecmd->block[1] = 0; + writecmd->block[0] = cpu_to_le32((u32)(lba&0xffffffff)); + writecmd->block[1] = cpu_to_le32((u32)((lba&0xffffffff00000000LL)>>32)); writecmd->count = cpu_to_le32(count<<9); writecmd->cid = cpu_to_le16(cid); writecmd->flags = 0; @@ -1096,7 +1247,7 @@ static int aac_write(struct scsi_cmnd * scsicmd, int cid) writecmd->command = cpu_to_le32(VM_CtHostWrite64); writecmd->cid = cpu_to_le16(cid); writecmd->sector_count = cpu_to_le16(count); - writecmd->block = cpu_to_le32(lba); + writecmd->block = cpu_to_le32((u32)(lba&0xffffffff)); writecmd->pad = 0; writecmd->flags = 0; @@ -1121,7 +1272,7 @@ static int aac_write(struct scsi_cmnd * scsicmd, int cid) writecmd = (struct aac_write *) fib_data(cmd_fibcontext); writecmd->command = cpu_to_le32(VM_CtBlockWrite); writecmd->cid = cpu_to_le32(cid); - writecmd->block = cpu_to_le32(lba); + writecmd->block = cpu_to_le32((u32)(lba&0xffffffff)); writecmd->count = cpu_to_le32(count * 512); writecmd->sg.count = cpu_to_le32(1); /* ->stable is not used - it did mean which type of write */ @@ -1310,11 +1461,18 @@ int aac_scsi_cmd(struct scsi_cmnd * scsicmd) */ if ((fsa_dev_ptr[cid].valid & 1) == 0) { switch (scsicmd->cmnd[0]) { + case SERVICE_ACTION_IN: + if (!(dev->raw_io_interface) || + !(dev->raw_io_64) || + ((scsicmd->cmnd[1] & 0x1f) != SAI_READ_CAPACITY_16)) + break; case INQUIRY: case READ_CAPACITY: case TEST_UNIT_READY: spin_unlock_irq(host->host_lock); probe_container(dev, cid); + if ((fsa_dev_ptr[cid].valid & 1) == 0) + fsa_dev_ptr[cid].valid = 0; spin_lock_irq(host->host_lock); if (fsa_dev_ptr[cid].valid == 0) { scsicmd->result = DID_NO_CONNECT << 16; @@ -1375,7 +1533,6 @@ int aac_scsi_cmd(struct scsi_cmnd * scsicmd) memset(&inq_data, 0, sizeof (struct inquiry_data)); inq_data.inqd_ver = 2; /* claim compliance to SCSI-2 */ - inq_data.inqd_dtq = 0x80; /* set RMB bit to one indicating that the medium is removable */ inq_data.inqd_rdf = 2; /* A response data format value of two indicates that the data shall be in the format specified in SCSI-2 */ inq_data.inqd_len = 31; /*Format for "pad2" is RelAdr | WBus32 | WBus16 | Sync | Linked |Reserved| CmdQue | SftRe */ @@ -1397,13 +1554,55 @@ int aac_scsi_cmd(struct scsi_cmnd * scsicmd) aac_internal_transfer(scsicmd, &inq_data, 0, sizeof(inq_data)); return aac_get_container_name(scsicmd, cid); } + case SERVICE_ACTION_IN: + if (!(dev->raw_io_interface) || + !(dev->raw_io_64) || + ((scsicmd->cmnd[1] & 0x1f) != SAI_READ_CAPACITY_16)) + break; + { + u64 capacity; + char cp[12]; + unsigned int offset = 0; + + dprintk((KERN_DEBUG "READ CAPACITY_16 command.\n")); + capacity = fsa_dev_ptr[cid].size - 1; + if (scsicmd->cmnd[13] > 12) { + offset = scsicmd->cmnd[13] - 12; + if (offset > sizeof(cp)) + break; + memset(cp, 0, offset); + aac_internal_transfer(scsicmd, cp, 0, offset); + } + cp[0] = (capacity >> 56) & 0xff; + cp[1] = (capacity >> 48) & 0xff; + cp[2] = (capacity >> 40) & 0xff; + cp[3] = (capacity >> 32) & 0xff; + cp[4] = (capacity >> 24) & 0xff; + cp[5] = (capacity >> 16) & 0xff; + cp[6] = (capacity >> 8) & 0xff; + cp[7] = (capacity >> 0) & 0xff; + cp[8] = 0; + cp[9] = 0; + cp[10] = 2; + cp[11] = 0; + aac_internal_transfer(scsicmd, cp, offset, sizeof(cp)); + + /* Do not cache partition table for arrays */ + scsicmd->device->removable = 1; + + scsicmd->result = DID_OK << 16 | COMMAND_COMPLETE << 8 | SAM_STAT_GOOD; + scsicmd->scsi_done(scsicmd); + + return 0; + } + case READ_CAPACITY: { u32 capacity; char cp[8]; dprintk((KERN_DEBUG "READ CAPACITY command.\n")); - if (fsa_dev_ptr[cid].size <= 0x100000000LL) + if (fsa_dev_ptr[cid].size <= 0x100000000ULL) capacity = fsa_dev_ptr[cid].size - 1; else capacity = (u32)-1; @@ -1417,6 +1616,8 @@ int aac_scsi_cmd(struct scsi_cmnd * scsicmd) cp[6] = 2; cp[7] = 0; aac_internal_transfer(scsicmd, cp, 0, sizeof(cp)); + /* Do not cache partition table for arrays */ + scsicmd->device->removable = 1; scsicmd->result = DID_OK << 16 | COMMAND_COMPLETE << 8 | SAM_STAT_GOOD; scsicmd->scsi_done(scsicmd); @@ -1497,6 +1698,8 @@ int aac_scsi_cmd(struct scsi_cmnd * scsicmd) { case READ_6: case READ_10: + case READ_12: + case READ_16: /* * Hack to keep track of ordinal number of the device that * corresponds to a container. Needed to convert @@ -1504,17 +1707,19 @@ int aac_scsi_cmd(struct scsi_cmnd * scsicmd) */ spin_unlock_irq(host->host_lock); - if (scsicmd->request->rq_disk) - memcpy(fsa_dev_ptr[cid].devname, - scsicmd->request->rq_disk->disk_name, - 8); - + if (scsicmd->request->rq_disk) + strlcpy(fsa_dev_ptr[cid].devname, + scsicmd->request->rq_disk->disk_name, + min(sizeof(fsa_dev_ptr[cid].devname), + sizeof(scsicmd->request->rq_disk->disk_name) + 1)); ret = aac_read(scsicmd, cid); spin_lock_irq(host->host_lock); return ret; case WRITE_6: case WRITE_10: + case WRITE_12: + case WRITE_16: spin_unlock_irq(host->host_lock); ret = aac_write(scsicmd, cid); spin_lock_irq(host->host_lock); @@ -1745,6 +1950,8 @@ static void aac_srb_callback(void *context, struct fib * fibptr) case WRITE_10: case READ_12: case WRITE_12: + case READ_16: + case WRITE_16: if(le32_to_cpu(srbreply->data_xfer_length) < scsicmd->underflow ) { printk(KERN_WARNING"aacraid: SCSI CMD underflow\n"); } else { @@ -1850,8 +2057,8 @@ static void aac_srb_callback(void *context, struct fib * fibptr) sizeof(scsicmd->sense_buffer) : le32_to_cpu(srbreply->sense_data_size); #ifdef AAC_DETAILED_STATUS_INFO - dprintk((KERN_WARNING "aac_srb_callback: check condition, status = %d len=%d\n", - le32_to_cpu(srbreply->status), len)); + printk(KERN_WARNING "aac_srb_callback: check condition, status = %d len=%d\n", + le32_to_cpu(srbreply->status), len); #endif memcpy(scsicmd->sense_buffer, srbreply->sense_data, len); diff --git a/trunk/drivers/scsi/aacraid/aacraid.h b/trunk/drivers/scsi/aacraid/aacraid.h index e40528185d48..4a99d2f000f4 100644 --- a/trunk/drivers/scsi/aacraid/aacraid.h +++ b/trunk/drivers/scsi/aacraid/aacraid.h @@ -1,6 +1,10 @@ #if (!defined(dprintk)) # define dprintk(x) #endif +/* eg: if (nblank(dprintk(x))) */ +#define _nblank(x) #x +#define nblank(x) _nblank(x)[0] + /*------------------------------------------------------------------------------ * D E F I N E S @@ -302,7 +306,6 @@ enum aac_queue_types { */ #define FsaNormal 1 -#define FsaHigh 2 /* * Define the FIB. The FIB is the where all the requested data and @@ -546,8 +549,6 @@ struct aac_queue { /* This is only valid for adapter to host command queues. */ spinlock_t *lock; /* Spinlock for this queue must take this lock before accessing the lock */ spinlock_t lockdata; /* Actual lock (used only on one side of the lock) */ - unsigned long SavedIrql; /* Previous IRQL when the spin lock is taken */ - u32 padding; /* Padding - FIXME - can remove I believe */ struct list_head cmdq; /* A queue of FIBs which need to be prcessed by the FS thread. This is */ /* only valid for command queues which receive entries from the adapter. */ struct list_head pendingq; /* A queue of outstanding fib's to the adapter. */ @@ -776,7 +777,9 @@ struct fsa_dev_info { u64 last; u64 size; u32 type; + u32 config_waiting_on; u16 queue_depth; + u8 config_needed; u8 valid; u8 ro; u8 locked; @@ -1012,6 +1015,7 @@ struct aac_dev /* macro side-effects BEWARE */ # define raw_io_interface \ init->InitStructRevision==cpu_to_le32(ADAPTER_INIT_STRUCT_REVISION_4) + u8 raw_io_64; u8 printf_enabled; }; @@ -1362,8 +1366,10 @@ struct aac_srb_reply #define VM_CtBlockVerify64 18 #define VM_CtHostRead64 19 #define VM_CtHostWrite64 20 +#define VM_DrvErrTblLog 21 +#define VM_NameServe64 22 -#define MAX_VMCOMMAND_NUM 21 /* used for sizing stats array - leave last */ +#define MAX_VMCOMMAND_NUM 23 /* used for sizing stats array - leave last */ /* * Descriptive information (eg, vital stats) @@ -1472,6 +1478,7 @@ struct aac_mntent { manager (eg, filesystem) */ __le32 altoid; /* != oid <==> snapshot or broken mirror exists */ + __le32 capacityhigh; }; #define FSCS_NOTCLEAN 0x0001 /* fsck is neccessary before mounting */ @@ -1707,6 +1714,7 @@ extern struct aac_common aac_config; #define AifCmdJobProgress 2 /* Progress report */ #define AifJobCtrZero 101 /* Array Zero progress */ #define AifJobStsSuccess 1 /* Job completes */ +#define AifJobStsRunning 102 /* Job running */ #define AifCmdAPIReport 3 /* Report from other user of API */ #define AifCmdDriverNotify 4 /* Notify host driver of event */ #define AifDenMorphComplete 200 /* A morph operation completed */ @@ -1777,6 +1785,7 @@ int fib_adapter_complete(struct fib * fibptr, unsigned short size); struct aac_driver_ident* aac_get_driver_ident(int devtype); int aac_get_adapter_info(struct aac_dev* dev); int aac_send_shutdown(struct aac_dev *dev); +int probe_container(struct aac_dev *dev, int cid); extern int numacb; extern int acbsize; extern char aac_driver_version[]; diff --git a/trunk/drivers/scsi/aacraid/comminit.c b/trunk/drivers/scsi/aacraid/comminit.c index 75abd0453289..59a341b2aedc 100644 --- a/trunk/drivers/scsi/aacraid/comminit.c +++ b/trunk/drivers/scsi/aacraid/comminit.c @@ -195,7 +195,7 @@ int aac_send_shutdown(struct aac_dev * dev) fibctx, sizeof(struct aac_close), FsaNormal, - 1, 1, + -2 /* Timeout silently */, 1, NULL, NULL); if (status == 0) @@ -313,8 +313,15 @@ struct aac_dev *aac_init_adapter(struct aac_dev *dev) dev->max_fib_size = sizeof(struct hw_fib); dev->sg_tablesize = host->sg_tablesize = (dev->max_fib_size - sizeof(struct aac_fibhdr) - - sizeof(struct aac_write) + sizeof(struct sgmap)) - / sizeof(struct sgmap); + - sizeof(struct aac_write) + sizeof(struct sgentry)) + / sizeof(struct sgentry); + dev->raw_io_64 = 0; + if ((!aac_adapter_sync_cmd(dev, GET_ADAPTER_PROPERTIES, + 0, 0, 0, 0, 0, 0, status+0, status+1, status+2, NULL, NULL)) && + (status[0] == 0x00000001)) { + if (status[1] & AAC_OPT_NEW_COMM_64) + dev->raw_io_64 = 1; + } if ((!aac_adapter_sync_cmd(dev, GET_COMM_PREFERRED_SETTINGS, 0, 0, 0, 0, 0, 0, status+0, status+1, status+2, status+3, status+4)) @@ -342,8 +349,8 @@ struct aac_dev *aac_init_adapter(struct aac_dev *dev) dev->max_fib_size = 512; dev->sg_tablesize = host->sg_tablesize = (512 - sizeof(struct aac_fibhdr) - - sizeof(struct aac_write) + sizeof(struct sgmap)) - / sizeof(struct sgmap); + - sizeof(struct aac_write) + sizeof(struct sgentry)) + / sizeof(struct sgentry); host->can_queue = AAC_NUM_IO_FIB; } else if (acbsize == 2048) { host->max_sectors = 512; diff --git a/trunk/drivers/scsi/aacraid/commsup.c b/trunk/drivers/scsi/aacraid/commsup.c index a1d303f03480..e4d543a474ae 100644 --- a/trunk/drivers/scsi/aacraid/commsup.c +++ b/trunk/drivers/scsi/aacraid/commsup.c @@ -39,7 +39,9 @@ #include #include #include +#include #include +#include #include "aacraid.h" @@ -269,40 +271,22 @@ static int aac_get_entry (struct aac_dev * dev, u32 qid, struct aac_entry **entr /* Interrupt Moderation, only interrupt for first two entries */ if (idx != le32_to_cpu(*(q->headers.consumer))) { if (--idx == 0) { - if (qid == AdapHighCmdQueue) - idx = ADAP_HIGH_CMD_ENTRIES; - else if (qid == AdapNormCmdQueue) + if (qid == AdapNormCmdQueue) idx = ADAP_NORM_CMD_ENTRIES; - else if (qid == AdapHighRespQueue) - idx = ADAP_HIGH_RESP_ENTRIES; - else if (qid == AdapNormRespQueue) + else idx = ADAP_NORM_RESP_ENTRIES; } if (idx != le32_to_cpu(*(q->headers.consumer))) *nonotify = 1; } - if (qid == AdapHighCmdQueue) { - if (*index >= ADAP_HIGH_CMD_ENTRIES) - *index = 0; - } else if (qid == AdapNormCmdQueue) { + if (qid == AdapNormCmdQueue) { if (*index >= ADAP_NORM_CMD_ENTRIES) *index = 0; /* Wrap to front of the Producer Queue. */ - } - else if (qid == AdapHighRespQueue) - { - if (*index >= ADAP_HIGH_RESP_ENTRIES) - *index = 0; - } - else if (qid == AdapNormRespQueue) - { + } else { if (*index >= ADAP_NORM_RESP_ENTRIES) *index = 0; /* Wrap to front of the Producer Queue. */ } - else { - printk("aacraid: invalid qid\n"); - BUG(); - } if ((*index + 1) == le32_to_cpu(*(q->headers.consumer))) { /* Queue is full */ printk(KERN_WARNING "Queue %d full, %u outstanding.\n", @@ -334,12 +318,8 @@ static int aac_queue_get(struct aac_dev * dev, u32 * index, u32 qid, struct hw_f { struct aac_entry * entry = NULL; int map = 0; - struct aac_queue * q = &dev->queues->queue[qid]; - - spin_lock_irqsave(q->lock, q->SavedIrql); - if (qid == AdapHighCmdQueue || qid == AdapNormCmdQueue) - { + if (qid == AdapNormCmdQueue) { /* if no entries wait for some if caller wants to */ while (!aac_get_entry(dev, qid, &entry, index, nonotify)) { @@ -350,9 +330,7 @@ static int aac_queue_get(struct aac_dev * dev, u32 * index, u32 qid, struct hw_f */ entry->size = cpu_to_le32(le16_to_cpu(hw_fib->header.Size)); map = 1; - } - else if (qid == AdapHighRespQueue || qid == AdapNormRespQueue) - { + } else { while(!aac_get_entry(dev, qid, &entry, index, nonotify)) { /* if no entries wait for some if caller wants to */ @@ -375,42 +353,6 @@ static int aac_queue_get(struct aac_dev * dev, u32 * index, u32 qid, struct hw_f return 0; } - -/** - * aac_insert_entry - insert a queue entry - * @dev: Adapter - * @index: Index of entry to insert - * @qid: Queue number - * @nonotify: Suppress adapter notification - * - * Gets the next free QE off the requested priorty adapter command - * queue and associates the Fib with the QE. The QE represented by - * index is ready to insert on the queue when this routine returns - * success. - */ - -static int aac_insert_entry(struct aac_dev * dev, u32 index, u32 qid, unsigned long nonotify) -{ - struct aac_queue * q = &dev->queues->queue[qid]; - - if(q == NULL) - BUG(); - *(q->headers.producer) = cpu_to_le32(index + 1); - spin_unlock_irqrestore(q->lock, q->SavedIrql); - - if (qid == AdapHighCmdQueue || - qid == AdapNormCmdQueue || - qid == AdapHighRespQueue || - qid == AdapNormRespQueue) - { - if (!nonotify) - aac_adapter_notify(dev, qid); - } - else - printk("Suprise insert!\n"); - return 0; -} - /* * Define the highest level of host to adapter communication routines. * These routines will support host to adapter FS commuication. These @@ -439,12 +381,13 @@ static int aac_insert_entry(struct aac_dev * dev, u32 index, u32 qid, unsigned l int fib_send(u16 command, struct fib * fibptr, unsigned long size, int priority, int wait, int reply, fib_callback callback, void * callback_data) { u32 index; - u32 qid; struct aac_dev * dev = fibptr->dev; unsigned long nointr = 0; struct hw_fib * hw_fib = fibptr->hw_fib; struct aac_queue * q; unsigned long flags = 0; + unsigned long qflags; + if (!(hw_fib->header.XferState & cpu_to_le32(HostOwned))) return -EBUSY; /* @@ -497,26 +440,8 @@ int fib_send(u16 command, struct fib * fibptr, unsigned long size, int priority * Get a queue entry connect the FIB to it and send an notify * the adapter a command is ready. */ - if (priority == FsaHigh) { - hw_fib->header.XferState |= cpu_to_le32(HighPriority); - qid = AdapHighCmdQueue; - } else { - hw_fib->header.XferState |= cpu_to_le32(NormalPriority); - qid = AdapNormCmdQueue; - } - q = &dev->queues->queue[qid]; + hw_fib->header.XferState |= cpu_to_le32(NormalPriority); - if(wait) - spin_lock_irqsave(&fibptr->event_lock, flags); - if(aac_queue_get( dev, &index, qid, hw_fib, 1, fibptr, &nointr)<0) - return -EWOULDBLOCK; - dprintk((KERN_DEBUG "fib_send: inserting a queue entry at index %d.\n",index)); - dprintk((KERN_DEBUG "Fib contents:.\n")); - dprintk((KERN_DEBUG " Command = %d.\n", hw_fib->header.Command)); - dprintk((KERN_DEBUG " XferState = %x.\n", hw_fib->header.XferState)); - dprintk((KERN_DEBUG " hw_fib va being sent=%p\n",fibptr->hw_fib)); - dprintk((KERN_DEBUG " hw_fib pa being sent=%lx\n",(ulong)fibptr->hw_fib_pa)); - dprintk((KERN_DEBUG " fib being sent=%p\n",fibptr)); /* * Fill in the Callback and CallbackContext if we are not * going to wait. @@ -525,22 +450,67 @@ int fib_send(u16 command, struct fib * fibptr, unsigned long size, int priority fibptr->callback = callback; fibptr->callback_data = callback_data; } - FIB_COUNTER_INCREMENT(aac_config.FibsSent); - list_add_tail(&fibptr->queue, &q->pendingq); - q->numpending++; fibptr->done = 0; fibptr->flags = 0; - if(aac_insert_entry(dev, index, qid, (nointr & aac_config.irq_mod)) < 0) - return -EWOULDBLOCK; + FIB_COUNTER_INCREMENT(aac_config.FibsSent); + + dprintk((KERN_DEBUG "fib_send: inserting a queue entry at index %d.\n",index)); + dprintk((KERN_DEBUG "Fib contents:.\n")); + dprintk((KERN_DEBUG " Command = %d.\n", hw_fib->header.Command)); + dprintk((KERN_DEBUG " XferState = %x.\n", hw_fib->header.XferState)); + dprintk((KERN_DEBUG " hw_fib va being sent=%p\n",fibptr->hw_fib)); + dprintk((KERN_DEBUG " hw_fib pa being sent=%lx\n",(ulong)fibptr->hw_fib_pa)); + dprintk((KERN_DEBUG " fib being sent=%p\n",fibptr)); + + q = &dev->queues->queue[AdapNormCmdQueue]; + + if(wait) + spin_lock_irqsave(&fibptr->event_lock, flags); + spin_lock_irqsave(q->lock, qflags); + aac_queue_get( dev, &index, AdapNormCmdQueue, hw_fib, 1, fibptr, &nointr); + + list_add_tail(&fibptr->queue, &q->pendingq); + q->numpending++; + *(q->headers.producer) = cpu_to_le32(index + 1); + spin_unlock_irqrestore(q->lock, qflags); + if (!(nointr & aac_config.irq_mod)) + aac_adapter_notify(dev, AdapNormCmdQueue); /* * If the caller wanted us to wait for response wait now. */ if (wait) { spin_unlock_irqrestore(&fibptr->event_lock, flags); - down(&fibptr->event_wait); + /* Only set for first known interruptable command */ + if (wait < 0) { + /* + * *VERY* Dangerous to time out a command, the + * assumption is made that we have no hope of + * functioning because an interrupt routing or other + * hardware failure has occurred. + */ + unsigned long count = 36000000L; /* 3 minutes */ + unsigned long qflags; + while (down_trylock(&fibptr->event_wait)) { + if (--count == 0) { + spin_lock_irqsave(q->lock, qflags); + q->numpending--; + list_del(&fibptr->queue); + spin_unlock_irqrestore(q->lock, qflags); + if (wait == -1) { + printk(KERN_ERR "aacraid: fib_send: first asynchronous command timed out.\n" + "Usually a result of a PCI interrupt routing problem;\n" + "update mother board BIOS or consider utilizing one of\n" + "the SAFE mode kernel options (acpi, apic etc)\n"); + } + return -ETIMEDOUT; + } + udelay(5); + } + } else + down(&fibptr->event_wait); if(fibptr->done == 0) BUG(); @@ -622,15 +592,9 @@ void aac_consumer_free(struct aac_dev * dev, struct aac_queue *q, u32 qid) case HostNormCmdQueue: notify = HostNormCmdNotFull; break; - case HostHighCmdQueue: - notify = HostHighCmdNotFull; - break; case HostNormRespQueue: notify = HostNormRespNotFull; break; - case HostHighRespQueue: - notify = HostHighRespNotFull; - break; default: BUG(); return; @@ -652,9 +616,13 @@ int fib_adapter_complete(struct fib * fibptr, unsigned short size) { struct hw_fib * hw_fib = fibptr->hw_fib; struct aac_dev * dev = fibptr->dev; + struct aac_queue * q; unsigned long nointr = 0; - if (hw_fib->header.XferState == 0) + unsigned long qflags; + + if (hw_fib->header.XferState == 0) { return 0; + } /* * If we plan to do anything check the structure type first. */ @@ -669,37 +637,21 @@ int fib_adapter_complete(struct fib * fibptr, unsigned short size) * send the completed cdb to the adapter. */ if (hw_fib->header.XferState & cpu_to_le32(SentFromAdapter)) { + u32 index; hw_fib->header.XferState |= cpu_to_le32(HostProcessed); - if (hw_fib->header.XferState & cpu_to_le32(HighPriority)) { - u32 index; - if (size) - { - size += sizeof(struct aac_fibhdr); - if (size > le16_to_cpu(hw_fib->header.SenderSize)) - return -EMSGSIZE; - hw_fib->header.Size = cpu_to_le16(size); - } - if(aac_queue_get(dev, &index, AdapHighRespQueue, hw_fib, 1, NULL, &nointr) < 0) { - return -EWOULDBLOCK; - } - if (aac_insert_entry(dev, index, AdapHighRespQueue, (nointr & (int)aac_config.irq_mod)) != 0) { - } - } else if (hw_fib->header.XferState & - cpu_to_le32(NormalPriority)) { - u32 index; - - if (size) { - size += sizeof(struct aac_fibhdr); - if (size > le16_to_cpu(hw_fib->header.SenderSize)) - return -EMSGSIZE; - hw_fib->header.Size = cpu_to_le16(size); - } - if (aac_queue_get(dev, &index, AdapNormRespQueue, hw_fib, 1, NULL, &nointr) < 0) - return -EWOULDBLOCK; - if (aac_insert_entry(dev, index, AdapNormRespQueue, (nointr & (int)aac_config.irq_mod)) != 0) - { - } + if (size) { + size += sizeof(struct aac_fibhdr); + if (size > le16_to_cpu(hw_fib->header.SenderSize)) + return -EMSGSIZE; + hw_fib->header.Size = cpu_to_le16(size); } + q = &dev->queues->queue[AdapNormRespQueue]; + spin_lock_irqsave(q->lock, qflags); + aac_queue_get(dev, &index, AdapNormRespQueue, hw_fib, 1, NULL, &nointr); + *(q->headers.producer) = cpu_to_le32(index + 1); + spin_unlock_irqrestore(q->lock, qflags); + if (!(nointr & (int)aac_config.irq_mod)) + aac_adapter_notify(dev, AdapNormRespQueue); } else { @@ -791,6 +743,268 @@ void aac_printf(struct aac_dev *dev, u32 val) memset(cp, 0, 256); } + +/** + * aac_handle_aif - Handle a message from the firmware + * @dev: Which adapter this fib is from + * @fibptr: Pointer to fibptr from adapter + * + * This routine handles a driver notify fib from the adapter and + * dispatches it to the appropriate routine for handling. + */ + +static void aac_handle_aif(struct aac_dev * dev, struct fib * fibptr) +{ + struct hw_fib * hw_fib = fibptr->hw_fib; + struct aac_aifcmd * aifcmd = (struct aac_aifcmd *)hw_fib->data; + int busy; + u32 container; + struct scsi_device *device; + enum { + NOTHING, + DELETE, + ADD, + CHANGE + } device_config_needed; + + /* Sniff for container changes */ + + if (!dev) + return; + container = (u32)-1; + + /* + * We have set this up to try and minimize the number of + * re-configures that take place. As a result of this when + * certain AIF's come in we will set a flag waiting for another + * type of AIF before setting the re-config flag. + */ + switch (le32_to_cpu(aifcmd->command)) { + case AifCmdDriverNotify: + switch (le32_to_cpu(((u32 *)aifcmd->data)[0])) { + /* + * Morph or Expand complete + */ + case AifDenMorphComplete: + case AifDenVolumeExtendComplete: + container = le32_to_cpu(((u32 *)aifcmd->data)[1]); + if (container >= dev->maximum_num_containers) + break; + + /* + * Find the Scsi_Device associated with the SCSI + * address. Make sure we have the right array, and if + * so set the flag to initiate a new re-config once we + * see an AifEnConfigChange AIF come through. + */ + + if ((dev != NULL) && (dev->scsi_host_ptr != NULL)) { + device = scsi_device_lookup(dev->scsi_host_ptr, + CONTAINER_TO_CHANNEL(container), + CONTAINER_TO_ID(container), + CONTAINER_TO_LUN(container)); + if (device) { + dev->fsa_dev[container].config_needed = CHANGE; + dev->fsa_dev[container].config_waiting_on = AifEnConfigChange; + scsi_device_put(device); + } + } + } + + /* + * If we are waiting on something and this happens to be + * that thing then set the re-configure flag. + */ + if (container != (u32)-1) { + if (container >= dev->maximum_num_containers) + break; + if (dev->fsa_dev[container].config_waiting_on == + le32_to_cpu(*(u32 *)aifcmd->data)) + dev->fsa_dev[container].config_waiting_on = 0; + } else for (container = 0; + container < dev->maximum_num_containers; ++container) { + if (dev->fsa_dev[container].config_waiting_on == + le32_to_cpu(*(u32 *)aifcmd->data)) + dev->fsa_dev[container].config_waiting_on = 0; + } + break; + + case AifCmdEventNotify: + switch (le32_to_cpu(((u32 *)aifcmd->data)[0])) { + /* + * Add an Array. + */ + case AifEnAddContainer: + container = le32_to_cpu(((u32 *)aifcmd->data)[1]); + if (container >= dev->maximum_num_containers) + break; + dev->fsa_dev[container].config_needed = ADD; + dev->fsa_dev[container].config_waiting_on = + AifEnConfigChange; + break; + + /* + * Delete an Array. + */ + case AifEnDeleteContainer: + container = le32_to_cpu(((u32 *)aifcmd->data)[1]); + if (container >= dev->maximum_num_containers) + break; + dev->fsa_dev[container].config_needed = DELETE; + dev->fsa_dev[container].config_waiting_on = + AifEnConfigChange; + break; + + /* + * Container change detected. If we currently are not + * waiting on something else, setup to wait on a Config Change. + */ + case AifEnContainerChange: + container = le32_to_cpu(((u32 *)aifcmd->data)[1]); + if (container >= dev->maximum_num_containers) + break; + if (dev->fsa_dev[container].config_waiting_on) + break; + dev->fsa_dev[container].config_needed = CHANGE; + dev->fsa_dev[container].config_waiting_on = + AifEnConfigChange; + break; + + case AifEnConfigChange: + break; + + } + + /* + * If we are waiting on something and this happens to be + * that thing then set the re-configure flag. + */ + if (container != (u32)-1) { + if (container >= dev->maximum_num_containers) + break; + if (dev->fsa_dev[container].config_waiting_on == + le32_to_cpu(*(u32 *)aifcmd->data)) + dev->fsa_dev[container].config_waiting_on = 0; + } else for (container = 0; + container < dev->maximum_num_containers; ++container) { + if (dev->fsa_dev[container].config_waiting_on == + le32_to_cpu(*(u32 *)aifcmd->data)) + dev->fsa_dev[container].config_waiting_on = 0; + } + break; + + case AifCmdJobProgress: + /* + * These are job progress AIF's. When a Clear is being + * done on a container it is initially created then hidden from + * the OS. When the clear completes we don't get a config + * change so we monitor the job status complete on a clear then + * wait for a container change. + */ + + if ((((u32 *)aifcmd->data)[1] == cpu_to_le32(AifJobCtrZero)) + && ((((u32 *)aifcmd->data)[6] == ((u32 *)aifcmd->data)[5]) + || (((u32 *)aifcmd->data)[4] == cpu_to_le32(AifJobStsSuccess)))) { + for (container = 0; + container < dev->maximum_num_containers; + ++container) { + /* + * Stomp on all config sequencing for all + * containers? + */ + dev->fsa_dev[container].config_waiting_on = + AifEnContainerChange; + dev->fsa_dev[container].config_needed = ADD; + } + } + if ((((u32 *)aifcmd->data)[1] == cpu_to_le32(AifJobCtrZero)) + && (((u32 *)aifcmd->data)[6] == 0) + && (((u32 *)aifcmd->data)[4] == cpu_to_le32(AifJobStsRunning))) { + for (container = 0; + container < dev->maximum_num_containers; + ++container) { + /* + * Stomp on all config sequencing for all + * containers? + */ + dev->fsa_dev[container].config_waiting_on = + AifEnContainerChange; + dev->fsa_dev[container].config_needed = DELETE; + } + } + break; + } + + device_config_needed = NOTHING; + for (container = 0; container < dev->maximum_num_containers; + ++container) { + if ((dev->fsa_dev[container].config_waiting_on == 0) + && (dev->fsa_dev[container].config_needed != NOTHING)) { + device_config_needed = + dev->fsa_dev[container].config_needed; + dev->fsa_dev[container].config_needed = NOTHING; + break; + } + } + if (device_config_needed == NOTHING) + return; + + /* + * If we decided that a re-configuration needs to be done, + * schedule it here on the way out the door, please close the door + * behind you. + */ + + busy = 0; + + + /* + * Find the Scsi_Device associated with the SCSI address, + * and mark it as changed, invalidating the cache. This deals + * with changes to existing device IDs. + */ + + if (!dev || !dev->scsi_host_ptr) + return; + /* + * force reload of disk info via probe_container + */ + if ((device_config_needed == CHANGE) + && (dev->fsa_dev[container].valid == 1)) + dev->fsa_dev[container].valid = 2; + if ((device_config_needed == CHANGE) || + (device_config_needed == ADD)) + probe_container(dev, container); + device = scsi_device_lookup(dev->scsi_host_ptr, + CONTAINER_TO_CHANNEL(container), + CONTAINER_TO_ID(container), + CONTAINER_TO_LUN(container)); + if (device) { + switch (device_config_needed) { + case DELETE: + scsi_remove_device(device); + break; + case CHANGE: + if (!dev->fsa_dev[container].valid) { + scsi_remove_device(device); + break; + } + scsi_rescan_device(&device->sdev_gendev); + + default: + break; + } + scsi_device_put(device); + } + if (device_config_needed == ADD) { + scsi_add_device(dev->scsi_host_ptr, + CONTAINER_TO_CHANNEL(container), + CONTAINER_TO_ID(container), + CONTAINER_TO_LUN(container)); + } + +} + /** * aac_command_thread - command processing thread * @dev: Adapter to monitor @@ -805,7 +1019,6 @@ int aac_command_thread(struct aac_dev * dev) { struct hw_fib *hw_fib, *hw_newfib; struct fib *fib, *newfib; - struct aac_queue_block *queues = dev->queues; struct aac_fib_context *fibctx; unsigned long flags; DECLARE_WAITQUEUE(wait, current); @@ -825,21 +1038,22 @@ int aac_command_thread(struct aac_dev * dev) * Let the DPC know it has a place to send the AIF's to. */ dev->aif_thread = 1; - add_wait_queue(&queues->queue[HostNormCmdQueue].cmdready, &wait); + add_wait_queue(&dev->queues->queue[HostNormCmdQueue].cmdready, &wait); set_current_state(TASK_INTERRUPTIBLE); + dprintk ((KERN_INFO "aac_command_thread start\n")); while(1) { - spin_lock_irqsave(queues->queue[HostNormCmdQueue].lock, flags); - while(!list_empty(&(queues->queue[HostNormCmdQueue].cmdq))) { + spin_lock_irqsave(dev->queues->queue[HostNormCmdQueue].lock, flags); + while(!list_empty(&(dev->queues->queue[HostNormCmdQueue].cmdq))) { struct list_head *entry; struct aac_aifcmd * aifcmd; set_current_state(TASK_RUNNING); - - entry = queues->queue[HostNormCmdQueue].cmdq.next; + + entry = dev->queues->queue[HostNormCmdQueue].cmdq.next; list_del(entry); - - spin_unlock_irqrestore(queues->queue[HostNormCmdQueue].lock, flags); + + spin_unlock_irqrestore(dev->queues->queue[HostNormCmdQueue].lock, flags); fib = list_entry(entry, struct fib, fiblink); /* * We will process the FIB here or pass it to a @@ -860,6 +1074,7 @@ int aac_command_thread(struct aac_dev * dev) aifcmd = (struct aac_aifcmd *) hw_fib->data; if (aifcmd->command == cpu_to_le32(AifCmdDriverNotify)) { /* Handle Driver Notify Events */ + aac_handle_aif(dev, fib); *(__le32 *)hw_fib->data = cpu_to_le32(ST_OK); fib_adapter_complete(fib, (u16)sizeof(u32)); } else { @@ -869,9 +1084,62 @@ int aac_command_thread(struct aac_dev * dev) u32 time_now, time_last; unsigned long flagv; - + unsigned num; + struct hw_fib ** hw_fib_pool, ** hw_fib_p; + struct fib ** fib_pool, ** fib_p; + + /* Sniff events */ + if ((aifcmd->command == + cpu_to_le32(AifCmdEventNotify)) || + (aifcmd->command == + cpu_to_le32(AifCmdJobProgress))) { + aac_handle_aif(dev, fib); + } + time_now = jiffies/HZ; + /* + * Warning: no sleep allowed while + * holding spinlock. We take the estimate + * and pre-allocate a set of fibs outside the + * lock. + */ + num = le32_to_cpu(dev->init->AdapterFibsSize) + / sizeof(struct hw_fib); /* some extra */ + spin_lock_irqsave(&dev->fib_lock, flagv); + entry = dev->fib_list.next; + while (entry != &dev->fib_list) { + entry = entry->next; + ++num; + } + spin_unlock_irqrestore(&dev->fib_lock, flagv); + hw_fib_pool = NULL; + fib_pool = NULL; + if (num + && ((hw_fib_pool = kmalloc(sizeof(struct hw_fib *) * num, GFP_KERNEL))) + && ((fib_pool = kmalloc(sizeof(struct fib *) * num, GFP_KERNEL)))) { + hw_fib_p = hw_fib_pool; + fib_p = fib_pool; + while (hw_fib_p < &hw_fib_pool[num]) { + if (!(*(hw_fib_p++) = kmalloc(sizeof(struct hw_fib), GFP_KERNEL))) { + --hw_fib_p; + break; + } + if (!(*(fib_p++) = kmalloc(sizeof(struct fib), GFP_KERNEL))) { + kfree(*(--hw_fib_p)); + break; + } + } + if ((num = hw_fib_p - hw_fib_pool) == 0) { + kfree(fib_pool); + fib_pool = NULL; + kfree(hw_fib_pool); + hw_fib_pool = NULL; + } + } else if (hw_fib_pool) { + kfree(hw_fib_pool); + hw_fib_pool = NULL; + } spin_lock_irqsave(&dev->fib_lock, flagv); entry = dev->fib_list.next; /* @@ -880,6 +1148,8 @@ int aac_command_thread(struct aac_dev * dev) * fib, and then set the event to wake up the * thread that is waiting for it. */ + hw_fib_p = hw_fib_pool; + fib_p = fib_pool; while (entry != &dev->fib_list) { /* * Extract the fibctx @@ -912,9 +1182,11 @@ int aac_command_thread(struct aac_dev * dev) * Warning: no sleep allowed while * holding spinlock */ - hw_newfib = kmalloc(sizeof(struct hw_fib), GFP_ATOMIC); - newfib = kmalloc(sizeof(struct fib), GFP_ATOMIC); - if (newfib && hw_newfib) { + if (hw_fib_p < &hw_fib_pool[num]) { + hw_newfib = *hw_fib_p; + *(hw_fib_p++) = NULL; + newfib = *fib_p; + *(fib_p++) = NULL; /* * Make the copy of the FIB */ @@ -929,15 +1201,11 @@ int aac_command_thread(struct aac_dev * dev) fibctx->count++; /* * Set the event to wake up the - * thread that will waiting. + * thread that is waiting. */ up(&fibctx->wait_sem); } else { printk(KERN_WARNING "aifd: didn't allocate NewFib.\n"); - if(newfib) - kfree(newfib); - if(hw_newfib) - kfree(hw_newfib); } entry = entry->next; } @@ -947,21 +1215,38 @@ int aac_command_thread(struct aac_dev * dev) *(__le32 *)hw_fib->data = cpu_to_le32(ST_OK); fib_adapter_complete(fib, sizeof(u32)); spin_unlock_irqrestore(&dev->fib_lock, flagv); + /* Free up the remaining resources */ + hw_fib_p = hw_fib_pool; + fib_p = fib_pool; + while (hw_fib_p < &hw_fib_pool[num]) { + if (*hw_fib_p) + kfree(*hw_fib_p); + if (*fib_p) + kfree(*fib_p); + ++fib_p; + ++hw_fib_p; + } + if (hw_fib_pool) + kfree(hw_fib_pool); + if (fib_pool) + kfree(fib_pool); } - spin_lock_irqsave(queues->queue[HostNormCmdQueue].lock, flags); kfree(fib); + spin_lock_irqsave(dev->queues->queue[HostNormCmdQueue].lock, flags); } /* * There are no more AIF's */ - spin_unlock_irqrestore(queues->queue[HostNormCmdQueue].lock, flags); + spin_unlock_irqrestore(dev->queues->queue[HostNormCmdQueue].lock, flags); schedule(); if(signal_pending(current)) break; set_current_state(TASK_INTERRUPTIBLE); } - remove_wait_queue(&queues->queue[HostNormCmdQueue].cmdready, &wait); + if (dev->queues) + remove_wait_queue(&dev->queues->queue[HostNormCmdQueue].cmdready, &wait); dev->aif_thread = 0; complete_and_exit(&dev->aif_completion, 0); + return 0; } diff --git a/trunk/drivers/scsi/aacraid/linit.c b/trunk/drivers/scsi/aacraid/linit.c index 4ff29d7f5825..de8490a92831 100644 --- a/trunk/drivers/scsi/aacraid/linit.c +++ b/trunk/drivers/scsi/aacraid/linit.c @@ -748,7 +748,8 @@ static int __devinit aac_probe_one(struct pci_dev *pdev, unique_id++; } - if (pci_enable_device(pdev)) + error = pci_enable_device(pdev); + if (error) goto out; if (pci_set_dma_mask(pdev, 0xFFFFFFFFULL) || @@ -772,6 +773,7 @@ static int __devinit aac_probe_one(struct pci_dev *pdev, shost->irq = pdev->irq; shost->base = pci_resource_start(pdev, 0); shost->unique_id = unique_id; + shost->max_cmd_len = 16; aac = (struct aac_dev *)shost->hostdata; aac->scsi_host_ptr = shost; @@ -799,7 +801,9 @@ static int __devinit aac_probe_one(struct pci_dev *pdev, goto out_free_fibs; aac->maximum_num_channels = aac_drivers[index].channels; - aac_get_adapter_info(aac); + error = aac_get_adapter_info(aac); + if (error < 0) + goto out_deinit; /* * Lets override negotiations and drop the maximum SG limit to 34 @@ -927,8 +931,8 @@ static int __init aac_init(void) printk(KERN_INFO "Adaptec %s driver (%s)\n", AAC_DRIVERNAME, aac_driver_version); - error = pci_module_init(&aac_pci_driver); - if (error) + error = pci_register_driver(&aac_pci_driver); + if (error < 0) return error; aac_cfg_major = register_chrdev( 0, "aac", &aac_cfg_fops); diff --git a/trunk/drivers/scsi/ahci.c b/trunk/drivers/scsi/ahci.c index 5ec866b00479..c2c8fa828e24 100644 --- a/trunk/drivers/scsi/ahci.c +++ b/trunk/drivers/scsi/ahci.c @@ -672,36 +672,17 @@ static irqreturn_t ahci_interrupt (int irq, void *dev_instance, struct pt_regs * for (i = 0; i < host_set->n_ports; i++) { struct ata_port *ap; + u32 tmp; - if (!