diff --git a/[refs] b/[refs] index dd7ae3ddbad7..a9a56a3fd8e9 100644 --- a/[refs] +++ b/[refs] @@ -1,2 +1,2 @@ --- -refs/heads/master: 2dc20a51dc1e2da6eae1182cfe4c4835fca26017 +refs/heads/master: 5283304bdcb4af47c46280ac21100d037d31fcf0 diff --git a/trunk/arch/arm/configs/ep93xx_defconfig b/trunk/arch/arm/configs/ep93xx_defconfig index 3b4802a849e4..2c6734460127 100644 --- a/trunk/arch/arm/configs/ep93xx_defconfig +++ b/trunk/arch/arm/configs/ep93xx_defconfig @@ -125,6 +125,7 @@ CONFIG_CRUNCH=y # # EP93xx Platforms # +CONFIG_MACH_ADSSPHERE=y CONFIG_MACH_EDB9302=y CONFIG_MACH_EDB9312=y CONFIG_MACH_EDB9315=y diff --git a/trunk/arch/arm/mach-ep93xx/Kconfig b/trunk/arch/arm/mach-ep93xx/Kconfig index e346b03cd921..fcc5e2500dc7 100644 --- a/trunk/arch/arm/mach-ep93xx/Kconfig +++ b/trunk/arch/arm/mach-ep93xx/Kconfig @@ -9,6 +9,12 @@ config CRUNCH comment "EP93xx Platforms" +config MACH_ADSSPHERE + bool "Support ADS Sphere" + help + Say 'Y' here if you want your kernel to support the ADS + Sphere board. + config MACH_EDB9302 bool "Support Cirrus Logic EDB9302" help diff --git a/trunk/arch/arm/mach-ep93xx/Makefile b/trunk/arch/arm/mach-ep93xx/Makefile index c2eb18b530c2..bd09cddbc52a 100644 --- a/trunk/arch/arm/mach-ep93xx/Makefile +++ b/trunk/arch/arm/mach-ep93xx/Makefile @@ -6,6 +6,7 @@ obj-m := obj-n := obj- := +obj-$(CONFIG_MACH_ADSSPHERE) += adssphere.o obj-$(CONFIG_MACH_EDB9302) += edb9302.o obj-$(CONFIG_MACH_EDB9312) += edb9312.o obj-$(CONFIG_MACH_EDB9315) += edb9315.o diff --git a/trunk/arch/arm/mach-ep93xx/adssphere.c b/trunk/arch/arm/mach-ep93xx/adssphere.c new file mode 100644 index 000000000000..ac5d5818eb7b --- /dev/null +++ b/trunk/arch/arm/mach-ep93xx/adssphere.c @@ -0,0 +1,91 @@ +/* + * arch/arm/mach-ep93xx/adssphere.c + * ADS Sphere support. + * + * Copyright (C) 2006 Lennert Buytenhek + * + * 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. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +static struct physmap_flash_data adssphere_flash_data = { + .width = 4, +}; + +static struct resource adssphere_flash_resource = { + .start = 0x60000000, + .end = 0x61ffffff, + .flags = IORESOURCE_MEM, +}; + +static struct platform_device adssphere_flash = { + .name = "physmap-flash", + .id = 0, + .dev = { + .platform_data = &adssphere_flash_data, + }, + .num_resources = 1, + .resource = &adssphere_flash_resource, +}; + +static struct ep93xx_eth_data adssphere_eth_data = { + .phy_id = 1, +}; + +static struct resource adssphere_eth_resource[] = { + { + .start = EP93XX_ETHERNET_PHYS_BASE, + .end = EP93XX_ETHERNET_PHYS_BASE + 0xffff, + .flags = IORESOURCE_MEM, + }, { + .start = IRQ_EP93XX_ETHERNET, + .end = IRQ_EP93XX_ETHERNET, + .flags = IORESOURCE_IRQ, + } +}; + +static struct platform_device adssphere_eth_device = { + .name = "ep93xx-eth", + .id = -1, + .dev = { + .platform_data = &adssphere_eth_data, + }, + .num_resources = 2, + .resource = adssphere_eth_resource, +}; + +static void __init adssphere_init_machine(void) +{ + ep93xx_init_devices(); + platform_device_register(&adssphere_flash); + + memcpy(adssphere_eth_data.dev_addr, + (void *)(EP93XX_ETHERNET_BASE + 0x50), 6); + platform_device_register(&adssphere_eth_device); +} + +MACHINE_START(ADSSPHERE, "ADS Sphere board") + /* Maintainer: Lennert Buytenhek */ + .phys_io = EP93XX_APB_PHYS_BASE, + .io_pg_offst = ((EP93XX_APB_VIRT_BASE) >> 18) & 0xfffc, + .boot_params = 0x00000100, + .map_io = ep93xx_map_io, + .init_irq = ep93xx_init_irq, + .timer = &ep93xx_timer, + .init_machine = adssphere_init_machine, +MACHINE_END diff --git a/trunk/arch/arm/mach-pxa/generic.c b/trunk/arch/arm/mach-pxa/generic.c index 6ae605857ca9..45fb2c3bcf82 100644 --- a/trunk/arch/arm/mach-pxa/generic.c +++ b/trunk/arch/arm/mach-pxa/generic.c @@ -25,10 +25,6 @@ #include #include -#include -#include -#include - #include #include #include @@ -44,62 +40,6 @@ #include "generic.h" -/* - * This is the PXA2xx sched_clock implementation. This has a resolution - * of at least 308ns and a maximum value that depends on the value of - * CLOCK_TICK_RATE. - * - * The return value is guaranteed to be monotonic in that range as - * long as there is always less than 582 seconds between successive - * calls to this function. - */ -unsigned long long sched_clock(void) -{ - unsigned long long v = cnt32_to_63(OSCR); - /* Note: top bit ov v needs cleared unless multiplier is even. */ - -#if CLOCK_TICK_RATE == 3686400 - /* 1E9 / 3686400 => 78125 / 288, max value = 32025597s (370 days). */ - /* The <<1 is used to get rid of tick.hi top bit */ - v *= 78125<<1; - do_div(v, 288<<1); -#elif CLOCK_TICK_RATE == 3250000 - /* 1E9 / 3250000 => 4000 / 13, max value = 709490156s (8211 days) */ - v *= 4000; - do_div(v, 13); -#elif CLOCK_TICK_RATE == 3249600 - /* 1E9 / 3249600 => 625000 / 2031, max value = 4541295s (52 days) */ - v *= 625000; - do_div(v, 2031); -#else -#warning "consider fixing sched_clock for your value of CLOCK_TICK_RATE" - /* - * 96-bit math to perform tick * NSEC_PER_SEC / CLOCK_TICK_RATE for - * any value of CLOCK_TICK_RATE. Max value is in the 80 thousand - * years range which is nice, but with higher computation cost. - */ - { - union { - unsigned long long val; - struct { unsigned long lo, hi; }; - } x; - unsigned long long y; - - x.val = v; - x.hi &= 0x7fffffff; - y = (unsigned long long)x.lo * NSEC_PER_SEC; - x.lo = y; - y = (y >> 32) + (unsigned long long)x.hi * NSEC_PER_SEC; - x.hi = do_div(y, CLOCK_TICK_RATE); - do_div(x.val, CLOCK_TICK_RATE); - x.hi += y; - v = x.val; - } -#endif - - return v; -} - /* * Handy function to set GPIO alternate functions */ diff --git a/trunk/arch/arm/mach-sa1100/generic.c b/trunk/arch/arm/mach-sa1100/generic.c index e510295c2580..4575f316e141 100644 --- a/trunk/arch/arm/mach-sa1100/generic.c +++ b/trunk/arch/arm/mach-sa1100/generic.c @@ -20,7 +20,6 @@ #include #include -#include #include #include #include @@ -119,21 +118,15 @@ EXPORT_SYMBOL(cpufreq_get); /* * This is the SA11x0 sched_clock implementation. This has - * a resolution of 271ns, and a maximum value of 32025597s (370 days). - * - * The return value is guaranteed to be monotonic in that range as - * long as there is always less than 582 seconds between successive - * calls to this function. - * + * a resolution of 271ns, and a maximum value of 1165s. * ( * 1E9 / 3686400 => * 78125 / 288) */ unsigned long long sched_clock(void) { - unsigned long long v = cnt32_to_63(OSCR); + unsigned long long v; - /* the <<1 gets rid of the cnt_32_to_63 top bit saving on a bic insn */ - v *= 78125<<1; - do_div(v, 288<<1); + v = (unsigned long long)OSCR * 78125; + do_div(v, 288); return v; } diff --git a/trunk/arch/arm/mach-versatile/core.c b/trunk/arch/arm/mach-versatile/core.c index 998b398df30e..3b8576111c16 100644 --- a/trunk/arch/arm/mach-versatile/core.c +++ b/trunk/arch/arm/mach-versatile/core.c @@ -27,7 +27,6 @@ #include #include -#include #include #include #include @@ -229,19 +228,14 @@ void __init versatile_map_io(void) /* * This is the Versatile sched_clock implementation. This has - * a resolution of 41.7ns, and a maximum value of about 35583 days. - * - * The return value is guaranteed to be monotonic in that range as - * long as there is always less than 89 seconds between successive - * calls to this function. + * a resolution of 41.7ns, and a maximum value of about 179s. */ unsigned long long sched_clock(void) { - unsigned long long v = cnt32_to_63(readl(VERSATILE_REFCOUNTER)); + unsigned long long v; - /* the <<1 gets rid of the cnt_32_to_63 top bit saving on a bic insn */ - v *= 125<<1; - do_div(v, 3<<1); + v = (unsigned long long)readl(VERSATILE_REFCOUNTER) * 125; + do_div(v, 3); return v; } diff --git a/trunk/include/asm-arm/cnt32_to_63.h b/trunk/include/asm-arm/cnt32_to_63.h deleted file mode 100644 index 480c873fa746..000000000000 --- a/trunk/include/asm-arm/cnt32_to_63.h +++ /dev/null @@ -1,78 +0,0 @@ -/* - * include/asm/cnt32_to_63.h -- extend a 32-bit counter to 63 bits - * - * Author: Nicolas Pitre - * Created: December 3, 2006 - * Copyright: MontaVista Software, Inc. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 - * as published by the Free Software Foundation. - */ - -#ifndef __INCLUDE_CNT32_TO_63_H__ -#define __INCLUDE_CNT32_TO_63_H__ - -#include -#include -#include - -/* - * Prototype: u64 cnt32_to_63(u32 cnt) - * Many hardware clock counters are only 32 bits wide and therefore have - * a relatively short period making wrap-arounds rather frequent. This - * is a problem when implementing sched_clock() for example, where a 64-bit - * non-wrapping monotonic value is expected to be returned. - * - * To overcome that limitation, let's extend a 32-bit counter to 63 bits - * in a completely lock free fashion. Bits 0 to 31 of the clock are provided - * by the hardware while bits 32 to 62 are stored in memory. The top bit in - * memory is used to synchronize with the hardware clock half-period. When - * the top bit of both counters (hardware and in memory) differ then the - * memory is updated with a new value, incrementing it when the hardware - * counter wraps around. - * - * Because a word store in memory is atomic then the incremented value will - * always be in synch with the top bit indicating to any potential concurrent - * reader if the value in memory is up to date or not with regards to the - * needed increment. And any race in updating the value in memory is harmless - * as the same value would simply be stored more than once. - * - * The only restriction for the algorithm to work properly is that this - * code must be executed at least once per each half period of the 32-bit - * counter to properly update the state bit in memory. This is usually not a - * problem in practice, but if it is then a kernel timer could be scheduled - * to manage for this code to be executed often enough. - * - * Note that the top bit (bit 63) in the returned value should be considered - * as garbage. It is not cleared here because callers are likely to use a - * multiplier on the returned value which can get rid of the top bit - * implicitly by making the multiplier even, therefore saving on a runtime - * clear-bit instruction. Otherwise caller must remember to clear the top - * bit explicitly. - */ - -/* this is used only to give gcc a clue about good code generation */ -typedef union { - struct { -#if defined(__LITTLE_ENDIAN) - u32 lo, hi; -#elif defined(__BIG_ENDIAN) - u32 hi, lo; -#endif - }; - u64 val; -} cnt32_to_63_t; - -#define cnt32_to_63(cnt_lo) \ -({ \ - static volatile u32 __m_cnt_hi = 0; \ - cnt32_to_63_t __x; \ - __x.hi = __m_cnt_hi; \ - __x.lo = (cnt_lo); \ - if (unlikely((s32)(__x.hi ^ __x.lo) < 0)) \ - __m_cnt_hi = __x.hi = (__x.hi ^ 0x80000000) + (__x.hi >> 31); \ - __x.val; \ -}) - -#endif diff --git a/trunk/include/asm-arm/div64.h b/trunk/include/asm-arm/div64.h index 37e0a96e8789..3682616804ca 100644 --- a/trunk/include/asm-arm/div64.h +++ b/trunk/include/asm-arm/div64.h @@ -27,7 +27,7 @@ #define __xh "r1" #endif -#define __do_div_asm(n, base) \ +#define do_div(n,base) \ ({ \ register unsigned int __base asm("r4") = base; \ register unsigned long long __n asm("r0") = n; \ @@ -45,182 +45,4 @@ __rem; \ }) -#if __GNUC__ < 4 - -/* - * gcc versions earlier than 4.0 are simply too problematic for the - * optimized implementation below. First there is gcc PR 15089 that - * tend to trig on more complex constructs, spurious .global __udivsi3 - * are inserted even if none of those symbols are referenced in the - * generated code, and those gcc versions are not able to do constant - * propagation on long long values anyway. - */ -#define do_div(n, base) __do_div_asm(n, base) - -#elif __GNUC__ >= 4 - -#include - -/* - * If the divisor happens to be constant, we determine the appropriate - * inverse at compile time to turn the division into a few inline - * multiplications instead which is much faster. And yet only if compiling - * for ARMv4 or higher (we need umull/umlal) and if the gcc version is - * sufficiently recent to perform proper long long constant propagation. - * (It is unfortunate that gcc doesn't perform all this internally.) - */ -#define do_div(n, base) \ -({ \ - unsigned int __r, __b = (base); \ - if (!__builtin_constant_p(__b) || __b == 0 || \ - (__LINUX_ARM_ARCH__ < 4 && (__b & (__b - 1)) != 0)) { \ - /* non-constant divisor (or zero): slow path */ \ - __r = __do_div_asm(n, __b); \ - } else if ((__b & (__b - 1)) == 0) { \ - /* Trivial: __b is constant and a power of 2 */ \ - /* gcc does the right thing with this code. */ \ - __r = n; \ - __r &= (__b - 1); \ - n /= __b; \ - } else { \ - /* Multiply by inverse of __b: n/b = n*(p/b)/p */ \ - /* We rely on the fact that most of this code gets */ \ - /* optimized away at compile time due to constant */ \ - /* propagation and only a couple inline assembly */ \ - /* instructions should remain. Better avoid any */ \ - /* code construct that might prevent that. */ \ - unsigned long long __res, __x, __t, __m, __n = n; \ - unsigned int __c, __p, __z = 0; \ - /* preserve low part of n for reminder computation */ \ - __r = __n; \ - /* determine number of bits to represent __b */ \ - __p = 1 << __div64_fls(__b); \ - /* compute __m = ((__p << 64) + __b - 1) / __b */ \ - __m = (~0ULL / __b) * __p; \ - __m += (((~0ULL % __b + 1) * __p) + __b - 1) / __b; \ - /* compute __res = __m*(~0ULL/__b*__b-1)/(__p << 64) */ \ - __x = ~0ULL / __b * __b - 1; \ - __res = (__m & 0xffffffff) * (__x & 0xffffffff); \ - __res >>= 32; \ - __res += (__m & 0xffffffff) * (__x >> 32); \ - __t = __res; \ - __res += (__x & 0xffffffff) * (__m >> 32); \ - __t = (__res < __t) ? (1ULL << 32) : 0; \ - __res = (__res >> 32) + __t; \ - __res += (__m >> 32) * (__x >> 32); \ - __res /= __p; \ - /* Now sanitize and optimize what we've