(irq_stat & (1 << i))) - continue; - + VPRINTK("port %u\n", i); ap = host_set->ports[i]; - if (ap) { + tmp = irq_stat & (1 << i); + if (tmp && ap) { struct ata_queued_cmd *qc; qc = ata_qc_from_tag(ap, ap->active_tag); - if (!ahci_host_intr(ap, qc)) - if (ata_ratelimit()) { - struct pci_dev *pdev = - to_pci_dev(ap->host_set->dev); - printk(KERN_WARNING - "ahci(%s): unhandled interrupt on port %u\n", - pci_name(pdev), i); - } - - VPRINTK("port %u\n", i); - } else { - VPRINTK("port %u (no irq)\n", i); - if (ata_ratelimit()) { - struct pci_dev *pdev = - to_pci_dev(ap->host_set->dev); - printk(KERN_WARNING - "ahci(%s): interrupt on disabled port %u\n", - pci_name(pdev), i); - } + if (ahci_host_intr(ap, qc)) + irq_ack |= (1 << i); } - - irq_ack |= (1 << i); } if (irq_ack) { diff --git a/trunk/drivers/scsi/aic7xxx/aic7770_osm.c b/trunk/drivers/scsi/aic7xxx/aic7770_osm.c index 70c5fb59c9ea..d754b3267863 100644 --- a/trunk/drivers/scsi/aic7xxx/aic7770_osm.c +++ b/trunk/drivers/scsi/aic7xxx/aic7770_osm.c @@ -112,6 +112,9 @@ aic7770_remove(struct device *dev) struct ahc_softc *ahc = dev_get_drvdata(dev); u_long s; + if (ahc->platform_data && ahc->platform_data->host) + scsi_remove_host(ahc->platform_data->host); + ahc_lock(ahc, &s); ahc_intr_enable(ahc, FALSE); ahc_unlock(ahc, &s); diff --git a/trunk/drivers/scsi/aic7xxx/aic79xx_osm.c b/trunk/drivers/scsi/aic7xxx/aic79xx_osm.c index 6b6d4e287793..95c285cc83e4 100644 --- a/trunk/drivers/scsi/aic7xxx/aic79xx_osm.c +++ b/trunk/drivers/scsi/aic7xxx/aic79xx_osm.c @@ -1192,11 +1192,6 @@ ahd_platform_free(struct ahd_softc *ahd) int i, j; if (ahd->platform_data != NULL) { - if (ahd->platform_data->host != NULL) { - scsi_remove_host(ahd->platform_data->host); - scsi_host_put(ahd->platform_data->host); - } - /* destroy all of the device and target objects */ for (i = 0; i < AHD_NUM_TARGETS; i++) { starget = ahd->platform_data->starget[i]; @@ -1226,6 +1221,9 @@ ahd_platform_free(struct ahd_softc *ahd) release_mem_region(ahd->platform_data->mem_busaddr, 0x1000); } + if (ahd->platform_data->host) + scsi_host_put(ahd->platform_data->host); + free(ahd->platform_data, M_DEVBUF); } } diff --git a/trunk/drivers/scsi/aic7xxx/aic79xx_osm_pci.c b/trunk/drivers/scsi/aic7xxx/aic79xx_osm_pci.c index 390b53852d4b..bf360ae021ab 100644 --- a/trunk/drivers/scsi/aic7xxx/aic79xx_osm_pci.c +++ b/trunk/drivers/scsi/aic7xxx/aic79xx_osm_pci.c @@ -95,6 +95,9 @@ ahd_linux_pci_dev_remove(struct pci_dev *pdev) struct ahd_softc *ahd = pci_get_drvdata(pdev); u_long s; + if (ahd->platform_data && ahd->platform_data->host) + scsi_remove_host(ahd->platform_data->host); + ahd_lock(ahd, &s); ahd_intr_enable(ahd, FALSE); ahd_unlock(ahd, &s); diff --git a/trunk/drivers/scsi/aic7xxx/aic7xxx_osm.c b/trunk/drivers/scsi/aic7xxx/aic7xxx_osm.c index 876d1de8480d..6ee1435d37fa 100644 --- a/trunk/drivers/scsi/aic7xxx/aic7xxx_osm.c +++ b/trunk/drivers/scsi/aic7xxx/aic7xxx_osm.c @@ -1209,11 +1209,6 @@ ahc_platform_free(struct ahc_softc *ahc) int i, j; if (ahc->platform_data != NULL) { - if (ahc->platform_data->host != NULL) { - scsi_remove_host(ahc->platform_data->host); - scsi_host_put(ahc->platform_data->host); - } - /* destroy all of the device and target objects */ for (i = 0; i < AHC_NUM_TARGETS; i++) { starget = ahc->platform_data->starget[i]; @@ -1242,6 +1237,9 @@ ahc_platform_free(struct ahc_softc *ahc) 0x1000); } + if (ahc->platform_data->host) + scsi_host_put(ahc->platform_data->host); + free(ahc->platform_data, M_DEVBUF); } } diff --git a/trunk/drivers/scsi/aic7xxx/aic7xxx_osm_pci.c b/trunk/drivers/scsi/aic7xxx/aic7xxx_osm_pci.c index 3ce77ddc889e..cb30d9c1153d 100644 --- a/trunk/drivers/scsi/aic7xxx/aic7xxx_osm_pci.c +++ b/trunk/drivers/scsi/aic7xxx/aic7xxx_osm_pci.c @@ -143,6 +143,9 @@ ahc_linux_pci_dev_remove(struct pci_dev *pdev) struct ahc_softc *ahc = pci_get_drvdata(pdev); u_long s; + if (ahc->platform_data && ahc->platform_data->host) + scsi_remove_host(ahc->platform_data->host); + ahc_lock(ahc, &s); ahc_intr_enable(ahc, FALSE); ahc_unlock(ahc, &s); diff --git a/trunk/drivers/scsi/hosts.c b/trunk/drivers/scsi/hosts.c index f2a72d33132c..02fe371b0ab8 100644 --- a/trunk/drivers/scsi/hosts.c +++ b/trunk/drivers/scsi/hosts.c @@ -176,6 +176,7 @@ void scsi_remove_host(struct Scsi_Host *shost) transport_unregister_device(&shost->shost_gendev); class_device_unregister(&shost->shost_classdev); device_del(&shost->shost_gendev); + scsi_proc_hostdir_rm(shost->hostt); } EXPORT_SYMBOL(scsi_remove_host); @@ -262,7 +263,6 @@ static void scsi_host_dev_release(struct device *dev) if (shost->work_q) destroy_workqueue(shost->work_q); - scsi_proc_hostdir_rm(shost->hostt); scsi_destroy_command_freelist(shost); kfree(shost->shost_data); diff --git a/trunk/drivers/scsi/libata-core.c b/trunk/drivers/scsi/libata-core.c index d568914c4344..e5b01997117a 100644 --- a/trunk/drivers/scsi/libata-core.c +++ b/trunk/drivers/scsi/libata-core.c @@ -48,7 +48,6 @@ #include #include #include -#include #include #include "scsi.h" #include "scsi_priv.h" @@ -63,7 +62,6 @@ static unsigned int ata_busy_sleep (struct ata_port *ap, unsigned long tmout_pat, unsigned long tmout); -static void ata_dev_init_params(struct ata_port *ap, struct ata_device *dev); static void ata_set_mode(struct ata_port *ap); static void ata_dev_set_xfermode(struct ata_port *ap, struct ata_device *dev); static unsigned int ata_get_mode_mask(struct ata_port *ap, int shift); @@ -71,6 +69,7 @@ static int fgb(u32 bitmap); static int ata_choose_xfer_mode(struct ata_port *ap, u8 *xfer_mode_out, unsigned int *xfer_shift_out); +static int ata_qc_complete_noop(struct ata_queued_cmd *qc, u8 drv_stat); static void __ata_qc_complete(struct ata_queued_cmd *qc); static unsigned int ata_unique_id = 1; @@ -1132,7 +1131,7 @@ static inline void ata_dump_id(struct ata_device *dev) static void ata_dev_identify(struct ata_port *ap, unsigned int device) { struct ata_device *dev = &ap->device[device]; - unsigned int major_version; + unsigned int i; u16 tmp; unsigned long xfer_modes; u8 status; @@ -1230,9 +1229,9 @@ static void ata_dev_identify(struct ata_port *ap, unsigned int device) * common ATA, ATAPI feature tests */ - /* we require DMA support (bits 8 of word 49) */ - if (!ata_id_has_dma(dev->id)) { - printk(KERN_DEBUG "ata%u: no dma\n", ap->id); + /* we require LBA and DMA support (bits 8 & 9 of word 49) */ + if (!ata_id_has_dma(dev->id) || !ata_id_has_lba(dev->id)) { + printk(KERN_DEBUG "ata%u: no dma/lba\n", ap->id); goto err_out_nosup; } @@ -1252,69 +1251,32 @@ static void ata_dev_identify(struct ata_port *ap, unsigned int device) if (!ata_id_is_ata(dev->id)) /* sanity check */ goto err_out_nosup; - /* get major version */ tmp = dev->id[ATA_ID_MAJOR_VER]; - for (major_version = 14; major_version >= 1; major_version--) - if (tmp & (1 << major_version)) + for (i = 14; i >= 1; i--) + if (tmp & (1 << i)) break; - /* - * The exact sequence expected by certain pre-ATA4 drives is: - * SRST RESET - * IDENTIFY - * INITIALIZE DEVICE PARAMETERS - * anything else.. - * Some drives were very specific about that exact sequence. - */ - if (major_version < 4 || (!ata_id_has_lba(dev->id))) - ata_dev_init_params(ap, dev); - - if (ata_id_has_lba(dev->id)) { - dev->flags |= ATA_DFLAG_LBA; - - if (ata_id_has_lba48(dev->id)) { - dev->flags |= ATA_DFLAG_LBA48; - dev->n_sectors = ata_id_u64(dev->id, 100); - } else { - dev->n_sectors = ata_id_u32(dev->id, 60); - } - - /* print device info to dmesg */ - printk(KERN_INFO "ata%u: dev %u ATA-%d, max %s, %Lu sectors:%s\n", - ap->id, device, - major_version, - ata_mode_string(xfer_modes), - (unsigned long long)dev->n_sectors, - dev->flags & ATA_DFLAG_LBA48 ? " LBA48" : " LBA"); - } else { - /* CHS */ - - /* Default translation */ - dev->cylinders = dev->id[1]; - dev->heads = dev->id[3]; - dev->sectors = dev->id[6]; - dev->n_sectors = dev->cylinders * dev->heads * dev->sectors; - - if (ata_id_current_chs_valid(dev->id)) { - /* Current CHS translation is valid. */ - dev->cylinders = dev->id[54]; - dev->heads = dev->id[55]; - dev->sectors = dev->id[56]; - - dev->n_sectors = ata_id_u32(dev->id, 57); - } - - /* print device info to dmesg */ - printk(KERN_INFO "ata%u: dev %u ATA-%d, max %s, %Lu sectors: CHS %d/%d/%d\n", - ap->id, device, - major_version, - ata_mode_string(xfer_modes), - (unsigned long long)dev->n_sectors, - (int)dev->cylinders, (int)dev->heads, (int)dev->sectors); + /* we require at least ATA-3 */ + if (i < 3) { + printk(KERN_DEBUG "ata%u: no ATA-3\n", ap->id); + goto err_out_nosup; + } + if (ata_id_has_lba48(dev->id)) { + dev->flags |= ATA_DFLAG_LBA48; + dev->n_sectors = ata_id_u64(dev->id, 100); + } else { + dev->n_sectors = ata_id_u32(dev->id, 60); } ap->host->max_cmd_len = 16; + + /* print device info to dmesg */ + printk(KERN_INFO "ata%u: dev %u ATA, max %s, %Lu sectors:%s\n", + ap->id, device, + ata_mode_string(xfer_modes), + (unsigned long long)dev->n_sectors, + dev->flags & ATA_DFLAG_LBA48 ? " lba48" : ""); } /* ATAPI-specific feature tests */ @@ -2181,54 +2143,6 @@ static void ata_dev_set_xfermode(struct ata_port *ap, struct ata_device *dev) DPRINTK("EXIT\n"); } -/** - * ata_dev_init_params - Issue INIT DEV PARAMS command - * @ap: Port associated with device @dev - * @dev: Device to which command will be sent - * - * LOCKING: - */ - -static void ata_dev_init_params(struct ata_port *ap, struct ata_device *dev) -{ - DECLARE_COMPLETION(wait); - struct ata_queued_cmd *qc; - int rc; - unsigned long flags; - u16 sectors = dev->id[6]; - u16 heads = dev->id[3]; - - /* Number of sectors per track 1-255. Number of heads 1-16 */ - if (sectors < 1 || sectors > 255 || heads < 1 || heads > 16) - return; - - /* set up init dev params taskfile */ - DPRINTK("init dev params \n"); - - qc = ata_qc_new_init(ap, dev); - BUG_ON(qc == NULL); - - qc->tf.command = ATA_CMD_INIT_DEV_PARAMS; - qc->tf.flags |= ATA_TFLAG_ISADDR | ATA_TFLAG_DEVICE; - qc->tf.protocol = ATA_PROT_NODATA; - qc->tf.nsect = sectors; - qc->tf.device |= (heads - 1) & 0x0f; /* max head = num. of heads - 1 */ - - qc->waiting = &wait; - qc->complete_fn = ata_qc_complete_noop; - - spin_lock_irqsave(&ap->host_set->lock, flags); - rc = ata_qc_issue(qc); - spin_unlock_irqrestore(&ap->host_set->lock, flags); - - if (rc) - ata_port_disable(ap); - else - wait_for_completion(&wait); - - DPRINTK("EXIT\n"); -} - /** * ata_sg_clean - Unmap DMA memory associated with command * @qc: Command containing DMA memory to be released @@ -2511,20 +2425,20 @@ void ata_poll_qc_complete(struct ata_queued_cmd *qc, u8 drv_stat) static unsigned long ata_pio_poll(struct ata_port *ap) { u8 status; - unsigned int poll_state = HSM_ST_UNKNOWN; - unsigned int reg_state = HSM_ST_UNKNOWN; - const unsigned int tmout_state = HSM_ST_TMOUT; - - switch (ap->hsm_task_state) { - case HSM_ST: - case HSM_ST_POLL: - poll_state = HSM_ST_POLL; - reg_state = HSM_ST; + unsigned int poll_state = PIO_ST_UNKNOWN; + unsigned int reg_state = PIO_ST_UNKNOWN; + const unsigned int tmout_state = PIO_ST_TMOUT; + + switch (ap->pio_task_state) { + case PIO_ST: + case PIO_ST_POLL: + poll_state = PIO_ST_POLL; + reg_state = PIO_ST; break; - case HSM_ST_LAST: - case HSM_ST_LAST_POLL: - poll_state = HSM_ST_LAST_POLL; - reg_state = HSM_ST_LAST; + case PIO_ST_LAST: + case PIO_ST_LAST_POLL: + poll_state = PIO_ST_LAST_POLL; + reg_state = PIO_ST_LAST; break; default: BUG(); @@ -2534,14 +2448,14 @@ static unsigned long ata_pio_poll(struct ata_port *ap) status = ata_chk_status(ap); if (status & ATA_BUSY) { if (time_after(jiffies, ap->pio_task_timeout)) { - ap->hsm_task_state = tmout_state; + ap->pio_task_state = tmout_state; return 0; } - ap->hsm_task_state = poll_state; + ap->pio_task_state = poll_state; return ATA_SHORT_PAUSE; } - ap->hsm_task_state = reg_state; + ap->pio_task_state = reg_state; return 0; } @@ -2566,14 +2480,14 @@ static int ata_pio_complete (struct ata_port *ap) * we enter, BSY will be cleared in a chk-status or two. If not, * the drive is probably seeking or something. Snooze for a couple * msecs, then chk-status again. If still busy, fall back to - * HSM_ST_POLL state. + * PIO_ST_POLL state. */ drv_stat = ata_busy_wait(ap, ATA_BUSY | ATA_DRQ, 10); if (drv_stat & (ATA_BUSY | ATA_DRQ)) { msleep(2); drv_stat = ata_busy_wait(ap, ATA_BUSY | ATA_DRQ, 10); if (drv_stat & (ATA_BUSY | ATA_DRQ)) { - ap->hsm_task_state = HSM_ST_LAST_POLL; + ap->pio_task_state = PIO_ST_LAST_POLL; ap->pio_task_timeout = jiffies + ATA_TMOUT_PIO; return 0; } @@ -2581,14 +2495,14 @@ static int ata_pio_complete (struct ata_port *ap) drv_stat = ata_wait_idle(ap); if (!ata_ok(drv_stat)) { - ap->hsm_task_state = HSM_ST_ERR; + ap->pio_task_state = PIO_ST_ERR; return 0; } qc = ata_qc_from_tag(ap, ap->active_tag); assert(qc != NULL); - ap->hsm_task_state = HSM_ST_IDLE; + ap->pio_task_state = PIO_ST_IDLE; ata_poll_qc_complete(qc, drv_stat); @@ -2748,7 +2662,7 @@ static void ata_pio_sector(struct ata_queued_cmd *qc) unsigned char *buf; if (qc->cursect == (qc->nsect - 1)) - ap->hsm_task_state = HSM_ST_LAST; + ap->pio_task_state = PIO_ST_LAST; page = sg[qc->cursg].page; offset = sg[qc->cursg].offset + qc->cursg_ofs * ATA_SECT_SIZE; @@ -2798,7 +2712,7 @@ static void __atapi_pio_bytes(struct ata_queued_cmd *qc, unsigned int bytes) unsigned int offset, count; if (qc->curbytes + bytes >= qc->nbytes) - ap->hsm_task_state = HSM_ST_LAST; + ap->pio_task_state = PIO_ST_LAST; next_sg: if (unlikely(qc->cursg >= qc->n_elem)) { @@ -2820,7 +2734,7 @@ static void __atapi_pio_bytes(struct ata_queued_cmd *qc, unsigned int bytes) for (i = 0; i < words; i++) ata_data_xfer(ap, (unsigned char*)pad_buf, 2, do_write); - ap->hsm_task_state = HSM_ST_LAST; + ap->pio_task_state = PIO_ST_LAST; return; } @@ -2901,7 +2815,7 @@ static void atapi_pio_bytes(struct ata_queued_cmd *qc) err_out: printk(KERN_INFO "ata%u: dev %u: ATAPI check failed\n", ap->id, dev->devno); - ap->hsm_task_state = HSM_ST_ERR; + ap->pio_task_state = PIO_ST_ERR; } /** @@ -2923,14 +2837,14 @@ static void ata_pio_block(struct ata_port *ap) * a chk-status or two. If not, the drive is probably seeking * or something. Snooze for a couple msecs, then * chk-status again. If still busy, fall back to - * HSM_ST_POLL state. + * PIO_ST_POLL state. */ status = ata_busy_wait(ap, ATA_BUSY, 5); if (status & ATA_BUSY) { msleep(2); status = ata_busy_wait(ap, ATA_BUSY, 10); if (status & ATA_BUSY) { - ap->hsm_task_state = HSM_ST_POLL; + ap->pio_task_state = PIO_ST_POLL; ap->pio_task_timeout = jiffies + ATA_TMOUT_PIO; return; } @@ -2942,7 +2856,7 @@ static void ata_pio_block(struct ata_port *ap) if (is_atapi_taskfile(&qc->tf)) { /* no more data to transfer or unsupported ATAPI command */ if ((status & ATA_DRQ) == 0) { - ap->hsm_task_state = HSM_ST_LAST; + ap->pio_task_state = PIO_ST_LAST; return; } @@ -2950,7 +2864,7 @@ static void ata_pio_block(struct ata_port *ap) } else { /* handle BSY=0, DRQ=0 as error */ if ((status & ATA_DRQ) == 0) { - ap->hsm_task_state = HSM_ST_ERR; + ap->pio_task_state = PIO_ST_ERR; return; } @@ -2970,7 +2884,7 @@ static void ata_pio_error(struct ata_port *ap) printk(KERN_WARNING "ata%u: PIO error, drv_stat 0x%x\n", ap->id, drv_stat); - ap->hsm_task_state = HSM_ST_IDLE; + ap->pio_task_state = PIO_ST_IDLE; ata_poll_qc_complete(qc, drv_stat | ATA_ERR); } @@ -2985,25 +2899,25 @@ static void ata_pio_task(void *_data) timeout = 0; qc_completed = 0; - switch (ap->hsm_task_state) { - case HSM_ST_IDLE: + switch (ap->pio_task_state) { + case PIO_ST_IDLE: return; - case HSM_ST: + case PIO_ST: ata_pio_block(ap); break; - case HSM_ST_LAST: + case PIO_ST_LAST: qc_completed = ata_pio_complete(ap); break; - case HSM_ST_POLL: - case HSM_ST_LAST_POLL: + case PIO_ST_POLL: + case PIO_ST_LAST_POLL: timeout = ata_pio_poll(ap); break; - case HSM_ST_TMOUT: - case HSM_ST_ERR: + case PIO_ST_TMOUT: + case PIO_ST_ERR: ata_pio_error(ap); return; } @@ -3014,6 +2928,52 @@ static void ata_pio_task(void *_data) goto fsm_start; } +static void atapi_request_sense(struct ata_port *ap, struct ata_device *dev, + struct scsi_cmnd *cmd) +{ + DECLARE_COMPLETION(wait); + struct ata_queued_cmd *qc; + unsigned long flags; + int rc; + + DPRINTK("ATAPI request sense\n"); + + qc = ata_qc_new_init(ap, dev); + BUG_ON(qc == NULL); + + /* FIXME: is this needed? */ + memset(cmd->sense_buffer, 0, sizeof(cmd->sense_buffer)); + + ata_sg_init_one(qc, cmd->sense_buffer, sizeof(cmd->sense_buffer)); + qc->dma_dir = DMA_FROM_DEVICE; + + memset(&qc->cdb, 0, ap->cdb_len); + qc->cdb[0] = REQUEST_SENSE; + qc->cdb[4] = SCSI_SENSE_BUFFERSIZE; + + qc->tf.flags |= ATA_TFLAG_ISADDR | ATA_TFLAG_DEVICE; + qc->tf.command = ATA_CMD_PACKET; + + qc->tf.protocol = ATA_PROT_ATAPI; + qc->tf.lbam = (8 * 1024) & 0xff; + qc->tf.lbah = (8 * 1024) >> 8; + qc->nbytes = SCSI_SENSE_BUFFERSIZE; + + qc->waiting = &wait; + qc->complete_fn = ata_qc_complete_noop; + + spin_lock_irqsave(&ap->host_set->lock, flags); + rc = ata_qc_issue(qc); + spin_unlock_irqrestore(&ap->host_set->lock, flags); + + if (rc) + ata_port_disable(ap); + else + wait_for_completion(&wait); + + DPRINTK("EXIT\n"); +} + /** * ata_qc_timeout - Handle timeout of queued command * @qc: Command that timed out @@ -3131,14 +3091,14 @@ void ata_eng_timeout(struct ata_port *ap) DPRINTK("ENTER\n"); qc = ata_qc_from_tag(ap, ap->active_tag); - if (qc) - ata_qc_timeout(qc); - else { + if (!qc) { printk(KERN_ERR "ata%u: BUG: timeout without command\n", ap->id); goto out; } + ata_qc_timeout(qc); + out: DPRINTK("EXIT\n"); } @@ -3196,18 +3156,14 @@ struct ata_queued_cmd *ata_qc_new_init(struct ata_port *ap, ata_tf_init(ap, &qc->tf, dev->devno); - if (dev->flags & ATA_DFLAG_LBA) { - qc->tf.flags |= ATA_TFLAG_LBA; - - if (dev->flags & ATA_DFLAG_LBA48) - qc->tf.flags |= ATA_TFLAG_LBA48; - } + if (dev->flags & ATA_DFLAG_LBA48) + qc->tf.flags |= ATA_TFLAG_LBA48; } return qc; } -int ata_qc_complete_noop(struct ata_queued_cmd *qc, u8 drv_stat) +static int ata_qc_complete_noop(struct ata_queued_cmd *qc, u8 drv_stat) { return 0; } @@ -3404,7 +3360,7 @@ int ata_qc_issue_prot(struct ata_queued_cmd *qc) case ATA_PROT_PIO: /* load tf registers, initiate polling pio */ ata_qc_set_polling(qc); ata_tf_to_host_nolock(ap, &qc->tf); - ap->hsm_task_state = HSM_ST; + ap->pio_task_state = PIO_ST; queue_work(ata_wq, &ap->pio_task); break; @@ -3630,7 +3586,7 @@ u8 ata_bmdma_status(struct ata_port *ap) void __iomem *mmio = (void __iomem *) ap->ioaddr.bmdma_addr; host_stat = readb(mmio + ATA_DMA_STATUS); } else - host_stat = inb(ap->ioaddr.bmdma_addr + ATA_DMA_STATUS); + host_stat = inb(ap->ioaddr.bmdma_addr + ATA_DMA_STATUS); return host_stat; } @@ -3850,7 +3806,7 @@ static void atapi_packet_task(void *_data) ata_data_xfer(ap, qc->cdb, ap->cdb_len, 1); /* PIO commands are handled by polling */ - ap->hsm_task_state = HSM_ST; + ap->pio_task_state = PIO_ST; queue_work(ata_wq, &ap->pio_task); } @@ -4157,7 +4113,7 @@ int ata_device_add(struct ata_probe_ent *ent) for (i = 0; i < count; i++) { struct ata_port *ap = host_set->ports[i]; - ata_scsi_scan_host(ap); + scsi_scan_host(ap->host); } dev_set_drvdata(dev, host_set); @@ -4317,87 +4273,85 @@ void ata_pci_host_stop (struct ata_host_set *host_set) * ata_pci_init_native_mode - Initialize native-mode driver * @pdev: pci device to be initialized * @port: array[2] of pointers to port info structures. - * @ports: bitmap of ports present * * Utility function which allocates and initializes an * ata_probe_ent structure for a standard dual-port * PIO-based IDE controller. The returned ata_probe_ent * structure can be passed to ata_device_add(). The returned * ata_probe_ent structure should then be freed with kfree(). - * - * The caller need only pass the address of the primary port, the - * secondary will be deduced automatically. If the device has non - * standard secondary port mappings this function can be called twice, - * once for each interface. */ struct ata_probe_ent * -ata_pci_init_native_mode(struct pci_dev *pdev, struct ata_port_info **port, int ports) +ata_pci_init_native_mode(struct pci_dev *pdev, struct ata_port_info **port) { struct ata_probe_ent *probe_ent = ata_probe_ent_alloc(pci_dev_to_dev(pdev), port[0]); - int p = 0; - if (!probe_ent) return NULL; + probe_ent->n_ports = 2; probe_ent->irq = pdev->irq; probe_ent->irq_flags = SA_SHIRQ; - if (ports & ATA_PORT_PRIMARY) { - probe_ent->port[p].cmd_addr = pci_resource_start(pdev, 0); - probe_ent->port[p].altstatus_addr = - probe_ent->port[p].ctl_addr = - pci_resource_start(pdev, 1) | ATA_PCI_CTL_OFS; - probe_ent->port[p].bmdma_addr = pci_resource_start(pdev, 4); - ata_std_ports(&probe_ent->port[p]); - p++; - } + probe_ent->port[0].cmd_addr = pci_resource_start(pdev, 0); + probe_ent->port[0].altstatus_addr = + probe_ent->port[0].ctl_addr = + pci_resource_start(pdev, 1) | ATA_PCI_CTL_OFS; + probe_ent->port[0].bmdma_addr = pci_resource_start(pdev, 4); - if (ports & ATA_PORT_SECONDARY) { - probe_ent->port[p].cmd_addr = pci_resource_start(pdev, 2); - probe_ent->port[p].altstatus_addr = - probe_ent->port[p].ctl_addr = - pci_resource_start(pdev, 3) | ATA_PCI_CTL_OFS; - probe_ent->port[p].bmdma_addr = pci_resource_start(pdev, 4) + 8; - ata_std_ports(&probe_ent->port[p]); - p++; - } + probe_ent->port[1].cmd_addr = pci_resource_start(pdev, 2); + probe_ent->port[1].altstatus_addr = + probe_ent->port[1].ctl_addr = + pci_resource_start(pdev, 3) | ATA_PCI_CTL_OFS; + probe_ent->port[1].bmdma_addr = pci_resource_start(pdev, 4) + 8; + + ata_std_ports(&probe_ent->port[0]); + ata_std_ports(&probe_ent->port[1]); - probe_ent->n_ports = p; return probe_ent; } -static struct ata_probe_ent *ata_pci_init_legacy_port(struct pci_dev *pdev, struct ata_port_info **port, int port_num) +static struct ata_probe_ent * +ata_pci_init_legacy_mode(struct pci_dev *pdev, struct ata_port_info **port, + struct ata_probe_ent **ppe2) { - struct ata_probe_ent *probe_ent; + struct ata_probe_ent *probe_ent, *probe_ent2; probe_ent = ata_probe_ent_alloc(pci_dev_to_dev(pdev), port[0]); if (!probe_ent) return NULL; + probe_ent2 = ata_probe_ent_alloc(pci_dev_to_dev(pdev), port[1]); + if (!probe_ent2) { + kfree(probe_ent); + return NULL; + } - - probe_ent->legacy_mode = 1; probe_ent->n_ports = 1; - probe_ent->hard_port_no = port_num; - - switch(port_num) - { - case 0: - probe_ent->irq = 14; - probe_ent->port[0].cmd_addr = 0x1f0; - probe_ent->port[0].altstatus_addr = - probe_ent->port[0].ctl_addr = 0x3f6; - break; - case 1: - probe_ent->irq = 15; - probe_ent->port[0].cmd_addr = 0x170; - probe_ent->port[0].altstatus_addr = - probe_ent->port[0].ctl_addr = 0x376; - break; - } - probe_ent->port[0].bmdma_addr = pci_resource_start(pdev, 4) + 8 * port_num; + probe_ent->irq = 14; + + probe_ent->hard_port_no = 0; + probe_ent->legacy_mode = 1; + + probe_ent2->n_ports = 1; + probe_ent2->irq = 15; + + probe_ent2->hard_port_no = 1; + probe_ent2->legacy_mode = 1; + + probe_ent->port[0].cmd_addr = 0x1f0; + probe_ent->port[0].altstatus_addr = + probe_ent->port[0].ctl_addr = 0x3f6; + probe_ent->port[0].bmdma_addr = pci_resource_start(pdev, 4); + + probe_ent2->port[0].cmd_addr = 0x170; + probe_ent2->port[0].altstatus_addr = + probe_ent2->port[0].ctl_addr = 0x376; + probe_ent2->port[0].bmdma_addr = pci_resource_start(pdev, 4)+8; + ata_std_ports(&probe_ent->port[0]); + ata_std_ports(&probe_ent2->port[0]); + + *ppe2 = probe_ent2; return probe_ent; } @@ -4426,7 +4380,7 @@ static struct ata_probe_ent *ata_pci_init_legacy_port(struct pci_dev *pdev, stru int ata_pci_init_one (struct pci_dev *pdev, struct ata_port_info **port_info, unsigned int n_ports) { - struct ata_probe_ent *probe_ent = NULL, *probe_ent2 = NULL; + struct ata_probe_ent *probe_ent, *probe_ent2 = NULL; struct ata_port_info *port[2]; u8 tmp8, mask; unsigned int legacy_mode = 0; @@ -4443,7 +4397,7 @@ int ata_pci_init_one (struct pci_dev *pdev, struct ata_port_info **port_info, if ((port[0]->host_flags & ATA_FLAG_NO_LEGACY) == 0 && (pdev->class >> 8) == PCI_CLASS_STORAGE_IDE) { - /* TODO: What if one channel is in native mode ... */ + /* TODO: support transitioning to native mode? */ pci_read_config_byte(pdev, PCI_CLASS_PROG, &tmp8); mask = (1 << 2) | (1 << 0); if ((tmp8 & mask) != mask) @@ -4451,20 +4405,11 @@ int ata_pci_init_one (struct pci_dev *pdev, struct ata_port_info **port_info, } /* FIXME... */ - if ((!legacy_mode) && (n_ports > 2)) { - printk(KERN_ERR "ata: BUG: native mode, n_ports > 2\n"); - n_ports = 2; - /* For now */ + if ((!legacy_mode) && (n_ports > 1)) { + printk(KERN_ERR "ata: BUG: native mode, n_ports > 1\n"); + return -EINVAL; } - /* FIXME: Really for ATA it isn't safe because the device may be - multi-purpose and we want to leave it alone if it was already - enabled. Secondly for shared use as Arjan says we want refcounting - - Checking dev->is_enabled is insufficient as this is not set at - boot for the primary video which is BIOS enabled - */ - rc = pci_enable_device(pdev); if (rc) return rc; @@ -4475,7 +4420,6 @@ int ata_pci_init_one (struct pci_dev *pdev, struct ata_port_info **port_info, goto err_out; } - /* FIXME: Should use platform specific mappers for legacy port ranges */ if (legacy_mode) { if (!request_region(0x1f0, 8, "libata")) { struct resource *conflict, res; @@ -4520,17 +4464,10 @@ int ata_pci_init_one (struct pci_dev *pdev, struct ata_port_info **port_info, goto err_out_regions; if (legacy_mode) { - if (legacy_mode & (1 << 0)) - probe_ent = ata_pci_init_legacy_port(pdev, port, 0); - if (legacy_mode & (1 << 1)) - probe_ent2 = ata_pci_init_legacy_port(pdev, port, 1); - } else { - if (n_ports == 2) - probe_ent = ata_pci_init_native_mode(pdev, port, ATA_PORT_PRIMARY | ATA_PORT_SECONDARY); - else - probe_ent = ata_pci_init_native_mode(pdev, port, ATA_PORT_PRIMARY); - } - if (!probe_ent && !probe_ent2) { + probe_ent = ata_pci_init_legacy_mode(pdev, port, &probe_ent2); + } else + probe_ent = ata_pci_init_native_mode(pdev, port); + if (!probe_ent) { rc = -ENOMEM; goto err_out_regions; } @@ -4642,27 +4579,6 @@ static void __exit ata_exit(void) module_init(ata_init); module_exit(ata_exit); -static unsigned long ratelimit_time; -static spinlock_t ata_ratelimit_lock = SPIN_LOCK_UNLOCKED; - -int ata_ratelimit(void) -{ - int rc; - unsigned long flags; - - spin_lock_irqsave(&ata_ratelimit_lock, flags); - - if (time_after(jiffies, ratelimit_time)) { - rc = 1; - ratelimit_time = jiffies + (HZ/5); - } else - rc = 0; - - spin_unlock_irqrestore(&ata_ratelimit_lock, flags); - - return rc; -} - /* * libata is essentially a library of internal helper functions for * low-level ATA host controller drivers. As such, the API/ABI is @@ -4704,7 +4620,6 @@ EXPORT_SYMBOL_GPL(sata_phy_reset); EXPORT_SYMBOL_GPL(__sata_phy_reset); EXPORT_SYMBOL_GPL(ata_bus_reset); EXPORT_SYMBOL_GPL(ata_port_disable); -EXPORT_SYMBOL_GPL(ata_ratelimit); EXPORT_SYMBOL_GPL(ata_scsi_ioctl); EXPORT_SYMBOL_GPL(ata_scsi_queuecmd); EXPORT_SYMBOL_GPL(ata_scsi_error); diff --git a/trunk/drivers/scsi/libata-scsi.c b/trunk/drivers/scsi/libata-scsi.c index 1c3a10fb3c44..104fd9a63e73 100644 --- a/trunk/drivers/scsi/libata-scsi.c +++ b/trunk/drivers/scsi/libata-scsi.c @@ -225,7 +225,7 @@ void ata_to_sense_error(struct ata_queued_cmd *qc, u8 drv_stat) }; int i = 0; - cmd->result = (DRIVER_SENSE << 24) | SAM_STAT_CHECK_CONDITION; + cmd->result = SAM_STAT_CHECK_CONDITION; /* * Is this an error we can process/parse @@ -435,21 +435,10 @@ static unsigned int ata_scsi_start_stop_xlat(struct ata_queued_cmd *qc, return 1; /* power conditions not supported */ if (scsicmd[4] & 0x1) { tf->nsect = 1; /* 1 sector, lba=0 */ - - if (qc->dev->flags & ATA_DFLAG_LBA) { - qc->tf.flags |= ATA_TFLAG_LBA; - - tf->lbah = 0x0; - tf->lbam = 0x0; - tf->lbal = 0x0; - tf->device |= ATA_LBA; - } else { - /* CHS */ - tf->lbal = 0x1; /* sect */ - tf->lbam = 0x0; /* cyl low */ - tf->lbah = 0x0; /* cyl high */ - } - + tf->lbah = 0x0; + tf->lbam = 0x0; + tf->lbal = 0x0; + tf->device |= ATA_LBA; tf->command = ATA_CMD_VERIFY; /* READ VERIFY */ } else { tf->nsect = 0; /* time period value (0 implies now) */ @@ -498,99 +487,6 @@ static unsigned int ata_scsi_flush_xlat(struct ata_queued_cmd *qc, u8 *scsicmd) return 0; } -/** - * scsi_6_lba_len - Get LBA and transfer length - * @scsicmd: SCSI command to translate - * - * Calculate LBA and transfer length for 6-byte commands. - * - * RETURNS: - * @plba: the LBA - * @plen: the transfer length - */ - -static void scsi_6_lba_len(u8 *scsicmd, u64 *plba, u32 *plen) -{ - u64 lba = 0; - u32 len = 0; - - VPRINTK("six-byte command\n"); - - lba |= ((u64)scsicmd[2]) << 8; - lba |= ((u64)scsicmd[3]); - - len |= ((u32)scsicmd[4]); - - *plba = lba; - *plen = len; -} - -/** - * scsi_10_lba_len - Get LBA and transfer length - * @scsicmd: SCSI command to translate - * - * Calculate LBA and transfer length for 10-byte commands. - * - * RETURNS: - * @plba: the LBA - * @plen: the transfer length - */ - -static void scsi_10_lba_len(u8 *scsicmd, u64 *plba, u32 *plen) -{ - u64 lba = 0; - u32 len = 0; - - VPRINTK("ten-byte command\n"); - - lba |= ((u64)scsicmd[2]) << 24; - lba |= ((u64)scsicmd[3]) << 16; - lba |= ((u64)scsicmd[4]) << 8; - lba |= ((u64)scsicmd[5]); - - len |= ((u32)scsicmd[7]) << 8; - len |= ((u32)scsicmd[8]); - - *plba = lba; - *plen = len; -} - -/** - * scsi_16_lba_len - Get LBA and transfer length - * @scsicmd: SCSI command to translate - * - * Calculate LBA and transfer length for 16-byte commands. - * - * RETURNS: - * @plba: the LBA - * @plen: the transfer length - */ - -static void scsi_16_lba_len(u8 *scsicmd, u64 *plba, u32 *plen) -{ - u64 lba = 0; - u32 len = 0; - - VPRINTK("sixteen-byte command\n"); - - lba |= ((u64)scsicmd[2]) << 56; - lba |= ((u64)scsicmd[3]) << 48; - lba |= ((u64)scsicmd[4]) << 40; - lba |= ((u64)scsicmd[5]) << 32; - lba |= ((u64)scsicmd[6]) << 24; - lba |= ((u64)scsicmd[7]) << 16; - lba |= ((u64)scsicmd[8]) << 8; - lba |= ((u64)scsicmd[9]); - - len |= ((u32)scsicmd[10]) << 24; - len |= ((u32)scsicmd[11]) << 16; - len |= ((u32)scsicmd[12]) << 8; - len |= ((u32)scsicmd[13]); - - *plba = lba; - *plen = len; -} - /** * ata_scsi_verify_xlat - Translate SCSI VERIFY command into an ATA one * @qc: Storage for translated ATA taskfile @@ -608,87 +504,78 @@ static void scsi_16_lba_len(u8 *scsicmd, u64 *plba, u32 *plen) static unsigned int ata_scsi_verify_xlat(struct ata_queued_cmd *qc, u8 *scsicmd) { struct ata_taskfile *tf = &qc->tf; - struct ata_device *dev = qc->dev; - unsigned int lba = tf->flags & ATA_TFLAG_LBA; unsigned int lba48 = tf->flags & ATA_TFLAG_LBA48; u64 dev_sectors = qc->dev->n_sectors; - u64 block; - u32 n_block; + u64 sect = 0; + u32 n_sect = 0; tf->flags |= ATA_TFLAG_ISADDR | ATA_TFLAG_DEVICE; tf->protocol = ATA_PROT_NODATA; + tf->device |= ATA_LBA; + + if (scsicmd[0] == VERIFY) { + sect |= ((u64)scsicmd[2]) << 24; + sect |= ((u64)scsicmd[3]) << 16; + sect |= ((u64)scsicmd[4]) << 8; + sect |= ((u64)scsicmd[5]); + + n_sect |= ((u32)scsicmd[7]) << 8; + n_sect |= ((u32)scsicmd[8]); + } + + else if (scsicmd[0] == VERIFY_16) { + sect |= ((u64)scsicmd[2]) << 56; + sect |= ((u64)scsicmd[3]) << 48; + sect |= ((u64)scsicmd[4]) << 40; + sect |= ((u64)scsicmd[5]) << 32; + sect |= ((u64)scsicmd[6]) << 24; + sect |= ((u64)scsicmd[7]) << 16; + sect |= ((u64)scsicmd[8]) << 8; + sect |= ((u64)scsicmd[9]); + + n_sect |= ((u32)scsicmd[10]) << 24; + n_sect |= ((u32)scsicmd[11]) << 16; + n_sect |= ((u32)scsicmd[12]) << 8; + n_sect |= ((u32)scsicmd[13]); + } - if (scsicmd[0] == VERIFY) - scsi_10_lba_len(scsicmd, &block, &n_block); - else if (scsicmd[0] == VERIFY_16) - scsi_16_lba_len(scsicmd, &block, &n_block); else return 1; - if (!n_block) + if (!n_sect) return 1; - if (block >= dev_sectors) + if (sect >= dev_sectors) return 1; - if ((block + n_block) > dev_sectors) + if ((sect + n_sect) > dev_sectors) return 1; if (lba48) { - if (n_block > (64 * 1024)) + if (n_sect > (64 * 1024)) return 1; } else { - if (n_block > 256) + if (n_sect > 256) return 1; } - if (lba) { - if (lba48) { - tf->command = ATA_CMD_VERIFY_EXT; - - tf->hob_nsect = (n_block >> 8) & 0xff; - - tf->hob_lbah = (block >> 40) & 0xff; - tf->hob_lbam = (block >> 32) & 0xff; - tf->hob_lbal = (block >> 24) & 0xff; - } else { - tf->command = ATA_CMD_VERIFY; - - tf->device |= (block >> 24) & 0xf; - } - - tf->nsect = n_block & 0xff; + if (lba48) { + tf->command = ATA_CMD_VERIFY_EXT; - tf->lbah = (block >> 16) & 0xff; - tf->lbam = (block >> 8) & 0xff; - tf->lbal = block & 0xff; + tf->hob_nsect = (n_sect >> 8) & 0xff; - tf->device |= ATA_LBA; + tf->hob_lbah = (sect >> 40) & 0xff; + tf->hob_lbam = (sect >> 32) & 0xff; + tf->hob_lbal = (sect >> 24) & 0xff; } else { - /* CHS */ - u32 sect, head, cyl, track; - - /* Convert LBA to CHS */ - track = (u32)block / dev->sectors; - cyl = track / dev->heads; - head = track % dev->heads; - sect = (u32)block % dev->sectors + 1; - - DPRINTK("block %u track %u cyl %u head %u sect %u\n", - (u32)block, track, cyl, head, sect); - - /* Check whether the converted CHS can fit. - Cylinder: 0-65535 - Head: 0-15 - Sector: 1-255*/ - if ((cyl >> 16) || (head >> 4) || (sect >> 8) || (!sect)) - return 1; - tf->command = ATA_CMD_VERIFY; - tf->nsect = n_block & 0xff; /* Sector count 0 means 256 sectors */ - tf->lbal = sect; - tf->lbam = cyl; - tf->lbah = cyl >> 8; - tf->device |= head; + + tf->device |= (sect >> 24) & 0xf; } + tf->nsect = n_sect & 0xff; + + tf->lbah = (sect >> 16) & 0xff; + tf->lbam = (sect >> 8) & 0xff; + tf->lbal = sect & 0xff; + return 0; } @@ -715,14 +602,11 @@ static unsigned int ata_scsi_verify_xlat(struct ata_queued_cmd *qc, u8 *scsicmd) static unsigned int ata_scsi_rw_xlat(struct ata_queued_cmd *qc, u8 *scsicmd) { struct ata_taskfile *tf = &qc->tf; - struct ata_device *dev = qc->dev; - unsigned int lba = tf->flags & ATA_TFLAG_LBA; unsigned int lba48 = tf->flags & ATA_TFLAG_LBA48; - u64 block; - u32 n_block; tf->flags |= ATA_TFLAG_ISADDR | ATA_TFLAG_DEVICE; tf->protocol = qc->dev->xfer_protocol; + tf->device |= ATA_LBA; if (scsicmd[0] == READ_10 || scsicmd[0] == READ_6 || scsicmd[0] == READ_16) { @@ -732,102 +616,90 @@ static unsigned int ata_scsi_rw_xlat(struct ata_queued_cmd *qc, u8 *scsicmd) tf->flags |= ATA_TFLAG_WRITE; } - /* Calculate the SCSI LBA and transfer length. */ - switch (scsicmd[0]) { - case READ_10: - case WRITE_10: - scsi_10_lba_len(scsicmd, &block, &n_block); - break; - case READ_6: - case WRITE_6: - scsi_6_lba_len(scsicmd, &block, &n_block); - - /* for 6-byte r/w commands, transfer length 0 - * means 256 blocks of data, not 0 block. - */ - if (!n_block) - n_block = 256; - break; - case READ_16: - case WRITE_16: - scsi_16_lba_len(scsicmd, &block, &n_block); - break; - default: - DPRINTK("no-byte command\n"); - return 1; - } - - /* Check and compose ATA command */ - if (!n_block) - /* For 10-byte and 16-byte SCSI R/W commands, transfer - * length 0 means transfer 0 block of data. - * However, for ATA R/W commands, sector count 0 means - * 256 or 65536 sectors, not 0 sectors as in SCSI. - */ - return 1; - - if (lba) { + if (scsicmd[0] == READ_10 || scsicmd[0] == WRITE_10) { if (lba48) { - /* The request -may- be too large for LBA48. */ - if ((block >> 48) || (n_block > 65536)) + tf->hob_nsect = scsicmd[7]; + tf->hob_lbal = scsicmd[2]; + + qc->nsect = ((unsigned int)scsicmd[7] << 8) | + scsicmd[8]; + } else { + /* if we don't support LBA48 addressing, the request + * -may- be too large. */ + if ((scsicmd[2] & 0xf0) || scsicmd[7]) return 1; - tf->hob_nsect = (n_block >> 8) & 0xff; + /* stores LBA27:24 in lower 4 bits of device reg */ + tf->device |= scsicmd[2]; - tf->hob_lbah = (block >> 40) & 0xff; - tf->hob_lbam = (block >> 32) & 0xff; - tf->hob_lbal = (block >> 24) & 0xff; - } else { - /* LBA28 */ + qc->nsect = scsicmd[8]; + } - /* The request -may- be too large for LBA28. */ - if ((block >> 28) || (n_block > 256)) - return 1; + tf->nsect = scsicmd[8]; + tf->lbal = scsicmd[5]; + tf->lbam = scsicmd[4]; + tf->lbah = scsicmd[3]; - tf->device |= (block >> 24) & 0xf; - } + VPRINTK("ten-byte command\n"); + if (qc->nsect == 0) /* we don't support length==0 cmds */ + return 1; + return 0; + } - qc->nsect = n_block; - tf->nsect = n_block & 0xff; + if (scsicmd[0] == READ_6 || scsicmd[0] == WRITE_6) { + qc->nsect = tf->nsect = scsicmd[4]; + if (!qc->nsect) { + qc->nsect = 256; + if (lba48) + tf->hob_nsect = 1; + } - tf->lbah = (block >> 16) & 0xff; - tf->lbam = (block >> 8) & 0xff; - tf->lbal = block & 0xff; + tf->lbal = scsicmd[3]; + tf->lbam = scsicmd[2]; + tf->lbah = scsicmd[1] & 0x1f; /* mask out reserved bits */ - tf->device |= ATA_LBA; - } else { - /* CHS */ - u32 sect, head, cyl, track; + VPRINTK("six-byte command\n"); + return 0; + } - /* The request -may- be too large for CHS addressing. */ - if ((block >> 28) || (n_block > 256)) + if (scsicmd[0] == READ_16 || scsicmd[0] == WRITE_16) { + /* rule out impossible LBAs and sector counts */ + if (scsicmd[2] || scsicmd[3] || scsicmd[10] || scsicmd[11]) return 1; - /* Convert LBA to CHS */ - track = (u32)block / dev->sectors; - cyl = track / dev->heads; - head = track % dev->heads; - sect = (u32)block % dev->sectors + 1; + if (lba48) { + tf->hob_nsect = scsicmd[12]; + tf->hob_lbal = scsicmd[6]; + tf->hob_lbam = scsicmd[5]; + tf->hob_lbah = scsicmd[4]; - DPRINTK("block %u track %u cyl %u head %u sect %u\n", - (u32)block, track, cyl, head, sect); + qc->nsect = ((unsigned int)scsicmd[12] << 8) | + scsicmd[13]; + } else { + /* once again, filter out impossible non-zero values */ + if (scsicmd[4] || scsicmd[5] || scsicmd[12] || + (scsicmd[6] & 0xf0)) + return 1; - /* Check whether the converted CHS can fit. - Cylinder: 0-65535 - Head: 0-15 - Sector: 1-255*/ - if ((cyl >> 16) || (head >> 4) || (sect >> 8) || (!sect)) - return 1; + /* stores LBA27:24 in lower 4 bits of device reg */ + tf->device |= scsicmd[6]; + + qc->nsect = scsicmd[13]; + } - qc->nsect = n_block; - tf->nsect = n_block & 0xff; /* Sector count 0 means 256 sectors */ - tf->lbal = sect; - tf->lbam = cyl; - tf->lbah = cyl >> 8; - tf->device |= head; + tf->nsect = scsicmd[13]; + tf->lbal = scsicmd[9]; + tf->lbam = scsicmd[8]; + tf->lbah = scsicmd[7]; + + VPRINTK("sixteen-byte command\n"); + if (qc->nsect == 0) /* we don't support length==0 cmds */ + return 1; + return 0; } - return 0; + DPRINTK("no-byte command\n"); + return 1; } static int ata_scsi_qc_complete(struct ata_queued_cmd *qc, u8 drv_stat) @@ -1374,20 +1246,10 @@ unsigned int ata_scsiop_read_cap(struct ata_scsi_args *args, u8 *rbuf, VPRINTK("ENTER\n"); - if (ata_id_has_lba(args->id)) { - if (ata_id_has_lba48(args->id)) - n_sectors = ata_id_u64(args->id, 100); - else - n_sectors = ata_id_u32(args->id, 60); - } else { - /* CHS default translation */ - n_sectors = args->id[1] * args->id[3] * args->id[6]; - - if (ata_id_current_chs_valid(args->id)) - /* CHS current translation */ - n_sectors = ata_id_u32(args->id, 57); - } - + if (ata_id_has_lba48(args->id)) + n_sectors = ata_id_u64(args->id, 100); + else + n_sectors = ata_id_u32(args->id, 60); n_sectors--; /* ATA TotalUserSectors - 1 */ if (args->cmd->cmnd[0] == READ_CAPACITY) { @@ -1468,7 +1330,7 @@ unsigned int ata_scsiop_report_luns(struct ata_scsi_args *args, u8 *rbuf, void ata_scsi_badcmd(struct scsi_cmnd *cmd, void (*done)(struct scsi_cmnd *), u8 asc, u8 ascq) { DPRINTK("ENTER\n"); - cmd->result = (DRIVER_SENSE << 24) | SAM_STAT_CHECK_CONDITION; + cmd->result = SAM_STAT_CHECK_CONDITION; cmd->sense_buffer[0] = 0x70; cmd->sense_buffer[2] = ILLEGAL_REQUEST; @@ -1479,79 +1341,19 @@ void ata_scsi_badcmd(struct scsi_cmnd *cmd, void (*done)(struct scsi_cmnd *), u8 done(cmd); } -void atapi_request_sense(struct ata_port *ap, struct ata_device *dev, - struct scsi_cmnd *cmd) -{ - DECLARE_COMPLETION(wait); - struct ata_queued_cmd *qc; - unsigned long flags; - int rc; - - DPRINTK("ATAPI request sense\n"); - - qc = ata_qc_new_init(ap, dev); - BUG_ON(qc == NULL); - - /* FIXME: is this needed? */ - memset(cmd->sense_buffer, 0, sizeof(cmd->sense_buffer)); - - ata_sg_init_one(qc, cmd->sense_buffer, sizeof(cmd->sense_buffer)); - qc->dma_dir = DMA_FROM_DEVICE; - - memset(&qc->cdb, 0, ap->cdb_len); - qc->cdb[0] = REQUEST_SENSE; - qc->cdb[4] = SCSI_SENSE_BUFFERSIZE; - - qc->tf.flags |= ATA_TFLAG_ISADDR | ATA_TFLAG_DEVICE; - qc->tf.command = ATA_CMD_PACKET; - - qc->tf.protocol = ATA_PROT_ATAPI; - qc->tf.lbam = (8 * 1024) & 0xff; - qc->tf.lbah = (8 * 1024) >> 8; - qc->nbytes = SCSI_SENSE_BUFFERSIZE; - - qc->waiting = &wait; - qc->complete_fn = ata_qc_complete_noop; - - spin_lock_irqsave(&ap->host_set->lock, flags); - rc = ata_qc_issue(qc); - spin_unlock_irqrestore(&ap->host_set->lock, flags); - - if (rc) - ata_port_disable(ap); - else - wait_for_completion(&wait); - - DPRINTK("EXIT\n"); -} - static int atapi_qc_complete(struct ata_queued_cmd *qc, u8 drv_stat) { struct scsi_cmnd *cmd = qc->scsicmd; - VPRINTK("ENTER, drv_stat == 0x%x\n", drv_stat); - - if (unlikely(drv_stat & (ATA_BUSY | ATA_DRQ))) - ata_to_sense_error(qc, drv_stat); - - else if (unlikely(drv_stat & ATA_ERR)) { + if (unlikely(drv_stat & (ATA_ERR | ATA_BUSY | ATA_DRQ))) { DPRINTK("request check condition\n"); - /* FIXME: command completion with check condition - * but no sense causes the error handler to run, - * which then issues REQUEST SENSE, fills in the sense - * buffer, and completes the command (for the second - * time). We need to issue REQUEST SENSE some other - * way, to avoid completing the command twice. - */ cmd->result = SAM_STAT_CHECK_CONDITION; qc->scsidone(cmd); return 1; - } - - else { + } else { u8 *scsicmd = cmd->cmnd; if (scsicmd[0] == INQUIRY) { @@ -1559,30 +1361,15 @@ static int atapi_qc_complete(struct ata_queued_cmd *qc, u8 drv_stat) unsigned int buflen; buflen = ata_scsi_rbuf_get(cmd, &buf); - - /* ATAPI devices typically report zero for their SCSI version, - * and sometimes deviate from the spec WRT response data - * format. If SCSI version is reported as zero like normal, - * then we make the following fixups: 1) Fake MMC-5 version, - * to indicate to the Linux scsi midlayer this is a modern - * device. 2) Ensure response data format / ATAPI information - * are always correct. - */ - /* FIXME: do we ever override EVPD pages and the like, with - * this code? - */ - if (buf[2] == 0) { - buf[2] = 0x5; - buf[3] = 0x32; - } - + buf[2] = 0x5; + buf[3] = (buf[3] & 0xf0) | 2; ata_scsi_rbuf_put(cmd, buf); } - cmd->result = SAM_STAT_GOOD; } qc->scsidone(cmd); + return 0; } /** @@ -1891,19 +1678,3 @@ void ata_scsi_simulate(u16 *id, } } -void ata_scsi_scan_host(struct ata_port *ap) -{ - struct ata_device *dev; - unsigned int i; - - if (ap->flags & ATA_FLAG_PORT_DISABLED) - return; - - for (i = 0; i < ATA_MAX_DEVICES; i++) { - dev = &ap->device[i]; - - if (ata_dev_present(dev)) - scsi_scan_target(&ap->host->shost_gendev, 0, i, 0, 0); - } -} - diff --git a/trunk/drivers/scsi/libata.h b/trunk/drivers/scsi/libata.h index a4b55dc9c698..d608b3a0f6fe 100644 --- a/trunk/drivers/scsi/libata.h +++ b/trunk/drivers/scsi/libata.h @@ -39,7 +39,6 @@ struct ata_scsi_args { /* libata-core.c */ extern int atapi_enabled; -extern int ata_qc_complete_noop(struct ata_queued_cmd *qc, u8 drv_stat); extern struct ata_queued_cmd *ata_qc_new_init(struct ata_port *ap, struct ata_device *dev); extern void ata_qc_free(struct ata_queued_cmd *qc); @@ -52,9 +51,6 @@ extern void swap_buf_le16(u16 *buf, unsigned int buf_words); /* libata-scsi.c */ -extern void atapi_request_sense(struct ata_port *ap, struct ata_device *dev, - struct scsi_cmnd *cmd); -extern void ata_scsi_scan_host(struct ata_port *ap); extern void ata_to_sense_error(struct ata_queued_cmd *qc, u8 drv_stat); extern int ata_scsi_error(struct Scsi_Host *host); extern unsigned int ata_scsiop_inq_std(struct ata_scsi_args *args, u8 *rbuf, diff --git a/trunk/drivers/scsi/lpfc/lpfc_attr.c b/trunk/drivers/scsi/lpfc/lpfc_attr.c index 86eaf6d408d5..acae7c48ef7d 100644 --- a/trunk/drivers/scsi/lpfc/lpfc_attr.c +++ b/trunk/drivers/scsi/lpfc/lpfc_attr.c @@ -973,10 +973,10 @@ lpfc_get_host_fabric_name (struct Scsi_Host *shost) if ((phba->fc_flag & FC_FABRIC) || ((phba->fc_topology == TOPOLOGY_LOOP) && (phba->fc_flag & FC_PUBLIC_LOOP))) - node_name = wwn_to_u64(phba->fc_fabparam.nodeName.wwn); + node_name = wwn_to_u64(phba->fc_fabparam.nodeName.u.wwn); else /* fabric is local port if there is no F/FL_Port */ - node_name = wwn_to_u64(phba->fc_nodename.wwn); + node_name = wwn_to_u64(phba->fc_nodename.u.wwn); spin_unlock_irq(shost->host_lock); @@ -1110,7 +1110,7 @@ lpfc_get_starget_node_name(struct scsi_target *starget) /* Search the mapped list for this target ID */ list_for_each_entry(ndlp, &phba->fc_nlpmap_list, nlp_listp) { if (starget->id == ndlp->nlp_sid) { - node_name = wwn_to_u64(ndlp->nlp_nodename.wwn); + node_name = wwn_to_u64(ndlp->nlp_nodename.u.wwn); break; } } @@ -1131,7 +1131,7 @@ lpfc_get_starget_port_name(struct scsi_target *starget) /* Search the mapped list for this target ID */ list_for_each_entry(ndlp, &phba->fc_nlpmap_list, nlp_listp) { if (starget->id == ndlp->nlp_sid) { - port_name = wwn_to_u64(ndlp->nlp_portname.wwn); + port_name = wwn_to_u64(ndlp->nlp_portname.u.wwn); break; } } diff --git a/trunk/drivers/scsi/lpfc/lpfc_hbadisc.c b/trunk/drivers/scsi/lpfc/lpfc_hbadisc.c index 4fb8eb0c84cf..56052f4510c3 100644 --- a/trunk/drivers/scsi/lpfc/lpfc_hbadisc.c +++ b/trunk/drivers/scsi/lpfc/lpfc_hbadisc.c @@ -1019,8 +1019,8 @@ lpfc_register_remote_port(struct lpfc_hba * phba, struct fc_rport_identifiers rport_ids; /* Remote port has reappeared. Re-register w/ FC transport */ - rport_ids.node_name = wwn_to_u64(ndlp->nlp_nodename.wwn); - rport_ids.port_name = wwn_to_u64(ndlp->nlp_portname.wwn); + rport_ids.node_name = wwn_to_u64(ndlp->nlp_nodename.u.wwn); + rport_ids.port_name = wwn_to_u64(ndlp->nlp_portname.u.wwn); rport_ids.port_id = ndlp->nlp_DID; rport_ids.roles = FC_RPORT_ROLE_UNKNOWN; if (ndlp->nlp_type & NLP_FCP_TARGET) diff --git a/trunk/drivers/scsi/lpfc/lpfc_hw.h b/trunk/drivers/scsi/lpfc/lpfc_hw.h index 047a87c26cc0..86c41981188b 100644 --- a/trunk/drivers/scsi/lpfc/lpfc_hw.h +++ b/trunk/drivers/scsi/lpfc/lpfc_hw.h @@ -280,9 +280,9 @@ struct lpfc_name { #define NAME_CCITT_GR_TYPE 0xE uint8_t IEEEextLsb; /* FC Word 0, bit 16:23, IEEE extended Lsb */ uint8_t IEEE[6]; /* FC IEEE address */ - }; + } s; uint8_t wwn[8]; - }; + } u; }; struct csp { diff --git a/trunk/drivers/scsi/lpfc/lpfc_init.c b/trunk/drivers/scsi/lpfc/lpfc_init.c index 454058f655db..0856ff7d3b33 100644 --- a/trunk/drivers/scsi/lpfc/lpfc_init.c +++ b/trunk/drivers/scsi/lpfc/lpfc_init.c @@ -285,7 +285,7 @@ lpfc_config_port_post(struct lpfc_hba * phba) if (phba->SerialNumber[0] == 0) { uint8_t *outptr; - outptr = (uint8_t *) & phba->fc_nodename.IEEE[0]; + outptr = &phba->fc_nodename.u.s.IEEE[0]; for (i = 0; i < 12; i++) { status = *outptr++; j = ((status & 0xf0) >> 4); @@ -1523,8 +1523,8 @@ lpfc_pci_probe_one(struct pci_dev *pdev, const struct pci_device_id *pid) * Must done after lpfc_sli_hba_setup() */ - fc_host_node_name(host) = wwn_to_u64(phba->fc_nodename.wwn); - fc_host_port_name(host) = wwn_to_u64(phba->fc_portname.wwn); + fc_host_node_name(host) = wwn_to_u64(phba->fc_nodename.u.wwn); + fc_host_port_name(host) = wwn_to_u64(phba->fc_portname.u.wwn); fc_host_supported_classes(host) = FC_COS_CLASS3; memset(fc_host_supported_fc4s(host), 0, diff --git a/trunk/drivers/scsi/megaraid.c b/trunk/drivers/scsi/megaraid.c index 6f308ebe3e79..61a6fd810bb4 100644 --- a/trunk/drivers/scsi/megaraid.c +++ b/trunk/drivers/scsi/megaraid.c @@ -621,8 +621,6 @@ mega_build_cmd(adapter_t *adapter, Scsi_Cmnd *cmd, int *busy) if(islogical) { switch (cmd->cmnd[0]) { case TEST_UNIT_READY: - memset(cmd->request_buffer, 0, cmd->request_bufflen); - #if MEGA_HAVE_CLUSTERING /* * Do we support clustering and is the support enabled @@ -652,11 +650,28 @@ mega_build_cmd(adapter_t *adapter, Scsi_Cmnd *cmd, int *busy) return NULL; #endif - case MODE_SENSE: + case MODE_SENSE: { + char *buf; + + if (cmd->use_sg) { + struct scatterlist *sg; + + sg = (struct scatterlist *)cmd->request_buffer; + buf = kmap_atomic(sg->page, KM_IRQ0) + + sg->offset; + } else + buf = cmd->request_buffer; memset(cmd->request_buffer, 0, cmd->cmnd[4]); + if (cmd->use_sg) { + struct scatterlist *sg; + + sg = (struct scatterlist *)cmd->request_buffer; + kunmap_atomic(buf - sg->offset, KM_IRQ0); + } cmd->result = (DID_OK << 16); cmd->scsi_done(cmd); return NULL; + } case READ_CAPACITY: case INQUIRY: @@ -1685,14 +1700,23 @@ mega_rundoneq (adapter_t *adapter) static void mega_free_scb(adapter_t *adapter, scb_t *scb) { + unsigned long length; + switch( scb->dma_type ) { case MEGA_DMA_TYPE_NONE: break; case MEGA_BULK_DATA: + if (scb->cmd->use_sg == 0) + length = scb->cmd->request_bufflen; + else { + struct scatterlist *sgl = + (struct scatterlist *)scb->cmd->request_buffer; + length = sgl->length; + } pci_unmap_page(adapter->dev, scb->dma_h_bulkdata, - scb->cmd->request_bufflen, scb->dma_direction); + length, scb->dma_direction); break; case MEGA_SGLIST: @@ -1741,6 +1765,7 @@ mega_build_sglist(adapter_t *adapter, scb_t *scb, u32 *buf, u32 *len) struct scatterlist *sgl; struct page *page; unsigned long offset; + unsigned int length; Scsi_Cmnd *cmd; int sgcnt; int idx; @@ -1748,14 +1773,23 @@ mega_build_sglist(adapter_t *adapter, scb_t *scb, u32 *buf, u32 *len) cmd = scb->cmd; /* Scatter-gather not used */ - if( !cmd->use_sg ) { - - page = virt_to_page(cmd->request_buffer); - offset = offset_in_page(cmd->request_buffer); + if( cmd->use_sg == 0 || (cmd->use_sg == 1 && + !adapter->has_64bit_addr)) { + + if (cmd->use_sg == 0) { + page = virt_to_page(cmd->request_buffer); + offset = offset_in_page(cmd->request_buffer); + length = cmd->request_bufflen; + } else { + sgl = (struct scatterlist *)cmd->request_buffer; + page = sgl->page; + offset = sgl->offset; + length = sgl->length; + } scb->dma_h_bulkdata = pci_map_page(adapter->dev, page, offset, - cmd->request_bufflen, + length, scb->dma_direction); scb->dma_type = MEGA_BULK_DATA; @@ -1765,14 +1799,14 @@ mega_build_sglist(adapter_t *adapter, scb_t *scb, u32 *buf, u32 *len) */ if( adapter->has_64bit_addr ) { scb->sgl64[0].address = scb->dma_h_bulkdata; - scb->sgl64[0].length = cmd->request_bufflen; + scb->sgl64[0].length = length; *buf = (u32)scb->sgl_dma_addr; - *len = (u32)cmd->request_bufflen; + *len = (u32)length; return 1; } else { *buf = (u32)scb->dma_h_bulkdata; - *len = (u32)cmd->request_bufflen; + *len = (u32)length; } return 0; } @@ -1791,27 +1825,23 @@ mega_build_sglist(adapter_t *adapter, scb_t *scb, u32 *buf, u32 *len) if( sgcnt > adapter->sglen ) BUG(); + *len = 0; + for( idx = 0; idx < sgcnt; idx++, sgl++ ) { if( adapter->has_64bit_addr ) { scb->sgl64[idx].address = sg_dma_address(sgl); - scb->sgl64[idx].length = sg_dma_len(sgl); + *len += scb->sgl64[idx].length = sg_dma_len(sgl); } else { scb->sgl[idx].address = sg_dma_address(sgl); - scb->sgl[idx].length = sg_dma_len(sgl); + *len += scb->sgl[idx].length = sg_dma_len(sgl); } } /* Reset pointer and length fields */ *buf = scb->sgl_dma_addr; - /* - * For passthru command, dataxferlen must be set, even for commands - * with a sg list - */ - *len = (u32)cmd->request_bufflen; - /* Return count of SG requests */ return sgcnt; } diff --git a/trunk/drivers/scsi/megaraid/Kconfig.megaraid b/trunk/drivers/scsi/megaraid/Kconfig.megaraid index 917d591d90b2..7363e12663ac 100644 --- a/trunk/drivers/scsi/megaraid/Kconfig.megaraid +++ b/trunk/drivers/scsi/megaraid/Kconfig.megaraid @@ -76,3 +76,12 @@ config MEGARAID_LEGACY To compile this driver as a module, choose M here: the module will be called megaraid endif + +config MEGARAID_SAS + tristate "LSI Logic MegaRAID SAS RAID Module" + depends on PCI && SCSI + help + Module for LSI Logic's SAS based RAID controllers. + To compile this driver as a module, choose 'm' here. + Module will be called megaraid_sas + diff --git a/trunk/drivers/scsi/megaraid/Makefile b/trunk/drivers/scsi/megaraid/Makefile index 6dd99f275722..f469915b97c3 100644 --- a/trunk/drivers/scsi/megaraid/Makefile +++ b/trunk/drivers/scsi/megaraid/Makefile @@ -1,2 +1,3 @@ obj-$(CONFIG_MEGARAID_MM) += megaraid_mm.o obj-$(CONFIG_MEGARAID_MAILBOX) += megaraid_mbox.o +obj-$(CONFIG_MEGARAID_SAS) += megaraid_sas.o diff --git a/trunk/drivers/scsi/megaraid/megaraid_sas.c b/trunk/drivers/scsi/megaraid/megaraid_sas.c new file mode 100644 index 000000000000..1b3148e842af --- /dev/null +++ b/trunk/drivers/scsi/megaraid/megaraid_sas.c @@ -0,0 +1,2805 @@ +/* + * + * Linux MegaRAID driver for SAS based RAID controllers + * + * Copyright (c) 2003-2005 LSI Logic Corporation. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + * FILE : megaraid_sas.c + * Version : v00.00.02.00-rc4 + * + * Authors: + * Sreenivas Bagalkote + * Sumant Patro + * + * List of supported controllers + * + * OEM Product Name VID DID SSVID SSID + * --- ------------ --- --- ---- ---- + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include "megaraid_sas.h" + +MODULE_LICENSE("GPL"); +MODULE_VERSION(MEGASAS_VERSION); +MODULE_AUTHOR("sreenivas.bagalkote@lsil.com"); +MODULE_DESCRIPTION("LSI Logic MegaRAID SAS Driver"); + +/* + * PCI ID table for all supported controllers + */ +static struct pci_device_id megasas_pci_table[] = { + + { + PCI_VENDOR_ID_LSI_LOGIC, + PCI_DEVICE_ID_LSI_SAS1064R, + PCI_ANY_ID, + PCI_ANY_ID, + }, + { + PCI_VENDOR_ID_DELL, + PCI_DEVICE_ID_DELL_PERC5, + PCI_ANY_ID, + PCI_ANY_ID, + }, + {0} /* Terminating entry */ +}; + +MODULE_DEVICE_TABLE(pci, megasas_pci_table); + +static int megasas_mgmt_majorno; +static struct megasas_mgmt_info megasas_mgmt_info; +static struct fasync_struct *megasas_async_queue; +static DECLARE_MUTEX(megasas_async_queue_mutex); + +/** + * megasas_get_cmd - Get a command from the free pool + * @instance: Adapter soft state + * + * Returns a free command from the pool + */ +static inline struct megasas_cmd *megasas_get_cmd(struct megasas_instance + *instance) +{ + unsigned long flags; + struct megasas_cmd *cmd = NULL; + + spin_lock_irqsave(&instance->cmd_pool_lock, flags); + + if (!list_empty(&instance->cmd_pool)) { + cmd = list_entry((&instance->cmd_pool)->next, + struct megasas_cmd, list); + list_del_init(&cmd->list); + } else { + printk(KERN_ERR "megasas: Command pool empty!\n"); + } + + spin_unlock_irqrestore(&instance->cmd_pool_lock, flags); + return cmd; +} + +/** + * megasas_return_cmd - Return a cmd to free command pool + * @instance: Adapter soft state + * @cmd: Command packet to be returned to free command pool + */ +static inline void +megasas_return_cmd(struct megasas_instance *instance, struct megasas_cmd *cmd) +{ + unsigned long flags; + + spin_lock_irqsave(&instance->cmd_pool_lock, flags); + + cmd->scmd = NULL; + list_add_tail(&cmd->list, &instance->cmd_pool); + + spin_unlock_irqrestore(&instance->cmd_pool_lock, flags); +} + +/** + * megasas_enable_intr - Enables interrupts + * @regs: MFI register set + */ +static inline void +megasas_enable_intr(struct megasas_register_set __iomem * regs) +{ + writel(1, &(regs)->outbound_intr_mask); + + /* Dummy readl to force pci flush */ + readl(®s->outbound_intr_mask); +} + +/** + * megasas_disable_intr - Disables interrupts + * @regs: MFI register set + */ +static inline void +megasas_disable_intr(struct megasas_register_set __iomem * regs) +{ + u32 mask = readl(®s->outbound_intr_mask) & (~0x00000001); + writel(mask, ®s->outbound_intr_mask); + + /* Dummy readl to force pci flush */ + readl(®s->outbound_intr_mask); +} + +/** + * megasas_issue_polled - Issues a polling command + * @instance: Adapter soft state + * @cmd: Command packet to be issued + * + * For polling, MFI requires the cmd_status to be set to 0xFF before posting. + */ +static int +megasas_issue_polled(struct megasas_instance *instance, struct megasas_cmd *cmd) +{ + int i; + u32 msecs = MFI_POLL_TIMEOUT_SECS * 1000; + + struct megasas_header *frame_hdr = &cmd->frame->hdr; + + frame_hdr->cmd_status = 0xFF; + frame_hdr->flags |= MFI_FRAME_DONT_POST_IN_REPLY_QUEUE; + + /* + * Issue the frame using inbound queue port + */ + writel(cmd->frame_phys_addr >> 3, + &instance->reg_set->inbound_queue_port); + + /* + * Wait for cmd_status to change + */ + for (i = 0; (i < msecs) && (frame_hdr->cmd_status == 0xff); i++) { + rmb(); + msleep(1); + } + + if (frame_hdr->cmd_status == 0xff) + return -ETIME; + + return 0; +} + +/** + * megasas_issue_blocked_cmd - Synchronous wrapper around regular FW cmds + * @instance: Adapter soft state + * @cmd: Command to be issued + * + * This function waits on an event for the command to be returned from ISR. + * Used to issue ioctl commands. + */ +static int +megasas_issue_blocked_cmd(struct megasas_instance *instance, + struct megasas_cmd *cmd) +{ + cmd->cmd_status = ENODATA; + + writel(cmd->frame_phys_addr >> 3, + &instance->reg_set->inbound_queue_port); + + wait_event(instance->int_cmd_wait_q, (cmd->cmd_status != ENODATA)); + + return 0; +} + +/** + * megasas_issue_blocked_abort_cmd - Aborts previously issued cmd + * @instance: Adapter soft state + * @cmd_to_abort: Previously issued cmd to be aborted + * + * MFI firmware can abort previously issued AEN comamnd (automatic event + * notification). The megasas_issue_blocked_abort_cmd() issues such abort + * cmd and blocks till it is completed. + */ +static int +megasas_issue_blocked_abort_cmd(struct megasas_instance *instance, + struct megasas_cmd *cmd_to_abort) +{ + struct megasas_cmd *cmd; + struct megasas_abort_frame *abort_fr; + + cmd = megasas_get_cmd(instance); + + if (!cmd) + return -1; + + abort_fr = &cmd->frame->abort; + + /* + * Prepare and issue the abort frame + */ + abort_fr->cmd = MFI_CMD_ABORT; + abort_fr->cmd_status = 0xFF; + abort_fr->flags = 0; + abort_fr->abort_context = cmd_to_abort->index; + abort_fr->abort_mfi_phys_addr_lo = cmd_to_abort->frame_phys_addr; + abort_fr->abort_mfi_phys_addr_hi = 0; + + cmd->sync_cmd = 1; + cmd->cmd_status = 0xFF; + + writel(cmd->frame_phys_addr >> 3, + &instance->reg_set->inbound_queue_port); + + /* + * Wait for this cmd to complete + */ + wait_event(instance->abort_cmd_wait_q, (cmd->cmd_status != 0xFF)); + + megasas_return_cmd(instance, cmd); + return 0; +} + +/** + * megasas_make_sgl32 - Prepares 32-bit SGL + * @instance: Adapter soft state + * @scp: SCSI command from the mid-layer + * @mfi_sgl: SGL to be filled in + * + * If successful, this function returns the number of SG elements. Otherwise, + * it returnes -1. + */ +static inline int +megasas_make_sgl32(struct megasas_instance *instance, struct scsi_cmnd *scp, + union megasas_sgl *mfi_sgl) +{ + int i; + int sge_count; + struct scatterlist *os_sgl; + + /* + * Return 0 if there is no data transfer + */ + if (!scp->request_buffer || !scp->request_bufflen) + return 0; + + if (!scp->use_sg) { + mfi_sgl->sge32[0].phys_addr = pci_map_single(instance->pdev, + scp-> + request_buffer, + scp-> + request_bufflen, + scp-> + sc_data_direction); + mfi_sgl->sge32[0].length = scp->request_bufflen; + + return 1; + } + + os_sgl = (struct scatterlist *)scp->request_buffer; + sge_count = pci_map_sg(instance->pdev, os_sgl, scp->use_sg, + scp->sc_data_direction); + + for (i = 0; i < sge_count; i++, os_sgl++) { + mfi_sgl->sge32[i].length = sg_dma_len(os_sgl); + mfi_sgl->sge32[i].phys_addr = sg_dma_address(os_sgl); + } + + return sge_count; +} + +/** + * megasas_make_sgl64 - Prepares 64-bit SGL + * @instance: Adapter soft state + * @scp: SCSI command from the mid-layer + * @mfi_sgl: SGL to be filled in + * + * If successful, this function returns the number of SG elements. Otherwise, + * it returnes -1. + */ +static inline int +megasas_make_sgl64(struct megasas_instance *instance, struct scsi_cmnd *scp, + union megasas_sgl *mfi_sgl) +{ + int i; + int sge_count; + struct scatterlist *os_sgl; + + /* + * Return 0 if there is no data transfer + */ + if (!scp->request_buffer || !scp->request_bufflen) + return 0; + + if (!scp->use_sg) { + mfi_sgl->sge64[0].phys_addr = pci_map_single(instance->pdev, + scp-> + request_buffer, + scp-> + request_bufflen, + scp-> + sc_data_direction); + + mfi_sgl->sge64[0].length = scp->request_bufflen; + + return 1; + } + + os_sgl = (struct scatterlist *)scp->request_buffer; + sge_count = pci_map_sg(instance->pdev, os_sgl, scp->use_sg, + scp->sc_data_direction); + + for (i = 0; i < sge_count; i++, os_sgl++) { + mfi_sgl->sge64[i].length = sg_dma_len(os_sgl); + mfi_sgl->sge64[i].phys_addr = sg_dma_address(os_sgl); + } + + return sge_count; +} + +/** + * megasas_build_dcdb - Prepares a direct cdb (DCDB) command + * @instance: Adapter soft state + * @scp: SCSI command + * @cmd: Command to be prepared in + * + * This function prepares CDB commands. These are typcially pass-through + * commands to the devices. + */ +static inline int +megasas_build_dcdb(struct megasas_instance *instance, struct scsi_cmnd *scp, + struct megasas_cmd *cmd) +{ + u32 sge_sz; + int sge_bytes; + u32 is_logical; + u32 device_id; + u16 flags = 0; + struct megasas_pthru_frame *pthru; + + is_logical = MEGASAS_IS_LOGICAL(scp); + device_id = MEGASAS_DEV_INDEX(instance, scp); + pthru = (struct megasas_pthru_frame *)cmd->frame; + + if (scp->sc_data_direction == PCI_DMA_TODEVICE) + flags = MFI_FRAME_DIR_WRITE; + else if (scp->sc_data_direction == PCI_DMA_FROMDEVICE) + flags = MFI_FRAME_DIR_READ; + else if (scp->sc_data_direction == PCI_DMA_NONE) + flags = MFI_FRAME_DIR_NONE; + + /* + * Prepare the DCDB frame + */ + pthru->cmd = (is_logical) ? MFI_CMD_LD_SCSI_IO : MFI_CMD_PD_SCSI_IO; + pthru->cmd_status = 0x0; + pthru->scsi_status = 0x0; + pthru->target_id = device_id; + pthru->lun = scp->device->lun; + pthru->cdb_len = scp->cmd_len; + pthru->timeout = 0; + pthru->flags = flags; + pthru->data_xfer_len = scp->request_bufflen; + + memcpy(pthru->cdb, scp->cmnd, scp->cmd_len); + + /* + * Construct SGL + */ + sge_sz = (IS_DMA64) ? sizeof(struct megasas_sge64) : + sizeof(struct megasas_sge32); + + if (IS_DMA64) { + pthru->flags |= MFI_FRAME_SGL64; + pthru->sge_count = megasas_make_sgl64(instance, scp, + &pthru->sgl); + } else + pthru->sge_count = megasas_make_sgl32(instance, scp, + &pthru->sgl); + + /* + * Sense info specific + */ + pthru->sense_len = SCSI_SENSE_BUFFERSIZE; + pthru->sense_buf_phys_addr_hi = 0; + pthru->sense_buf_phys_addr_lo = cmd->sense_phys_addr; + + sge_bytes = sge_sz * pthru->sge_count; + + /* + * Compute the total number of frames this command consumes. FW uses + * this number to pull sufficient number of frames from host memory. + */ + cmd->frame_count = (sge_bytes / MEGAMFI_FRAME_SIZE) + + ((sge_bytes % MEGAMFI_FRAME_SIZE) ? 1 : 0) + 1; + + if (cmd->frame_count > 7) + cmd->frame_count = 8; + + return cmd->frame_count; +} + +/** + * megasas_build_ldio - Prepares IOs to logical devices + * @instance: Adapter soft state + * @scp: SCSI command + * @cmd: Command to to be prepared + * + * Frames (and accompanying SGLs) for regular SCSI IOs use this function. + */ +static inline int +megasas_build_ldio(struct megasas_instance *instance, struct scsi_cmnd *scp, + struct megasas_cmd *cmd) +{ + u32 sge_sz; + int sge_bytes; + u32 device_id; + u8 sc = scp->cmnd[0]; + u16 flags = 0; + struct megasas_io_frame *ldio; + + device_id = MEGASAS_DEV_INDEX(instance, scp); + ldio = (struct megasas_io_frame *)cmd->frame; + + if (scp->sc_data_direction == PCI_DMA_TODEVICE) + flags = MFI_FRAME_DIR_WRITE; + else if (scp->sc_data_direction == PCI_DMA_FROMDEVICE) + flags = MFI_FRAME_DIR_READ; + + /* + * Preare the Logical IO frame: 2nd bit is zero for all read cmds + */ + ldio->cmd = (sc & 0x02) ? MFI_CMD_LD_WRITE : MFI_CMD_LD_READ; + ldio->cmd_status = 0x0; + ldio->scsi_status = 0x0; + ldio->target_id = device_id; + ldio->timeout = 0; + ldio->reserved_0 = 0; + ldio->pad_0 = 0; + ldio->flags = flags; + ldio->start_lba_hi = 0; + ldio->access_byte = (scp->cmd_len != 6) ? scp->cmnd[1] : 0; + + /* + * 6-byte READ(0x08) or WRITE(0x0A) cdb + */ + if (scp->cmd_len == 6) { + ldio->lba_count = (u32) scp->cmnd[4]; + ldio->start_lba_lo = ((u32) scp->cmnd[1] << 16) | + ((u32) scp->cmnd[2] << 8) | (u32) scp->cmnd[3]; + + ldio->start_lba_lo &= 0x1FFFFF; + } + + /* + * 10-byte READ(0x28) or WRITE(0x2A) cdb + */ + else if (scp->cmd_len == 10) { + ldio->lba_count = (u32) scp->cmnd[8] | + ((u32) scp->cmnd[7] << 8); + ldio->start_lba_lo = ((u32) scp->cmnd[2] << 24) | + ((u32) scp->cmnd[3] << 16) | + ((u32) scp->cmnd[4] << 8) | (u32) scp->cmnd[5]; + } + + /* + * 12-byte READ(0xA8) or WRITE(0xAA) cdb + */ + else if (scp->cmd_len == 12) { + ldio->lba_count = ((u32) scp->cmnd[6] << 24) | + ((u32) scp->cmnd[7] << 16) | + ((u32) scp->cmnd[8] << 8) | (u32) scp->cmnd[9]; + + ldio->start_lba_lo = ((u32) scp->cmnd[2] << 24) | + ((u32) scp->cmnd[3] << 16) | + ((u32) scp->cmnd[4] << 8) | (u32) scp->cmnd[5]; + } + + /* + * 16-byte READ(0x88) or WRITE(0x8A) cdb + */ + else if (scp->cmd_len == 16) { + ldio->lba_count = ((u32) scp->cmnd[10] << 24) | + ((u32) scp->cmnd[11] << 16) | + ((u32) scp->cmnd[12] << 8) | (u32) scp->cmnd[13]; + + ldio->start_lba_lo = ((u32) scp->cmnd[6] << 24) | + ((u32) scp->cmnd[7] << 16) | + ((u32) scp->cmnd[8] << 8) | (u32) scp->cmnd[9]; + + ldio->start_lba_hi = ((u32) scp->cmnd[2] << 24) | + ((u32) scp->cmnd[3] << 16) | + ((u32) scp->cmnd[4] << 8) | (u32) scp->cmnd[5]; + + } + + /* + * Construct SGL + */ + sge_sz = (IS_DMA64) ? sizeof(struct megasas_sge64) : + sizeof(struct megasas_sge32); + + if (IS_DMA64) { + ldio->flags |= MFI_FRAME_SGL64; + ldio->sge_count = megasas_make_sgl64(instance, scp, &ldio->sgl); + } else + ldio->sge_count = megasas_make_sgl32(instance, scp, &ldio->sgl); + + /* + * Sense info specific + */ + ldio->sense_len = SCSI_SENSE_BUFFERSIZE; + ldio->sense_buf_phys_addr_hi = 0; + ldio->sense_buf_phys_addr_lo = cmd->sense_phys_addr; + + sge_bytes = sge_sz * ldio->sge_count; + + cmd->frame_count = (sge_bytes / MEGAMFI_FRAME_SIZE) + + ((sge_bytes % MEGAMFI_FRAME_SIZE) ? 1 : 0) + 1; + + if (cmd->frame_count > 7) + cmd->frame_count = 8; + + return cmd->frame_count; +} + +/** + * megasas_build_cmd - Prepares a command packet + * @instance: Adapter soft state + * @scp: SCSI command + * @frame_count: [OUT] Number of frames used to prepare this command + */ +static inline struct megasas_cmd *megasas_build_cmd(struct megasas_instance + *instance, + struct scsi_cmnd *scp, + int *frame_count) +{ + u32 logical_cmd; + struct megasas_cmd *cmd; + + /* + * Find out if this is logical or physical drive command. + */ + logical_cmd = MEGASAS_IS_LOGICAL(scp); + + /* + * Logical drive command + */ + if (logical_cmd) { + + if (scp->device->id >= MEGASAS_MAX_LD) { + scp->result = DID_BAD_TARGET << 16; + return NULL; + } + + switch (scp->cmnd[0]) { + + case READ_10: + case WRITE_10: + case READ_12: + case WRITE_12: + case READ_6: + case WRITE_6: + case READ_16: + case WRITE_16: + /* + * Fail for LUN > 0 + */ + if (scp->device->lun) { + scp->result = DID_BAD_TARGET << 16; + return NULL; + } + + cmd = megasas_get_cmd(instance); + + if (!cmd) { + scp->result = DID_IMM_RETRY << 16; + return NULL; + } + + *frame_count = megasas_build_ldio(instance, scp, cmd); + + if (!(*frame_count)) { + megasas_return_cmd(instance, cmd); + return NULL; + } + + return cmd; + + default: + /* + * Fail for LUN > 0 + */ + if (scp->device->lun) { + scp->result = DID_BAD_TARGET << 16; + return NULL; + } + + cmd = megasas_get_cmd(instance); + + if (!cmd) { + scp->result = DID_IMM_RETRY << 16; + return NULL; + } + + *frame_count = megasas_build_dcdb(instance, scp, cmd); + + if (!(*frame_count)) { + megasas_return_cmd(instance, cmd); + return NULL; + } + + return cmd; + } + } else { + cmd = megasas_get_cmd(instance); + + if (!cmd) { + scp->result = DID_IMM_RETRY << 16; + return NULL; + } + + *frame_count = megasas_build_dcdb(instance, scp, cmd); + + if (!(*frame_count)) { + megasas_return_cmd(instance, cmd); + return NULL; + } + + return cmd; + } + + return NULL; +} + +/** + * megasas_queue_command - Queue entry point + * @scmd: SCSI command to be queued + * @done: Callback entry point + */ +static int +megasas_queue_command(struct scsi_cmnd *scmd, void (*done) (struct scsi_cmnd *)) +{ + u32 frame_count; + unsigned long flags; + struct megasas_cmd *cmd; + struct megasas_instance *instance; + + instance = (struct megasas_instance *) + scmd->device->host->hostdata; + scmd->scsi_done = done; + scmd->result = 0; + + cmd = megasas_build_cmd(instance, scmd, &frame_count); + + if (!cmd) { + done(scmd); + return 0; + } + + cmd->scmd = scmd; + scmd->SCp.ptr = (char *)cmd; + scmd->SCp.sent_command = jiffies; + + /* + * Issue the command to the FW + */ + spin_lock_irqsave(&instance->instance_lock, flags); + instance->fw_outstanding++; + spin_unlock_irqrestore(&instance->instance_lock, flags); + + writel(((cmd->frame_phys_addr >> 3) | (cmd->frame_count - 1)), + &instance->reg_set->inbound_queue_port); + + return 0; +} + +/** + * megasas_wait_for_outstanding - Wait for all outstanding cmds + * @instance: Adapter soft state + * + * This function waits for upto MEGASAS_RESET_WAIT_TIME seconds for FW to + * complete all its outstanding commands. Returns error if one or more IOs + * are pending after this time period. It also marks the controller dead. + */ +static int megasas_wait_for_outstanding(struct megasas_instance *instance) +{ + int i; + u32 wait_time = MEGASAS_RESET_WAIT_TIME; + + for (i = 0; i < wait_time; i++) { + + if (!instance->fw_outstanding) + break; + + if (!(i % MEGASAS_RESET_NOTICE_INTERVAL)) { + printk(KERN_NOTICE "megasas: [%2d]waiting for %d " + "commands to complete\n", i, + instance->fw_outstanding); + } + + msleep(1000); + } + + if (instance->fw_outstanding) { + instance->hw_crit_error = 1; + return FAILED; + } + + return SUCCESS; +} + +/** + * megasas_generic_reset - Generic reset routine + * @scmd: Mid-layer SCSI command + * + * This routine implements a generic reset handler for device, bus and host + * reset requests. Device, bus and host specific reset handlers can use this + * function after they do their specific tasks. + */ +static int megasas_generic_reset(struct scsi_cmnd *scmd) +{ + int ret_val; + struct megasas_instance *instance; + + instance = (struct megasas_instance *)scmd->device->host->hostdata; + + printk(KERN_NOTICE "megasas: RESET -%ld cmd=%x \n", + scmd->serial_number, scmd->cmnd[0], scmd->device->channel, + scmd->device->id, scmd->device->lun); + + if (instance->hw_crit_error) { + printk(KERN_ERR "megasas: cannot recover from previous reset " + "failures\n"); + return FAILED; + } + + spin_unlock(scmd->device->host->host_lock); + + ret_val = megasas_wait_for_outstanding(instance); + + if (ret_val == SUCCESS) + printk(KERN_NOTICE "megasas: reset successful \n"); + else + printk(KERN_ERR "megasas: failed to do reset\n"); + + spin_lock(scmd->device->host->host_lock); + + return ret_val; +} + +static enum scsi_eh_timer_return megasas_reset_timer(struct scsi_cmnd *scmd) +{ + unsigned long seconds; + + if (scmd->SCp.ptr) { + seconds = (jiffies - scmd->SCp.sent_command) / HZ; + + if (seconds < 90) { + return EH_RESET_TIMER; + } else { + return EH_NOT_HANDLED; + } + } + + return EH_HANDLED; +} + +/** + * megasas_reset_device - Device reset handler entry point + */ +static int megasas_reset_device(struct scsi_cmnd *scmd) +{ + int ret; + + /* + * First wait for all commands to complete + */ + ret = megasas_generic_reset(scmd); + + return ret; +} + +/** + * megasas_reset_bus_host - Bus & host reset handler entry point + */ +static int megasas_reset_bus_host(struct scsi_cmnd *scmd) +{ + int ret; + + /* + * Frist wait for all commands to complete + */ + ret = megasas_generic_reset(scmd); + + return ret; +} + +/** + * megasas_service_aen - Processes an event notification + * @instance: Adapter soft state + * @cmd: AEN command completed by the ISR + * + * For AEN, driver sends a command down to FW that is held by the FW till an + * event occurs. When an event of interest occurs, FW completes the command + * that it was previously holding. + * + * This routines sends SIGIO signal to processes that have registered with the + * driver for AEN. + */ +static void +megasas_service_aen(struct megasas_instance *instance, struct megasas_cmd *cmd) +{ + /* + * Don't signal app if it is just an aborted previously registered aen + */ + if (!cmd->abort_aen) + kill_fasync(&megasas_async_queue, SIGIO, POLL_IN); + else + cmd->abort_aen = 0; + + instance->aen_cmd = NULL; + megasas_return_cmd(instance, cmd); +} + +/* + * Scsi host template for megaraid_sas driver + */ +static struct scsi_host_template megasas_template = { + + .module = THIS_MODULE, + .name = "LSI Logic SAS based MegaRAID driver", + .proc_name = "megaraid_sas", + .queuecommand = megasas_queue_command, + .eh_device_reset_handler = megasas_reset_device, + .eh_bus_reset_handler = megasas_reset_bus_host, + .eh_host_reset_handler = megasas_reset_bus_host, + .eh_timed_out = megasas_reset_timer, + .use_clustering = ENABLE_CLUSTERING, +}; + +/** + * megasas_complete_int_cmd - Completes an internal command + * @instance: Adapter soft state + * @cmd: Command to be completed + * + * The megasas_issue_blocked_cmd() function waits for a command to complete + * after it issues a command. This function wakes up that waiting routine by + * calling wake_up() on the wait queue. + */ +static void +megasas_complete_int_cmd(struct megasas_instance *instance, + struct megasas_cmd *cmd) +{ + cmd->cmd_status = cmd->frame->io.cmd_status; + + if (cmd->cmd_status == ENODATA) { + cmd->cmd_status = 0; + } + wake_up(&instance->int_cmd_wait_q); +} + +/** + * megasas_complete_abort - Completes aborting a command + * @instance: Adapter soft state + * @cmd: Cmd that was issued to abort another cmd + * + * The megasas_issue_blocked_abort_cmd() function waits on abort_cmd_wait_q + * after it issues an abort on a previously issued command. This function + * wakes up all functions waiting on the same wait queue. + */ +static void +megasas_complete_abort(struct megasas_instance *instance, + struct megasas_cmd *cmd) +{ + if (cmd->sync_cmd) { + cmd->sync_cmd = 0; + cmd->cmd_status = 0; + wake_up(&instance->abort_cmd_wait_q); + } + + return; +} + +/** + * megasas_unmap_sgbuf - Unmap SG buffers + * @instance: Adapter soft state + * @cmd: Completed command + */ +static inline void +megasas_unmap_sgbuf(struct megasas_instance *instance, struct megasas_cmd *cmd) +{ + dma_addr_t buf_h; + u8 opcode; + + if (cmd->scmd->use_sg) { + pci_unmap_sg(instance->pdev, cmd->scmd->request_buffer, + cmd->scmd->use_sg, cmd->scmd->sc_data_direction); + return; + } + + if (!cmd->scmd->request_bufflen) + return; + + opcode = cmd->frame->hdr.cmd; + + if ((opcode == MFI_CMD_LD_READ) || (opcode == MFI_CMD_LD_WRITE)) { + if (IS_DMA64) + buf_h = cmd->frame->io.sgl.sge64[0].phys_addr; + else + buf_h = cmd->frame->io.sgl.sge32[0].phys_addr; + } else { + if (IS_DMA64) + buf_h = cmd->frame->pthru.sgl.sge64[0].phys_addr; + else + buf_h = cmd->frame->pthru.sgl.sge32[0].phys_addr; + } + + pci_unmap_single(instance->pdev, buf_h, cmd->scmd->request_bufflen, + cmd->scmd->sc_data_direction); + return; +} + +/** + * megasas_complete_cmd - Completes a command + * @instance: Adapter soft state + * @cmd: Command to be completed + * @alt_status: If non-zero, use this value as status to + * SCSI mid-layer instead of the value returned + * by the FW. This should be used if caller wants + * an alternate status (as in the case of aborted + * commands) + */ +static inline void +megasas_complete_cmd(struct megasas_instance *instance, struct megasas_cmd *cmd, + u8 alt_status) +{ + int exception = 0; + struct megasas_header *hdr = &cmd->frame->hdr; + unsigned long flags; + + if (cmd->scmd) { + cmd->scmd->SCp.ptr = (char *)0; + } + + switch (hdr->cmd) { + + case MFI_CMD_PD_SCSI_IO: + case MFI_CMD_LD_SCSI_IO: + + /* + * MFI_CMD_PD_SCSI_IO and MFI_CMD_LD_SCSI_IO could have been + * issued either through an IO path or an IOCTL path. If it + * was via IOCTL, we will send it to internal completion. + */ + if (cmd->sync_cmd) { + cmd->sync_cmd = 0; + megasas_complete_int_cmd(instance, cmd); + break; + } + + /* + * Don't export physical disk devices to mid-layer. + */ + if (!MEGASAS_IS_LOGICAL(cmd->scmd) && + (hdr->cmd_status == MFI_STAT_OK) && + (cmd->scmd->cmnd[0] == INQUIRY)) { + + if (((*(u8 *) cmd->scmd->request_buffer) & 0x1F) == + TYPE_DISK) { + cmd->scmd->result = DID_BAD_TARGET << 16; + exception = 1; + } + } + + case MFI_CMD_LD_READ: + case MFI_CMD_LD_WRITE: + + if (alt_status) { + cmd->scmd->result = alt_status << 16; + exception = 1; + } + + if (exception) { + + spin_lock_irqsave(&instance->instance_lock, flags); + instance->fw_outstanding--; + spin_unlock_irqrestore(&instance->instance_lock, flags); + + megasas_unmap_sgbuf(instance, cmd); + cmd->scmd->scsi_done(cmd->scmd); + megasas_return_cmd(instance, cmd); + + break; + } + + switch (hdr->cmd_status) { + + case MFI_STAT_OK: + cmd->scmd->result = DID_OK << 16; + break; + + case MFI_STAT_SCSI_IO_FAILED: + case MFI_STAT_LD_INIT_IN_PROGRESS: + cmd->scmd->result = + (DID_ERROR << 16) | hdr->scsi_status; + break; + + case MFI_STAT_SCSI_DONE_WITH_ERROR: + + cmd->scmd->result = (DID_OK << 16) | hdr->scsi_status; + + if (hdr->scsi_status == SAM_STAT_CHECK_CONDITION) { + memset(cmd->scmd->sense_buffer, 0, + SCSI_SENSE_BUFFERSIZE); + memcpy(cmd->scmd->sense_buffer, cmd->sense, + hdr->sense_len); + + cmd->scmd->result |= DRIVER_SENSE << 24; + } + + break; + + case MFI_STAT_LD_OFFLINE: + case MFI_STAT_DEVICE_NOT_FOUND: + cmd->scmd->result = DID_BAD_TARGET << 16; + break; + + default: + printk(KERN_DEBUG "megasas: MFI FW status %#x\n", + hdr->cmd_status); + cmd->scmd->result = DID_ERROR << 16; + break; + } + + spin_lock_irqsave(&instance->instance_lock, flags); + instance->fw_outstanding--; + spin_unlock_irqrestore(&instance->instance_lock, flags); + + megasas_unmap_sgbuf(instance, cmd); + cmd->scmd->scsi_done(cmd->scmd); + megasas_return_cmd(instance, cmd); + + break; + + case MFI_CMD_SMP: + case MFI_CMD_STP: + case MFI_CMD_DCMD: + + /* + * See if got an event notification + */ + if (cmd->frame->dcmd.opcode == MR_DCMD_CTRL_EVENT_WAIT) + megasas_service_aen(instance, cmd); + else + megasas_complete_int_cmd(instance, cmd); + + break; + + case MFI_CMD_ABORT: + /* + * Cmd issued to abort another cmd returned + */ + megasas_complete_abort(instance, cmd); + break; + + default: + printk("megasas: Unknown command completed! [0x%X]\n", + hdr->cmd); + break; + } +} + +/** + * megasas_deplete_reply_queue - Processes all completed commands + * @instance: Adapter soft state + * @alt_status: Alternate status to be returned to + * SCSI mid-layer instead of the status + * returned by the FW + */ +static inline int +megasas_deplete_reply_queue(struct megasas_instance *instance, u8 alt_status) +{ + u32 status; + u32 producer; + u32 consumer; + u32 context; + struct megasas_cmd *cmd; + + /* + * Check if it is our interrupt + */ + status = readl(&instance->reg_set->outbound_intr_status); + + if (!(status & MFI_OB_INTR_STATUS_MASK)) { + return IRQ_NONE; + } + + /* + * Clear the interrupt by writing back the same value + */ + writel(status, &instance->reg_set->outbound_intr_status); + + producer = *instance->producer; + consumer = *instance->consumer; + + while (consumer != producer) { + context = instance->reply_queue[consumer]; + + cmd = instance->cmd_list[context]; + + megasas_complete_cmd(instance, cmd, alt_status); + + consumer++; + if (consumer == (instance->max_fw_cmds + 1)) { + consumer = 0; + } + } + + *instance->consumer = producer; + + return IRQ_HANDLED; +} + +/** + * megasas_isr - isr entry point + */ +static irqreturn_t megasas_isr(int irq, void *devp, struct pt_regs *regs) +{ + return megasas_deplete_reply_queue((struct megasas_instance *)devp, + DID_OK); +} + +/** + * megasas_transition_to_ready - Move the FW to READY state + * @reg_set: MFI register set + * + * During the initialization, FW passes can potentially be in any one of + * several possible states. If the FW in operational, waiting-for-handshake + * states, driver must take steps to bring it to ready state. Otherwise, it + * has to wait for the ready state. + */ +static int +megasas_transition_to_ready(struct megasas_register_set __iomem * reg_set) +{ + int i; + u8 max_wait; + u32 fw_state; + u32 cur_state; + + fw_state = readl(®_set->outbound_msg_0) & MFI_STATE_MASK; + + while (fw_state != MFI_STATE_READY) { + + printk(KERN_INFO "megasas: Waiting for FW to come to ready" + " state\n"); + switch (fw_state) { + + case MFI_STATE_FAULT: + + printk(KERN_DEBUG "megasas: FW in FAULT state!!\n"); + return -ENODEV; + + case MFI_STATE_WAIT_HANDSHAKE: + /* + * Set the CLR bit in inbound doorbell + */ + writel(MFI_INIT_CLEAR_HANDSHAKE, + ®_set->inbound_doorbell); + + max_wait = 2; + cur_state = MFI_STATE_WAIT_HANDSHAKE; + break; + + case MFI_STATE_OPERATIONAL: + /* + * Bring it to READY state; assuming max wait 2 secs + */ + megasas_disable_intr(reg_set); + writel(MFI_INIT_READY, ®_set->inbound_doorbell); + + max_wait = 10; + cur_state = MFI_STATE_OPERATIONAL; + break; + + case MFI_STATE_UNDEFINED: + /* + * This state should not last for more than 2 seconds + */ + max_wait = 2; + cur_state = MFI_STATE_UNDEFINED; + break; + + case MFI_STATE_BB_INIT: + max_wait = 2; + cur_state = MFI_STATE_BB_INIT; + break; + + case MFI_STATE_FW_INIT: + max_wait = 20; + cur_state = MFI_STATE_FW_INIT; + break; + + case MFI_STATE_FW_INIT_2: + max_wait = 20; + cur_state = MFI_STATE_FW_INIT_2; + break; + + case MFI_STATE_DEVICE_SCAN: + max_wait = 20; + cur_state = MFI_STATE_DEVICE_SCAN; + break; + + case MFI_STATE_FLUSH_CACHE: + max_wait = 20; + cur_state = MFI_STATE_FLUSH_CACHE; + break; + + default: + printk(KERN_DEBUG "megasas: Unknown state 0x%x\n", + fw_state); + return -ENODEV; + } + + /* + * The cur_state should not last for more than max_wait secs + */ + for (i = 0; i < (max_wait * 1000); i++) { + fw_state = MFI_STATE_MASK & + readl(®_set->outbound_msg_0); + + if (fw_state == cur_state) { + msleep(1); + } else + break; + } + + /* + * Return error if fw_state hasn't changed after max_wait + */ + if (fw_state == cur_state) { + printk(KERN_DEBUG "FW state [%d] hasn't changed " + "in %d secs\n", fw_state, max_wait); + return -ENODEV; + } + }; + + return 0; +} + +/** + * megasas_teardown_frame_pool - Destroy the cmd frame DMA pool + * @instance: Adapter soft state + */ +static void megasas_teardown_frame_pool(struct megasas_instance *instance) +{ + int i; + u32 max_cmd = instance->max_fw_cmds; + struct megasas_cmd *cmd; + + if (!instance->frame_dma_pool) + return; + + /* + * Return all frames to pool + */ + for (i = 0; i < max_cmd; i++) { + + cmd = instance->cmd_list[i]; + + if (cmd->frame) + pci_pool_free(instance->frame_dma_pool, cmd->frame, + cmd->frame_phys_addr); + + if (cmd->sense) + pci_pool_free(instance->sense_dma_pool, cmd->frame, + cmd->sense_phys_addr); + } + + /* + * Now destroy the pool itself + */ + pci_pool_destroy(instance->frame_dma_pool); + pci_pool_destroy(instance->sense_dma_pool); + + instance->frame_dma_pool = NULL; + instance->sense_dma_pool = NULL; +} + +/** + * megasas_create_frame_pool - Creates DMA pool for cmd frames + * @instance: Adapter soft state + * + * Each command packet has an embedded DMA memory buffer that is used for + * filling MFI frame and the SG list that immediately follows the frame. This + * function creates those DMA memory buffers for each command packet by using + * PCI pool facility. + */ +static int megasas_create_frame_pool(struct megasas_instance *instance) +{ + int i; + u32 max_cmd; + u32 sge_sz; + u32 sgl_sz; + u32 total_sz; + u32 frame_count; + struct megasas_cmd *cmd; + + max_cmd = instance->max_fw_cmds; + + /* + * Size of our frame is 64 bytes for MFI frame, followed by max SG + * elements and finally SCSI_SENSE_BUFFERSIZE bytes for sense buffer + */ + sge_sz = (IS_DMA64) ? sizeof(struct megasas_sge64) : + sizeof(struct megasas_sge32); + + /* + * Calculated the number of 64byte frames required for SGL + */ + sgl_sz = sge_sz * instance->max_num_sge; + frame_count = (sgl_sz + MEGAMFI_FRAME_SIZE - 1) / MEGAMFI_FRAME_SIZE; + + /* + * We need one extra frame for the MFI command + */ + frame_count++; + + total_sz = MEGAMFI_FRAME_SIZE * frame_count; + /* + * Use DMA pool facility provided by PCI layer + */ + instance->frame_dma_pool = pci_pool_create("megasas frame pool", + instance->pdev, total_sz, 64, + 0); + + if (!instance->frame_dma_pool) { + printk(KERN_DEBUG "megasas: failed to setup frame pool\n"); + return -ENOMEM; + } + + instance->sense_dma_pool = pci_pool_create("megasas sense pool", + instance->pdev, 128, 4, 0); + + if (!instance->sense_dma_pool) { + printk(KERN_DEBUG "megasas: failed to setup sense pool\n"); + + pci_pool_destroy(instance->frame_dma_pool); + instance->frame_dma_pool = NULL; + + return -ENOMEM; + } + + /* + * Allocate and attach a frame to each of the commands in cmd_list. + * By making cmd->index as the context instead of the &cmd, we can + * always use 32bit context regardless of the architecture + */ + for (i = 0; i < max_cmd; i++) { + + cmd = instance->cmd_list[i]; + + cmd->frame = pci_pool_alloc(instance->frame_dma_pool, + GFP_KERNEL, &cmd->frame_phys_addr); + + cmd->sense = pci_pool_alloc(instance->sense_dma_pool, + GFP_KERNEL, &cmd->sense_phys_addr); + + /* + * megasas_teardown_frame_pool() takes care of freeing + * whatever has been allocated + */ + if (!cmd->frame || !cmd->sense) { + printk(KERN_DEBUG "megasas: pci_pool_alloc failed \n"); + megasas_teardown_frame_pool(instance); + return -ENOMEM; + } + + cmd->frame->io.context = cmd->index; + } + + return 0; +} + +/** + * megasas_free_cmds - Free all the cmds in the free cmd pool + * @instance: Adapter soft state + */ +static void megasas_free_cmds(struct megasas_instance *instance) +{ + int i; + /* First free the MFI frame pool */ + megasas_teardown_frame_pool(instance); + + /* Free all the commands in the cmd_list */ + for (i = 0; i < instance->max_fw_cmds; i++) + kfree(instance->cmd_list[i]); + + /* Free the cmd_list buffer itself */ + kfree(instance->cmd_list); + instance->cmd_list = NULL; + + INIT_LIST_HEAD(&instance->cmd_pool); +} + +/** + * megasas_alloc_cmds - Allocates the command packets + * @instance: Adapter soft state + * + * Each command that is issued to the FW, whether IO commands from the OS or + * internal commands like IOCTLs, are wrapped in local data structure called + * megasas_cmd. The frame embedded in this megasas_cmd is actually issued to + * the FW. + * + * Each frame has a 32-bit field called context (tag). This context is used + * to get back the megasas_cmd from the frame when a frame gets completed in + * the ISR. Typically the address of the megasas_cmd itself would be used as + * the context. But we wanted to keep the differences between 32 and 64 bit + * systems to the mininum. We always use 32 bit integers for the context. In + * this driver, the 32 bit values are the indices into an array cmd_list. + * This array is used only to look up the megasas_cmd given the context. The + * free commands themselves are maintained in a linked list called cmd_pool. + */ +static int megasas_alloc_cmds(struct megasas_instance *instance) +{ + int i; + int j; + u32 max_cmd; + struct megasas_cmd *cmd; + + max_cmd = instance->max_fw_cmds; + + /* + * instance->cmd_list is an array of struct megasas_cmd pointers. + * Allocate the dynamic array first and then allocate individual + * commands. + */ + instance->cmd_list = kmalloc(sizeof(struct megasas_cmd *) * max_cmd, + GFP_KERNEL); + + if (!instance->cmd_list) { + printk(KERN_DEBUG "megasas: out of memory\n"); + return -ENOMEM; + } + + memset(instance->cmd_list, 0, sizeof(struct megasas_cmd *) * max_cmd); + + for (i = 0; i < max_cmd; i++) { + instance->cmd_list[i] = kmalloc(sizeof(struct megasas_cmd), + GFP_KERNEL); + + if (!instance->cmd_list[i]) { + + for (j = 0; j < i; j++) + kfree(instance->cmd_list[j]); + + kfree(instance->cmd_list); + instance->cmd_list = NULL; + + return -ENOMEM; + } + } + + /* + * Add all the commands to command pool (instance->cmd_pool) + */ + for (i = 0; i < max_cmd; i++) { + cmd = instance->cmd_list[i]; + memset(cmd, 0, sizeof(struct megasas_cmd)); + cmd->index = i; + cmd->instance = instance; + + list_add_tail(&cmd->list, &instance->cmd_pool); + } + + /* + * Create a frame pool and assign one frame to each cmd + */ + if (megasas_create_frame_pool(instance)) { + printk(KERN_DEBUG "megasas: Error creating frame DMA pool\n"); + megasas_free_cmds(instance); + } + + return 0; +} + +/** + * megasas_get_controller_info - Returns FW's controller structure + * @instance: Adapter soft state + * @ctrl_info: Controller information structure + * + * Issues an internal command (DCMD) to get the FW's controller structure. + * This information is mainly used to find out the maximum IO transfer per + * command supported by the FW. + */ +static int +megasas_get_ctrl_info(struct megasas_instance *instance, + struct megasas_ctrl_info *ctrl_info) +{ + int ret = 0; + struct megasas_cmd *cmd; + struct megasas_dcmd_frame *dcmd; + struct megasas_ctrl_info *ci; + dma_addr_t ci_h = 0; + + cmd = megasas_get_cmd(instance); + + if (!cmd) { + printk(KERN_DEBUG "megasas: Failed to get a free cmd\n"); + return -ENOMEM; + } + + dcmd = &cmd->frame->dcmd; + + ci = pci_alloc_consistent(instance->pdev, + sizeof(struct megasas_ctrl_info), &ci_h); + + if (!ci) { + printk(KERN_DEBUG "Failed to alloc mem for ctrl info\n"); + megasas_return_cmd(instance, cmd); + return -ENOMEM; + } + + memset(ci, 0, sizeof(*ci)); + memset(dcmd->mbox.b, 0, MFI_MBOX_SIZE); + + dcmd->cmd = MFI_CMD_DCMD; + dcmd->cmd_status = 0xFF; + dcmd->sge_count = 1; + dcmd->flags = MFI_FRAME_DIR_READ; + dcmd->timeout = 0; + dcmd->data_xfer_len = sizeof(struct megasas_ctrl_info); + dcmd->opcode = MR_DCMD_CTRL_GET_INFO; + dcmd->sgl.sge32[0].phys_addr = ci_h; + dcmd->sgl.sge32[0].length = sizeof(struct megasas_ctrl_info); + + if (!megasas_issue_polled(instance, cmd)) { + ret = 0; + memcpy(ctrl_info, ci, sizeof(struct megasas_ctrl_info)); + } else { + ret = -1; + } + + pci_free_consistent(instance->pdev, sizeof(struct megasas_ctrl_info), + ci, ci_h); + + megasas_return_cmd(instance, cmd); + return ret; +} + +/** + * megasas_init_mfi - Initializes the FW + * @instance: Adapter soft state + * + * This is the main function for initializing MFI firmware. + */ +static int megasas_init_mfi(struct megasas_instance *instance) +{ + u32 context_sz; + u32 reply_q_sz; + u32 max_sectors_1; + u32 max_sectors_2; + struct megasas_register_set __iomem *reg_set; + + struct megasas_cmd *cmd; + struct megasas_ctrl_info *ctrl_info; + + struct megasas_init_frame *init_frame; + struct megasas_init_queue_info *initq_info; + dma_addr_t init_frame_h; + dma_addr_t initq_info_h; + + /* + * Map the message registers + */ + instance->base_addr = pci_resource_start(instance->pdev, 0); + + if (pci_request_regions(instance->pdev, "megasas: LSI Logic")) { + printk(KERN_DEBUG "megasas: IO memory region busy!\n"); + return -EBUSY; + } + + instance->reg_set = ioremap_nocache(instance->base_addr, 8192); + + if (!instance->reg_set) { + printk(KERN_DEBUG "megasas: Failed to map IO mem\n"); + goto fail_ioremap; + } + + reg_set = instance->reg_set; + + /* + * We expect the FW state to be READY + */ + if (megasas_transition_to_ready(instance->reg_set)) + goto fail_ready_state; + + /* + * Get various operational parameters from status register + */ + instance->max_fw_cmds = readl(®_set->outbound_msg_0) & 0x00FFFF; + instance->max_num_sge = (readl(®_set->outbound_msg_0) & 0xFF0000) >> + 0x10; + /* + * Create a pool of commands + */ + if (megasas_alloc_cmds(instance)) + goto fail_alloc_cmds; + + /* + * Allocate memory for reply queue. Length of reply queue should + * be _one_ more than the maximum commands handled by the firmware. + * + * Note: When FW completes commands, it places corresponding contex + * values in this circular reply queue. This circular queue is a fairly + * typical producer-consumer queue. FW is the producer (of completed + * commands) and the driver is the consumer. + */ + context_sz = sizeof(u32); + reply_q_sz = context_sz * (instance->max_fw_cmds + 1); + + instance->reply_queue = pci_alloc_consistent(instance->pdev, + reply_q_sz, + &instance->reply_queue_h); + + if (!instance->reply_queue) { + printk(KERN_DEBUG "megasas: Out of DMA mem for reply queue\n"); + goto fail_reply_queue; + } + + /* + * Prepare a init frame. Note the init frame points to queue info + * structure. Each frame has SGL allocated after first 64 bytes. For + * this frame - since we don't need any SGL - we use SGL's space as + * queue info structure + * + * We will not get a NULL command below. We just created the pool. + */ + cmd = megasas_get_cmd(instance); + + init_frame = (struct megasas_init_frame *)cmd->frame; + initq_info = (struct megasas_init_queue_info *) + ((unsigned long)init_frame + 64); + + init_frame_h = cmd->frame_phys_addr; + initq_info_h = init_frame_h + 64; + + memset(init_frame, 0, MEGAMFI_FRAME_SIZE); + memset(initq_info, 0, sizeof(struct megasas_init_queue_info)); + + initq_info->reply_queue_entries = instance->max_fw_cmds + 1; + initq_info->reply_queue_start_phys_addr_lo = instance->reply_queue_h; + + initq_info->producer_index_phys_addr_lo = instance->producer_h; + initq_info->consumer_index_phys_addr_lo = instance->consumer_h; + + init_frame->cmd = MFI_CMD_INIT; + init_frame->cmd_status = 0xFF; + init_frame->queue_info_new_phys_addr_lo = initq_info_h; + + init_frame->data_xfer_len = sizeof(struct megasas_init_queue_info); + + /* + * Issue the init frame in polled mode + */ + if (megasas_issue_polled(instance, cmd)) { + printk(KERN_DEBUG "megasas: Failed to init firmware\n"); + goto fail_fw_init; + } + + megasas_return_cmd(instance, cmd); + + ctrl_info = kmalloc(sizeof(struct megasas_ctrl_info), GFP_KERNEL); + + /* + * Compute the max allowed sectors per IO: The controller info has two + * limits on max sectors. Driver should use the minimum of these two. + * + * 1 << stripe_sz_ops.min = max sectors per strip + * + * Note that older firmwares ( < FW ver 30) didn't report information + * to calculate max_sectors_1. So the number ended up as zero always. + */ + if (ctrl_info && !megasas_get_ctrl_info(instance, ctrl_info)) { + + max_sectors_1 = (1 << ctrl_info->stripe_sz_ops.min) * + ctrl_info->max_strips_per_io; + max_sectors_2 = ctrl_info->max_request_size; + + instance->max_sectors_per_req = (max_sectors_1 < max_sectors_2) + ? max_sectors_1 : max_sectors_2; + } else + instance->max_sectors_per_req = instance->max_num_sge * + PAGE_SIZE / 512; + + kfree(ctrl_info); + + return 0; + + fail_fw_init: + megasas_return_cmd(instance, cmd); + + pci_free_consistent(instance->pdev, reply_q_sz, + instance->reply_queue, instance->reply_queue_h); + fail_reply_queue: + megasas_free_cmds(instance); + + fail_alloc_cmds: + fail_ready_state: + iounmap(instance->reg_set); + + fail_ioremap: + pci_release_regions(instance->pdev); + + return -EINVAL; +} + +/** + * megasas_release_mfi - Reverses the FW initialization + * @intance: Adapter soft state + */ +static void megasas_release_mfi(struct megasas_instance *instance) +{ + u32 reply_q_sz = sizeof(u32) * (instance->max_fw_cmds + 1); + + pci_free_consistent(instance->pdev, reply_q_sz, + instance->reply_queue, instance->reply_queue_h); + + megasas_free_cmds(instance); + + iounmap(instance->reg_set); + + pci_release_regions(instance->pdev); +} + +/** + * megasas_get_seq_num - Gets latest event sequence numbers + * @instance: Adapter soft state + * @eli: FW event log sequence numbers information + * + * FW maintains a log of all events in a non-volatile area. Upper layers would + * usually find out the latest sequence number of the events, the seq number at + * the boot etc. They would "read" all the events below the latest seq number + * by issuing a direct fw cmd (DCMD). For the future events (beyond latest seq + * number), they would subsribe to AEN (asynchronous event notification) and + * wait for the events to happen. + */ +static int +megasas_get_seq_num(struct megasas_instance *instance, + struct megasas_evt_log_info *eli) +{ + struct megasas_cmd *cmd; + struct megasas_dcmd_frame *dcmd; + struct megasas_evt_log_info *el_info; + dma_addr_t el_info_h = 0; + + cmd = megasas_get_cmd(instance); + + if (!cmd) { + return -ENOMEM; + } + + dcmd = &cmd->frame->dcmd; + el_info = pci_alloc_consistent(instance->pdev, + sizeof(struct megasas_evt_log_info), + &el_info_h); + + if (!el_info) { + megasas_return_cmd(instance, cmd); + return -ENOMEM; + } + + memset(el_info, 0, sizeof(*el_info)); + memset(dcmd->mbox.b, 0, MFI_MBOX_SIZE); + + dcmd->cmd = MFI_CMD_DCMD; + dcmd->cmd_status = 0x0; + dcmd->sge_count = 1; + dcmd->flags = MFI_FRAME_DIR_READ; + dcmd->timeout = 0; + dcmd->data_xfer_len = sizeof(struct megasas_evt_log_info); + dcmd->opcode = MR_DCMD_CTRL_EVENT_GET_INFO; + dcmd->sgl.sge32[0].phys_addr = el_info_h; + dcmd->sgl.sge32[0].length = sizeof(struct megasas_evt_log_info); + + megasas_issue_blocked_cmd(instance, cmd); + + /* + * Copy the data back into callers buffer + */ + memcpy(eli, el_info, sizeof(struct megasas_evt_log_info)); + + pci_free_consistent(instance->pdev, sizeof(struct megasas_evt_log_info), + el_info, el_info_h); + + megasas_return_cmd(instance, cmd); + + return 0; +} + +/** + * megasas_register_aen - Registers for asynchronous event notification + * @instance: Adapter soft state + * @seq_num: The starting sequence number + * @class_locale: Class of the event + * + * This function subscribes for AEN for events beyond the @seq_num. It requests + * to be notified if and only if the event is of type @class_locale + */ +static int +megasas_register_aen(struct megasas_instance *instance, u32 seq_num, + u32 class_locale_word) +{ + int ret_val; + struct megasas_cmd *cmd; + struct megasas_dcmd_frame *dcmd; + union megasas_evt_class_locale curr_aen; + union megasas_evt_class_locale prev_aen; + + /* + * If there an AEN pending already (aen_cmd), check if the + * class_locale of that pending AEN is inclusive of the new + * AEN request we currently have. If it is, then we don't have + * to do anything. In other words, whichever events the current + * AEN request is subscribing to, have already been subscribed + * to. + * + * If the old_cmd is _not_ inclusive, then we have to abort + * that command, form a class_locale that is superset of both + * old and current and re-issue to the FW + */ + + curr_aen.word = class_locale_word; + + if (instance->aen_cmd) { + + prev_aen.word = instance->aen_cmd->frame->dcmd.mbox.w[1]; + + /* + * A class whose enum value is smaller is inclusive of all + * higher values. If a PROGRESS (= -1) was previously + * registered, then a new registration requests for higher + * classes need not be sent to FW. They are automatically + * included. + * + * Locale numbers don't have such hierarchy. They are bitmap + * values + */ + if ((prev_aen.members.class <= curr_aen.members.class) && + !((prev_aen.members.locale & curr_aen.members.locale) ^ + curr_aen.members.locale)) { + /* + * Previously issued event registration includes + * current request. Nothing to do. + */ + return 0; + } else { + curr_aen.members.locale |= prev_aen.members.locale; + + if (prev_aen.members.class < curr_aen.members.class) + curr_aen.members.class = prev_aen.members.class; + + instance->aen_cmd->abort_aen = 1; + ret_val = megasas_issue_blocked_abort_cmd(instance, + instance-> + aen_cmd); + + if (ret_val) { + printk(KERN_DEBUG "megasas: Failed to abort " + "previous AEN command\n"); + return ret_val; + } + } + } + + cmd = megasas_get_cmd(instance); + + if (!cmd) + return -ENOMEM; + + dcmd = &cmd->frame->dcmd; + + memset(instance->evt_detail, 0, sizeof(struct megasas_evt_detail)); + + /* + * Prepare DCMD for aen registration + */ + memset(dcmd->mbox.b, 0, MFI_MBOX_SIZE); + + dcmd->cmd = MFI_CMD_DCMD; + dcmd->cmd_status = 0x0; + dcmd->sge_count = 1; + dcmd->flags = MFI_FRAME_DIR_READ; + dcmd->timeout = 0; + dcmd->data_xfer_len = sizeof(struct megasas_evt_detail); + dcmd->opcode = MR_DCMD_CTRL_EVENT_WAIT; + dcmd->mbox.w[0] = seq_num; + dcmd->mbox.w[1] = curr_aen.word; + dcmd->sgl.sge32[0].phys_addr = (u32) instance->evt_detail_h; + dcmd->sgl.sge32[0].length = sizeof(struct megasas_evt_detail); + + /* + * Store reference to the cmd used to register for AEN. When an + * application wants us to register for AEN, we have to abort this + * cmd and re-register with a new EVENT LOCALE supplied by that app + */ + instance->aen_cmd = cmd; + + /* + * Issue the aen registration frame + */ + writel(cmd->frame_phys_addr >> 3, + &instance->reg_set->inbound_queue_port); + + return 0; +} + +/** + * megasas_start_aen - Subscribes to AEN during driver load time + * @instance: Adapter soft state + */ +static int megasas_start_aen(struct megasas_instance *instance) +{ + struct megasas_evt_log_info eli; + union megasas_evt_class_locale class_locale; + + /* + * Get the latest sequence number from FW + */ + memset(&eli, 0, sizeof(eli)); + + if (megasas_get_seq_num(instance, &eli)) + return -1; + + /* + * Register AEN with FW for latest sequence number plus 1 + */ + class_locale.members.reserved = 0; + class_locale.members.locale = MR_EVT_LOCALE_ALL; + class_locale.members.class = MR_EVT_CLASS_DEBUG; + + return megasas_register_aen(instance, eli.newest_seq_num + 1, + class_locale.word); +} + +/** + * megasas_io_attach - Attaches this driver to SCSI mid-layer + * @instance: Adapter soft state + */ +static int megasas_io_attach(struct megasas_instance *instance) +{ + struct Scsi_Host *host = instance->host; + + /* + * Export parameters required by SCSI mid-layer + */ + host->irq = instance->pdev->irq; + host->unique_id = instance->unique_id; + host->can_queue = instance->max_fw_cmds - MEGASAS_INT_CMDS; + host->this_id = instance->init_id; + host->sg_tablesize = instance->max_num_sge; + host->max_sectors = instance->max_sectors_per_req; + host->cmd_per_lun = 128; + host->max_channel = MEGASAS_MAX_CHANNELS - 1; + host->max_id = MEGASAS_MAX_DEV_PER_CHANNEL; + host->max_lun = MEGASAS_MAX_LUN; + + /* + * Notify the mid-layer about the new controller + */ + if (scsi_add_host(host, &instance->pdev->dev)) { + printk(KERN_DEBUG "megasas: scsi_add_host failed\n"); + return -ENODEV; + } + + /* + * Trigger SCSI to scan our drives + */ + scsi_scan_host(host); + return 0; +} + +/** + * megasas_probe_one - PCI hotplug entry point + * @pdev: PCI device structure + * @id: PCI ids of supported hotplugged adapter + */ +static int __devinit +megasas_probe_one(struct pci_dev *pdev, const struct pci_device_id *id) +{ + int rval; + struct Scsi_Host *host; + struct megasas_instance *instance; + + /* + * Announce PCI information + */ + printk(KERN_INFO "megasas: %#4.04x:%#4.04x:%#4.04x:%#4.04x: ", + pdev->vendor, pdev->device, pdev->subsystem_vendor, + pdev->subsystem_device); + + printk("bus %d:slot %d:func %d\n", + pdev->bus->number, PCI_SLOT(pdev->devfn), PCI_FUNC(pdev->devfn)); + + /* + * PCI prepping: enable device set bus mastering and dma mask + */ + rval = pci_enable_device(pdev); + + if (rval) { + return rval; + } + + pci_set_master(pdev); + + /* + * All our contollers are capable of performing 64-bit DMA + */ + if (IS_DMA64) { + if (pci_set_dma_mask(pdev, DMA_64BIT_MASK) != 0) { + + if (pci_set_dma_mask(pdev, DMA_32BIT_MASK) != 0) + goto fail_set_dma_mask; + } + } else { + if (pci_set_dma_mask(pdev, DMA_32BIT_MASK) != 0) + goto fail_set_dma_mask; + } + + host = scsi_host_alloc(&megasas_template, + sizeof(struct megasas_instance)); + + if (!host) { + printk(KERN_DEBUG "megasas: scsi_host_alloc failed\n"); + goto fail_alloc_instance; + } + + instance = (struct megasas_instance *)host->hostdata; + memset(instance, 0, sizeof(*instance)); + + instance->producer = pci_alloc_consistent(pdev, sizeof(u32), + &instance->producer_h); + instance->consumer = pci_alloc_consistent(pdev, sizeof(u32), + &instance->consumer_h); + + if (!instance->producer || !instance->consumer) { + printk(KERN_DEBUG "megasas: Failed to allocate memory for " + "producer, consumer\n"); + goto fail_alloc_dma_buf; + } + + *instance->producer = 0; + *instance->consumer = 0; + + instance->evt_detail = pci_alloc_consistent(pdev, + sizeof(struct + megasas_evt_detail), + &instance->evt_detail_h); + + if (!instance->evt_detail) { + printk(KERN_DEBUG "megasas: Failed to allocate memory for " + "event detail structure\n"); + goto fail_alloc_dma_buf; + } + + /* + * Initialize locks and queues + */ + INIT_LIST_HEAD(&instance->cmd_pool); + + init_waitqueue_head(&instance->int_cmd_wait_q); + init_waitqueue_head(&instance->abort_cmd_wait_q); + + spin_lock_init(&instance->cmd_pool_lock); + spin_lock_init(&instance->instance_lock); + + sema_init(&instance->aen_mutex, 1); + sema_init(&instance->ioctl_sem, MEGASAS_INT_CMDS); + + /* + * Initialize PCI related and misc parameters + */ + instance->pdev = pdev; + instance->host = host; + instance->unique_id = pdev->bus->number << 8 | pdev->devfn; + instance->init_id = MEGASAS_DEFAULT_INIT_ID; + + /* + * Initialize MFI Firmware + */ + if (megasas_init_mfi(instance)) + goto fail_init_mfi; + + /* + * Register IRQ + */ + if (request_irq(pdev->irq, megasas_isr, SA_SHIRQ, "megasas", instance)) { + printk(KERN_DEBUG "megasas: Failed to register IRQ\n"); + goto fail_irq; + } + + megasas_enable_intr(instance->reg_set); + + /* + * Store instance in PCI softstate + */ + pci_set_drvdata(pdev, instance); + + /* + * Add this controller to megasas_mgmt_info structure so that it + * can be exported to management applications + */ + megasas_mgmt_info.count++; + megasas_mgmt_info.instance[megasas_mgmt_info.max_index] = instance; + megasas_mgmt_info.max_index++; + + /* + * Initiate AEN (Asynchronous Event Notification) + */ + if (megasas_start_aen(instance)) { + printk(KERN_DEBUG "megasas: start aen failed\n"); + goto fail_start_aen; + } + + /* + * Register with SCSI mid-layer + */ + if (megasas_io_attach(instance)) + goto fail_io_attach; + + return 0; + + fail_start_aen: + fail_io_attach: + megasas_mgmt_info.count--; + megasas_mgmt_info.instance[megasas_mgmt_info.max_index] = NULL; + megasas_mgmt_info.max_index--; + + pci_set_drvdata(pdev, NULL); + megasas_disable_intr(instance->reg_set); + free_irq(instance->pdev->irq, instance); + + megasas_release_mfi(instance); + + fail_irq: + fail_init_mfi: + fail_alloc_dma_buf: + if (instance->evt_detail) + pci_free_consistent(pdev, sizeof(struct megasas_evt_detail), + instance->evt_detail, + instance->evt_detail_h); + + if (instance->producer) + pci_free_consistent(pdev, sizeof(u32), instance->producer, + instance->producer_h); + if (instance->consumer) + pci_free_consistent(pdev, sizeof(u32), instance->consumer, + instance->consumer_h); + scsi_host_put(host); + + fail_alloc_instance: + fail_set_dma_mask: + pci_disable_device(pdev); + + return -ENODEV; +} + +/** + * megasas_flush_cache - Requests FW to flush all its caches + * @instance: Adapter soft state + */ +static void megasas_flush_cache(struct megasas_instance *instance) +{ + struct megasas_cmd *cmd; + struct megasas_dcmd_frame *dcmd; + + cmd = megasas_get_cmd(instance); + + if (!cmd) + return; + + dcmd = &cmd->frame->dcmd; + + memset(dcmd->mbox.b, 0, MFI_MBOX_SIZE); + + dcmd->cmd = MFI_CMD_DCMD; + dcmd->cmd_status = 0x0; + dcmd->sge_count = 0; + dcmd->flags = MFI_FRAME_DIR_NONE; + dcmd->timeout = 0; + dcmd->data_xfer_len = 0; + dcmd->opcode = MR_DCMD_CTRL_CACHE_FLUSH; + dcmd->mbox.b[0] = MR_FLUSH_CTRL_CACHE | MR_FLUSH_DISK_CACHE; + + megasas_issue_blocked_cmd(instance, cmd); + + megasas_return_cmd(instance, cmd); + + return; +} + +/** + * megasas_shutdown_controller - Instructs FW to shutdown the controller + * @instance: Adapter soft state + */ +static void megasas_shutdown_controller(struct megasas_instance *instance) +{ + struct megasas_cmd *cmd; + struct megasas_dcmd_frame *dcmd; + + cmd = megasas_get_cmd(instance); + + if (!cmd) + return; + + if (instance->aen_cmd) + megasas_issue_blocked_abort_cmd(instance, instance->aen_cmd); + + dcmd = &cmd->frame->dcmd; + + memset(dcmd->mbox.b, 0, MFI_MBOX_SIZE); + + dcmd->cmd = MFI_CMD_DCMD; + dcmd->cmd_status = 0x0; + dcmd->sge_count = 0; + dcmd->flags = MFI_FRAME_DIR_NONE; + dcmd->timeout = 0; + dcmd->data_xfer_len = 0; + dcmd->opcode = MR_DCMD_CTRL_SHUTDOWN; + + megasas_issue_blocked_cmd(instance, cmd); + + megasas_return_cmd(instance, cmd); + + return; +} + +/** + * megasas_detach_one - PCI hot"un"plug entry point + * @pdev: PCI device structure + */ +static void megasas_detach_one(struct pci_dev *pdev) +{ + int i; + struct Scsi_Host *host; + struct megasas_instance *instance; + + instance = pci_get_drvdata(pdev); + host = instance->host; + + scsi_remove_host(instance->host); + megasas_flush_cache(instance); + megasas_shutdown_controller(instance); + + /* + * Take the instance off the instance array. Note that we will not + * decrement the max_index. We let this array be sparse array + */ + for (i = 0; i < megasas_mgmt_info.max_index; i++) { + if (megasas_mgmt_info.instance[i] == instance) { + megasas_mgmt_info.count--; + megasas_mgmt_info.instance[i] = NULL; + + break; + } + } + + pci_set_drvdata(instance->pdev, NULL); + + megasas_disable_intr(instance->reg_set); + + free_irq(instance->pdev->irq, instance); + + megasas_release_mfi(instance); + + pci_free_consistent(pdev, sizeof(struct megasas_evt_detail), + instance->evt_detail, instance->evt_detail_h); + + pci_free_consistent(pdev, sizeof(u32), instance->producer, + instance->producer_h); + + pci_free_consistent(pdev, sizeof(u32), instance->consumer, + instance->consumer_h); + + scsi_host_put(host); + + pci_set_drvdata(pdev, NULL); + + pci_disable_device(pdev); + + return; +} + +/** + * megasas_shutdown - Shutdown entry point + * @device: Generic device structure + */ +static void megasas_shutdown(struct pci_dev *pdev) +{ + struct megasas_instance *instance = pci_get_drvdata(pdev); + megasas_flush_cache(instance); +} + +/** + * megasas_mgmt_open - char node "open" entry point + */ +static int megasas_mgmt_open(struct inode *inode, struct file *filep) +{ + /* + * Allow only those users with admin rights + */ + if (!capable(CAP_SYS_ADMIN)) + return -EACCES; + + return 0; +} + +/** + * megasas_mgmt_release - char node "release" entry point + */ +static int megasas_mgmt_release(struct inode *inode, struct file *filep) +{ + filep->private_data = NULL; + fasync_helper(-1, filep, 0, &megasas_async_queue); + + return 0; +} + +/** + * megasas_mgmt_fasync - Async notifier registration from applications + * + * This function adds the calling process to a driver global queue. When an + * event occurs, SIGIO will be sent to all processes in this queue. + */ +static int megasas_mgmt_fasync(int fd, struct file *filep, int mode) +{ + int rc; + + down(&megasas_async_queue_mutex); + + rc = fasync_helper(fd, filep, mode, &megasas_async_queue); + + up(&megasas_async_queue_mutex); + + if (rc >= 0) { + /* For sanity check when we get ioctl */ + filep->private_data = filep; + return 0; + } + + printk(KERN_DEBUG "megasas: fasync_helper failed [%d]\n", rc); + + return rc; +} + +/** + * megasas_mgmt_fw_ioctl - Issues management ioctls to FW + * @instance: Adapter soft state + * @argp: User's ioctl packet + */ +static int +megasas_mgmt_fw_ioctl(struct megasas_instance *instance, + struct megasas_iocpacket __user * user_ioc, + struct megasas_iocpacket *ioc) +{ + struct megasas_sge32 *kern_sge32; + struct megasas_cmd *cmd; + void *kbuff_arr[MAX_IOCTL_SGE]; + dma_addr_t buf_handle = 0; + int error = 0, i; + void *sense = NULL; + dma_addr_t sense_handle; + u32 *sense_ptr; + + memset(kbuff_arr, 0, sizeof(kbuff_arr)); + + if (ioc->sge_count > MAX_IOCTL_SGE) { + printk(KERN_DEBUG "megasas: SGE count [%d] > max limit [%d]\n", + ioc->sge_count, MAX_IOCTL_SGE); + return -EINVAL; + } + + cmd = megasas_get_cmd(instance); + if (!cmd) { + printk(KERN_DEBUG "megasas: Failed to get a cmd packet\n"); + return -ENOMEM; + } + + /* + * User's IOCTL packet has 2 frames (maximum). Copy those two + * frames into our cmd's frames. cmd->frame's context will get + * overwritten when we copy from user's frames. So set that value + * alone separately + */ + memcpy(cmd->frame, ioc->frame.raw, 2 * MEGAMFI_FRAME_SIZE); + cmd->frame->hdr.context = cmd->index; + + /* + * The management interface between applications and the fw uses + * MFI frames. E.g, RAID configuration changes, LD property changes + * etc are accomplishes through different kinds of MFI frames. The + * driver needs to care only about substituting user buffers with + * kernel buffers in SGLs. The location of SGL is embedded in the + * struct iocpacket itself. + */ + kern_sge32 = (struct megasas_sge32 *) + ((unsigned long)cmd->frame + ioc->sgl_off); + + /* + * For each user buffer, create a mirror buffer and copy in + */ + for (i = 0; i < ioc->sge_count; i++) { + kbuff_arr[i] = pci_alloc_consistent(instance->pdev, + ioc->sgl[i].iov_len, + &buf_handle); + if (!kbuff_arr[i]) { + printk(KERN_DEBUG "megasas: Failed to alloc " + "kernel SGL buffer for IOCTL \n"); + error = -ENOMEM; + goto out; + } + + /* + * We don't change the dma_coherent_mask, so + * pci_alloc_consistent only returns 32bit addresses + */ + kern_sge32[i].phys_addr = (u32) buf_handle; + kern_sge32[i].length = ioc->sgl[i].iov_len; + + /* + * We created a kernel buffer corresponding to the + * user buffer. Now copy in from the user buffer + */ + if (copy_from_user(kbuff_arr[i], ioc->sgl[i].iov_base, + (u32) (ioc->sgl[i].iov_len))) { + error = -EFAULT; + goto out; + } + } + + if (ioc->sense_len) { + sense = pci_alloc_consistent(instance->pdev, ioc->sense_len, + &sense_handle); + if (!sense) { + error = -ENOMEM; + goto out; + } + + sense_ptr = + (u32 *) ((unsigned long)cmd->frame + ioc->sense_off); + *sense_ptr = sense_handle; + } + + /* + * Set the sync_cmd flag so that the ISR knows not to complete this + * cmd to the SCSI mid-layer + */ + cmd->sync_cmd = 1; + megasas_issue_blocked_cmd(instance, cmd); + cmd->sync_cmd = 0; + + /* + * copy out the kernel buffers to user buffers + */ + for (i = 0; i < ioc->sge_count; i++) { + if (copy_to_user(ioc->sgl[i].iov_base, kbuff_arr[i], + ioc->sgl[i].iov_len)) { + error = -EFAULT; + goto out; + } + } + + /* + * copy out the sense + */ + if (ioc->sense_len) { + /* + * sense_ptr points to the location that has the user + * sense buffer address + */ + sense_ptr = (u32 *) ((unsigned long)ioc->frame.raw + + ioc->sense_off); + + if (copy_to_user((void __user *)((unsigned long)(*sense_ptr)), + sense, ioc->sense_len)) { + error = -EFAULT; + goto out; + } + } + + /* + * copy the status codes returned by the fw + */ + if (copy_to_user(&user_ioc->frame.hdr.cmd_status, + &cmd->frame->hdr.cmd_status, sizeof(u8))) { + printk(KERN_DEBUG "megasas: Error copying out cmd_status\n"); + error = -EFAULT; + } + + out: + if (sense) { + pci_free_consistent(instance->pdev, ioc->sense_len, + sense, sense_handle); + } + + for (i = 0; i < ioc->sge_count && kbuff_arr[i]; i++) { + pci_free_consistent(instance->pdev, + kern_sge32[i].length, + kbuff_arr[i], kern_sge32[i].phys_addr); + } + + megasas_return_cmd(instance, cmd); + return error; +} + +static struct megasas_instance *megasas_lookup_instance(u16 host_no) +{ + int i; + + for (i = 0; i < megasas_mgmt_info.max_index; i++) { + + if ((megasas_mgmt_info.instance[i]) && + (megasas_mgmt_info.instance[i]->host->host_no == host_no)) + return megasas_mgmt_info.instance[i]; + } + + return NULL; +} + +static int megasas_mgmt_ioctl_fw(struct file *file, unsigned long arg) +{ + struct megasas_iocpacket __user *user_ioc = + (struct megasas_iocpacket __user *)arg; + struct megasas_iocpacket *ioc; + struct megasas_instance *instance; + int error; + + ioc = kmalloc(sizeof(*ioc), GFP_KERNEL); + if (!ioc) + return -ENOMEM; + + if (copy_from_user(ioc, user_ioc, sizeof(*ioc))) { + error = -EFAULT; + goto out_kfree_ioc; + } + + instance = megasas_lookup_instance(ioc->host_no); + if (!instance) { + error = -ENODEV; + goto out_kfree_ioc; + } + + /* + * We will allow only MEGASAS_INT_CMDS number of parallel ioctl cmds + */ + if (down_interruptible(&instance->ioctl_sem)) { + error = -ERESTARTSYS; + goto out_kfree_ioc; + } + error = megasas_mgmt_fw_ioctl(instance, user_ioc, ioc); + up(&instance->ioctl_sem); + + out_kfree_ioc: + kfree(ioc); + return error; +} + +static int megasas_mgmt_ioctl_aen(struct file *file, unsigned long arg) +{ + struct megasas_instance *instance; + struct megasas_aen aen; + int error; + + if (file->private_data != file) { + printk(KERN_DEBUG "megasas: fasync_helper was not " + "called first\n"); + return -EINVAL; + } + + if (copy_from_user(&aen, (void __user *)arg, sizeof(aen))) + return -EFAULT; + + instance = megasas_lookup_instance(aen.host_no); + + if (!instance) + return -ENODEV; + + down(&instance->aen_mutex); + error = megasas_register_aen(instance, aen.seq_num, + aen.class_locale_word); + up(&instance->aen_mutex); + return error; +} + +/** + * megasas_mgmt_ioctl - char node ioctl entry point + */ +static long +megasas_mgmt_ioctl(struct file *file, unsigned int cmd, unsigned long arg) +{ + switch (cmd) { + case MEGASAS_IOC_FIRMWARE: + return megasas_mgmt_ioctl_fw(file, arg); + + case MEGASAS_IOC_GET_AEN: + return megasas_mgmt_ioctl_aen(file, arg); + } + + return -ENOTTY; +} + +#ifdef CONFIG_COMPAT +static int megasas_mgmt_compat_ioctl_fw(struct file *file, unsigned long arg) +{ + struct compat_megasas_iocpacket __user *cioc = + (struct compat_megasas_iocpacket __user *)arg; + struct megasas_iocpacket __user *ioc = + compat_alloc_user_space(sizeof(struct megasas_iocpacket)); + int i; + int error = 0; + + clear_user(ioc, sizeof(*ioc)); + + if (copy_in_user(&ioc->host_no, &cioc->host_no, sizeof(u16)) || + copy_in_user(&ioc->sgl_off, &cioc->sgl_off, sizeof(u32)) || + copy_in_user(&ioc->sense_off, &cioc->sense_off, sizeof(u32)) || + copy_in_user(&ioc->sense_len, &cioc->sense_len, sizeof(u32)) || + copy_in_user(ioc->frame.raw, cioc->frame.raw, 128) || + copy_in_user(&ioc->sge_count, &cioc->sge_count, sizeof(u32))) + return -EFAULT; + + for (i = 0; i < MAX_IOCTL_SGE; i++) { + compat_uptr_t ptr; + + if (get_user(ptr, &cioc->sgl[i].iov_base) || + put_user(compat_ptr(ptr), &ioc->sgl[i].iov_base) || + copy_in_user(&ioc->sgl[i].iov_len, + &cioc->sgl[i].iov_len, sizeof(compat_size_t))) + return -EFAULT; + } + + error = megasas_mgmt_ioctl_fw(file, (unsigned long)ioc); + + if (copy_in_user(&cioc->frame.hdr.cmd_status, + &ioc->frame.hdr.cmd_status, sizeof(u8))) { + printk(KERN_DEBUG "megasas: error copy_in_user cmd_status\n"); + return -EFAULT; + } + return error; +} + +static long +megasas_mgmt_compat_ioctl(struct file *file, unsigned int cmd, + unsigned long arg) +{ + switch (cmd) { + case MEGASAS_IOC_FIRMWARE:{ + return megasas_mgmt_compat_ioctl_fw(file, arg); + } + case MEGASAS_IOC_GET_AEN: + return megasas_mgmt_ioctl_aen(file, arg); + } + + return -ENOTTY; +} +#endif + +/* + * File operations structure for management interface + */ +static struct file_operations megasas_mgmt_fops = { + .owner = THIS_MODULE, + .open = megasas_mgmt_open, + .release = megasas_mgmt_release, + .fasync = megasas_mgmt_fasync, + .unlocked_ioctl = megasas_mgmt_ioctl, +#ifdef CONFIG_COMPAT + .compat_ioctl = megasas_mgmt_compat_ioctl, +#endif +}; + +/* + * PCI hotplug support registration structure + */ +static struct pci_driver megasas_pci_driver = { + + .name = "megaraid_sas", + .id_table = megasas_pci_table, + .probe = megasas_probe_one, + .remove = __devexit_p(megasas_detach_one), + .shutdown = megasas_shutdown, +}; + +/* + * Sysfs driver attributes + */ +static ssize_t megasas_sysfs_show_version(struct device_driver *dd, char *buf) +{ + return snprintf(buf, strlen(MEGASAS_VERSION) + 2, "%s\n", + MEGASAS_VERSION); +} + +static DRIVER_ATTR(version, S_IRUGO, megasas_sysfs_show_version, NULL); + +static ssize_t +megasas_sysfs_show_release_date(struct device_driver *dd, char *buf) +{ + return snprintf(buf, strlen(MEGASAS_RELDATE) + 2, "%s\n", + MEGASAS_RELDATE); +} + +static DRIVER_ATTR(release_date, S_IRUGO, megasas_sysfs_show_release_date, + NULL); + +/** + * megasas_init - Driver load entry point + */ +static int __init megasas_init(void) +{ + int rval; + + /* + * Announce driver version and other information + */ + printk(KERN_INFO "megasas: %s %s\n", MEGASAS_VERSION, + MEGASAS_EXT_VERSION); + + memset(&megasas_mgmt_info, 0, sizeof(megasas_mgmt_info)); + + /* + * Register character device node + */ + rval = register_chrdev(0, "megaraid_sas_ioctl", &megasas_mgmt_fops); + + if (rval < 0) { + printk(KERN_DEBUG "megasas: failed to open device node\n"); + return rval; + } + + megasas_mgmt_majorno = rval; + + /* + * Register ourselves as PCI hotplug module + */ + rval = pci_module_init(&megasas_pci_driver); + + if (rval) { + printk(KERN_DEBUG "megasas: PCI hotplug regisration failed \n"); + unregister_chrdev(megasas_mgmt_majorno, "megaraid_sas_ioctl"); + } + + driver_create_file(&megasas_pci_driver.driver, &driver_attr_version); + driver_create_file(&megasas_pci_driver.driver, + &driver_attr_release_date); + + return rval; +} + +/** + * megasas_exit - Driver unload entry point + */ +static void __exit megasas_exit(void) +{ + driver_remove_file(&megasas_pci_driver.driver, &driver_attr_version); + driver_remove_file(&megasas_pci_driver.driver, + &driver_attr_release_date); + + pci_unregister_driver(&megasas_pci_driver); + unregister_chrdev(megasas_mgmt_majorno, "megaraid_sas_ioctl"); +} + +module_init(megasas_init); +module_exit(megasas_exit); diff --git a/trunk/drivers/scsi/megaraid/megaraid_sas.h b/trunk/drivers/scsi/megaraid/megaraid_sas.h new file mode 100644 index 000000000000..eaec9d531424 --- /dev/null +++ b/trunk/drivers/scsi/megaraid/megaraid_sas.h @@ -0,0 +1,1142 @@ +/* + * + * Linux MegaRAID driver for SAS based RAID controllers + * + * Copyright (c) 2003-2005 LSI Logic Corporation. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + * FILE : megaraid_sas.h + */ + +#ifndef LSI_MEGARAID_SAS_H +#define LSI_MEGARAID_SAS_H + +/** + * MegaRAID SAS Driver meta data + */ +#define MEGASAS_VERSION "00.00.02.00-rc4" +#define MEGASAS_RELDATE "Sep 16, 2005" +#define MEGASAS_EXT_VERSION "Fri Sep 16 12:37:08 EDT 2005" + +/* + * ===================================== + * MegaRAID SAS MFI firmware definitions + * ===================================== + */ + +/* + * MFI stands for MegaRAID SAS FW Interface. This is just a moniker for + * protocol between the software and firmware. Commands are issued using + * "message frames" + */ + +/** + * FW posts its state in upper 4 bits of outbound_msg_0 register + */ +#define MFI_STATE_MASK 0xF0000000 +#define MFI_STATE_UNDEFINED 0x00000000 +#define MFI_STATE_BB_INIT 0x10000000 +#define MFI_STATE_FW_INIT 0x40000000 +#define MFI_STATE_WAIT_HANDSHAKE 0x60000000 +#define MFI_STATE_FW_INIT_2 0x70000000 +#define MFI_STATE_DEVICE_SCAN 0x80000000 +#define MFI_STATE_FLUSH_CACHE 0xA0000000 +#define MFI_STATE_READY 0xB0000000 +#define MFI_STATE_OPERATIONAL 0xC0000000 +#define MFI_STATE_FAULT 0xF0000000 + +#define MEGAMFI_FRAME_SIZE 64 + +/** + * During FW init, clear pending cmds & reset state using inbound_msg_0 + * + * ABORT : Abort all pending cmds + * READY : Move from OPERATIONAL to READY state; discard queue info + * MFIMODE : Discard (possible) low MFA posted in 64-bit mode (??) + * CLR_HANDSHAKE: FW is waiting for HANDSHAKE from BIOS or Driver + */ +#define MFI_INIT_ABORT 0x00000000 +#define MFI_INIT_READY 0x00000002 +#define MFI_INIT_MFIMODE 0x00000004 +#define MFI_INIT_CLEAR_HANDSHAKE 0x00000008 +#define MFI_RESET_FLAGS MFI_INIT_READY|MFI_INIT_MFIMODE + +/** + * MFI frame flags + */ +#define MFI_FRAME_POST_IN_REPLY_QUEUE 0x0000 +#define MFI_FRAME_DONT_POST_IN_REPLY_QUEUE 0x0001 +#define MFI_FRAME_SGL32 0x0000 +#define MFI_FRAME_SGL64 0x0002 +#define MFI_FRAME_SENSE32 0x0000 +#define MFI_FRAME_SENSE64 0x0004 +#define MFI_FRAME_DIR_NONE 0x0000 +#define MFI_FRAME_DIR_WRITE 0x0008 +#define MFI_FRAME_DIR_READ 0x0010 +#define MFI_FRAME_DIR_BOTH 0x0018 + +/** + * Definition for cmd_status + */ +#define MFI_CMD_STATUS_POLL_MODE 0xFF + +/** + * MFI command opcodes + */ +#define MFI_CMD_INIT 0x00 +#define MFI_CMD_LD_READ 0x01 +#define MFI_CMD_LD_WRITE 0x02 +#define MFI_CMD_LD_SCSI_IO 0x03 +#define MFI_CMD_PD_SCSI_IO 0x04 +#define MFI_CMD_DCMD 0x05 +#define MFI_CMD_ABORT 0x06 +#define MFI_CMD_SMP 0x07 +#define MFI_CMD_STP 0x08 + +#define MR_DCMD_CTRL_GET_INFO 0x01010000 + +#define MR_DCMD_CTRL_CACHE_FLUSH 0x01101000 +#define MR_FLUSH_CTRL_CACHE 0x01 +#define MR_FLUSH_DISK_CACHE 0x02 + +#define MR_DCMD_CTRL_SHUTDOWN 0x01050000 +#define MR_ENABLE_DRIVE_SPINDOWN 0x01 + +#define MR_DCMD_CTRL_EVENT_GET_INFO 0x01040100 +#define MR_DCMD_CTRL_EVENT_GET 0x01040300 +#define MR_DCMD_CTRL_EVENT_WAIT 0x01040500 +#define MR_DCMD_LD_GET_PROPERTIES 0x03030000 + +#define MR_DCMD_CLUSTER 0x08000000 +#define MR_DCMD_CLUSTER_RESET_ALL 0x08010100 +#define MR_DCMD_CLUSTER_RESET_LD 0x08010200 + +/** + * MFI command completion codes + */ +enum MFI_STAT { + MFI_STAT_OK = 0x00, + MFI_STAT_INVALID_CMD = 0x01, + MFI_STAT_INVALID_DCMD = 0x02, + MFI_STAT_INVALID_PARAMETER = 0x03, + MFI_STAT_INVALID_SEQUENCE_NUMBER = 0x04, + MFI_STAT_ABORT_NOT_POSSIBLE = 0x05, + MFI_STAT_APP_HOST_CODE_NOT_FOUND = 0x06, + MFI_STAT_APP_IN_USE = 0x07, + MFI_STAT_APP_NOT_INITIALIZED = 0x08, + MFI_STAT_ARRAY_INDEX_INVALID = 0x09, + MFI_STAT_ARRAY_ROW_NOT_EMPTY = 0x0a, + MFI_STAT_CONFIG_RESOURCE_CONFLICT = 0x0b, + MFI_STAT_DEVICE_NOT_FOUND = 0x0c, + MFI_STAT_DRIVE_TOO_SMALL = 0x0d, + MFI_STAT_FLASH_ALLOC_FAIL = 0x0e, + MFI_STAT_FLASH_BUSY = 0x0f, + MFI_STAT_FLASH_ERROR = 0x10, + MFI_STAT_FLASH_IMAGE_BAD = 0x11, + MFI_STAT_FLASH_IMAGE_INCOMPLETE = 0x12, + MFI_STAT_FLASH_NOT_OPEN = 0x13, + MFI_STAT_FLASH_NOT_STARTED = 0x14, + MFI_STAT_FLUSH_FAILED = 0x15, + MFI_STAT_HOST_CODE_NOT_FOUNT = 0x16, + MFI_STAT_LD_CC_IN_PROGRESS = 0x17, + MFI_STAT_LD_INIT_IN_PROGRESS = 0x18, + MFI_STAT_LD_LBA_OUT_OF_RANGE = 0x19, + MFI_STAT_LD_MAX_CONFIGURED = 0x1a, + MFI_STAT_LD_NOT_OPTIMAL = 0x1b, + MFI_STAT_LD_RBLD_IN_PROGRESS = 0x1c, + MFI_STAT_LD_RECON_IN_PROGRESS = 0x1d, + MFI_STAT_LD_WRONG_RAID_LEVEL = 0x1e, + MFI_STAT_MAX_SPARES_EXCEEDED = 0x1f, + MFI_STAT_MEMORY_NOT_AVAILABLE = 0x20, + MFI_STAT_MFC_HW_ERROR = 0x21, + MFI_STAT_NO_HW_PRESENT = 0x22, + MFI_STAT_NOT_FOUND = 0x23, + MFI_STAT_NOT_IN_ENCL = 0x24, + MFI_STAT_PD_CLEAR_IN_PROGRESS = 0x25, + MFI_STAT_PD_TYPE_WRONG = 0x26, + MFI_STAT_PR_DISABLED = 0x27, + MFI_STAT_ROW_INDEX_INVALID = 0x28, + MFI_STAT_SAS_CONFIG_INVALID_ACTION = 0x29, + MFI_STAT_SAS_CONFIG_INVALID_DATA = 0x2a, + MFI_STAT_SAS_CONFIG_INVALID_PAGE = 0x2b, + MFI_STAT_SAS_CONFIG_INVALID_TYPE = 0x2c, + MFI_STAT_SCSI_DONE_WITH_ERROR = 0x2d, + MFI_STAT_SCSI_IO_FAILED = 0x2e, + MFI_STAT_SCSI_RESERVATION_CONFLICT = 0x2f, + MFI_STAT_SHUTDOWN_FAILED = 0x30, + MFI_STAT_TIME_NOT_SET = 0x31, + MFI_STAT_WRONG_STATE = 0x32, + MFI_STAT_LD_OFFLINE = 0x33, + MFI_STAT_PEER_NOTIFICATION_REJECTED = 0x34, + MFI_STAT_PEER_NOTIFICATION_FAILED = 0x35, + MFI_STAT_RESERVATION_IN_PROGRESS = 0x36, + MFI_STAT_I2C_ERRORS_DETECTED = 0x37, + MFI_STAT_PCI_ERRORS_DETECTED = 0x38, + + MFI_STAT_INVALID_STATUS = 0xFF +}; + +/* + * Number of mailbox bytes in DCMD message frame + */ +#define MFI_MBOX_SIZE 12 + +enum MR_EVT_CLASS { + + MR_EVT_CLASS_DEBUG = -2, + MR_EVT_CLASS_PROGRESS = -1, + MR_EVT_CLASS_INFO = 0, + MR_EVT_CLASS_WARNING = 1, + MR_EVT_CLASS_CRITICAL = 2, + MR_EVT_CLASS_FATAL = 3, + MR_EVT_CLASS_DEAD = 4, + +}; + +enum MR_EVT_LOCALE { + + MR_EVT_LOCALE_LD = 0x0001, + MR_EVT_LOCALE_PD = 0x0002, + MR_EVT_LOCALE_ENCL = 0x0004, + MR_EVT_LOCALE_BBU = 0x0008, + MR_EVT_LOCALE_SAS = 0x0010, + MR_EVT_LOCALE_CTRL = 0x0020, + MR_EVT_LOCALE_CONFIG = 0x0040, + MR_EVT_LOCALE_CLUSTER = 0x0080, + MR_EVT_LOCALE_ALL = 0xffff, + +}; + +enum MR_EVT_ARGS { + + MR_EVT_ARGS_NONE, + MR_EVT_ARGS_CDB_SENSE, + MR_EVT_ARGS_LD, + MR_EVT_ARGS_LD_COUNT, + MR_EVT_ARGS_LD_LBA, + MR_EVT_ARGS_LD_OWNER, + MR_EVT_ARGS_LD_LBA_PD_LBA, + MR_EVT_ARGS_LD_PROG, + MR_EVT_ARGS_LD_STATE, + MR_EVT_ARGS_LD_STRIP, + MR_EVT_ARGS_PD, + MR_EVT_ARGS_PD_ERR, + MR_EVT_ARGS_PD_LBA, + MR_EVT_ARGS_PD_LBA_LD, + MR_EVT_ARGS_PD_PROG, + MR_EVT_ARGS_PD_STATE, + MR_EVT_ARGS_PCI, + MR_EVT_ARGS_RATE, + MR_EVT_ARGS_STR, + MR_EVT_ARGS_TIME, + MR_EVT_ARGS_ECC, + +}; + +/* + * SAS controller properties + */ +struct megasas_ctrl_prop { + + u16 seq_num; + u16 pred_fail_poll_interval; + u16 intr_throttle_count; + u16 intr_throttle_timeouts; + u8 rebuild_rate; + u8 patrol_read_rate; + u8 bgi_rate; + u8 cc_rate; + u8 recon_rate; + u8 cache_flush_interval; + u8 spinup_drv_count; + u8 spinup_delay; + u8 cluster_enable; + u8 coercion_mode; + u8 alarm_enable; + u8 disable_auto_rebuild; + u8 disable_battery_warn; + u8 ecc_bucket_size; + u16 ecc_bucket_leak_rate; + u8 restore_hotspare_on_insertion; + u8 expose_encl_devices; + u8 reserved[38]; + +} __attribute__ ((packed)); + +/* + * SAS controller information + */ +struct megasas_ctrl_info { + + /* + * PCI device information + */ + struct { + + u16 vendor_id; + u16 device_id; + u16 sub_vendor_id; + u16 sub_device_id; + u8 reserved[24]; + + } __attribute__ ((packed)) pci; + + /* + * Host interface information + */ + struct { + + u8 PCIX:1; + u8 PCIE:1; + u8 iSCSI:1; + u8 SAS_3G:1; + u8 reserved_0:4; + u8 reserved_1[6]; + u8 port_count; + u64 port_addr[8]; + + } __attribute__ ((packed)) host_interface; + + /* + * Device (backend) interface information + */ + struct { + + u8 SPI:1; + u8 SAS_3G:1; + u8 SATA_1_5G:1; + u8 SATA_3G:1; + u8 reserved_0:4; + u8 reserved_1[6]; + u8 port_count; + u64 port_addr[8]; + + } __attribute__ ((packed)) device_interface; + + /* + * List of components residing in flash. All str are null terminated + */ + u32 image_check_word; + u32 image_component_count; + + struct { + + char name[8]; + char version[32]; + char build_date[16]; + char built_time[16]; + + } __attribute__ ((packed)) image_component[8]; + + /* + * List of flash components that have been flashed on the card, but + * are not in use, pending reset of the adapter. This list will be + * empty if a flash operation has not occurred. All stings are null + * terminated + */ + u32 pending_image_component_count; + + struct { + + char name[8]; + char version[32]; + char build_date[16]; + char build_time[16]; + + } __attribute__ ((packed)) pending_image_component[8]; + + u8 max_arms; + u8 max_spans; + u8 max_arrays; + u8 max_lds; + + char product_name[80]; + char serial_no[32]; + + /* + * Other physical/controller/operation information. Indicates the + * presence of the hardware + */ + struct { + + u32 bbu:1; + u32 alarm:1; + u32 nvram:1; + u32 uart:1; + u32 reserved:28; + + } __attribute__ ((packed)) hw_present; + + u32 current_fw_time; + + /* + * Maximum data transfer sizes + */ + u16 max_concurrent_cmds; + u16 max_sge_count; + u32 max_request_size; + + /* + * Logical and physical device counts + */ + u16 ld_present_count; + u16 ld_degraded_count; + u16 ld_offline_count; + + u16 pd_present_count; + u16 pd_disk_present_count; + u16 pd_disk_pred_failure_count; + u16 pd_disk_failed_count; + + /* + * Memory size information + */ + u16 nvram_size; + u16 memory_size; + u16 flash_size; + + /* + * Error counters + */ + u16 mem_correctable_error_count; + u16 mem_uncorrectable_error_count; + + /* + * Cluster information + */ + u8 cluster_permitted; + u8 cluster_active; + + /* + * Additional max data transfer sizes + */ + u16 max_strips_per_io; + + /* + * Controller capabilities structures + */ + struct { + + u32 raid_level_0:1; + u32 raid_level_1:1; + u32 raid_level_5:1; + u32 raid_level_1E:1; + u32 raid_level_6:1; + u32 reserved:27; + + } __attribute__ ((packed)) raid_levels; + + struct { + + u32 rbld_rate:1; + u32 cc_rate:1; + u32 bgi_rate:1; + u32 recon_rate:1; + u32 patrol_rate:1; + u32 alarm_control:1; + u32 cluster_supported:1; + u32 bbu:1; + u32 spanning_allowed:1; + u32 dedicated_hotspares:1; + u32 revertible_hotspares:1; + u32 foreign_config_import:1; + u32 self_diagnostic:1; + u32 mixed_redundancy_arr:1; + u32 global_hot_spares:1; + u32 reserved:17; + + } __attribute__ ((packed)) adapter_operations; + + struct { + + u32 read_policy:1; + u32 write_policy:1; + u32 io_policy:1; + u32 access_policy:1; + u32 disk_cache_policy:1; + u32 reserved:27; + + } __attribute__ ((packed)) ld_operations; + + struct { + + u8 min; + u8 max; + u8 reserved[2]; + + } __attribute__ ((packed)) stripe_sz_ops; + + struct { + + u32 force_online:1; + u32 force_offline:1; + u32 force_rebuild:1; + u32 reserved:29; + + } __attribute__ ((packed)) pd_operations; + + struct { + + u32 ctrl_supports_sas:1; + u32 ctrl_supports_sata:1; + u32 allow_mix_in_encl:1; + u32 allow_mix_in_ld:1; + u32 allow_sata_in_cluster:1; + u32 reserved:27; + + } __attribute__ ((packed)) pd_mix_support; + + /* + * Define ECC single-bit-error bucket information + */ + u8 ecc_bucket_count; + u8 reserved_2[11]; + + /* + * Include the controller properties (changeable items) + */ + struct megasas_ctrl_prop properties; + + /* + * Define FW pkg version (set in envt v'bles on OEM basis) + */ + char package_version[0x60]; + + u8 pad[0x800 - 0x6a0]; + +} __attribute__ ((packed)); + +/* + * =============================== + * MegaRAID SAS driver definitions + * =============================== + */ +#define MEGASAS_MAX_PD_CHANNELS 2 +#define MEGASAS_MAX_LD_CHANNELS 2 +#define MEGASAS_MAX_CHANNELS (MEGASAS_MAX_PD_CHANNELS + \ + MEGASAS_MAX_LD_CHANNELS) +#define MEGASAS_MAX_DEV_PER_CHANNEL 128 +#define MEGASAS_DEFAULT_INIT_ID -1 +#define MEGASAS_MAX_LUN 8 +#define MEGASAS_MAX_LD 64 + +/* + * When SCSI mid-layer calls driver's reset routine, driver waits for + * MEGASAS_RESET_WAIT_TIME seconds for all outstanding IO to complete. Note + * that the driver cannot _actually_ abort or reset pending commands. While + * it is waiting for the commands to complete, it prints a diagnostic message + * every MEGASAS_RESET_NOTICE_INTERVAL seconds + */ +#define MEGASAS_RESET_WAIT_TIME 180 +#define MEGASAS_RESET_NOTICE_INTERVAL 5 + +#define MEGASAS_IOCTL_CMD 0 + +/* + * FW reports the maximum of number of commands that it can accept (maximum + * commands that can be outstanding) at any time. The driver must report a + * lower number to the mid layer because it can issue a few internal commands + * itself (E.g, AEN, abort cmd, IOCTLs etc). The number of commands it needs + * is shown below + */ +#define MEGASAS_INT_CMDS 32 + +/* + * FW can accept both 32 and 64 bit SGLs. We want to allocate 32/64 bit + * SGLs based on the size of dma_addr_t + */ +#define IS_DMA64 (sizeof(dma_addr_t) == 8) + +#define MFI_OB_INTR_STATUS_MASK 0x00000002 +#define MFI_POLL_TIMEOUT_SECS 10 + +struct megasas_register_set { + + u32 reserved_0[4]; /*0000h */ + + u32 inbound_msg_0; /*0010h */ + u32 inbound_msg_1; /*0014h */ + u32 outbound_msg_0; /*0018h */ + u32 outbound_msg_1; /*001Ch */ + + u32 inbound_doorbell; /*0020h */ + u32 inbound_intr_status; /*0024h */ + u32 inbound_intr_mask; /*0028h */ + + u32 outbound_doorbell; /*002Ch */ + u32 outbound_intr_status; /*0030h */ + u32 outbound_intr_mask; /*0034h */ + + u32 reserved_1[2]; /*0038h */ + + u32 inbound_queue_port; /*0040h */ + u32 outbound_queue_port; /*0044h */ + + u32 reserved_2; /*004Ch */ + + u32 index_registers[1004]; /*0050h */ + +} __attribute__ ((packed)); + +struct megasas_sge32 { + + u32 phys_addr; + u32 length; + +} __attribute__ ((packed)); + +struct megasas_sge64 { + + u64 phys_addr; + u32 length; + +} __attribute__ ((packed)); + +union megasas_sgl { + + struct megasas_sge32 sge32[1]; + struct megasas_sge64 sge64[1]; + +} __attribute__ ((packed)); + +struct megasas_header { + + u8 cmd; /*00h */ + u8 sense_len; /*01h */ + u8 cmd_status; /*02h */ + u8 scsi_status; /*03h */ + + u8 target_id; /*04h */ + u8 lun; /*05h */ + u8 cdb_len; /*06h */ + u8 sge_count; /*07h */ + + u32 context; /*08h */ + u32 pad_0; /*0Ch */ + + u16 flags; /*10h */ + u16 timeout; /*12h */ + u32 data_xferlen; /*14h */ + +} __attribute__ ((packed)); + +union megasas_sgl_frame { + + struct megasas_sge32 sge32[8]; + struct megasas_sge64 sge64[5]; + +} __attribute__ ((packed)); + +struct megasas_init_frame { + + u8 cmd; /*00h */ + u8 reserved_0; /*01h */ + u8 cmd_status; /*02h */ + + u8 reserved_1; /*03h */ + u32 reserved_2; /*04h */ + + u32 context; /*08h */ + u32 pad_0; /*0Ch */ + + u16 flags; /*10h */ + u16 reserved_3; /*12h */ + u32 data_xfer_len; /*14h */ + + u32 queue_info_new_phys_addr_lo; /*18h */ + u32 queue_info_new_phys_addr_hi; /*1Ch */ + u32 queue_info_old_phys_addr_lo; /*20h */ + u32 queue_info_old_phys_addr_hi; /*24h */ + + u32 reserved_4[6]; /*28h */ + +} __attribute__ ((packed)); + +struct megasas_init_queue_info { + + u32 init_flags; /*00h */ + u32 reply_queue_entries; /*04h */ + + u32 reply_queue_start_phys_addr_lo; /*08h */ + u32 reply_queue_start_phys_addr_hi; /*0Ch */ + u32 producer_index_phys_addr_lo; /*10h */ + u32 producer_index_phys_addr_hi; /*14h */ + u32 consumer_index_phys_addr_lo; /*18h */ + u32 consumer_index_phys_addr_hi; /*1Ch */ + +} __attribute__ ((packed)); + +struct megasas_io_frame { + + u8 cmd; /*00h */ + u8 sense_len; /*01h */ + u8 cmd_status; /*02h */ + u8 scsi_status; /*03h */ + + u8 target_id; /*04h */ + u8 access_byte; /*05h */ + u8 reserved_0; /*06h */ + u8 sge_count; /*07h */ + + u32 context; /*08h */ + u32 pad_0; /*0Ch */ + + u16 flags; /*10h */ + u16 timeout; /*12h */ + u32 lba_count; /*14h */ + + u32 sense_buf_phys_addr_lo; /*18h */ + u32 sense_buf_phys_addr_hi; /*1Ch */ + + u32 start_lba_lo; /*20h */ + u32 start_lba_hi; /*24h */ + + union megasas_sgl sgl; /*28h */ + +} __attribute__ ((packed)); + +struct megasas_pthru_frame { + + u8 cmd; /*00h */ + u8 sense_len; /*01h */ + u8 cmd_status; /*02h */ + u8 scsi_status; /*03h */ + + u8 target_id; /*04h */ + u8 lun; /*05h */ + u8 cdb_len; /*06h */ + u8 sge_count; /*07h */ + + u32 context; /*08h */ + u32 pad_0; /*0Ch */ + + u16 flags; /*10h */ + u16 timeout; /*12h */ + u32 data_xfer_len; /*14h */ + + u32 sense_buf_phys_addr_lo; /*18h */ + u32 sense_buf_phys_addr_hi; /*1Ch */ + + u8 cdb[16]; /*20h */ + union megasas_sgl sgl; /*30h */ + +} __attribute__ ((packed)); + +struct megasas_dcmd_frame { + + u8 cmd; /*00h */ + u8 reserved_0; /*01h */ + u8 cmd_status; /*02h */ + u8 reserved_1[4]; /*03h */ + u8 sge_count; /*07h */ + + u32 context; /*08h */ + u32 pad_0; /*0Ch */ + + u16 flags; /*10h */ + u16 timeout; /*12h */ + + u32 data_xfer_len; /*14h */ + u32 opcode; /*18h */ + + union { /*1Ch */ + u8 b[12]; + u16 s[6]; + u32 w[3]; + } mbox; + + union megasas_sgl sgl; /*28h */ + +} __attribute__ ((packed)); + +struct megasas_abort_frame { + + u8 cmd; /*00h */ + u8 reserved_0; /*01h */ + u8 cmd_status; /*02h */ + + u8 reserved_1; /*03h */ + u32 reserved_2; /*04h */ + + u32 context; /*08h */ + u32 pad_0; /*0Ch */ + + u16 flags; /*10h */ + u16 reserved_3; /*12h */ + u32 reserved_4; /*14h */ + + u32 abort_context; /*18h */ + u32 pad_1; /*1Ch */ + + u32 abort_mfi_phys_addr_lo; /*20h */ + u32 abort_mfi_phys_addr_hi; /*24h */ + + u32 reserved_5[6]; /*28h */ + +} __attribute__ ((packed)); + +struct megasas_smp_frame { + + u8 cmd; /*00h */ + u8 reserved_1; /*01h */ + u8 cmd_status; /*02h */ + u8 connection_status; /*03h */ + + u8 reserved_2[3]; /*04h */ + u8 sge_count; /*07h */ + + u32 context; /*08h */ + u32 pad_0; /*0Ch */ + + u16 flags; /*10h */ + u16 timeout; /*12h */ + + u32 data_xfer_len; /*14h */ + u64 sas_addr; /*18h */ + + union { + struct megasas_sge32 sge32[2]; /* [0]: resp [1]: req */ + struct megasas_sge64 sge64[2]; /* [0]: resp [1]: req */ + } sgl; + +} __attribute__ ((packed)); + +struct megasas_stp_frame { + + u8 cmd; /*00h */ + u8 reserved_1; /*01h */ + u8 cmd_status; /*02h */ + u8 reserved_2; /*03h */ + + u8 target_id; /*04h */ + u8 reserved_3[2]; /*05h */ + u8 sge_count; /*07h */ + + u32 context; /*08h */ + u32 pad_0; /*0Ch */ + + u16 flags; /*10h */ + u16 timeout; /*12h */ + + u32 data_xfer_len; /*14h */ + + u16 fis[10]; /*18h */ + u32 stp_flags; + + union { + struct megasas_sge32 sge32[2]; /* [0]: resp [1]: data */ + struct megasas_sge64 sge64[2]; /* [0]: resp [1]: data */ + } sgl; + +} __attribute__ ((packed)); + +union megasas_frame { + + struct megasas_header hdr; + struct megasas_init_frame init; + struct megasas_io_frame io; + struct megasas_pthru_frame pthru; + struct megasas_dcmd_frame dcmd; + struct megasas_abort_frame abort; + struct megasas_smp_frame smp; + struct megasas_stp_frame stp; + + u8 raw_bytes[64]; +}; + +struct megasas_cmd; + +union megasas_evt_class_locale { + + struct { + u16 locale; + u8 reserved; + s8 class; + } __attribute__ ((packed)) members; + + u32 word; + +} __attribute__ ((packed)); + +struct megasas_evt_log_info { + u32 newest_seq_num; + u32 oldest_seq_num; + u32 clear_seq_num; + u32 shutdown_seq_num; + u32 boot_seq_num; + +} __attribute__ ((packed)); + +struct megasas_progress { + + u16 progress; + u16 elapsed_seconds; + +} __attribute__ ((packed)); + +struct megasas_evtarg_ld { + + u16 target_id; + u8 ld_index; + u8 reserved; + +} __attribute__ ((packed)); + +struct megasas_evtarg_pd { + u16 device_id; + u8 encl_index; + u8 slot_number; + +} __attribute__ ((packed)); + +struct megasas_evt_detail { + + u32 seq_num; + u32 time_stamp; + u32 code; + union megasas_evt_class_locale cl; + u8 arg_type; + u8 reserved1[15]; + + union { + struct { + struct megasas_evtarg_pd pd; + u8 cdb_length; + u8 sense_length; + u8 reserved[2]; + u8 cdb[16]; + u8 sense[64]; + } __attribute__ ((packed)) cdbSense; + + struct megasas_evtarg_ld ld; + + struct { + struct megasas_evtarg_ld ld; + u64 count; + } __attribute__ ((packed)) ld_count; + + struct { + u64 lba; + struct megasas_evtarg_ld ld; + } __attribute__ ((packed)) ld_lba; + + struct { + struct megasas_evtarg_ld ld; + u32 prevOwner; + u32 newOwner; + } __attribute__ ((packed)) ld_owner; + + struct { + u64 ld_lba; + u64 pd_lba; + struct megasas_evtarg_ld ld; + struct megasas_evtarg_pd pd; + } __attribute__ ((packed)) ld_lba_pd_lba; + + struct { + struct megasas_evtarg_ld ld; + struct megasas_progress prog; + } __attribute__ ((packed)) ld_prog; + + struct { + struct megasas_evtarg_ld ld; + u32 prev_state; + u32 new_state; + } __attribute__ ((packed)) ld_state; + + struct { + u64 strip; + struct megasas_evtarg_ld ld; + } __attribute__ ((packed)) ld_strip; + + struct megasas_evtarg_pd pd; + + struct { + struct megasas_evtarg_pd pd; + u32 err; + } __attribute__ ((packed)) pd_err; + + struct { + u64 lba; + struct megasas_evtarg_pd pd; + } __attribute__ ((packed)) pd_lba; + + struct { + u64 lba; + struct megasas_evtarg_pd pd; + struct megasas_evtarg_ld ld; + } __attribute__ ((packed)) pd_lba_ld; + + struct { + struct megasas_evtarg_pd pd; + struct megasas_progress prog; + } __attribute__ ((packed)) pd_prog; + + struct { + struct megasas_evtarg_pd pd; + u32 prevState; + u32 newState; + } __attribute__ ((packed)) pd_state; + + struct { + u16 vendorId; + u16 deviceId; + u16 subVendorId; + u16 subDeviceId; + } __attribute__ ((packed)) pci; + + u32 rate; + char str[96]; + + struct { + u32 rtc; + u32 elapsedSeconds; + } __attribute__ ((packed)) time; + + struct { + u32 ecar; + u32 elog; + char str[64]; + } __attribute__ ((packed)) ecc; + + u8 b[96]; + u16 s[48]; + u32 w[24]; + u64 d[12]; + } args; + + char description[128]; + +} __attribute__ ((packed)); + +struct megasas_instance { + + u32 *producer; + dma_addr_t producer_h; + u32 *consumer; + dma_addr_t consumer_h; + + u32 *reply_queue; + dma_addr_t reply_queue_h; + + unsigned long base_addr; + struct megasas_register_set __iomem *reg_set; + + s8 init_id; + u8 reserved[3]; + + u16 max_num_sge; + u16 max_fw_cmds; + u32 max_sectors_per_req; + + struct megasas_cmd **cmd_list; + struct list_head cmd_pool; + spinlock_t cmd_pool_lock; + struct dma_pool *frame_dma_pool; + struct dma_pool *sense_dma_pool; + + struct megasas_evt_detail *evt_detail; + dma_addr_t evt_detail_h; + struct megasas_cmd *aen_cmd; + struct semaphore aen_mutex; + struct semaphore ioctl_sem; + + struct Scsi_Host *host; + + wait_queue_head_t int_cmd_wait_q; + wait_queue_head_t abort_cmd_wait_q; + + struct pci_dev *pdev; + u32 unique_id; + + u32 fw_outstanding; + u32 hw_crit_error; + spinlock_t instance_lock; +}; + +#define MEGASAS_IS_LOGICAL(scp) \ + (scp->device->channel < MEGASAS_MAX_PD_CHANNELS) ? 