got. */ \ - if (~0ULL % (__b / (__b & -__b)) == 0) { \ - /* those cases can be simplified with: */ \ - __n /= (__b & -__b); \ - __m = ~0ULL / (__b / (__b & -__b)); \ - __p = 1; \ - __c = 1; \ - } else if (__res != __x / __b) { \ - /* We can't get away without a correction */ \ - /* to compensate for bit truncation errors. */ \ - /* To avoid it we'd need an additional bit */ \ - /* to represent __m which would overflow it. */ \ - /* Instead we do m=p/b and n/b=(n*m+m)/p. */ \ - __c = 1; \ - /* Compute __m = (__p << 64) / __b */ \ - __m = (~0ULL / __b) * __p; \ - __m += ((~0ULL % __b + 1) * __p) / __b; \ - } else { \ - /* Reduce __m/__p, and try to clear bit 31 */ \ - /* of __m when possible otherwise that'll */ \ - /* need extra overflow handling later. */ \ - unsigned int __bits = -(__m & -__m); \ - __bits |= __m >> 32; \ - __bits = (~__bits) << 1; \ - /* If __bits == 0 then setting bit 31 is */ \ - /* unavoidable. Simply apply the maximum */ \ - /* possible reduction in that case. */ \ - /* Otherwise the MSB of __bits indicates the */ \ - /* best reduction we should apply. */ \ - if (!__bits) { \ - __p /= (__m & -__m); \ - __m /= (__m & -__m); \ - } else { \ - __p >>= __div64_fls(__bits); \ - __m >>= __div64_fls(__bits); \ - } \ - /* No correction needed. */ \ - __c = 0; \ - } \ - /* Now we have a combination of 2 conditions: */ \ - /* 1) whether or not we need a correction (__c), and */ \ - /* 2) whether or not there might be an overflow in */ \ - /* the cross product (__m & ((1<<63) | (1<<31))) */ \ - /* Select the best insn combination to perform the */ \ - /* actual __m * __n / (__p << 64) operation. */ \ - if (!__c) { \ - asm ( "umull %Q0, %R0, %1, %Q2\n\t" \ - "mov %Q0, #0" \ - : "=&r" (__res) \ - : "r" (__m), "r" (__n) \ - : "cc" ); \ - } else if (!(__m & ((1ULL << 63) | (1ULL << 31)))) { \ - __res = __m; \ - asm ( "umlal %Q0, %R0, %Q1, %Q2\n\t" \ - "mov %Q0, #0" \ - : "+r" (__res) \ - : "r" (__m), "r" (__n) \ - : "cc" ); \ - } else { \ - asm ( "umull %Q0, %R0, %Q1, %Q2\n\t" \ - "cmn %Q0, %Q1\n\t" \ - "adcs %R0, %R0, %R1\n\t" \ - "adc %Q0, %3, #0" \ - : "=&r" (__res) \ - : "r" (__m), "r" (__n), "r" (__z) \ - : "cc" ); \ - } \ - if (!(__m & ((1ULL << 63) | (1ULL << 31)))) { \ - asm ( "umlal %R0, %Q0, %R1, %Q2\n\t" \ - "umlal %R0, %Q0, %Q1, %R2\n\t" \ - "mov %R0, #0\n\t" \ - "umlal %Q0, %R0, %R1, %R2" \ - : "+r" (__res) \ - : "r" (__m), "r" (__n) \ - : "cc" ); \ - } else { \ - asm ( "umlal %R0, %Q0, %R2, %Q3\n\t" \ - "umlal %R0, %1, %Q2, %R3\n\t" \ - "mov %R0, #0\n\t" \ - "adds %Q0, %1, %Q0\n\t" \ - "adc %R0, %R0, #0\n\t" \ - "umlal %Q0, %R0, %R2, %R3" \ - : "+r" (__res), "+r" (__z) \ - : "r" (__m), "r" (__n) \ - : "cc" ); \ - } \ - __res /= __p; \ - /* The reminder can be computed with 32-bit regs */ \ - /* only, and gcc is good at that. */ \ - { \ - unsigned int __res0 = __res; \ - unsigned int __b0 = __b; \ - __r -= __res0 * __b0; \ - } \ - /* BUG_ON(__r >= __b || __res * __b + __r != n); */ \ - n = __res; \ - } \ - __r; \ -}) - -/* our own fls implementation to make sure constant propagation is fine */ -#define __div64_fls(bits) \ -({ \ - unsigned int __left = (bits), __nr = 0; \ - if (__left & 0xffff0000) __nr += 16, __left >>= 16; \ - if (__left & 0x0000ff00) __nr += 8, __left >>= 8; \ - if (__left & 0x000000f0) __nr += 4, __left >>= 4; \ - if (__left & 0x0000000c) __nr += 2, __left >>= 2; \ - if (__left & 0x00000002) __nr += 1; \ - __nr; \ -}) - -#endif - #endif