0 : 1 + +#define MEGASAS_DEV_INDEX(inst, scp) \ + ((scp->device->channel % 2) * MEGASAS_MAX_DEV_PER_CHANNEL) + \ + scp->device->id + +struct megasas_cmd { + + union megasas_frame *frame; + dma_addr_t frame_phys_addr; + u8 *sense; + dma_addr_t sense_phys_addr; + + u32 index; + u8 sync_cmd; + u8 cmd_status; + u16 abort_aen; + + struct list_head list; + struct scsi_cmnd *scmd; + struct megasas_instance *instance; + u32 frame_count; +}; + +#define MAX_MGMT_ADAPTERS 1024 +#define MAX_IOCTL_SGE 16 + +struct megasas_iocpacket { + + u16 host_no; + u16 __pad1; + u32 sgl_off; + u32 sge_count; + u32 sense_off; + u32 sense_len; + union { + u8 raw[128]; + struct megasas_header hdr; + } frame; + + struct iovec sgl[MAX_IOCTL_SGE]; + +} __attribute__ ((packed)); + +struct megasas_aen { + u16 host_no; + u16 __pad1; + u32 seq_num; + u32 class_locale_word; +} __attribute__ ((packed)); + +#ifdef CONFIG_COMPAT +struct compat_megasas_iocpacket { + u16 host_no; + u16 __pad1; + u32 sgl_off; + u32 sge_count; + u32 sense_off; + u32 sense_len; + union { + u8 raw[128]; + struct megasas_header hdr; + } frame; + struct compat_iovec sgl[MAX_IOCTL_SGE]; +} __attribute__ ((packed)); + +#define MEGASAS_IOC_FIRMWARE _IOWR('M', 1, struct compat_megasas_iocpacket) +#else +#define MEGASAS_IOC_FIRMWARE _IOWR('M', 1, struct megasas_iocpacket) +#endif + +#define MEGASAS_IOC_GET_AEN _IOW('M', 3, struct megasas_aen) + +struct megasas_mgmt_info { + + u16 count; + struct megasas_instance *instance[MAX_MGMT_ADAPTERS]; + int max_index; +}; + +#endif /*LSI_MEGARAID_SAS_H */ diff --git a/trunk/drivers/scsi/qla2xxx/qla_rscn.c b/trunk/drivers/scsi/qla2xxx/qla_rscn.c index bdc3bc74bbe1..1eba98828636 100644 --- a/trunk/drivers/scsi/qla2xxx/qla_rscn.c +++ b/trunk/drivers/scsi/qla2xxx/qla_rscn.c @@ -330,6 +330,8 @@ qla2x00_update_login_fcport(scsi_qla_host_t *ha, struct mbx_entry *mbxstat, fcport->flags &= ~FCF_FAILOVER_NEEDED; fcport->iodesc_idx_sent = IODESC_INVALID_INDEX; atomic_set(&fcport->state, FCS_ONLINE); + if (fcport->rport) + fc_remote_port_unblock(fcport->rport); } diff --git a/trunk/drivers/scsi/sata_mv.c b/trunk/drivers/scsi/sata_mv.c index d457f5673476..ea76fe44585e 100644 --- a/trunk/drivers/scsi/sata_mv.c +++ b/trunk/drivers/scsi/sata_mv.c @@ -35,7 +35,7 @@ #include #define DRV_NAME "sata_mv" -#define DRV_VERSION "0.24" +#define DRV_VERSION "0.12" enum { /* BAR's are enumerated in terms of pci_resource_start() terms */ @@ -55,61 +55,31 @@ enum { MV_SATAHC_ARBTR_REG_SZ = MV_MINOR_REG_AREA_SZ, /* arbiter */ MV_PORT_REG_SZ = MV_MINOR_REG_AREA_SZ, - MV_USE_Q_DEPTH = ATA_DEF_QUEUE, + MV_Q_CT = 32, + MV_CRQB_SZ = 32, + MV_CRPB_SZ = 8, - MV_MAX_Q_DEPTH = 32, - MV_MAX_Q_DEPTH_MASK = MV_MAX_Q_DEPTH - 1, - - /* CRQB needs alignment on a 1KB boundary. Size == 1KB - * CRPB needs alignment on a 256B boundary. Size == 256B - * SG count of 176 leads to MV_PORT_PRIV_DMA_SZ == 4KB - * ePRD (SG) entries need alignment on a 16B boundary. Size == 16B - */ - MV_CRQB_Q_SZ = (32 * MV_MAX_Q_DEPTH), - MV_CRPB_Q_SZ = (8 * MV_MAX_Q_DEPTH), - MV_MAX_SG_CT = 176, - MV_SG_TBL_SZ = (16 * MV_MAX_SG_CT), - MV_PORT_PRIV_DMA_SZ = (MV_CRQB_Q_SZ + MV_CRPB_Q_SZ + MV_SG_TBL_SZ), - - /* Our DMA boundary is determined by an ePRD being unable to handle - * anything larger than 64KB - */ - MV_DMA_BOUNDARY = 0xffffU, + MV_DMA_BOUNDARY = 0xffffffffU, + SATAHC_MASK = (~(MV_SATAHC_REG_SZ - 1)), MV_PORTS_PER_HC = 4, /* == (port / MV_PORTS_PER_HC) to determine HC from 0-7 port */ MV_PORT_HC_SHIFT = 2, - /* == (port % MV_PORTS_PER_HC) to determine hard port from 0-7 port */ + /* == (port % MV_PORTS_PER_HC) to determine port from 0-7 port */ MV_PORT_MASK = 3, /* Host Flags */ MV_FLAG_DUAL_HC = (1 << 30), /* two SATA Host Controllers */ MV_FLAG_IRQ_COALESCE = (1 << 29), /* IRQ coalescing capability */ - MV_FLAG_GLBL_SFT_RST = (1 << 28), /* Global Soft Reset support */ - MV_COMMON_FLAGS = (ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY | - ATA_FLAG_SATA_RESET | ATA_FLAG_MMIO), - MV_6XXX_FLAGS = (MV_FLAG_IRQ_COALESCE | - MV_FLAG_GLBL_SFT_RST), + MV_FLAG_BDMA = (1 << 28), /* Basic DMA */ chip_504x = 0, chip_508x = 1, chip_604x = 2, chip_608x = 3, - CRQB_FLAG_READ = (1 << 0), - CRQB_TAG_SHIFT = 1, - CRQB_CMD_ADDR_SHIFT = 8, - CRQB_CMD_CS = (0x2 << 11), - CRQB_CMD_LAST = (1 << 15), - - CRPB_FLAG_STATUS_SHIFT = 8, - - EPRD_FLAG_END_OF_TBL = (1 << 31), - /* PCI interface registers */ - PCI_COMMAND_OFS = 0xc00, - PCI_MAIN_CMD_STS_OFS = 0xd30, STOP_PCI_MASTER = (1 << 2), PCI_MASTER_EMPTY = (1 << 3), @@ -141,13 +111,20 @@ enum { HC_CFG_OFS = 0, HC_IRQ_CAUSE_OFS = 0x14, - CRPB_DMA_DONE = (1 << 0), /* shift by port # */ + CRBP_DMA_DONE = (1 << 0), /* shift by port # */ HC_IRQ_COAL = (1 << 4), /* IRQ coalescing */ DEV_IRQ = (1 << 8), /* shift by port # */ /* Shadow block registers */ - SHD_BLK_OFS = 0x100, - SHD_CTL_AST_OFS = 0x20, /* ofs from SHD_BLK_OFS */ + SHD_PIO_DATA_OFS = 0x100, + SHD_FEA_ERR_OFS = 0x104, + SHD_SECT_CNT_OFS = 0x108, + SHD_LBA_L_OFS = 0x10C, + SHD_LBA_M_OFS = 0x110, + SHD_LBA_H_OFS = 0x114, + SHD_DEV_HD_OFS = 0x118, + SHD_CMD_STA_OFS = 0x11C, + SHD_CTL_AST_OFS = 0x120, /* SATA registers */ SATA_STATUS_OFS = 0x300, /* ctrl, err regs follow status */ @@ -155,11 +132,6 @@ enum { /* Port registers */ EDMA_CFG_OFS = 0, - EDMA_CFG_Q_DEPTH = 0, /* queueing disabled */ - EDMA_CFG_NCQ = (1 << 5), - EDMA_CFG_NCQ_GO_ON_ERR = (1 << 14), /* continue on error */ - EDMA_CFG_RD_BRST_EXT = (1 << 11), /* read burst 512B */ - EDMA_CFG_WR_BUFF_LEN = (1 << 13), /* write buffer 512B */ EDMA_ERR_IRQ_CAUSE_OFS = 0x8, EDMA_ERR_IRQ_MASK_OFS = 0xc, @@ -189,85 +161,33 @@ enum { EDMA_ERR_LNK_DATA_TX | EDMA_ERR_TRANS_PROTO), - EDMA_REQ_Q_BASE_HI_OFS = 0x10, - EDMA_REQ_Q_IN_PTR_OFS = 0x14, /* also contains BASE_LO */ - EDMA_REQ_Q_BASE_LO_MASK = 0xfffffc00U, - - EDMA_REQ_Q_OUT_PTR_OFS = 0x18, - EDMA_REQ_Q_PTR_SHIFT = 5, - - EDMA_RSP_Q_BASE_HI_OFS = 0x1c, - EDMA_RSP_Q_IN_PTR_OFS = 0x20, - EDMA_RSP_Q_OUT_PTR_OFS = 0x24, /* also contains BASE_LO */ - EDMA_RSP_Q_BASE_LO_MASK = 0xffffff00U, - EDMA_RSP_Q_PTR_SHIFT = 3, - EDMA_CMD_OFS = 0x28, EDMA_EN = (1 << 0), EDMA_DS = (1 << 1), ATA_RST = (1 << 2), - /* Host private flags (hp_flags) */ - MV_HP_FLAG_MSI = (1 << 0), - - /* Port private flags (pp_flags) */ - MV_PP_FLAG_EDMA_EN = (1 << 0), - MV_PP_FLAG_EDMA_DS_ACT = (1 << 1), -}; - -/* Command ReQuest Block: 32B */ -struct mv_crqb { - u32 sg_addr; - u32 sg_addr_hi; - u16 ctrl_flags; - u16 ata_cmd[11]; -}; - -/* Command ResPonse Block: 8B */ -struct mv_crpb { - u16 id; - u16 flags; - u32 tmstmp; -}; + /* BDMA is 6xxx part only */ + BDMA_CMD_OFS = 0x224, + BDMA_START = (1 << 0), -/* EDMA Physical Region Descriptor (ePRD); A.K.A. SG */ -struct mv_sg { - u32 addr; - u32 flags_size; - u32 addr_hi; - u32 reserved; + MV_UNDEF = 0, }; struct mv_port_priv { - struct mv_crqb *crqb; - dma_addr_t crqb_dma; - struct mv_crpb *crpb; - dma_addr_t crpb_dma; - struct mv_sg *sg_tbl; - dma_addr_t sg_tbl_dma; - - unsigned req_producer; /* cp of req_in_ptr */ - unsigned rsp_consumer; /* cp of rsp_out_ptr */ - u32 pp_flags; + }; struct mv_host_priv { - u32 hp_flags; + }; static void mv_irq_clear(struct ata_port *ap); static u32 mv_scr_read(struct ata_port *ap, unsigned int sc_reg_in); static void mv_scr_write(struct ata_port *ap, unsigned int sc_reg_in, u32 val); -static u8 mv_check_err(struct ata_port *ap); static void mv_phy_reset(struct ata_port *ap); -static void mv_host_stop(struct ata_host_set *host_set); -static int mv_port_start(struct ata_port *ap); -static void mv_port_stop(struct ata_port *ap); -static void mv_qc_prep(struct ata_queued_cmd *qc); -static int mv_qc_issue(struct ata_queued_cmd *qc); +static int mv_master_reset(void __iomem *mmio_base); static irqreturn_t mv_interrupt(int irq, void *dev_instance, struct pt_regs *regs); -static void mv_eng_timeout(struct ata_port *ap); static int mv_init_one(struct pci_dev *pdev, const struct pci_device_id *ent); static Scsi_Host_Template mv_sht = { @@ -276,13 +196,13 @@ static Scsi_Host_Template mv_sht = { .ioctl = ata_scsi_ioctl, .queuecommand = ata_scsi_queuecmd, .eh_strategy_handler = ata_scsi_error, - .can_queue = MV_USE_Q_DEPTH, + .can_queue = ATA_DEF_QUEUE, .this_id = ATA_SHT_THIS_ID, - .sg_tablesize = MV_MAX_SG_CT, + .sg_tablesize = MV_UNDEF, .max_sectors = ATA_MAX_SECTORS, .cmd_per_lun = ATA_SHT_CMD_PER_LUN, .emulated = ATA_SHT_EMULATED, - .use_clustering = ATA_SHT_USE_CLUSTERING, + .use_clustering = MV_UNDEF, .proc_name = DRV_NAME, .dma_boundary = MV_DMA_BOUNDARY, .slave_configure = ata_scsi_slave_config, @@ -296,16 +216,15 @@ static struct ata_port_operations mv_ops = { .tf_load = ata_tf_load, .tf_read = ata_tf_read, .check_status = ata_check_status, - .check_err = mv_check_err, .exec_command = ata_exec_command, .dev_select = ata_std_dev_select, .phy_reset = mv_phy_reset, - .qc_prep = mv_qc_prep, - .qc_issue = mv_qc_issue, + .qc_prep = ata_qc_prep, + .qc_issue = ata_qc_issue_prot, - .eng_timeout = mv_eng_timeout, + .eng_timeout = ata_eng_timeout, .irq_handler = mv_interrupt, .irq_clear = mv_irq_clear, @@ -313,39 +232,46 @@ static struct ata_port_operations mv_ops = { .scr_read = mv_scr_read, .scr_write = mv_scr_write, - .port_start = mv_port_start, - .port_stop = mv_port_stop, - .host_stop = mv_host_stop, + .port_start = ata_port_start, + .port_stop = ata_port_stop, + .host_stop = ata_host_stop, }; static struct ata_port_info mv_port_info[] = { { /* chip_504x */ .sht = &mv_sht, - .host_flags = MV_COMMON_FLAGS, - .pio_mask = 0x1f, /* pio0-4 */ - .udma_mask = 0, /* 0x7f (udma0-6 disabled for now) */ + .host_flags = (ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY | + ATA_FLAG_SATA_RESET | ATA_FLAG_MMIO), + .pio_mask = 0x1f, /* pio4-0 */ + .udma_mask = 0, /* 0x7f (udma6-0 disabled for now) */ .port_ops = &mv_ops, }, { /* chip_508x */ .sht = &mv_sht, - .host_flags = (MV_COMMON_FLAGS | MV_FLAG_DUAL_HC), - .pio_mask = 0x1f, /* pio0-4 */ - .udma_mask = 0, /* 0x7f (udma0-6 disabled for now) */ + .host_flags = (ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY | + ATA_FLAG_SATA_RESET | ATA_FLAG_MMIO | + MV_FLAG_DUAL_HC), + .pio_mask = 0x1f, /* pio4-0 */ + .udma_mask = 0, /* 0x7f (udma6-0 disabled for now) */ .port_ops = &mv_ops, }, { /* chip_604x */ .sht = &mv_sht, - .host_flags = (MV_COMMON_FLAGS | MV_6XXX_FLAGS), - .pio_mask = 0x1f, /* pio0-4 */ - .udma_mask = 0x7f, /* udma0-6 */ + .host_flags = (ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY | + ATA_FLAG_SATA_RESET | ATA_FLAG_MMIO | + MV_FLAG_IRQ_COALESCE | MV_FLAG_BDMA), + .pio_mask = 0x1f, /* pio4-0 */ + .udma_mask = 0, /* 0x7f (udma6-0 disabled for now) */ .port_ops = &mv_ops, }, { /* chip_608x */ .sht = &mv_sht, - .host_flags = (MV_COMMON_FLAGS | MV_6XXX_FLAGS | - MV_FLAG_DUAL_HC), - .pio_mask = 0x1f, /* pio0-4 */ - .udma_mask = 0x7f, /* udma0-6 */ + .host_flags = (ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY | + ATA_FLAG_SATA_RESET | ATA_FLAG_MMIO | + MV_FLAG_IRQ_COALESCE | MV_FLAG_DUAL_HC | + MV_FLAG_BDMA), + .pio_mask = 0x1f, /* pio4-0 */ + .udma_mask = 0, /* 0x7f (udma6-0 disabled for now) */ .port_ops = &mv_ops, }, }; @@ -380,6 +306,12 @@ static inline void writelfl(unsigned long data, void __iomem *addr) (void) readl(addr); /* flush to avoid PCI posted write */ } +static inline void __iomem *mv_port_addr_to_hc_base(void __iomem *port_mmio) +{ + return ((void __iomem *)((unsigned long)port_mmio & + (unsigned long)SATAHC_MASK)); +} + static inline void __iomem *mv_hc_base(void __iomem *base, unsigned int hc) { return (base + MV_SATAHC0_REG_BASE + (hc * MV_SATAHC_REG_SZ)); @@ -397,150 +329,24 @@ static inline void __iomem *mv_ap_base(struct ata_port *ap) return mv_port_base(ap->host_set->mmio_base, ap->port_no); } -static inline int mv_get_hc_count(unsigned long hp_flags) +static inline int mv_get_hc_count(unsigned long flags) { - return ((hp_flags & MV_FLAG_DUAL_HC) ? 2 : 1); + return ((flags & MV_FLAG_DUAL_HC) ? 2 : 1); } -static void mv_irq_clear(struct ata_port *ap) -{ -} - -/** - * mv_start_dma - Enable eDMA engine - * @base: port base address - * @pp: port private data - * - * Verify the local cache of the eDMA state is accurate with an - * assert. - * - * LOCKING: - * Inherited from caller. - */ -static void mv_start_dma(void __iomem *base, struct mv_port_priv *pp) -{ - if (!(MV_PP_FLAG_EDMA_EN & pp->pp_flags)) { - writelfl(EDMA_EN, base + EDMA_CMD_OFS); - pp->pp_flags |= MV_PP_FLAG_EDMA_EN; - } - assert(EDMA_EN & readl(base + EDMA_CMD_OFS)); -} - -/** - * mv_stop_dma - Disable eDMA engine - * @ap: ATA channel to manipulate - * - * Verify the local cache of the eDMA state is accurate with an - * assert. - * - * LOCKING: - * Inherited from caller. - */ -static void mv_stop_dma(struct ata_port *ap) +static inline int mv_is_edma_active(struct ata_port *ap) { void __iomem *port_mmio = mv_ap_base(ap); - struct mv_port_priv *pp = ap->private_data; - u32 reg; - int i; - - if (MV_PP_FLAG_EDMA_EN & pp->pp_flags) { - /* Disable EDMA if active. The disable bit auto clears. - */ - writelfl(EDMA_DS, port_mmio + EDMA_CMD_OFS); - pp->pp_flags &= ~MV_PP_FLAG_EDMA_EN; - } else { - assert(!(EDMA_EN & readl(port_mmio + EDMA_CMD_OFS))); - } - - /* now properly wait for the eDMA to stop */ - for (i = 1000; i > 0; i--) { - reg = readl(port_mmio + EDMA_CMD_OFS); - if (!(EDMA_EN & reg)) { - break; - } - udelay(100); - } - - if (EDMA_EN & reg) { - printk(KERN_ERR "ata%u: Unable to stop eDMA\n", ap->id); - /* FIXME: Consider doing a reset here to recover */ - } + return (EDMA_EN & readl(port_mmio + EDMA_CMD_OFS)); } -#ifdef ATA_DEBUG -static void mv_dump_mem(void __iomem *start, unsigned bytes) +static inline int mv_port_bdma_capable(struct ata_port *ap) { - int b, w; - for (b = 0; b < bytes; ) { - DPRINTK("%p: ", start + b); - for (w = 0; b < bytes && w < 4; w++) { - printk("%08x ",readl(start + b)); - b += sizeof(u32); - } - printk("\n"); - } + return (ap->flags & MV_FLAG_BDMA); } -#endif -static void mv_dump_pci_cfg(struct pci_dev *pdev, unsigned bytes) -{ -#ifdef ATA_DEBUG - int b, w; - u32 dw; - for (b = 0; b < bytes; ) { - DPRINTK("%02x: ", b); - for (w = 0; b < bytes && w < 4; w++) { - (void) pci_read_config_dword(pdev,b,&dw); - printk("%08x ",dw); - b += sizeof(u32); - } - printk("\n"); - } -#endif -} -static void mv_dump_all_regs(void __iomem *mmio_base, int port, - struct pci_dev *pdev) +static void mv_irq_clear(struct ata_port *ap) { -#ifdef ATA_DEBUG - void __iomem *hc_base = mv_hc_base(mmio_base, - port >> MV_PORT_HC_SHIFT); - void __iomem *port_base; - int start_port, num_ports, p, start_hc, num_hcs, hc; - - if (0 > port) { - start_hc = start_port = 0; - num_ports = 8; /* shld be benign for 4 port devs */ - num_hcs = 2; - } else { - start_hc = port >> MV_PORT_HC_SHIFT; - start_port = port; - num_ports = num_hcs = 1; - } - DPRINTK("All registers for port(s) %u-%u:\n", start_port, - num_ports > 1 ? num_ports - 1 : start_port); - - if (NULL != pdev) { - DPRINTK("PCI config space regs:\n"); - mv_dump_pci_cfg(pdev, 0x68); - } - DPRINTK("PCI regs:\n"); - mv_dump_mem(mmio_base+0xc00, 0x3c); - mv_dump_mem(mmio_base+0xd00, 0x34); - mv_dump_mem(mmio_base+0xf00, 0x4); - mv_dump_mem(mmio_base+0x1d00, 0x6c); - for (hc = start_hc; hc < start_hc + num_hcs; hc++) { - hc_base = mv_hc_base(mmio_base, port >> MV_PORT_HC_SHIFT); - DPRINTK("HC regs (HC %i):\n", hc); - mv_dump_mem(hc_base, 0x1c); - } - for (p = start_port; p < start_port + num_ports; p++) { - port_base = mv_port_base(mmio_base, p); - DPRINTK("EDMA regs (port %i):\n",p); - mv_dump_mem(port_base, 0x54); - DPRINTK("SATA regs (port %i):\n",p); - mv_dump_mem(port_base+0x300, 0x60); - } -#endif } static unsigned int mv_scr_offset(unsigned int sc_reg_in) @@ -583,37 +389,30 @@ static void mv_scr_write(struct ata_port *ap, unsigned int sc_reg_in, u32 val) } } -/** - * mv_global_soft_reset - Perform the 6xxx global soft reset - * @mmio_base: base address of the HBA - * - * This routine only applies to 6xxx parts. - * - * LOCKING: - * Inherited from caller. - */ -static int mv_global_soft_reset(void __iomem *mmio_base) +static int mv_master_reset(void __iomem *mmio_base) { void __iomem *reg = mmio_base + PCI_MAIN_CMD_STS_OFS; int i, rc = 0; u32 t; + VPRINTK("ENTER\n"); + /* Following procedure defined in PCI "main command and status * register" table. */ t = readl(reg); writel(t | STOP_PCI_MASTER, reg); - for (i = 0; i < 1000; i++) { - udelay(1); + for (i = 0; i < 100; i++) { + msleep(10); t = readl(reg); if (PCI_MASTER_EMPTY & t) { break; } } if (!(PCI_MASTER_EMPTY & t)) { - printk(KERN_ERR DRV_NAME ": PCI master won't flush\n"); - rc = 1; + printk(KERN_ERR DRV_NAME "PCI master won't flush\n"); + rc = 1; /* broken HW? */ goto done; } @@ -626,411 +425,48 @@ static int mv_global_soft_reset(void __iomem *mmio_base) } while (!(GLOB_SFT_RST & t) && (i-- > 0)); if (!(GLOB_SFT_RST & t)) { - printk(KERN_ERR DRV_NAME ": can't set global reset\n"); - rc = 1; + printk(KERN_ERR DRV_NAME "can't set global reset\n"); + rc = 1; /* broken HW? */ goto done; } - /* clear reset and *reenable the PCI master* (not mentioned in spec) */ + /* clear reset */ i = 5; do { - writel(t & ~(GLOB_SFT_RST | STOP_PCI_MASTER), reg); + writel(t & ~GLOB_SFT_RST, reg); t = readl(reg); udelay(1); } while ((GLOB_SFT_RST & t) && (i-- > 0)); if (GLOB_SFT_RST & t) { - printk(KERN_ERR DRV_NAME ": can't clear global reset\n"); - rc = 1; - } -done: - return rc; -} - -/** - * mv_host_stop - Host specific cleanup/stop routine. - * @host_set: host data structure - * - * Disable ints, cleanup host memory, call general purpose - * host_stop. - * - * LOCKING: - * Inherited from caller. - */ -static void mv_host_stop(struct ata_host_set *host_set) -{ - struct mv_host_priv *hpriv = host_set->private_data; - struct pci_dev *pdev = to_pci_dev(host_set->dev); - - if (hpriv->hp_flags & MV_HP_FLAG_MSI) { - pci_disable_msi(pdev); - } else { - pci_intx(pdev, 0); - } - kfree(hpriv); - ata_host_stop(host_set); -} - -/** - * mv_port_start - Port specific init/start routine. - * @ap: ATA channel to manipulate - * - * Allocate and point to DMA memory, init port private memory, - * zero indices. - * - * LOCKING: - * Inherited from caller. - */ -static int mv_port_start(struct ata_port *ap) -{ - struct device *dev = ap->host_set->dev; - struct mv_port_priv *pp; - void __iomem *port_mmio = mv_ap_base(ap); - void *mem; - dma_addr_t mem_dma; - - pp = kmalloc(sizeof(*pp), GFP_KERNEL); - if (!pp) { - return -ENOMEM; - } - memset(pp, 0, sizeof(*pp)); - - mem = dma_alloc_coherent(dev, MV_PORT_PRIV_DMA_SZ, &mem_dma, - GFP_KERNEL); - if (!mem) { - kfree(pp); - return -ENOMEM; - } - memset(mem, 0, MV_PORT_PRIV_DMA_SZ); - - /* First item in chunk of DMA memory: - * 32-slot command request table (CRQB), 32 bytes each in size - */ - pp->crqb = mem; - pp->crqb_dma = mem_dma; - mem += MV_CRQB_Q_SZ; - mem_dma += MV_CRQB_Q_SZ; - - /* Second item: - * 32-slot command response table (CRPB), 8 bytes each in size - */ - pp->crpb = mem; - pp->crpb_dma = mem_dma; - mem += MV_CRPB_Q_SZ; - mem_dma += MV_CRPB_Q_SZ; - - /* Third item: - * Table of scatter-gather descriptors (ePRD), 16 bytes each - */ - pp->sg_tbl = mem; - pp->sg_tbl_dma = mem_dma; - - writelfl(EDMA_CFG_Q_DEPTH | EDMA_CFG_RD_BRST_EXT | - EDMA_CFG_WR_BUFF_LEN, port_mmio + EDMA_CFG_OFS); - - writel((pp->crqb_dma >> 16) >> 16, port_mmio + EDMA_REQ_Q_BASE_HI_OFS); - writelfl(pp->crqb_dma & EDMA_REQ_Q_BASE_LO_MASK, - port_mmio + EDMA_REQ_Q_IN_PTR_OFS); - - writelfl(0, port_mmio + EDMA_REQ_Q_OUT_PTR_OFS); - writelfl(0, port_mmio + EDMA_RSP_Q_IN_PTR_OFS); - - writel((pp->crpb_dma >> 16) >> 16, port_mmio + EDMA_RSP_Q_BASE_HI_OFS); - writelfl(pp->crpb_dma & EDMA_RSP_Q_BASE_LO_MASK, - port_mmio + EDMA_RSP_Q_OUT_PTR_OFS); - - pp->req_producer = pp->rsp_consumer = 0; - - /* Don't turn on EDMA here...do it before DMA commands only. Else - * we'll be unable to send non-data, PIO, etc due to restricted access - * to shadow regs. - */ - ap->private_data = pp; - return 0; -} - -/** - * mv_port_stop - Port specific cleanup/stop routine. - * @ap: ATA channel to manipulate - * - * Stop DMA, cleanup port memory. - * - * LOCKING: - * This routine uses the host_set lock to protect the DMA stop. - */ -static void mv_port_stop(struct ata_port *ap) -{ - struct device *dev = ap->host_set->dev; - struct mv_port_priv *pp = ap->private_data; - unsigned long flags; - - spin_lock_irqsave(&ap->host_set->lock, flags); - mv_stop_dma(ap); - spin_unlock_irqrestore(&ap->host_set->lock, flags); - - ap->private_data = NULL; - dma_free_coherent(dev, MV_PORT_PRIV_DMA_SZ, pp->crpb, pp->crpb_dma); - kfree(pp); -} - -/** - * mv_fill_sg - Fill out the Marvell ePRD (scatter gather) entries - * @qc: queued command whose SG list to source from - * - * Populate the SG list and mark the last entry. - * - * LOCKING: - * Inherited from caller. - */ -static void mv_fill_sg(struct ata_queued_cmd *qc) -{ - struct mv_port_priv *pp = qc->ap->private_data; - unsigned int i; - - for (i = 0; i < qc->n_elem; i++) { - u32 sg_len; - dma_addr_t addr; - - addr = sg_dma_address(&qc->sg[i]); - sg_len = sg_dma_len(&qc->sg[i]); - - pp->sg_tbl[i].addr = cpu_to_le32(addr & 0xffffffff); - pp->sg_tbl[i].addr_hi = cpu_to_le32((addr >> 16) >> 16); - assert(0 == (sg_len & ~MV_DMA_BOUNDARY)); - pp->sg_tbl[i].flags_size = cpu_to_le32(sg_len); - } - if (0 < qc->n_elem) { - pp->sg_tbl[qc->n_elem - 1].flags_size |= EPRD_FLAG_END_OF_TBL; - } -} - -static inline unsigned mv_inc_q_index(unsigned *index) -{ - *index = (*index + 1) & MV_MAX_Q_DEPTH_MASK; - return *index; -} - -static inline void mv_crqb_pack_cmd(u16 *cmdw, u8 data, u8 addr, unsigned last) -{ - *cmdw = data | (addr << CRQB_CMD_ADDR_SHIFT) | CRQB_CMD_CS | - (last ? CRQB_CMD_LAST : 0); -} - -/** - * mv_qc_prep - Host specific command preparation. - * @qc: queued command to prepare - * - * This routine simply redirects to the general purpose routine - * if command is not DMA. Else, it handles prep of the CRQB - * (command request block), does some sanity checking, and calls - * the SG load routine. - * - * LOCKING: - * Inherited from caller. - */ -static void mv_qc_prep(struct ata_queued_cmd *qc) -{ - struct ata_port *ap = qc->ap; - struct mv_port_priv *pp = ap->private_data; - u16 *cw; - struct ata_taskfile *tf; - u16 flags = 0; - - if (ATA_PROT_DMA != qc->tf.protocol) { - return; + printk(KERN_ERR DRV_NAME "can't clear global reset\n"); + rc = 1; /* broken HW? */ } - /* the req producer index should be the same as we remember it */ - assert(((readl(mv_ap_base(qc->ap) + EDMA_REQ_Q_IN_PTR_OFS) >> - EDMA_REQ_Q_PTR_SHIFT) & MV_MAX_Q_DEPTH_MASK) == - pp->req_producer); - - /* Fill in command request block - */ - if (!(qc->tf.flags & ATA_TFLAG_WRITE)) { - flags |= CRQB_FLAG_READ; - } - assert(MV_MAX_Q_DEPTH > qc->tag); - flags |= qc->tag << CRQB_TAG_SHIFT; - - pp->crqb[pp->req_producer].sg_addr = - cpu_to_le32(pp->sg_tbl_dma & 0xffffffff); - pp->crqb[pp->req_producer].sg_addr_hi = - cpu_to_le32((pp->sg_tbl_dma >> 16) >> 16); - pp->crqb[pp->req_producer].ctrl_flags = cpu_to_le16(flags); - - cw = &pp->crqb[pp->req_producer].ata_cmd[0]; - tf = &qc->tf; - - /* Sadly, the CRQB cannot accomodate all registers--there are - * only 11 bytes...so we must pick and choose required - * registers based on the command. So, we drop feature and - * hob_feature for [RW] DMA commands, but they are needed for - * NCQ. NCQ will drop hob_nsect. - */ - switch (tf->command) { - case ATA_CMD_READ: - case ATA_CMD_READ_EXT: - case ATA_CMD_WRITE: - case ATA_CMD_WRITE_EXT: - mv_crqb_pack_cmd(cw++, tf->hob_nsect, ATA_REG_NSECT, 0); - break; -#ifdef LIBATA_NCQ /* FIXME: remove this line when NCQ added */ - case ATA_CMD_FPDMA_READ: - case ATA_CMD_FPDMA_WRITE: - mv_crqb_pack_cmd(cw++, tf->hob_feature, ATA_REG_FEATURE, 0); - mv_crqb_pack_cmd(cw++, tf->feature, ATA_REG_FEATURE, 0); - break; -#endif /* FIXME: remove this line when NCQ added */ - default: - /* The only other commands EDMA supports in non-queued and - * non-NCQ mode are: [RW] STREAM DMA and W DMA FUA EXT, none - * of which are defined/used by Linux. If we get here, this - * driver needs work. - * - * FIXME: modify libata to give qc_prep a return value and - * return error here. - */ - BUG_ON(tf->command); - break; - } - mv_crqb_pack_cmd(cw++, tf->nsect, ATA_REG_NSECT, 0); - mv_crqb_pack_cmd(cw++, tf->hob_lbal, ATA_REG_LBAL, 0); - mv_crqb_pack_cmd(cw++, tf->lbal, ATA_REG_LBAL, 0); - mv_crqb_pack_cmd(cw++, tf->hob_lbam, ATA_REG_LBAM, 0); - mv_crqb_pack_cmd(cw++, tf->lbam, ATA_REG_LBAM, 0); - mv_crqb_pack_cmd(cw++, tf->hob_lbah, ATA_REG_LBAH, 0); - mv_crqb_pack_cmd(cw++, tf->lbah, ATA_REG_LBAH, 0); - mv_crqb_pack_cmd(cw++, tf->device, ATA_REG_DEVICE, 0); - mv_crqb_pack_cmd(cw++, tf->command, ATA_REG_CMD, 1); /* last */ - - if (!(qc->flags & ATA_QCFLAG_DMAMAP)) { - return; - } - mv_fill_sg(qc); -} - -/** - * mv_qc_issue - Initiate a command to the host - * @qc: queued command to start - * - * This routine simply redirects to the general purpose routine - * if command is not DMA. Else, it sanity checks our local - * caches of the request producer/consumer indices then enables - * DMA and bumps the request producer index. - * - * LOCKING: - * Inherited from caller. - */ -static int mv_qc_issue(struct ata_queued_cmd *qc) -{ - void __iomem *port_mmio = mv_ap_base(qc->ap); - struct mv_port_priv *pp = qc->ap->private_data; - u32 in_ptr; - - if (ATA_PROT_DMA != qc->tf.protocol) { - /* We're about to send a non-EDMA capable command to the - * port. Turn off EDMA so there won't be problems accessing - * shadow block, etc registers. - */ - mv_stop_dma(qc->ap); - return ata_qc_issue_prot(qc); - } - - in_ptr = readl(port_mmio + EDMA_REQ_Q_IN_PTR_OFS); - - /* the req producer index should be the same as we remember it */ - assert(((in_ptr >> EDMA_REQ_Q_PTR_SHIFT) & MV_MAX_Q_DEPTH_MASK) == - pp->req_producer); - /* until we do queuing, the queue should be empty at this point */ - assert(((in_ptr >> EDMA_REQ_Q_PTR_SHIFT) & MV_MAX_Q_DEPTH_MASK) == - ((readl(port_mmio + EDMA_REQ_Q_OUT_PTR_OFS) >> - EDMA_REQ_Q_PTR_SHIFT) & MV_MAX_Q_DEPTH_MASK)); - - mv_inc_q_index(&pp->req_producer); /* now incr producer index */ - - mv_start_dma(port_mmio, pp); - - /* and write the request in pointer to kick the EDMA to life */ - in_ptr &= EDMA_REQ_Q_BASE_LO_MASK; - in_ptr |= pp->req_producer << EDMA_REQ_Q_PTR_SHIFT; - writelfl(in_ptr, port_mmio + EDMA_REQ_Q_IN_PTR_OFS); - - return 0; -} - -/** - * mv_get_crpb_status - get status from most recently completed cmd - * @ap: ATA channel to manipulate - * - * This routine is for use when the port is in DMA mode, when it - * will be using the CRPB (command response block) method of - * returning command completion information. We assert indices - * are good, grab status, and bump the response consumer index to - * prove that we're up to date. - * - * LOCKING: - * Inherited from caller. - */ -static u8 mv_get_crpb_status(struct ata_port *ap) -{ - void __iomem *port_mmio = mv_ap_base(ap); - struct mv_port_priv *pp = ap->private_data; - u32 out_ptr; - - out_ptr = readl(port_mmio + EDMA_RSP_Q_OUT_PTR_OFS); - - /* the response consumer index should be the same as we remember it */ - assert(((out_ptr >> EDMA_RSP_Q_PTR_SHIFT) & MV_MAX_Q_DEPTH_MASK) == - pp->rsp_consumer); - - /* increment our consumer index... */ - pp->rsp_consumer = mv_inc_q_index(&pp->rsp_consumer); - - /* and, until we do NCQ, there should only be 1 CRPB waiting */ - assert(((readl(port_mmio + EDMA_RSP_Q_IN_PTR_OFS) >> - EDMA_RSP_Q_PTR_SHIFT) & MV_MAX_Q_DEPTH_MASK) == - pp->rsp_consumer); - - /* write out our inc'd consumer index so EDMA knows we're caught up */ - out_ptr &= EDMA_RSP_Q_BASE_LO_MASK; - out_ptr |= pp->rsp_consumer << EDMA_RSP_Q_PTR_SHIFT; - writelfl(out_ptr, port_mmio + EDMA_RSP_Q_OUT_PTR_OFS); - - /* Return ATA status register for completed CRPB */ - return (pp->crpb[pp->rsp_consumer].flags >> CRPB_FLAG_STATUS_SHIFT); + done: + VPRINTK("EXIT, rc = %i\n", rc); + return rc; } -/** - * mv_err_intr - Handle error interrupts on the port - * @ap: ATA channel to manipulate - * - * In most cases, just clear the interrupt and move on. However, - * some cases require an eDMA reset, which is done right before - * the COMRESET in mv_phy_reset(). The SERR case requires a - * clear of pending errors in the SATA SERROR register. Finally, - * if the port disabled DMA, update our cached copy to match. - * - * LOCKING: - * Inherited from caller. - */ static void mv_err_intr(struct ata_port *ap) { - void __iomem *port_mmio = mv_ap_base(ap); + void __iomem *port_mmio; u32 edma_err_cause, serr = 0; + /* bug here b/c we got an err int on a port we don't know about, + * so there's no way to clear it + */ + BUG_ON(NULL == ap); + port_mmio = mv_ap_base(ap); + edma_err_cause = readl(port_mmio + EDMA_ERR_IRQ_CAUSE_OFS); if (EDMA_ERR_SERR & edma_err_cause) { serr = scr_read(ap, SCR_ERROR); scr_write_flush(ap, SCR_ERROR, serr); } - if (EDMA_ERR_SELF_DIS & edma_err_cause) { - struct mv_port_priv *pp = ap->private_data; - pp->pp_flags &= ~MV_PP_FLAG_EDMA_EN; - } - DPRINTK(KERN_ERR "ata%u: port error; EDMA err cause: 0x%08x " - "SERR: 0x%08x\n", ap->id, edma_err_cause, serr); + DPRINTK("port %u error; EDMA err cause: 0x%08x SERR: 0x%08x\n", + ap->port_no, edma_err_cause, serr); /* Clear EDMA now that SERR cleanup done */ writelfl(0, port_mmio + EDMA_ERR_IRQ_CAUSE_OFS); @@ -1041,21 +477,7 @@ static void mv_err_intr(struct ata_port *ap) } } -/** - * mv_host_intr - Handle all interrupts on the given host controller - * @host_set: host specific structure - * @relevant: port error bits relevant to this host controller - * @hc: which host controller we're to look at - * - * Read then write clear the HC interrupt status then walk each - * port connected to the HC and see if it needs servicing. Port - * success ints are reported in the HC interrupt status reg, the - * port error ints are reported in the higher level main - * interrupt status register and thus are passed in via the - * 'relevant' argument. - * - * LOCKING: - * Inherited from caller. +/* Handle any outstanding interrupts in a single SATAHC */ static void mv_host_intr(struct ata_host_set *host_set, u32 relevant, unsigned int hc) @@ -1065,8 +487,8 @@ static void mv_host_intr(struct ata_host_set *host_set, u32 relevant, struct ata_port *ap; struct ata_queued_cmd *qc; u32 hc_irq_cause; - int shift, port, port0, hard_port, handled; - u8 ata_status = 0; + int shift, port, port0, hard_port; + u8 ata_status; if (hc == 0) { port0 = 0; @@ -1077,7 +499,7 @@ static void mv_host_intr(struct ata_host_set *host_set, u32 relevant, /* we'll need the HC success int register in most cases */ hc_irq_cause = readl(hc_mmio + HC_IRQ_CAUSE_OFS); if (hc_irq_cause) { - writelfl(~hc_irq_cause, hc_mmio + HC_IRQ_CAUSE_OFS); + writelfl(0, hc_mmio + HC_IRQ_CAUSE_OFS); } VPRINTK("ENTER, hc%u relevant=0x%08x HC IRQ cause=0x%08x\n", @@ -1086,38 +508,35 @@ static void mv_host_intr(struct ata_host_set *host_set, u32 relevant, for (port = port0; port < port0 + MV_PORTS_PER_HC; port++) { ap = host_set->ports[port]; hard_port = port & MV_PORT_MASK; /* range 0-3 */ - handled = 0; /* ensure ata_status is set if handled++ */ + ata_status = 0xffU; - if ((CRPB_DMA_DONE << hard_port) & hc_irq_cause) { - /* new CRPB on the queue; just one at a time until NCQ - */ - ata_status = mv_get_crpb_status(ap); - handled++; - } else if ((DEV_IRQ << hard_port) & hc_irq_cause) { - /* received ATA IRQ; read the status reg to clear INTRQ + if (((CRBP_DMA_DONE | DEV_IRQ) << hard_port) & hc_irq_cause) { + BUG_ON(NULL == ap); + /* rcv'd new resp, basic DMA complete, or ATA IRQ */ + /* This is needed to clear the ATA INTRQ. + * FIXME: don't read the status reg in EDMA mode! */ ata_status = readb((void __iomem *) ap->ioaddr.status_addr); - handled++; } - shift = port << 1; /* (port * 2) */ + shift = port * 2; if (port >= MV_PORTS_PER_HC) { shift++; /* skip bit 8 in the HC Main IRQ reg */ } if ((PORT0_ERR << shift) & relevant) { mv_err_intr(ap); - /* OR in ATA_ERR to ensure libata knows we took one */ + /* FIXME: smart to OR in ATA_ERR? */ ata_status = readb((void __iomem *) ap->ioaddr.status_addr) | ATA_ERR; - handled++; } - if (handled && ap) { + if (ap) { qc = ata_qc_from_tag(ap, ap->active_tag); if (NULL != qc) { VPRINTK("port %u IRQ found for qc, " "ata_status 0x%x\n", port,ata_status); + BUG_ON(0xffU == ata_status); /* mark qc status appropriately */ ata_qc_complete(qc, ata_status); } @@ -1126,30 +545,17 @@ static void mv_host_intr(struct ata_host_set *host_set, u32 relevant, VPRINTK("EXIT\n"); } -/** - * mv_interrupt - - * @irq: unused - * @dev_instance: private data; in this case the host structure - * @regs: unused - * - * Read the read only register to determine if any host - * controllers have pending interrupts. If so, call lower level - * routine to handle. Also check for PCI errors which are only - * reported here. - * - * LOCKING: - * This routine holds the host_set lock while processing pending - * interrupts. - */ static irqreturn_t mv_interrupt(int irq, void *dev_instance, struct pt_regs *regs) { struct ata_host_set *host_set = dev_instance; unsigned int hc, handled = 0, n_hcs; - void __iomem *mmio = host_set->mmio_base; + void __iomem *mmio; u32 irq_stat; + mmio = host_set->mmio_base; irq_stat = readl(mmio + HC_MAIN_IRQ_CAUSE_OFS); + n_hcs = mv_get_hc_count(host_set->ports[0]->flags); /* check the cases where we either have nothing pending or have read * a bogus register value which can indicate HW removal or PCI fault @@ -1158,105 +564,64 @@ static irqreturn_t mv_interrupt(int irq, void *dev_instance, return IRQ_NONE; } - n_hcs = mv_get_hc_count(host_set->ports[0]->flags); spin_lock(&host_set->lock); for (hc = 0; hc < n_hcs; hc++) { u32 relevant = irq_stat & (HC0_IRQ_PEND << (hc * HC_SHIFT)); if (relevant) { mv_host_intr(host_set, relevant, hc); - handled++; + handled = 1; } } if (PCI_ERR & irq_stat) { - printk(KERN_ERR DRV_NAME ": PCI ERROR; PCI IRQ cause=0x%08x\n", - readl(mmio + PCI_IRQ_CAUSE_OFS)); - - DPRINTK("All regs @ PCI error\n"); - mv_dump_all_regs(mmio, -1, to_pci_dev(host_set->dev)); - - writelfl(0, mmio + PCI_IRQ_CAUSE_OFS); - handled++; + /* FIXME: these are all masked by default, but still need + * to recover from them properly. + */ } + spin_unlock(&host_set->lock); return IRQ_RETVAL(handled); } -/** - * mv_check_err - Return the error shadow register to caller. - * @ap: ATA channel to manipulate - * - * Marvell requires DMA to be stopped before accessing shadow - * registers. So we do that, then return the needed register. - * - * LOCKING: - * Inherited from caller. FIXME: protect mv_stop_dma with lock? - */ -static u8 mv_check_err(struct ata_port *ap) -{ - mv_stop_dma(ap); /* can't read shadow regs if DMA on */ - return readb((void __iomem *) ap->ioaddr.error_addr); -} - -/** - * mv_phy_reset - Perform eDMA reset followed by COMRESET - * @ap: ATA channel to manipulate - * - * Part of this is taken from __sata_phy_reset and modified to - * not sleep since this routine gets called from interrupt level. - * - * LOCKING: - * Inherited from caller. This is coded to safe to call at - * interrupt level, i.e. it does not sleep. - */ static void mv_phy_reset(struct ata_port *ap) { void __iomem *port_mmio = mv_ap_base(ap); struct ata_taskfile tf; struct ata_device *dev = &ap->device[0]; - unsigned long timeout; + u32 edma = 0, bdma; VPRINTK("ENTER, port %u, mmio 0x%p\n", ap->port_no, port_mmio); - mv_stop_dma(ap); + edma = readl(port_mmio + EDMA_CMD_OFS); + if (EDMA_EN & edma) { + /* disable EDMA if active */ + edma &= ~EDMA_EN; + writelfl(edma | EDMA_DS, port_mmio + EDMA_CMD_OFS); + udelay(1); + } else if (mv_port_bdma_capable(ap) && + (bdma = readl(port_mmio + BDMA_CMD_OFS)) & BDMA_START) { + /* disable BDMA if active */ + writelfl(bdma & ~BDMA_START, port_mmio + BDMA_CMD_OFS); + } - writelfl(ATA_RST, port_mmio + EDMA_CMD_OFS); + writelfl(edma | ATA_RST, port_mmio + EDMA_CMD_OFS); udelay(25); /* allow reset propagation */ /* Spec never mentions clearing the bit. Marvell's driver does * clear the bit, however. */ - writelfl(0, port_mmio + EDMA_CMD_OFS); + writelfl(edma & ~ATA_RST, port_mmio + EDMA_CMD_OFS); - VPRINTK("S-regs after ATA_RST: SStat 0x%08x SErr 0x%08x " - "SCtrl 0x%08x\n", mv_scr_read(ap, SCR_STATUS), - mv_scr_read(ap, SCR_ERROR), mv_scr_read(ap, SCR_CONTROL)); + VPRINTK("Done. Now calling __sata_phy_reset()\n"); /* proceed to init communications via the scr_control reg */ - scr_write_flush(ap, SCR_CONTROL, 0x301); - mdelay(1); - scr_write_flush(ap, SCR_CONTROL, 0x300); - timeout = jiffies + (HZ * 1); - do { - mdelay(10); - if ((scr_read(ap, SCR_STATUS) & 0xf) != 1) - break; - } while (time_before(jiffies, timeout)); + __sata_phy_reset(ap); - VPRINTK("S-regs after PHY wake: SStat 0x%08x SErr 0x%08x " - "SCtrl 0x%08x\n", mv_scr_read(ap, SCR_STATUS), - mv_scr_read(ap, SCR_ERROR), mv_scr_read(ap, SCR_CONTROL)); - - if (sata_dev_present(ap)) { - ata_port_probe(ap); - } else { - printk(KERN_INFO "ata%u: no device found (phy stat %08x)\n", - ap->id, scr_read(ap, SCR_STATUS)); - ata_port_disable(ap); + if (ap->flags & ATA_FLAG_PORT_DISABLED) { + VPRINTK("Port disabled pre-sig. Exiting.\n"); return; } - ap->cbl = ATA_CBL_SATA; tf.lbah = readb((void __iomem *) ap->ioaddr.lbah_addr); tf.lbam = readb((void __iomem *) ap->ioaddr.lbam_addr); @@ -1271,118 +636,37 @@ static void mv_phy_reset(struct ata_port *ap) VPRINTK("EXIT\n"); } -/** - * mv_eng_timeout - Routine called by libata when SCSI times out I/O - * @ap: ATA channel to manipulate - * - * Intent is to clear all pending error conditions, reset the - * chip/bus, fail the command, and move on. - * - * LOCKING: - * This routine holds the host_set lock while failing the command. - */ -static void mv_eng_timeout(struct ata_port *ap) -{ - struct ata_queued_cmd *qc; - unsigned long flags; - - printk(KERN_ERR "ata%u: Entering mv_eng_timeout\n",ap->id); - DPRINTK("All regs @ start of eng_timeout\n"); - mv_dump_all_regs(ap->host_set->mmio_base, ap->port_no, - to_pci_dev(ap->host_set->dev)); - - qc = ata_qc_from_tag(ap, ap->active_tag); - printk(KERN_ERR "mmio_base %p ap %p qc %p scsi_cmnd %p &cmnd %p\n", - ap->host_set->mmio_base, ap, qc, qc->scsicmd, - &qc->scsicmd->cmnd); - - mv_err_intr(ap); - mv_phy_reset(ap); - - if (!qc) { - printk(KERN_ERR "ata%u: BUG: timeout without command\n", - ap->id); - } else { - /* hack alert! We cannot use the supplied completion - * function from inside the ->eh_strategy_handler() thread. - * libata is the only user of ->eh_strategy_handler() in - * any kernel, so the default scsi_done() assumes it is - * not being called from the SCSI EH. - */ - spin_lock_irqsave(&ap->host_set->lock, flags); - qc->scsidone = scsi_finish_command; - ata_qc_complete(qc, ATA_ERR); - spin_unlock_irqrestore(&ap->host_set->lock, flags); - } -} - -/** - * mv_port_init - Perform some early initialization on a single port. - * @port: libata data structure storing shadow register addresses - * @port_mmio: base address of the port - * - * Initialize shadow register mmio addresses, clear outstanding - * interrupts on the port, and unmask interrupts for the future - * start of the port. - * - * LOCKING: - * Inherited from caller. - */ -static void mv_port_init(struct ata_ioports *port, void __iomem *port_mmio) +static void mv_port_init(struct ata_ioports *port, unsigned long base) { - unsigned long shd_base = (unsigned long) port_mmio + SHD_BLK_OFS; - unsigned serr_ofs; - - /* PIO related setup - */ - port->data_addr = shd_base + (sizeof(u32) * ATA_REG_DATA); - port->error_addr = - port->feature_addr = shd_base + (sizeof(u32) * ATA_REG_ERR); - port->nsect_addr = shd_base + (sizeof(u32) * ATA_REG_NSECT); - port->lbal_addr = shd_base + (sizeof(u32) * ATA_REG_LBAL); - port->lbam_addr = shd_base + (sizeof(u32) * ATA_REG_LBAM); - port->lbah_addr = shd_base + (sizeof(u32) * ATA_REG_LBAH); - port->device_addr = shd_base + (sizeof(u32) * ATA_REG_DEVICE); - port->status_addr = - port->command_addr = shd_base + (sizeof(u32) * ATA_REG_STATUS); - /* special case: control/altstatus doesn't have ATA_REG_ address */ - port->altstatus_addr = port->ctl_addr = shd_base + SHD_CTL_AST_OFS; - - /* unused: */ + /* PIO related setup */ + port->data_addr = base + SHD_PIO_DATA_OFS; + port->error_addr = port->feature_addr = base + SHD_FEA_ERR_OFS; + port->nsect_addr = base + SHD_SECT_CNT_OFS; + port->lbal_addr = base + SHD_LBA_L_OFS; + port->lbam_addr = base + SHD_LBA_M_OFS; + port->lbah_addr = base + SHD_LBA_H_OFS; + port->device_addr = base + SHD_DEV_HD_OFS; + port->status_addr = port->command_addr = base + SHD_CMD_STA_OFS; + port->altstatus_addr = port->ctl_addr = base + SHD_CTL_AST_OFS; + /* unused */ port->cmd_addr = port->bmdma_addr = port->scr_addr = 0; - /* Clear any currently outstanding port interrupt conditions */ - serr_ofs = mv_scr_offset(SCR_ERROR); - writelfl(readl(port_mmio + serr_ofs), port_mmio + serr_ofs); - writelfl(0, port_mmio + EDMA_ERR_IRQ_CAUSE_OFS); - /* unmask all EDMA error interrupts */ - writelfl(~0, port_mmio + EDMA_ERR_IRQ_MASK_OFS); + writel(~0, (void __iomem *)base + EDMA_ERR_IRQ_MASK_OFS); VPRINTK("EDMA cfg=0x%08x EDMA IRQ err cause/mask=0x%08x/0x%08x\n", - readl(port_mmio + EDMA_CFG_OFS), - readl(port_mmio + EDMA_ERR_IRQ_CAUSE_OFS), - readl(port_mmio + EDMA_ERR_IRQ_MASK_OFS)); + readl((void __iomem *)base + EDMA_CFG_OFS), + readl((void __iomem *)base + EDMA_ERR_IRQ_CAUSE_OFS), + readl((void __iomem *)base + EDMA_ERR_IRQ_MASK_OFS)); } -/** - * mv_host_init - Perform some early initialization of the host. - * @probe_ent: early data struct representing the host - * - * If possible, do an early global reset of the host. Then do - * our port init and clear/unmask all/relevant host interrupts. - * - * LOCKING: - * Inherited from caller. - */ static int mv_host_init(struct ata_probe_ent *probe_ent) { int rc = 0, n_hc, port, hc; void __iomem *mmio = probe_ent->mmio_base; void __iomem *port_mmio; - if ((MV_FLAG_GLBL_SFT_RST & probe_ent->host_flags) && - mv_global_soft_reset(probe_ent->mmio_base)) { + if (mv_master_reset(probe_ent->mmio_base)) { rc = 1; goto done; } @@ -1392,27 +676,17 @@ static int mv_host_init(struct ata_probe_ent *probe_ent) for (port = 0; port < probe_ent->n_ports; port++) { port_mmio = mv_port_base(mmio, port); - mv_port_init(&probe_ent->port[port], port_mmio); + mv_port_init(&probe_ent->port[port], (unsigned long)port_mmio); } for (hc = 0; hc < n_hc; hc++) { - void __iomem *hc_mmio = mv_hc_base(mmio, hc); - - VPRINTK("HC%i: HC config=0x%08x HC IRQ cause " - "(before clear)=0x%08x\n", hc, - readl(hc_mmio + HC_CFG_OFS), - readl(hc_mmio + HC_IRQ_CAUSE_OFS)); - - /* Clear any currently outstanding hc interrupt conditions */ - writelfl(0, hc_mmio + HC_IRQ_CAUSE_OFS); + VPRINTK("HC%i: HC config=0x%08x HC IRQ cause=0x%08x\n", hc, + readl(mv_hc_base(mmio, hc) + HC_CFG_OFS), + readl(mv_hc_base(mmio, hc) + HC_IRQ_CAUSE_OFS)); } - /* Clear any currently outstanding host interrupt conditions */ - writelfl(0, mmio + PCI_IRQ_CAUSE_OFS); - - /* and unmask interrupt generation for host regs */ - writelfl(PCI_UNMASK_ALL_IRQS, mmio + PCI_IRQ_MASK_OFS); - writelfl(~HC_MAIN_MASKED_IRQS, mmio + HC_MAIN_IRQ_MASK_OFS); + writel(~HC_MAIN_MASKED_IRQS, mmio + HC_MAIN_IRQ_MASK_OFS); + writel(PCI_UNMASK_ALL_IRQS, mmio + PCI_IRQ_MASK_OFS); VPRINTK("HC MAIN IRQ cause/mask=0x%08x/0x%08x " "PCI int cause/mask=0x%08x/0x%08x\n", @@ -1420,53 +694,11 @@ static int mv_host_init(struct ata_probe_ent *probe_ent) readl(mmio + HC_MAIN_IRQ_MASK_OFS), readl(mmio + PCI_IRQ_CAUSE_OFS), readl(mmio + PCI_IRQ_MASK_OFS)); -done: - return rc; -} - -/** - * mv_print_info - Dump key info to kernel log for perusal. - * @probe_ent: early data struct representing the host - * - * FIXME: complete this. - * - * LOCKING: - * Inherited from caller. - */ -static void mv_print_info(struct ata_probe_ent *probe_ent) -{ - struct pci_dev *pdev = to_pci_dev(probe_ent->dev); - struct mv_host_priv *hpriv = probe_ent->private_data; - u8 rev_id, scc; - const char *scc_s; - /* Use this to determine the HW stepping of the chip so we know - * what errata to workaround - */ - pci_read_config_byte(pdev, PCI_REVISION_ID, &rev_id); - - pci_read_config_byte(pdev, PCI_CLASS_DEVICE, &scc); - if (scc == 0) - scc_s = "SCSI"; - else if (scc == 0x01) - scc_s = "RAID"; - else - scc_s = "unknown"; - - printk(KERN_INFO DRV_NAME - "(%s) %u slots %u ports %s mode IRQ via %s\n", - pci_name(pdev), (unsigned)MV_MAX_Q_DEPTH, probe_ent->n_ports, - scc_s, (MV_HP_FLAG_MSI & hpriv->hp_flags) ? "MSI" : "INTx"); + done: + return rc; } -/** - * mv_init_one - handle a positive probe of a Marvell host - * @pdev: PCI device found - * @ent: PCI device ID entry for the matched host - * - * LOCKING: - * Inherited from caller. - */ static int mv_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) { static int printed_version = 0; @@ -1474,12 +706,16 @@ static int mv_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) struct mv_host_priv *hpriv; unsigned int board_idx = (unsigned int)ent->driver_data; void __iomem *mmio_base; - int pci_dev_busy = 0, rc; + int pci_dev_busy = 0; + int rc; if (!printed_version++) { - printk(KERN_INFO DRV_NAME " version " DRV_VERSION "\n"); + printk(KERN_DEBUG DRV_NAME " version " DRV_VERSION "\n"); } + VPRINTK("ENTER for PCI Bus:Slot.Func=%u:%u.%u\n", pdev->bus->number, + PCI_SLOT(pdev->devfn), PCI_FUNC(pdev->devfn)); + rc = pci_enable_device(pdev); if (rc) { return rc; @@ -1491,6 +727,8 @@ static int mv_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) goto err_out; } + pci_intx(pdev, 1); + probe_ent = kmalloc(sizeof(*probe_ent), GFP_KERNEL); if (probe_ent == NULL) { rc = -ENOMEM; @@ -1501,7 +739,8 @@ static int mv_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) probe_ent->dev = pci_dev_to_dev(pdev); INIT_LIST_HEAD(&probe_ent->node); - mmio_base = pci_iomap(pdev, MV_PRIMARY_BAR, 0); + mmio_base = ioremap_nocache(pci_resource_start(pdev, MV_PRIMARY_BAR), + pci_resource_len(pdev, MV_PRIMARY_BAR)); if (mmio_base == NULL) { rc = -ENOMEM; goto err_out_free_ent; @@ -1530,40 +769,37 @@ static int mv_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) if (rc) { goto err_out_hpriv; } +/* mv_print_info(probe_ent); */ - /* Enable interrupts */ - if (pci_enable_msi(pdev) == 0) { - hpriv->hp_flags |= MV_HP_FLAG_MSI; - } else { - pci_intx(pdev, 1); - } - - mv_dump_pci_cfg(pdev, 0x68); - mv_print_info(probe_ent); - - if (ata_device_add(probe_ent) == 0) { - rc = -ENODEV; /* No devices discovered */ - goto err_out_dev_add; + { + int b, w; + u32 dw[4]; /* hold a line of 16b */ + VPRINTK("PCI config space:\n"); + for (b = 0; b < 0x40; ) { + for (w = 0; w < 4; w++) { + (void) pci_read_config_dword(pdev,b,&dw[w]); + b += sizeof(*dw); + } + VPRINTK("%08x %08x %08x %08x\n", + dw[0],dw[1],dw[2],dw[3]); + } } + /* FIXME: check ata_device_add return value */ + ata_device_add(probe_ent); kfree(probe_ent); + return 0; -err_out_dev_add: - if (MV_HP_FLAG_MSI & hpriv->hp_flags) { - pci_disable_msi(pdev); - } else { - pci_intx(pdev, 0); - } -err_out_hpriv: + err_out_hpriv: kfree(hpriv); -err_out_iounmap: - pci_iounmap(pdev, mmio_base); -err_out_free_ent: + err_out_iounmap: + iounmap(mmio_base); + err_out_free_ent: kfree(probe_ent); -err_out_regions: + err_out_regions: pci_release_regions(pdev); -err_out: + err_out: if (!pci_dev_busy) { pci_disable_device(pdev); } diff --git a/trunk/drivers/scsi/sata_nv.c b/trunk/drivers/scsi/sata_nv.c index 749ff92d8c63..c05653c7779d 100644 --- a/trunk/drivers/scsi/sata_nv.c +++ b/trunk/drivers/scsi/sata_nv.c @@ -405,7 +405,7 @@ static int nv_init_one (struct pci_dev *pdev, const struct pci_device_id *ent) rc = -ENOMEM; ppi = &nv_port_info; - probe_ent = ata_pci_init_native_mode(pdev, &ppi, ATA_PORT_PRIMARY | ATA_PORT_SECONDARY); + probe_ent = ata_pci_init_native_mode(pdev, &ppi); if (!probe_ent) goto err_out_regions; diff --git a/trunk/drivers/scsi/sata_promise.c b/trunk/drivers/scsi/sata_promise.c index def7e0d9dacb..538ad727bd2e 100644 --- a/trunk/drivers/scsi/sata_promise.c +++ b/trunk/drivers/scsi/sata_promise.c @@ -438,11 +438,11 @@ static inline unsigned int pdc_host_intr( struct ata_port *ap, break; default: - ap->stats.idle_irq++; - break; + ap->stats.idle_irq++; + break; } - return handled; + return handled; } static void pdc_irq_clear(struct ata_port *ap) diff --git a/trunk/drivers/scsi/sata_sis.c b/trunk/drivers/scsi/sata_sis.c index 0761a3234fcf..b227e51d12f4 100644 --- a/trunk/drivers/scsi/sata_sis.c +++ b/trunk/drivers/scsi/sata_sis.c @@ -263,7 +263,7 @@ static int sis_init_one (struct pci_dev *pdev, const struct pci_device_id *ent) goto err_out_regions; ppi = &sis_port_info; - probe_ent = ata_pci_init_native_mode(pdev, &ppi, ATA_PORT_PRIMARY | ATA_PORT_SECONDARY); + probe_ent = ata_pci_init_native_mode(pdev, &ppi); if (!probe_ent) { rc = -ENOMEM; goto err_out_regions; diff --git a/trunk/drivers/scsi/sata_uli.c b/trunk/drivers/scsi/sata_uli.c index 9c06f2abe7f7..4c9fb8b71be1 100644 --- a/trunk/drivers/scsi/sata_uli.c +++ b/trunk/drivers/scsi/sata_uli.c @@ -202,7 +202,7 @@ static int uli_init_one (struct pci_dev *pdev, const struct pci_device_id *ent) goto err_out_regions; ppi = &uli_port_info; - probe_ent = ata_pci_init_native_mode(pdev, &ppi, ATA_PORT_PRIMARY | ATA_PORT_SECONDARY); + probe_ent = ata_pci_init_native_mode(pdev, &ppi); if (!probe_ent) { rc = -ENOMEM; goto err_out_regions; diff --git a/trunk/drivers/scsi/sata_via.c b/trunk/drivers/scsi/sata_via.c index 565872479b9a..128b996b07b7 100644 --- a/trunk/drivers/scsi/sata_via.c +++ b/trunk/drivers/scsi/sata_via.c @@ -212,7 +212,7 @@ static struct ata_probe_ent *vt6420_init_probe_ent(struct pci_dev *pdev) struct ata_probe_ent *probe_ent; struct ata_port_info *ppi = &svia_port_info; - probe_ent = ata_pci_init_native_mode(pdev, &ppi, ATA_PORT_PRIMARY | ATA_PORT_SECONDARY); + probe_ent = ata_pci_init_native_mode(pdev, &ppi); if (!probe_ent) return NULL; diff --git a/trunk/drivers/scsi/scsi_scan.c b/trunk/drivers/scsi/scsi_scan.c index fcf9f6cbb142..327c5d7e5bd2 100644 --- a/trunk/drivers/scsi/scsi_scan.c +++ b/trunk/drivers/scsi/scsi_scan.c @@ -587,6 +587,7 @@ static int scsi_probe_lun(struct scsi_device *sdev, char *inq_result, if (sdev->scsi_level >= 2 || (sdev->scsi_level == 1 && (inq_result[3] & 0x0f) == 1)) sdev->scsi_level++; + sdev->sdev_target->scsi_level = sdev->scsi_level; return 0; } @@ -771,6 +772,15 @@ static int scsi_add_lun(struct scsi_device *sdev, char *inq_result, int *bflags) return SCSI_SCAN_LUN_PRESENT; } +static inline void scsi_destroy_sdev(struct scsi_device *sdev) +{ + if (sdev->host->hostt->slave_destroy) + sdev->host->hostt->slave_destroy(sdev); + transport_destroy_device(&sdev->sdev_gendev); + put_device(&sdev->sdev_gendev); +} + + /** * scsi_probe_and_add_lun - probe a LUN, if a LUN is found add it * @starget: pointer to target device structure @@ -803,9 +813,9 @@ static int scsi_probe_and_add_lun(struct scsi_target *starget, * The rescan flag is used as an optimization, the first scan of a * host adapter calls into here with rescan == 0. */ - if (rescan) { - sdev = scsi_device_lookup_by_target(starget, lun); - if (sdev) { + sdev = scsi_device_lookup_by_target(starget, lun); + if (sdev) { + if (rescan || sdev->sdev_state != SDEV_CREATED) { SCSI_LOG_SCAN_BUS(3, printk(KERN_INFO "scsi scan: device exists on %s\n", sdev->sdev_gendev.bus_id)); @@ -820,9 +830,9 @@ static int scsi_probe_and_add_lun(struct scsi_target *starget, sdev->model); return SCSI_SCAN_LUN_PRESENT; } - } - - sdev = scsi_alloc_sdev(starget, lun, hostdata); + scsi_device_put(sdev); + } else + sdev = scsi_alloc_sdev(starget, lun, hostdata); if (!sdev) goto out; @@ -877,12 +887,8 @@ static int scsi_probe_and_add_lun(struct scsi_target *starget, res = SCSI_SCAN_NO_RESPONSE; } } - } else { - if (sdev->host->hostt->slave_destroy) - sdev->host->hostt->slave_destroy(sdev); - transport_destroy_device(&sdev->sdev_gendev); - put_device(&sdev->sdev_gendev); - } + } else + scsi_destroy_sdev(sdev); out: return res; } @@ -1054,7 +1060,7 @@ EXPORT_SYMBOL(int_to_scsilun); * 0: scan completed (or no memory, so further scanning is futile) * 1: no report lun scan, or not configured **/ -static int scsi_report_lun_scan(struct scsi_device *sdev, int bflags, +static int scsi_report_lun_scan(struct scsi_target *starget, int bflags, int rescan) { char devname[64]; @@ -1067,7 +1073,8 @@ static int scsi_report_lun_scan(struct scsi_device *sdev, int bflags, struct scsi_lun *lunp, *lun_data; u8 *data; struct scsi_sense_hdr sshdr; - struct scsi_target *starget = scsi_target(sdev); + struct scsi_device *sdev; + struct Scsi_Host *shost = dev_to_shost(&starget->dev); /* * Only support SCSI-3 and up devices if BLIST_NOREPORTLUN is not set. @@ -1075,15 +1082,23 @@ static int scsi_report_lun_scan(struct scsi_device *sdev, int bflags, * support more than 8 LUNs. */ if ((bflags & BLIST_NOREPORTLUN) || - sdev->scsi_level < SCSI_2 || - (sdev->scsi_level < SCSI_3 && - (!(bflags & BLIST_REPORTLUN2) || sdev->host->max_lun <= 8)) ) + starget->scsi_level < SCSI_2 || + (starget->scsi_level < SCSI_3 && + (!(bflags & BLIST_REPORTLUN2) || shost->max_lun <= 8)) ) return 1; if (bflags & BLIST_NOLUN) return 0; + if (!(sdev = scsi_device_lookup_by_target(starget, 0))) { + sdev = scsi_alloc_sdev(starget, 0, NULL); + if (!sdev) + return 0; + if (scsi_device_get(sdev)) + return 0; + } + sprintf(devname, "host %d channel %d id %d", - sdev->host->host_no, sdev->channel, sdev->id); + shost->host_no, sdev->channel, sdev->id); /* * Allocate enough to hold the header (the same size as one scsi_lun) @@ -1098,8 +1113,10 @@ static int scsi_report_lun_scan(struct scsi_device *sdev, int bflags, length = (max_scsi_report_luns + 1) * sizeof(struct scsi_lun); lun_data = kmalloc(length, GFP_ATOMIC | (sdev->host->unchecked_isa_dma ? __GFP_DMA : 0)); - if (!lun_data) + if (!lun_data) { + printk(ALLOC_FAILURE_MSG, __FUNCTION__); goto out; + } scsi_cmd[0] = REPORT_LUNS; @@ -1201,10 +1218,6 @@ static int scsi_report_lun_scan(struct scsi_device *sdev, int bflags, for (i = 0; i < sizeof(struct scsi_lun); i++) printk("%02x", data[i]); printk(" has a LUN larger than currently supported.\n"); - } else if (lun == 0) { - /* - * LUN 0 has already been scanned. - */ } else if (lun > sdev->host->max_lun) { printk(KERN_WARNING "scsi: %s lun%d has a LUN larger" " than allowed by the host adapter\n", @@ -1227,13 +1240,13 @@ static int scsi_report_lun_scan(struct scsi_device *sdev, int bflags, } kfree(lun_data); - return 0; - out: - /* - * We are out of memory, don't try scanning any further. - */ - printk(ALLOC_FAILURE_MSG, __FUNCTION__); + scsi_device_put(sdev); + if (sdev->sdev_state == SDEV_CREATED) + /* + * the sdev we used didn't appear in the report luns scan + */ + scsi_destroy_sdev(sdev); return 0; } @@ -1299,7 +1312,6 @@ static void __scsi_scan_target(struct device *parent, unsigned int channel, struct Scsi_Host *shost = dev_to_shost(parent); int bflags = 0; int res; - struct scsi_device *sdev = NULL; struct scsi_target *starget; if (shost->this_id == id) @@ -1325,27 +1337,16 @@ static void __scsi_scan_target(struct device *parent, unsigned int channel, * Scan LUN 0, if there is some response, scan further. Ideally, we * would not configure LUN 0 until all LUNs are scanned. */ - res = scsi_probe_and_add_lun(starget, 0, &bflags, &sdev, rescan, NULL); - if (res == SCSI_SCAN_LUN_PRESENT) { - if (scsi_report_lun_scan(sdev, bflags, rescan) != 0) + res = scsi_probe_and_add_lun(starget, 0, &bflags, NULL, rescan, NULL); + if (res == SCSI_SCAN_LUN_PRESENT || res == SCSI_SCAN_TARGET_PRESENT) { + if (scsi_report_lun_scan(starget, bflags, rescan) != 0) /* * The REPORT LUN did not scan the target, * do a sequential scan. */ scsi_sequential_lun_scan(starget, bflags, - res, sdev->scsi_level, rescan); - } else if (res == SCSI_SCAN_TARGET_PRESENT) { - /* - * There's a target here, but lun 0 is offline so we - * can't use the report_lun scan. Fall back to a - * sequential lun scan with a bflags of SPARSELUN and - * a default scsi level of SCSI_2 - */ - scsi_sequential_lun_scan(starget, BLIST_SPARSELUN, - SCSI_SCAN_TARGET_PRESENT, SCSI_2, rescan); + res, starget->scsi_level, rescan); } - if (sdev) - scsi_device_put(sdev); out_reap: /* now determine if the target has any children at all @@ -1542,10 +1543,7 @@ void scsi_free_host_dev(struct scsi_device *sdev) { BUG_ON(sdev->id != sdev->host->this_id); - if (sdev->host->hostt->slave_destroy) - sdev->host->hostt->slave_destroy(sdev); - transport_destroy_device(&sdev->sdev_gendev); - put_device(&sdev->sdev_gendev); + scsi_destroy_sdev(sdev); } EXPORT_SYMBOL(scsi_free_host_dev); diff --git a/trunk/drivers/scsi/scsi_transport_sas.c b/trunk/drivers/scsi/scsi_transport_sas.c index ff724bbe6611..1d145d2f9a38 100644 --- a/trunk/drivers/scsi/scsi_transport_sas.c +++ b/trunk/drivers/scsi/scsi_transport_sas.c @@ -628,17 +628,16 @@ sas_rphy_delete(struct sas_rphy *rphy) struct Scsi_Host *shost = dev_to_shost(parent->dev.parent); struct sas_host_attrs *sas_host = to_sas_host_attrs(shost); - transport_destroy_device(&rphy->dev); + scsi_remove_target(dev); - scsi_remove_target(&rphy->dev); + transport_remove_device(dev); + device_del(dev); + transport_destroy_device(dev); spin_lock(&sas_host->lock); list_del(&rphy->list); spin_unlock(&sas_host->lock); - transport_remove_device(dev); - device_del(dev); - transport_destroy_device(dev); put_device(&parent->dev); } EXPORT_SYMBOL(sas_rphy_delete); diff --git a/trunk/include/linux/ata.h b/trunk/include/linux/ata.h index ecb7346d0c16..a5b74efab067 100644 --- a/trunk/include/linux/ata.h +++ b/trunk/include/linux/ata.h @@ -132,7 +132,6 @@ enum { ATA_CMD_PACKET = 0xA0, ATA_CMD_VERIFY = 0x40, ATA_CMD_VERIFY_EXT = 0x42, - ATA_CMD_INIT_DEV_PARAMS = 0x91, /* SETFEATURES stuff */ SETFEATURES_XFER = 0x03, @@ -147,14 +146,14 @@ enum { XFER_MW_DMA_2 = 0x22, XFER_MW_DMA_1 = 0x21, XFER_MW_DMA_0 = 0x20, - XFER_SW_DMA_2 = 0x12, - XFER_SW_DMA_1 = 0x11, - XFER_SW_DMA_0 = 0x10, XFER_PIO_4 = 0x0C, XFER_PIO_3 = 0x0B, XFER_PIO_2 = 0x0A, XFER_PIO_1 = 0x09, XFER_PIO_0 = 0x08, + XFER_SW_DMA_2 = 0x12, + XFER_SW_DMA_1 = 0x11, + XFER_SW_DMA_0 = 0x10, XFER_PIO_SLOW = 0x00, /* ATAPI stuff */ @@ -182,7 +181,6 @@ enum { ATA_TFLAG_ISADDR = (1 << 1), /* enable r/w to nsect/lba regs */ ATA_TFLAG_DEVICE = (1 << 2), /* enable r/w to device reg */ ATA_TFLAG_WRITE = (1 << 3), /* data dir: host->dev==1 (write) */ - ATA_TFLAG_LBA = (1 << 4), /* enable LBA */ }; enum ata_tf_protocols { @@ -252,18 +250,6 @@ struct ata_taskfile { ((u64) (id)[(n) + 1] << 16) | \ ((u64) (id)[(n) + 0]) ) -static inline int ata_id_current_chs_valid(u16 *id) -{ - /* For ATA-1 devices, if the INITIALIZE DEVICE PARAMETERS command - has not been issued to the device then the values of - id[54] to id[56] are vendor specific. */ - return (id[53] & 0x01) && /* Current translation valid */ - id[54] && /* cylinders in current translation */ - id[55] && /* heads in current translation */ - id[55] <= 16 && - id[56]; /* sectors in current translation */ -} - static inline int atapi_cdb_len(u16 *dev_id) { u16 tmp = dev_id[0] & 0x3; diff --git a/trunk/include/linux/inetdevice.h b/trunk/include/linux/inetdevice.h index 7e1e15f934f3..fd7af86151b1 100644 --- a/trunk/include/linux/inetdevice.h +++ b/trunk/include/linux/inetdevice.h @@ -142,13 +142,21 @@ static __inline__ int bad_mask(u32 mask, u32 addr) #define endfor_ifa(in_dev) } +static inline struct in_device *__in_dev_get_rcu(const struct net_device *dev) +{ + struct in_device *in_dev = dev->ip_ptr; + if (in_dev) + in_dev = rcu_dereference(in_dev); + return in_dev; +} + static __inline__ struct in_device * in_dev_get(const struct net_device *dev) { struct in_device *in_dev; rcu_read_lock(); - in_dev = dev->ip_ptr; + in_dev = __in_dev_get_rcu(dev); if (in_dev) atomic_inc(&in_dev->refcnt); rcu_read_unlock(); @@ -156,7 +164,7 @@ in_dev_get(const struct net_device *dev) } static __inline__ struct in_device * -__in_dev_get(const struct net_device *dev) +__in_dev_get_rtnl(const struct net_device *dev) { return (struct in_device*)dev->ip_ptr; } diff --git a/trunk/include/linux/ipv6.h b/trunk/include/linux/ipv6.h index bb6f88e14061..e0b922785d98 100644 --- a/trunk/include/linux/ipv6.h +++ b/trunk/include/linux/ipv6.h @@ -372,8 +372,9 @@ static inline struct raw6_sock *raw6_sk(const struct sock *sk) #define inet_v6_ipv6only(__sk) 0 #endif /* defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) */ -#define INET6_MATCH(__sk, __saddr, __daddr, __ports, __dif) \ - (((*((__u32 *)&(inet_sk(__sk)->dport))) == (__ports)) && \ +#define INET6_MATCH(__sk, __hash, __saddr, __daddr, __ports, __dif)\ + (((__sk)->sk_hash == (__hash)) && \ + ((*((__u32 *)&(inet_sk(__sk)->dport))) == (__ports)) && \ ((__sk)->sk_family == AF_INET6) && \ ipv6_addr_equal(&inet6_sk(__sk)->daddr, (__saddr)) && \ ipv6_addr_equal(&inet6_sk(__sk)->rcv_saddr, (__daddr)) && \ diff --git a/trunk/include/linux/libata.h b/trunk/include/linux/libata.h index 7929cfc9318d..ceee1fc42c60 100644 --- a/trunk/include/linux/libata.h +++ b/trunk/include/linux/libata.h @@ -97,7 +97,6 @@ enum { ATA_DFLAG_LBA48 = (1 << 0), /* device supports LBA48 */ ATA_DFLAG_PIO = (1 << 1), /* device currently in PIO mode */ ATA_DFLAG_LOCK_SECTORS = (1 << 2), /* don't adjust max_sectors */ - ATA_DFLAG_LBA = (1 << 3), /* device supports LBA */ ATA_DEV_UNKNOWN = 0, /* unknown device */ ATA_DEV_ATA = 1, /* ATA device */ @@ -155,21 +154,17 @@ enum { ATA_SHIFT_UDMA = 0, ATA_SHIFT_MWDMA = 8, ATA_SHIFT_PIO = 11, - - /* Masks for port functions */ - ATA_PORT_PRIMARY = (1 << 0), - ATA_PORT_SECONDARY = (1 << 1), }; -enum hsm_task_states { - HSM_ST_UNKNOWN, - HSM_ST_IDLE, - HSM_ST_POLL, - HSM_ST_TMOUT, - HSM_ST, - HSM_ST_LAST, - HSM_ST_LAST_POLL, - HSM_ST_ERR, +enum pio_task_states { + PIO_ST_UNKNOWN, + PIO_ST_IDLE, + PIO_ST_POLL, + PIO_ST_TMOUT, + PIO_ST, + PIO_ST_LAST, + PIO_ST_LAST_POLL, + PIO_ST_ERR, }; /* forward declarations */ @@ -287,11 +282,6 @@ struct ata_device { u8 xfer_protocol; /* taskfile xfer protocol */ u8 read_cmd; /* opcode to use on read */ u8 write_cmd; /* opcode to use on write */ - - /* for CHS addressing */ - u16 cylinders; /* Number of cylinders */ - u16 heads; /* Number of heads */ - u16 sectors; /* Number of sectors per track */ }; struct ata_port { @@ -329,7 +319,7 @@ struct ata_port { struct work_struct packet_task; struct work_struct pio_task; - unsigned int hsm_task_state; + unsigned int pio_task_state; unsigned long pio_task_timeout; void *private_data; @@ -410,8 +400,6 @@ extern int ata_scsi_queuecmd(struct scsi_cmnd *cmd, void (*done)(struct scsi_cmn extern int ata_scsi_error(struct Scsi_Host *host); extern int ata_scsi_release(struct Scsi_Host *host); extern unsigned int ata_host_intr(struct ata_port *ap, struct ata_queued_cmd *qc); -extern int ata_ratelimit(void); - /* * Default driver ops implementations */ @@ -464,7 +452,7 @@ struct pci_bits { extern void ata_pci_host_stop (struct ata_host_set *host_set); extern struct ata_probe_ent * -ata_pci_init_native_mode(struct pci_dev *pdev, struct ata_port_info **port, int portmask); +ata_pci_init_native_mode(struct pci_dev *pdev, struct ata_port_info **port); extern int pci_test_config_bits(struct pci_dev *pdev, struct pci_bits *bits); #endif /* CONFIG_PCI */ diff --git a/trunk/include/linux/pci_ids.h b/trunk/include/linux/pci_ids.h index eb36fd293b41..f74ed9462475 100644 --- a/trunk/include/linux/pci_ids.h +++ b/trunk/include/linux/pci_ids.h @@ -185,6 +185,7 @@ #define PCI_DEVICE_ID_LSI_61C102 0x0901 #define PCI_DEVICE_ID_LSI_63C815 0x1000 #define PCI_DEVICE_ID_LSI_SAS1064 0x0050 +#define PCI_DEVICE_ID_LSI_SAS1064R 0x0411 #define PCI_DEVICE_ID_LSI_SAS1066 0x005E #define PCI_DEVICE_ID_LSI_SAS1068 0x0054 #define PCI_DEVICE_ID_LSI_SAS1064A 0x005C @@ -560,6 +561,7 @@ #define PCI_VENDOR_ID_DELL 0x1028 #define PCI_DEVICE_ID_DELL_RACIII 0x0008 #define PCI_DEVICE_ID_DELL_RAC4 0x0012 +#define PCI_DEVICE_ID_DELL_PERC5 0x0015 #define PCI_VENDOR_ID_MATROX 0x102B #define PCI_DEVICE_ID_MATROX_MGA_2 0x0518 diff --git a/trunk/include/linux/skbuff.h b/trunk/include/linux/skbuff.h index 2741c0c55e83..466c879f82b8 100644 --- a/trunk/include/linux/skbuff.h +++ b/trunk/include/linux/skbuff.h @@ -155,8 +155,6 @@ struct skb_shared_info { #define SKB_DATAREF_SHIFT 16 #define SKB_DATAREF_MASK ((1 << SKB_DATAREF_SHIFT) - 1) -extern struct timeval skb_tv_base; - struct skb_timeval { u32 off_sec; u32 off_usec; @@ -175,7 +173,7 @@ enum { * @prev: Previous buffer in list * @list: List we are on * @sk: Socket we are owned by - * @tstamp: Time we arrived stored as offset to skb_tv_base + * @tstamp: Time we arrived * @dev: Device we arrived on/are leaving by * @input_dev: Device we arrived on * @h: Transport layer header @@ -1255,10 +1253,6 @@ static inline void skb_get_timestamp(const struct sk_buff *skb, struct timeval * { stamp->tv_sec = skb->tstamp.off_sec; stamp->tv_usec = skb->tstamp.off_usec; - if (skb->tstamp.off_sec) { - stamp->tv_sec += skb_tv_base.tv_sec; - stamp->tv_usec += skb_tv_base.tv_usec; - } } /** @@ -1272,8 +1266,8 @@ static inline void skb_get_timestamp(const struct sk_buff *skb, struct timeval * */ static inline void skb_set_timestamp(struct sk_buff *skb, const struct timeval *stamp) { - skb->tstamp.off_sec = stamp->tv_sec - skb_tv_base.tv_sec; - skb->tstamp.off_usec = stamp->tv_usec - skb_tv_base.tv_usec; + skb->tstamp.off_sec = stamp->tv_sec; + skb->tstamp.off_usec = stamp->tv_usec; } extern void __net_timestamp(struct sk_buff *skb); diff --git a/trunk/include/linux/tc_ematch/tc_em_meta.h b/trunk/include/linux/tc_ematch/tc_em_meta.h index 081b1ee8516e..e21937cf91d0 100644 --- a/trunk/include/linux/tc_ematch/tc_em_meta.h +++ b/trunk/include/linux/tc_ematch/tc_em_meta.h @@ -71,7 +71,7 @@ enum TCF_META_ID_SK_SNDBUF, TCF_META_ID_SK_ALLOCS, TCF_META_ID_SK_ROUTE_CAPS, - TCF_META_ID_SK_HASHENT, + TCF_META_ID_SK_HASH, TCF_META_ID_SK_LINGERTIME, TCF_META_ID_SK_ACK_BACKLOG, TCF_META_ID_SK_MAX_ACK_BACKLOG, diff --git a/trunk/include/net/inet6_hashtables.h b/trunk/include/net/inet6_hashtables.h index 03df3b157960..5a2beed5a770 100644 --- a/trunk/include/net/inet6_hashtables.h +++ b/trunk/include/net/inet6_hashtables.h @@ -26,19 +26,18 @@ struct inet_hashinfo; /* I have no idea if this is a good hash for v6 or not. -DaveM */ -static inline int inet6_ehashfn(const struct in6_addr *laddr, const u16 lport, - const struct in6_addr *faddr, const u16 fport, - const int ehash_size) +static inline unsigned int inet6_ehashfn(const struct in6_addr *laddr, const u16 lport, + const struct in6_addr *faddr, const u16 fport) { - int hashent = (lport ^ fport); + unsigned int hashent = (lport ^ fport); hashent ^= (laddr->s6_addr32[3] ^ faddr->s6_addr32[3]); hashent ^= hashent >> 16; hashent ^= hashent >> 8; - return (hashent & (ehash_size - 1)); + return hashent; } -static inline int inet6_sk_ehashfn(const struct sock *sk, const int ehash_size) +static inline int inet6_sk_ehashfn(const struct sock *sk) { const struct inet_sock *inet = inet_sk(sk); const struct ipv6_pinfo *np = inet6_sk(sk); @@ -46,7 +45,7 @@ static inline int inet6_sk_ehashfn(const struct sock *sk, const int ehash_size) const struct in6_addr *faddr = &np->daddr; const __u16 lport = inet->num; const __u16 fport = inet->dport; - return inet6_ehashfn(laddr, lport, faddr, fport, ehash_size); + return inet6_ehashfn(laddr, lport, faddr, fport); } /* @@ -69,14 +68,14 @@ static inline struct sock * /* Optimize here for direct hit, only listening connections can * have wildcards anyways. */ - const int hash = inet6_ehashfn(daddr, hnum, saddr, sport, - hashinfo->ehash_size); - struct inet_ehash_bucket *head = &hashinfo->ehash[hash]; + unsigned int hash = inet6_ehashfn(daddr, hnum, saddr, sport); + struct inet_ehash_bucket *head = inet_ehash_bucket(hashinfo, hash); + prefetch(head->chain.first); read_lock(&head->lock); sk_for_each(sk, node, &head->chain) { /* For IPV6 do the cheaper port and family tests first. */ - if (INET6_MATCH(sk, saddr, daddr, ports, dif)) + if (INET6_MATCH(sk, hash, saddr, daddr, ports, dif)) goto hit; /* You sunk my battleship! */ } /* Must check for a TIME_WAIT'er before going to listener hash. */ diff --git a/trunk/include/net/inet_hashtables.h b/trunk/include/net/inet_hashtables.h index 646b6ea7fe26..35f49e65e295 100644 --- a/trunk/include/net/inet_hashtables.h +++ b/trunk/include/net/inet_hashtables.h @@ -108,7 +108,7 @@ struct inet_hashinfo { struct inet_bind_hashbucket *bhash; int bhash_size; - int ehash_size; + unsigned int ehash_size; /* All sockets in TCP_LISTEN state will be in here. This is the only * table where wildcard'd TCP sockets can exist. Hash function here @@ -130,17 +130,16 @@ struct inet_hashinfo { int port_rover; }; -static inline int inet_ehashfn(const __u32 laddr, const __u16 lport, - const __u32 faddr, const __u16 fport, - const int ehash_size) +static inline unsigned int inet_ehashfn(const __u32 laddr, const __u16 lport, + const __u32 faddr, const __u16 fport) { - int h = (laddr ^ lport) ^ (faddr ^ fport); + unsigned int h = (laddr ^ lport) ^ (faddr ^ fport); h ^= h >> 16; h ^= h >> 8; - return h & (ehash_size - 1); + return h; } -static inline int inet_sk_ehashfn(const struct sock *sk, const int ehash_size) +static inline int inet_sk_ehashfn(const struct sock *sk) { const struct inet_sock *inet = inet_sk(sk); const __u32 laddr = inet->rcv_saddr; @@ -148,7 +147,14 @@ static inline int inet_sk_ehashfn(const struct sock *sk, const int ehash_size) const __u32 faddr = inet->daddr; const __u16 fport = inet->dport; - return inet_ehashfn(laddr, lport, faddr, fport, ehash_size); + return inet_ehashfn(laddr, lport, faddr, fport); +} + +static inline struct inet_ehash_bucket *inet_ehash_bucket( + struct inet_hashinfo *hashinfo, + unsigned int hash) +{ + return &hashinfo->ehash[hash & (hashinfo->ehash_size - 1)]; } extern struct inet_bind_bucket * @@ -235,9 +241,11 @@ static inline void __inet_hash(struct inet_hashinfo *hashinfo, lock = &hashinfo->lhash_lock; inet_listen_wlock(hashinfo); } else { - sk->sk_hashent = inet_sk_ehashfn(sk, hashinfo->ehash_size); - list = &hashinfo->ehash[sk->sk_hashent].chain; - lock = &hashinfo->ehash[sk->sk_hashent].lock; + struct inet_ehash_bucket *head; + sk->sk_hash = inet_sk_ehashfn(sk); + head = inet_ehash_bucket(hashinfo, sk->sk_hash); + list = &head->chain; + lock = &head->lock; write_lock(lock); } __sk_add_node(sk, list); @@ -268,9 +276,8 @@ static inline void inet_unhash(struct inet_hashinfo *hashinfo, struct sock *sk) inet_listen_wlock(hashinfo); lock = &hashinfo->lhash_lock; } else { - struct inet_ehash_bucket *head = &hashinfo->ehash[sk->sk_hashent]; - lock = &head->lock; - write_lock_bh(&head->lock); + lock = &inet_ehash_bucket(hashinfo, sk->sk_hash)->lock; + write_lock_bh(lock); } if (__sk_del_node_init(sk)) @@ -337,23 +344,27 @@ static inline struct sock * #define INET_ADDR_COOKIE(__name, __saddr, __daddr) \ const __u64 __name = (((__u64)(__daddr)) << 32) | ((__u64)(__saddr)); #endif /* __BIG_ENDIAN */ -#define INET_MATCH(__sk, __cookie, __saddr, __daddr, __ports, __dif)\ - (((*((__u64 *)&(inet_sk(__sk)->daddr))) == (__cookie)) && \ +#define INET_MATCH(__sk, __hash, __cookie, __saddr, __daddr, __ports, __dif)\ + (((__sk)->sk_hash == (__hash)) && \ + ((*((__u64 *)&(inet_sk(__sk)->daddr))) == (__cookie)) && \ ((*((__u32 *)&(inet_sk(__sk)->dport))) == (__ports)) && \ (!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif)))) -#define INET_TW_MATCH(__sk, __cookie, __saddr, __daddr, __ports, __dif)\ - (((*((__u64 *)&(inet_twsk(__sk)->tw_daddr))) == (__cookie)) && \ +#define INET_TW_MATCH(__sk, __hash, __cookie, __saddr, __daddr, __ports, __dif)\ + (((__sk)->sk_hash == (__hash)) && \ + ((*((__u64 *)&(inet_twsk(__sk)->tw_daddr))) == (__cookie)) && \ ((*((__u32 *)&(inet_twsk(__sk)->tw_dport))) == (__ports)) && \ (!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif)))) #else /* 32-bit arch */ #define INET_ADDR_COOKIE(__name, __saddr, __daddr) -#define INET_MATCH(__sk, __cookie, __saddr, __daddr, __ports, __dif) \ - ((inet_sk(__sk)->daddr == (__saddr)) && \ +#define INET_MATCH(__sk, __hash, __cookie, __saddr, __daddr, __ports, __dif) \ + (((__sk)->sk_hash == (__hash)) && \ + (inet_sk(__sk)->daddr == (__saddr)) && \ (inet_sk(__sk)->rcv_saddr == (__daddr)) && \ ((*((__u32 *)&(inet_sk(__sk)->dport))) == (__ports)) && \ (!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif)))) -#define INET_TW_MATCH(__sk, __cookie, __saddr, __daddr, __ports, __dif) \ - ((inet_twsk(__sk)->tw_daddr == (__saddr)) && \ +#define INET_TW_MATCH(__sk, __hash,__cookie, __saddr, __daddr, __ports, __dif) \ + (((__sk)->sk_hash == (__hash)) && \ + (inet_twsk(__sk)->tw_daddr == (__saddr)) && \ (inet_twsk(__sk)->tw_rcv_saddr == (__daddr)) && \ ((*((__u32 *)&(inet_twsk(__sk)->tw_dport))) == (__ports)) && \ (!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif)))) @@ -378,18 +389,19 @@ static inline struct sock * /* Optimize here for direct hit, only listening connections can * have wildcards anyways. */ - const int hash = inet_ehashfn(daddr, hnum, saddr, sport, hashinfo->ehash_size); - struct inet_ehash_bucket *head = &hashinfo->ehash[hash]; + unsigned int hash = inet_ehashfn(daddr, hnum, saddr, sport); + struct inet_ehash_bucket *head = inet_ehash_bucket(hashinfo, hash); + prefetch(head->chain.first); read_lock(&head->lock); sk_for_each(sk, node, &head->chain) { - if (INET_MATCH(sk, acookie, saddr, daddr, ports, dif)) + if (INET_MATCH(sk, hash, acookie, saddr, daddr, ports, dif)) goto hit; /* You sunk my battleship! */ } /* Must check for a TIME_WAIT'er before going to listener hash. */ sk_for_each(sk, node, &(head + hashinfo->ehash_size)->chain) { - if (INET_TW_MATCH(sk, acookie, saddr, daddr, ports, dif)) + if (INET_TW_MATCH(sk, hash, acookie, saddr, daddr, ports, dif)) goto hit; } sk = NULL; diff --git a/trunk/include/net/inet_timewait_sock.h b/trunk/include/net/inet_timewait_sock.h index 3b070352e869..4ade56ef3a4d 100644 --- a/trunk/include/net/inet_timewait_sock.h +++ b/trunk/include/net/inet_timewait_sock.h @@ -112,6 +112,7 @@ struct inet_timewait_sock { #define tw_node __tw_common.skc_node #define tw_bind_node __tw_common.skc_bind_node #define tw_refcnt __tw_common.skc_refcnt +#define tw_hash __tw_common.skc_hash #define tw_prot __tw_common.skc_prot volatile unsigned char tw_substate; /* 3 bits hole, try to pack */ @@ -126,7 +127,6 @@ struct inet_timewait_sock { /* And these are ours. */ __u8 tw_ipv6only:1; /* 31 bits hole, try to pack */ - int tw_hashent; int tw_timeout; unsigned long tw_ttd; struct inet_bind_bucket *tw_tb; diff --git a/trunk/include/net/sock.h b/trunk/include/net/sock.h index 8c48fbecb7cf..b6440805c420 100644 --- a/trunk/include/net/sock.h +++ b/trunk/include/net/sock.h @@ -99,6 +99,7 @@ struct proto; * @skc_node: main hash linkage for various protocol lookup tables * @skc_bind_node: bind hash linkage for various protocol lookup tables * @skc_refcnt: reference count + * @skc_hash: hash value used with various protocol lookup tables * @skc_prot: protocol handlers inside a network family * * This is the minimal network layer representation of sockets, the header @@ -112,6 +113,7 @@ struct sock_common { struct hlist_node skc_node; struct hlist_node skc_bind_node; atomic_t skc_refcnt; + unsigned int skc_hash; struct proto *skc_prot; }; @@ -139,7 +141,6 @@ struct sock_common { * @sk_no_check: %SO_NO_CHECK setting, wether or not checkup packets * @sk_route_caps: route capabilities (e.g. %NETIF_F_TSO) * @sk_lingertime: %SO_LINGER l_linger setting - * @sk_hashent: hash entry in several tables (e.g. inet_hashinfo.ehash) * @sk_backlog: always used with the per-socket spinlock held * @sk_callback_lock: used with the callbacks in the end of this struct * @sk_error_queue: rarely used @@ -186,6 +187,7 @@ struct sock { #define sk_node __sk_common.skc_node #define sk_bind_node __sk_common.skc_bind_node #define sk_refcnt __sk_common.skc_refcnt +#define sk_hash __sk_common.skc_hash #define sk_prot __sk_common.skc_prot unsigned char sk_shutdown : 2, sk_no_check : 2, @@ -208,7 +210,6 @@ struct sock { unsigned int sk_allocation; int sk_sndbuf; int sk_route_caps; - int sk_hashent; unsigned long sk_flags; unsigned long sk_lingertime; /* diff --git a/trunk/include/scsi/scsi_device.h b/trunk/include/scsi/scsi_device.h index c0e4c67d836f..7ece05666feb 100644 --- a/trunk/include/scsi/scsi_device.h +++ b/trunk/include/scsi/scsi_device.h @@ -163,6 +163,7 @@ struct scsi_target { unsigned int id; /* target id ... replace * scsi_device.id eventually */ unsigned long create:1; /* signal that it needs to be added */ + char scsi_level; void *hostdata; /* available to low-level driver */ unsigned long starget_data[0]; /* for the transport */ /* starget_data must be the last element!!!! */ diff --git a/trunk/net/atm/clip.c b/trunk/net/atm/clip.c index 28dab55a4387..4f54c9a5e84a 100644 --- a/trunk/net/atm/clip.c +++ b/trunk/net/atm/clip.c @@ -310,7 +310,7 @@ static int clip_constructor(struct neighbour *neigh) if (neigh->type != RTN_UNICAST) return -EINVAL; rcu_read_lock(); - in_dev = rcu_dereference(__in_dev_get(dev)); + in_dev = __in_dev_get_rcu(dev); if (!in_dev) { rcu_read_unlock(); return -EINVAL; diff --git a/trunk/net/atm/common.c b/trunk/net/atm/common.c index 801a5813ec60..63feea49fb13 100644 --- a/trunk/net/atm/common.c +++ b/trunk/net/atm/common.c @@ -46,7 +46,7 @@ static void __vcc_insert_socket(struct sock *sk) struct atm_vcc *vcc = atm_sk(sk); struct hlist_head *head = &vcc_hash[vcc->vci & (VCC_HTABLE_SIZE - 1)]; - sk->sk_hashent = vcc->vci & (VCC_HTABLE_SIZE - 1); + sk->sk_hash = vcc->vci & (VCC_HTABLE_SIZE - 1); sk_add_node(sk, head); } diff --git a/trunk/net/core/netpoll.c b/trunk/net/core/netpoll.c index 5265dfd69928..802fe11efad0 100644 --- a/trunk/net/core/netpoll.c +++ b/trunk/net/core/netpoll.c @@ -703,7 +703,7 @@ int netpoll_setup(struct netpoll *np) if (!np->local_ip) { rcu_read_lock(); - in_dev = __in_dev_get(ndev); + in_dev = __in_dev_get_rcu(ndev); if (!in_dev || !in_dev->ifa_list) { rcu_read_unlock(); diff --git a/trunk/net/core/pktgen.c b/trunk/net/core/pktgen.c index b7f2d65a614f..5f043d346694 100644 --- a/trunk/net/core/pktgen.c +++ b/trunk/net/core/pktgen.c @@ -1667,13 +1667,12 @@ static void pktgen_setup_inject(struct pktgen_dev *pkt_dev) struct in_device *in_dev; rcu_read_lock(); - in_dev = __in_dev_get(pkt_dev->odev); + in_dev = __in_dev_get_rcu(pkt_dev->odev); if (in_dev) { if (in_dev->ifa_list) { pkt_dev->saddr_min = in_dev->ifa_list->ifa_address; pkt_dev->saddr_max = pkt_dev->saddr_min; } - __in_dev_put(in_dev); } rcu_read_unlock(); } diff --git a/trunk/net/core/skbuff.c b/trunk/net/core/skbuff.c index f80a28785610..0e9431b59fb2 100644 --- a/trunk/net/core/skbuff.c +++ b/trunk/net/core/skbuff.c @@ -71,8 +71,6 @@ static kmem_cache_t *skbuff_head_cache __read_mostly; static kmem_cache_t *skbuff_fclone_cache __read_mostly; -struct timeval __read_mostly skb_tv_base; - /* * Keep out-of-line to prevent kernel bloat. * __builtin_return_address is not used because it is not always @@ -1708,8 +1706,6 @@ void __init skb_init(void) NULL, NULL); if (!skbuff_fclone_cache) panic("cannot create skbuff cache"); - - do_gettimeofday(&skb_tv_base); } EXPORT_SYMBOL(___pskb_trim); @@ -1743,4 +1739,3 @@ EXPORT_SYMBOL(skb_prepare_seq_read); EXPORT_SYMBOL(skb_seq_read); EXPORT_SYMBOL(skb_abort_seq_read); EXPORT_SYMBOL(skb_find_text); -EXPORT_SYMBOL(skb_tv_base); diff --git a/trunk/net/dccp/ipv4.c b/trunk/net/dccp/ipv4.c index 40fe6afacde6..ae088d1347af 100644 --- a/trunk/net/dccp/ipv4.c +++ b/trunk/net/dccp/ipv4.c @@ -62,27 +62,27 @@ static int __dccp_v4_check_established(struct sock *sk, const __u16 lport, const int dif = sk->sk_bound_dev_if; INET_ADDR_COOKIE(acookie, saddr, daddr) const __u32 ports = INET_COMBINED_PORTS(inet->dport, lport); - const int hash = inet_ehashfn(daddr, lport, saddr, inet->dport, - dccp_hashinfo.ehash_size); - struct inet_ehash_bucket *head = &dccp_hashinfo.ehash[hash]; + unsigned int hash = inet_ehashfn(daddr, lport, saddr, inet->dport); + struct inet_ehash_bucket *head = inet_ehash_bucket(&dccp_hashinfo, hash); const struct sock *sk2; const struct hlist_node *node; struct inet_timewait_sock *tw; + prefetch(head->chain.first); write_lock(&head->lock); /* Check TIME-WAIT sockets first. */ sk_for_each(sk2, node, &(head + dccp_hashinfo.ehash_size)->chain) { tw = inet_twsk(sk2); - if (INET_TW_MATCH(sk2, acookie, saddr, daddr, ports, dif)) + if (INET_TW_MATCH(sk2, hash, acookie, saddr, daddr, ports, dif)) goto not_unique; } tw = NULL; /* And established part... */ sk_for_each(sk2, node, &head->chain) { - if (INET_MATCH(sk2, acookie, saddr, daddr, ports, dif)) + if (INET_MATCH(sk2, hash, acookie, saddr, daddr, ports, dif)) goto not_unique; } @@ -90,7 +90,7 @@ static int __dccp_v4_check_established(struct sock *sk, const __u16 lport, * in hash table socket with a funny identity. */ inet->num = lport; inet->sport = htons(lport); - sk->sk_hashent = hash; + sk->sk_hash = hash; BUG_TRAP(sk_unhashed(sk)); __sk_add_node(sk, &head->chain); sock_prot_inc_use(sk->sk_prot); diff --git a/trunk/net/econet/af_econet.c b/trunk/net/econet/af_econet.c index 4a62093eb343..34fdac51df96 100644 --- a/trunk/net/econet/af_econet.c +++ b/trunk/net/econet/af_econet.c @@ -406,7 +406,7 @@ static int econet_sendmsg(struct kiocb *iocb, struct socket *sock, unsigned long network = 0; rcu_read_lock(); - idev = __in_dev_get(dev); + idev = __in_dev_get_rcu(dev); if (idev) { if (idev->ifa_list) network = ntohl(idev->ifa_list->ifa_address) & diff --git a/trunk/net/ipv4/arp.c b/trunk/net/ipv4/arp.c index 8bf312bdea13..b425748f02d7 100644 --- a/trunk/net/ipv4/arp.c +++ b/trunk/net/ipv4/arp.c @@ -241,7 +241,7 @@ static int arp_constructor(struct neighbour *neigh) neigh->type = inet_addr_type(addr); rcu_read_lock(); - in_dev = rcu_dereference(__in_dev_get(dev)); + in_dev = __in_dev_get_rcu(dev); if (in_dev == NULL) { rcu_read_unlock(); return -EINVAL; @@ -697,12 +697,6 @@ void arp_send(int type, int ptype, u32 dest_ip, arp_xmit(skb); } -static void parp_redo(struct sk_buff *skb) -{ - nf_reset(skb); - arp_rcv(skb, skb->dev, NULL, skb->dev); -} - /* * Process an arp request. */ @@ -922,6 +916,11 @@ static int arp_process(struct sk_buff *skb) return 0; } +static void parp_redo(struct sk_buff *skb) +{ + arp_process(skb); +} + /* * Receive an arp request from the device layer. @@ -990,8 +989,8 @@ static int arp_req_set(struct arpreq *r, struct net_device * dev) ipv4_devconf.proxy_arp = 1; return 0; } - if (__in_dev_get(dev)) { - __in_dev_get(dev)->cnf.proxy_arp = 1; + if (__in_dev_get_rtnl(dev)) { + __in_dev_get_rtnl(dev)->cnf.proxy_arp = 1; return 0; } return -ENXIO; @@ -1096,8 +1095,8 @@ static int arp_req_delete(struct arpreq *r, struct net_device * dev) ipv4_devconf.proxy_arp = 0; return 0; } - if (__in_dev_get(dev)) { - __in_dev_get(dev)->cnf.proxy_arp = 0; + if (__in_dev_get_rtnl(dev)) { + __in_dev_get_rtnl(dev)->cnf.proxy_arp = 0; return 0; } return -ENXIO; diff --git a/trunk/net/ipv4/devinet.c b/trunk/net/ipv4/devinet.c index ba2895ae8151..74f2207e131a 100644 --- a/trunk/net/ipv4/devinet.c +++ b/trunk/net/ipv4/devinet.c @@ -351,7 +351,7 @@ static int inet_insert_ifa(struct in_ifaddr *ifa) static int inet_set_ifa(struct net_device *dev, struct in_ifaddr *ifa) { - struct in_device *in_dev = __in_dev_get(dev); + struct in_device *in_dev = __in_dev_get_rtnl(dev); ASSERT_RTNL(); @@ -449,7 +449,7 @@ static int inet_rtm_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg goto out; rc = -ENOBUFS; - if ((in_dev = __in_dev_get(dev)) == NULL) { + if ((in_dev = __in_dev_get_rtnl(dev)) == NULL) { in_dev = inetdev_init(dev); if (!in_dev) goto out; @@ -584,7 +584,7 @@ int devinet_ioctl(unsigned int cmd, void __user *arg) if (colon) *colon = ':'; - if ((in_dev = __in_dev_get(dev)) != NULL) { + if ((in_dev = __in_dev_get_rtnl(dev)) != NULL) { if (tryaddrmatch) { /* Matthias Andree */ /* compare label and address (4.4BSD style) */ @@ -748,7 +748,7 @@ int devinet_ioctl(unsigned int cmd, void __user *arg) static int inet_gifconf(struct net_device *dev, char __user *buf, int len) { - struct in_device *in_dev = __in_dev_get(dev); + struct in_device *in_dev = __in_dev_get_rtnl(dev); struct in_ifaddr *ifa; struct ifreq ifr; int done = 0; @@ -791,7 +791,7 @@ u32 inet_select_addr(const struct net_device *dev, u32 dst, int scope) struct in_device *in_dev; rcu_read_lock(); - in_dev = __in_dev_get(dev); + in_dev = __in_dev_get_rcu(dev); if (!in_dev) goto no_in_dev; @@ -818,7 +818,7 @@ u32 inet_select_addr(const struct net_device *dev, u32 dst, int scope) read_lock(&dev_base_lock); rcu_read_lock(); for (dev = dev_base; dev; dev = dev->next) { - if ((in_dev = __in_dev_get(dev)) == NULL) + if ((in_dev = __in_dev_get_rcu(dev)) == NULL) continue; for_primary_ifa(in_dev) { @@ -887,7 +887,7 @@ u32 inet_confirm_addr(const struct net_device *dev, u32 dst, u32 local, int scop if (dev) { rcu_read_lock(); - if ((in_dev = __in_dev_get(dev))) + if ((in_dev = __in_dev_get_rcu(dev))) addr = confirm_addr_indev(in_dev, dst, local, scope); rcu_read_unlock(); @@ -897,7 +897,7 @@ u32 inet_confirm_addr(const struct net_device *dev, u32 dst, u32 local, int scop read_lock(&dev_base_lock); rcu_read_lock(); for (dev = dev_base; dev; dev = dev->next) { - if ((in_dev = __in_dev_get(dev))) { + if ((in_dev = __in_dev_get_rcu(dev))) { addr = confirm_addr_indev(in_dev, dst, local, scope); if (addr) break; @@ -957,7 +957,7 @@ static int inetdev_event(struct notifier_block *this, unsigned long event, void *ptr) { struct net_device *dev = ptr; - struct in_device *in_dev = __in_dev_get(dev); + struct in_device *in_dev = __in_dev_get_rtnl(dev); ASSERT_RTNL(); @@ -1078,7 +1078,7 @@ static int inet_dump_ifaddr(struct sk_buff *skb, struct netlink_callback *cb) if (idx > s_idx) s_ip_idx = 0; rcu_read_lock(); - if ((in_dev = __in_dev_get(dev)) == NULL) { + if ((in_dev = __in_dev_get_rcu(dev)) == NULL) { rcu_read_unlock(); continue; } @@ -1149,7 +1149,7 @@ void inet_forward_change(void) for (dev = dev_base; dev; dev = dev->next) { struct in_device *in_dev; rcu_read_lock(); - in_dev = __in_dev_get(dev); + in_dev = __in_dev_get_rcu(dev); if (in_dev) in_dev->cnf.forwarding = on; rcu_read_unlock(); diff --git a/trunk/net/ipv4/fib_frontend.c b/trunk/net/ipv4/fib_frontend.c index 4e1379f71269..e61bc7177eb1 100644 --- a/trunk/net/ipv4/fib_frontend.c +++ b/trunk/net/ipv4/fib_frontend.c @@ -173,7 +173,7 @@ int fib_validate_source(u32 src, u32 dst, u8 tos, int oif, no_addr = rpf = 0; rcu_read_lock(); - in_dev = __in_dev_get(dev); + in_dev = __in_dev_get_rcu(dev); if (in_dev) { no_addr = in_dev->ifa_list == NULL; rpf = IN_DEV_RPFILTER(in_dev); @@ -607,7 +607,7 @@ static int fib_inetaddr_event(struct notifier_block *this, unsigned long event, static int fib_netdev_event(struct notifier_block *this, unsigned long event, void *ptr) { struct net_device *dev = ptr; - struct in_device *in_dev = __in_dev_get(dev); + struct in_device *in_dev = __in_dev_get_rtnl(dev); if (event == NETDEV_UNREGISTER) { fib_disable_ip(dev, 2); diff --git a/trunk/net/ipv4/fib_semantics.c b/trunk/net/ipv4/fib_semantics.c index d41219e8037c..186f20c4a45e 100644 --- a/trunk/net/ipv4/fib_semantics.c +++ b/trunk/net/ipv4/fib_semantics.c @@ -1087,7 +1087,7 @@ fib_convert_rtentry(int cmd, struct nlmsghdr *nl, struct rtmsg *rtm, rta->rta_oif = &dev->ifindex; if (colon) { struct in_ifaddr *ifa; - struct in_device *in_dev = __in_dev_get(dev); + struct in_device *in_dev = __in_dev_get_rtnl(dev); if (!in_dev) return -ENODEV; *colon = ':'; @@ -1268,7 +1268,7 @@ int fib_sync_up(struct net_device *dev) } if (nh->nh_dev == NULL || !(nh->nh_dev->flags&IFF_UP)) continue; - if (nh->nh_dev != dev || __in_dev_get(dev) == NULL) + if (nh->nh_dev != dev || !__in_dev_get_rtnl(dev)) continue; alive++; spin_lock_bh(&fib_multipath_lock); diff --git a/trunk/net/ipv4/icmp.c b/trunk/net/ipv4/icmp.c index 24eb56ae1b5a..90dca711ac9f 100644 --- a/trunk/net/ipv4/icmp.c +++ b/trunk/net/ipv4/icmp.c @@ -188,7 +188,7 @@ struct icmp_err icmp_err_convert[] = { /* Control parameters for ECHO replies. */ int sysctl_icmp_echo_ignore_all; -int sysctl_icmp_echo_ignore_broadcasts; +int sysctl_icmp_echo_ignore_broadcasts = 1; /* Control parameter - ignore bogus broadcast responses? */ int sysctl_icmp_ignore_bogus_error_responses; diff --git a/trunk/net/ipv4/igmp.c b/trunk/net/ipv4/igmp.c index 70c44e4c3ceb..8b6d3939e1e6 100644 --- a/trunk/net/ipv4/igmp.c +++ b/trunk/net/ipv4/igmp.c @@ -1323,7 +1323,7 @@ static struct in_device * ip_mc_find_dev(struct ip_mreqn *imr) } if (dev) { imr->imr_ifindex = dev->ifindex; - idev = __in_dev_get(dev); + idev = __in_dev_get_rtnl(dev); } return idev; } diff --git a/trunk/net/ipv4/inet_timewait_sock.c b/trunk/net/ipv4/inet_timewait_sock.c index 4d1502a49852..f9076ef3a1a8 100644 --- a/trunk/net/ipv4/inet_timewait_sock.c +++ b/trunk/net/ipv4/inet_timewait_sock.c @@ -20,7 +20,7 @@ void __inet_twsk_kill(struct inet_timewait_sock *tw, struct inet_hashinfo *hashi struct inet_bind_hashbucket *bhead; struct inet_bind_bucket *tb; /* Unlink from established hashes. */ - struct inet_ehash_bucket *ehead = &hashinfo->ehash[tw->tw_hashent]; + struct inet_ehash_bucket *ehead = inet_ehash_bucket(hashinfo, tw->tw_hash); write_lock(&ehead->lock); if (hlist_unhashed(&tw->tw_node)) { @@ -60,7 +60,7 @@ void __inet_twsk_hashdance(struct inet_timewait_sock *tw, struct sock *sk, { const struct inet_sock *inet = inet_sk(sk); const struct inet_connection_sock *icsk = inet_csk(sk); - struct inet_ehash_bucket *ehead = &hashinfo->ehash[sk->sk_hashent]; + struct inet_ehash_bucket *ehead = inet_ehash_bucket(hashinfo, sk->sk_hash); struct inet_bind_hashbucket *bhead; /* Step 1: Put TW into bind hash. Original socket stays there too. Note, that any socket with inet->num != 0 MUST be bound in @@ -106,7 +106,7 @@ struct inet_timewait_sock *inet_twsk_alloc(const struct sock *sk, const int stat tw->tw_dport = inet->dport; tw->tw_family = sk->sk_family; tw->tw_reuse = sk->sk_reuse; - tw->tw_hashent = sk->sk_hashent; + tw->tw_hash = sk->sk_hash; tw->tw_ipv6only = 0; tw->tw_prot = sk->sk_prot_creator; atomic_set(&tw->tw_refcnt, 1); diff --git a/trunk/net/ipv4/ip_gre.c b/trunk/net/ipv4/ip_gre.c index f0d5740d7e22..896ce3f8f53a 100644 --- a/trunk/net/ipv4/ip_gre.c +++ b/trunk/net/ipv4/ip_gre.c @@ -1104,10 +1104,10 @@ static int ipgre_open(struct net_device *dev) return -EADDRNOTAVAIL; dev = rt->u.dst.dev; ip_rt_put(rt); - if (__in_dev_get(dev) == NULL) + if (__in_dev_get_rtnl(dev) == NULL) return -EADDRNOTAVAIL; t->mlink = dev->ifindex; - ip_mc_inc_group(__in_dev_get(dev), t->parms.iph.daddr); + ip_mc_inc_group(__in_dev_get_rtnl(dev), t->parms.iph.daddr); } return 0; } diff --git a/trunk/net/ipv4/ipmr.c b/trunk/net/ipv4/ipmr.c index 9dbf5909f3a6..302b7eb507c9 100644 --- a/trunk/net/ipv4/ipmr.c +++ b/trunk/net/ipv4/ipmr.c @@ -149,7 +149,7 @@ struct net_device *ipmr_new_tunnel(struct vifctl *v) if (err == 0 && (dev = __dev_get_by_name(p.name)) != NULL) { dev->flags |= IFF_MULTICAST; - in_dev = __in_dev_get(dev); + in_dev = __in_dev_get_rtnl(dev); if (in_dev == NULL && (in_dev = inetdev_init(dev)) == NULL) goto failure; in_dev->cnf.rp_filter = 0; @@ -278,7 +278,7 @@ static int vif_delete(int vifi) dev_set_allmulti(dev, -1); - if ((in_dev = __in_dev_get(dev)) != NULL) { + if ((in_dev = __in_dev_get_rtnl(dev)) != NULL) { in_dev->cnf.mc_forwarding--; ip_rt_multicast_event(in_dev); } @@ -421,7 +421,7 @@ static int vif_add(struct vifctl *vifc, int mrtsock) return -EINVAL; } - if ((in_dev = __in_dev_get(dev)) == NULL) + if ((in_dev = __in_dev_get_rtnl(dev)) == NULL) return -EADDRNOTAVAIL; in_dev->cnf.mc_forwarding++; dev_set_allmulti(dev, +1); diff --git a/trunk/net/ipv4/netfilter/ip_conntrack_netbios_ns.c b/trunk/net/ipv4/netfilter/ip_conntrack_netbios_ns.c index 577bac22dcc6..186646eb249f 100644 --- a/trunk/net/ipv4/netfilter/ip_conntrack_netbios_ns.c +++ b/trunk/net/ipv4/netfilter/ip_conntrack_netbios_ns.c @@ -58,7 +58,7 @@ static int help(struct sk_buff **pskb, goto out; rcu_read_lock(); - in_dev = __in_dev_get(rt->u.dst.dev); + in_dev = __in_dev_get_rcu(rt->u.dst.dev); if (in_dev != NULL) { for_primary_ifa(in_dev) { if (ifa->ifa_broadcast == iph->daddr) { diff --git a/trunk/net/ipv4/netfilter/ip_queue.c b/trunk/net/ipv4/netfilter/ip_queue.c index d54f14d926f6..36339eb39e17 100644 --- a/trunk/net/ipv4/netfilter/ip_queue.c +++ b/trunk/net/ipv4/netfilter/ip_queue.c @@ -240,8 +240,8 @@ ipq_build_packet_message(struct ipq_queue_entry *entry, int *errp) pmsg->packet_id = (unsigned long )entry; pmsg->data_len = data_len; - pmsg->timestamp_sec = skb_tv_base.tv_sec + entry->skb->tstamp.off_sec; - pmsg->timestamp_usec = skb_tv_base.tv_usec + entry->skb->tstamp.off_usec; + pmsg->timestamp_sec = entry->skb->tstamp.off_sec; + pmsg->timestamp_usec = entry->skb->tstamp.off_usec; pmsg->mark = entry->skb->nfmark; pmsg->hook = entry->info->hook; pmsg->hw_protocol = entry->skb->protocol; diff --git a/trunk/net/ipv4/netfilter/ipt_REDIRECT.c b/trunk/net/ipv4/netfilter/ipt_REDIRECT.c index 715cb613405c..5245bfd33d52 100644 --- a/trunk/net/ipv4/netfilter/ipt_REDIRECT.c +++ b/trunk/net/ipv4/netfilter/ipt_REDIRECT.c @@ -93,7 +93,7 @@ redirect_target(struct sk_buff **pskb, newdst = 0; rcu_read_lock(); - indev = __in_dev_get((*pskb)->dev); + indev = __in_dev_get_rcu((*pskb)->dev); if (indev && (ifa = indev->ifa_list)) newdst = ifa->ifa_local; rcu_read_unlock(); diff --git a/trunk/net/ipv4/netfilter/ipt_ULOG.c b/trunk/net/ipv4/netfilter/ipt_ULOG.c index e2c14f3cb2fc..2883ccd8a91d 100644 --- a/trunk/net/ipv4/netfilter/ipt_ULOG.c +++ b/trunk/net/ipv4/netfilter/ipt_ULOG.c @@ -225,8 +225,8 @@ static void ipt_ulog_packet(unsigned int hooknum, /* copy hook, prefix, timestamp, payload, etc. */ pm->data_len = copy_len; - pm->timestamp_sec = skb_tv_base.tv_sec + skb->tstamp.off_sec; - pm->timestamp_usec = skb_tv_base.tv_usec + skb->tstamp.off_usec; + pm->timestamp_sec = skb->tstamp.off_sec; + pm->timestamp_usec = skb->tstamp.off_usec; pm->mark = skb->nfmark; pm->hook = hooknum; if (prefix != NULL) diff --git a/trunk/net/ipv4/route.c b/trunk/net/ipv4/route.c index 8549f26e2495..381dd6a6aebb 100644 --- a/trunk/net/ipv4/route.c +++ b/trunk/net/ipv4/route.c @@ -2128,7 +2128,7 @@ int ip_route_input(struct sk_buff *skb, u32 daddr, u32 saddr, struct in_device *in_dev; rcu_read_lock(); - if ((in_dev = __in_dev_get(dev)) != NULL) { + if ((in_dev = __in_dev_get_rcu(dev)) != NULL) { int our = ip_check_mc(in_dev, daddr, saddr, skb->nh.iph->protocol); if (our @@ -2443,7 +2443,9 @@ static int ip_route_output_slow(struct rtable **rp, const struct flowi *oldflp) err = -ENODEV; if (dev_out == NULL) goto out; - if (__in_dev_get(dev_out) == NULL) { + + /* RACE: Check return value of inet_select_addr instead. */ + if (__in_dev_get_rtnl(dev_out) == NULL) { dev_put(dev_out); goto out; /* Wrong error code */ } diff --git a/trunk/net/ipv4/tcp_ipv4.c b/trunk/net/ipv4/tcp_ipv4.c index 13dfb391cdf1..c85819d8474b 100644 --- a/trunk/net/ipv4/tcp_ipv4.c +++ b/trunk/net/ipv4/tcp_ipv4.c @@ -130,19 +130,20 @@ static int __tcp_v4_check_established(struct sock *sk, __u16 lport, int dif = sk->sk_bound_dev_if; INET_ADDR_COOKIE(acookie, saddr, daddr) const __u32 ports = INET_COMBINED_PORTS(inet->dport, lport); - const int hash = inet_ehashfn(daddr, lport, saddr, inet->dport, tcp_hashinfo.ehash_size); - struct inet_ehash_bucket *head = &tcp_hashinfo.ehash[hash]; + unsigned int hash = inet_ehashfn(daddr, lport, saddr, inet->dport); + struct inet_ehash_bucket *head = inet_ehash_bucket(&tcp_hashinfo, hash); struct sock *sk2; const struct hlist_node *node; struct inet_timewait_sock *tw; + prefetch(head->chain.first); write_lock(&head->lock); /* Check TIME-WAIT sockets first. */ sk_for_each(sk2, node, &(head + tcp_hashinfo.ehash_size)->chain) { tw = inet_twsk(sk2); - if (INET_TW_MATCH(sk2, acookie, saddr, daddr, ports, dif)) { + if (INET_TW_MATCH(sk2, hash, acookie, saddr, daddr, ports, dif)) { const struct tcp_timewait_sock *tcptw = tcp_twsk(sk2); struct tcp_sock *tp = tcp_sk(sk); @@ -179,7 +180,7 @@ static int __tcp_v4_check_established(struct sock *sk, __u16 lport, /* And established part... */ sk_for_each(sk2, node, &head->chain) { - if (INET_MATCH(sk2, acookie, saddr, daddr, ports, dif)) + if (INET_MATCH(sk2, hash, acookie, saddr, daddr, ports, dif)) goto not_unique; } @@ -188,7 +189,7 @@ static int __tcp_v4_check_established(struct sock *sk, __u16 lport, * in hash table socket with a funny identity. */ inet->num = lport; inet->sport = htons(lport); - sk->sk_hashent = hash; + sk->sk_hash = hash; BUG_TRAP(sk_unhashed(sk)); __sk_add_node(sk, &head->chain); sock_prot_inc_use(sk->sk_prot); diff --git a/trunk/net/ipv6/addrconf.c b/trunk/net/ipv6/addrconf.c index 4e509e52fbc1..a970b4727ce8 100644 --- a/trunk/net/ipv6/addrconf.c +++ b/trunk/net/ipv6/addrconf.c @@ -1806,7 +1806,7 @@ static void sit_add_v4_addrs(struct inet6_dev *idev) } for (dev = dev_base; dev != NULL; dev = dev->next) { - struct in_device * in_dev = __in_dev_get(dev); + struct in_device * in_dev = __in_dev_get_rtnl(dev); if (in_dev && (dev->flags & IFF_UP)) { struct in_ifaddr * ifa; diff --git a/trunk/net/ipv6/ip6_output.c b/trunk/net/ipv6/ip6_output.c index 2f589f24c093..563b442ffab8 100644 --- a/trunk/net/ipv6/ip6_output.c +++ b/trunk/net/ipv6/ip6_output.c @@ -666,7 +666,7 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) */ fh->nexthdr = nexthdr; fh->reserved = 0; - if (frag_id) { + if (!frag_id) { ipv6_select_ident(skb, fh); frag_id = fh->identification; } else diff --git a/trunk/net/ipv6/netfilter/ip6_queue.c b/trunk/net/ipv6/netfilter/ip6_queue.c index aa11cf366efa..5027bbe6415e 100644 --- a/trunk/net/ipv6/netfilter/ip6_queue.c +++ b/trunk/net/ipv6/netfilter/ip6_queue.c @@ -238,8 +238,8 @@ ipq_build_packet_message(struct ipq_queue_entry *entry, int *errp) pmsg->packet_id = (unsigned long )entry; pmsg->data_len = data_len; - pmsg->timestamp_sec = skb_tv_base.tv_sec + entry->skb->tstamp.off_sec; - pmsg->timestamp_usec = skb_tv_base.tv_usec + entry->skb->tstamp.off_usec; + pmsg->timestamp_sec = entry->skb->tstamp.off_sec; + pmsg->timestamp_usec = entry->skb->tstamp.off_usec; pmsg->mark = entry->skb->nfmark; pmsg->hook = entry->info->hook; pmsg->hw_protocol = entry->skb->protocol; diff --git a/trunk/net/ipv6/tcp_ipv6.c b/trunk/net/ipv6/tcp_ipv6.c index 80643e6b346b..d693cb988b78 100644 --- a/trunk/net/ipv6/tcp_ipv6.c +++ b/trunk/net/ipv6/tcp_ipv6.c @@ -209,9 +209,11 @@ static __inline__ void __tcp_v6_hash(struct sock *sk) lock = &tcp_hashinfo.lhash_lock; inet_listen_wlock(&tcp_hashinfo); } else { - sk->sk_hashent = inet6_sk_ehashfn(sk, tcp_hashinfo.ehash_size); - list = &tcp_hashinfo.ehash[sk->sk_hashent].chain; - lock = &tcp_hashinfo.ehash[sk->sk_hashent].lock; + unsigned int hash; + sk->sk_hash = hash = inet6_sk_ehashfn(sk); + hash &= (tcp_hashinfo.ehash_size - 1); + list = &tcp_hashinfo.ehash[hash].chain; + lock = &tcp_hashinfo.ehash[hash].lock; write_lock(lock); } @@ -322,13 +324,13 @@ static int __tcp_v6_check_established(struct sock *sk, const __u16 lport, const struct in6_addr *saddr = &np->daddr; const int dif = sk->sk_bound_dev_if; const u32 ports = INET_COMBINED_PORTS(inet->dport, lport); - const int hash = inet6_ehashfn(daddr, inet->num, saddr, inet->dport, - tcp_hashinfo.ehash_size); - struct inet_ehash_bucket *head = &tcp_hashinfo.ehash[hash]; + unsigned int hash = inet6_ehashfn(daddr, inet->num, saddr, inet->dport); + struct inet_ehash_bucket *head = inet_ehash_bucket(&tcp_hashinfo, hash); struct sock *sk2; const struct hlist_node *node; struct inet_timewait_sock *tw; + prefetch(head->chain.first); write_lock(&head->lock); /* Check TIME-WAIT sockets first. */ @@ -365,14 +367,14 @@ static int __tcp_v6_check_established(struct sock *sk, const __u16 lport, /* And established part... */ sk_for_each(sk2, node, &head->chain) { - if (INET6_MATCH(sk2, saddr, daddr, ports, dif)) + if (INET6_MATCH(sk2, hash, saddr, daddr, ports, dif)) goto not_unique; } unique: BUG_TRAP(sk_unhashed(sk)); __sk_add_node(sk, &head->chain); - sk->sk_hashent = hash; + sk->sk_hash = hash; sock_prot_inc_use(sk->sk_prot); write_unlock(&head->lock); diff --git a/trunk/net/ipv6/udp.c b/trunk/net/ipv6/udp.c index 6001948600f3..e4cad11f284a 100644 --- a/trunk/net/ipv6/udp.c +++ b/trunk/net/ipv6/udp.c @@ -852,10 +852,16 @@ static int udpv6_sendmsg(struct kiocb *iocb, struct sock *sk, else if (!corkreq) err = udp_v6_push_pending_frames(sk, up); - if (dst && connected) - ip6_dst_store(sk, dst, - ipv6_addr_equal(&fl->fl6_dst, &np->daddr) ? - &np->daddr : NULL); + if (dst) { + if (connected) { + ip6_dst_store(sk, dst, + ipv6_addr_equal(&fl->fl6_dst, &np->daddr) ? + &np->daddr : NULL); + } else { + dst_release(dst); + } + } + if (err > 0) err = np->recverr ? net_xmit_errno(err) : 0; release_sock(sk); diff --git a/trunk/net/irda/irlan/irlan_eth.c b/trunk/net/irda/irlan/irlan_eth.c index 071cd2cefd8a..953e255d2bc8 100644 --- a/trunk/net/irda/irlan/irlan_eth.c +++ b/trunk/net/irda/irlan/irlan_eth.c @@ -310,7 +310,7 @@ void irlan_eth_send_gratuitous_arp(struct net_device *dev) #ifdef CONFIG_INET IRDA_DEBUG(4, "IrLAN: Sending gratuitous ARP\n"); rcu_read_lock(); - in_dev = __in_dev_get(dev); + in_dev = __in_dev_get_rcu(dev); if (in_dev == NULL) goto out; if (in_dev->ifa_list) diff --git a/trunk/net/netfilter/nfnetlink_log.c b/trunk/net/netfilter/nfnetlink_log.c index ff5601ceedcb..efcd10f996ba 100644 --- a/trunk/net/netfilter/nfnetlink_log.c +++ b/trunk/net/netfilter/nfnetlink_log.c @@ -494,8 +494,8 @@ __build_packet_message(struct nfulnl_instance *inst, if (skb->tstamp.off_sec) { struct nfulnl_msg_packet_timestamp ts; - ts.sec = cpu_to_be64(skb_tv_base.tv_sec + skb->tstamp.off_sec); - ts.usec = cpu_to_be64(skb_tv_base.tv_usec + skb->tstamp.off_usec); + ts.sec = cpu_to_be64(skb->tstamp.off_sec); + ts.usec = cpu_to_be64(skb->tstamp.off_usec); NFA_PUT(inst->skb, NFULA_TIMESTAMP, sizeof(ts), &ts); } diff --git a/trunk/net/netfilter/nfnetlink_queue.c b/trunk/net/netfilter/nfnetlink_queue.c index f81fe8c52e99..eaa44c49567b 100644 --- a/trunk/net/netfilter/nfnetlink_queue.c +++ b/trunk/net/netfilter/nfnetlink_queue.c @@ -492,8 +492,8 @@ nfqnl_build_packet_message(struct nfqnl_instance *queue, if (entry->skb->tstamp.off_sec) { struct nfqnl_msg_packet_timestamp ts; - ts.sec = cpu_to_be64(skb_tv_base.tv_sec + entry->skb->tstamp.off_sec); - ts.usec = cpu_to_be64(skb_tv_base.tv_usec + entry->skb->tstamp.off_usec); + ts.sec = cpu_to_be64(entry->skb->tstamp.off_sec); + ts.usec = cpu_to_be64(entry->skb->tstamp.off_usec); NFA_PUT(skb, NFQA_TIMESTAMP, sizeof(ts), &ts); } diff --git a/trunk/net/packet/af_packet.c b/trunk/net/packet/af_packet.c index 6a67a87384cc..499ae3df4a44 100644 --- a/trunk/net/packet/af_packet.c +++ b/trunk/net/packet/af_packet.c @@ -654,8 +654,8 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, struct packe __net_timestamp(skb); sock_enable_timestamp(sk); } - h->tp_sec = skb_tv_base.tv_sec + skb->tstamp.off_sec; - h->tp_usec = skb_tv_base.tv_usec + skb->tstamp.off_usec; + h->tp_sec = skb->tstamp.off_sec; + h->tp_usec = skb->tstamp.off_usec; sll = (struct sockaddr_ll*)((u8*)h + TPACKET_ALIGN(sizeof(*h))); sll->sll_halen = 0; diff --git a/trunk/net/sched/em_meta.c b/trunk/net/sched/em_meta.c index 00eae5f9a01a..cf68a59fdc5a 100644 --- a/trunk/net/sched/em_meta.c +++ b/trunk/net/sched/em_meta.c @@ -393,10 +393,10 @@ META_COLLECTOR(int_sk_route_caps) dst->value = skb->sk->sk_route_caps; } -META_COLLECTOR(int_sk_hashent) +META_COLLECTOR(int_sk_hash) { SKIP_NONLOCAL(skb); - dst->value = skb->sk->sk_hashent; + dst->value = skb->sk->sk_hash; } META_COLLECTOR(int_sk_lingertime) @@ -515,7 +515,7 @@ static struct meta_ops __meta_ops[TCF_META_TYPE_MAX+1][TCF_META_ID_MAX+1] = { [META_ID(SK_FORWARD_ALLOCS)] = META_FUNC(int_sk_fwd_alloc), [META_ID(SK_ALLOCS)] = META_FUNC(int_sk_alloc), [META_ID(SK_ROUTE_CAPS)] = META_FUNC(int_sk_route_caps), - [META_ID(SK_HASHENT)] = META_FUNC(int_sk_hashent), + [META_ID(SK_HASH)] = META_FUNC(int_sk_hash), [META_ID(SK_LINGERTIME)] = META_FUNC(int_sk_lingertime), [META_ID(SK_ACK_BACKLOG)] = META_FUNC(int_sk_ack_bl), [META_ID(SK_MAX_ACK_BACKLOG)] = META_FUNC(int_sk_max_ack_bl), diff --git a/trunk/net/sctp/protocol.c b/trunk/net/sctp/protocol.c index e7025be77691..f01d1c9002a1 100644 --- a/trunk/net/sctp/protocol.c +++ b/trunk/net/sctp/protocol.c @@ -147,7 +147,7 @@ static void sctp_v4_copy_addrlist(struct list_head *addrlist, struct sctp_sockaddr_entry *addr; rcu_read_lock(); - if ((in_dev = __in_dev_get(dev)) == NULL) { + if ((in_dev = __in_dev_get_rcu(dev)) == NULL) { rcu_read_unlock(); return; } diff --git a/trunk/net/sysctl_net.c b/trunk/net/sysctl_net.c index c5241fcbb966..55538f6b60ff 100644 --- a/trunk/net/sysctl_net.c +++ b/trunk/net/sysctl_net.c @@ -16,6 +16,8 @@ #include #include +#include + #ifdef CONFIG_INET #include #endif