From 00a99339f0a30b593d06f31499c4bbd965979567 Mon Sep 17 00:00:00 2001 From: Vineet Gupta Date: Fri, 7 Sep 2018 15:13:10 -0700 Subject: [PATCH 1/5] ARCv2: build: use mcpu=hs38 iso generic mcpu=archs helps gcc with better instruction selections such as 64-bit multiply MPYD before ------ 82c34b58 : 82c34b58: ld r2,[0x83068d00] 82c34b60: add_s r2,r2,0x7530 82c34b66: mov_s r0,0x989680 82c34b6c: mpymu r5,r2,r0 82c34b70: mpy r4,r2,r0 82c34b74: mov_s r0,r4 82c34b76: j_s.d [blink] 82c34b78: mov_s r1,r5 82c34b7a: nop_s after ------ 82c34b7c : 82c34b7c: ld r0,[0x83064d00] 82c34b84: add_s r0,r0,0x7530 82c34b8a: mpydu r0,r0,0x989680 82c34b92: j_s [blink] Signed-off-by: Vineet Gupta --- arch/arc/Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arc/Makefile b/arch/arc/Makefile index 99cce77ab98f2..dcc0f0cb0ca5d 100644 --- a/arch/arc/Makefile +++ b/arch/arc/Makefile @@ -18,7 +18,7 @@ KBUILD_DEFCONFIG := nsim_700_defconfig cflags-y += -fno-common -pipe -fno-builtin -mmedium-calls -D__linux__ cflags-$(CONFIG_ISA_ARCOMPACT) += -mA7 -cflags-$(CONFIG_ISA_ARCV2) += -mcpu=archs +cflags-$(CONFIG_ISA_ARCV2) += -mcpu=hs38 is_700 = $(shell $(CC) -dM -E - < /dev/null | grep -q "ARC700" && echo 1 || echo 0) From 615f64458ad890ef94abc879a66d8b27236e733a Mon Sep 17 00:00:00 2001 From: Alexey Brodkin Date: Thu, 13 Sep 2018 23:24:28 +0300 Subject: [PATCH 2/5] ARC: build: Get rid of toolchain check This check is very naive: we simply test if GCC invoked without "-mcpu=XXX" has ARC700 define set. In that case we think that GCC was built with "--with-cpu=arc700" and has libgcc built for ARC700. Otherwise if ARC700 is not defined we think that everythng was built for ARCv2. But in reality our life is much more interesting. 1. Regardless of GCC configuration (i.e. what we pass in "--with-cpu" it may generate code for any ARC core). 2. libgcc might be built with explicitly specified "--mcpu=YYY" That's exactly what happens in case of multilibbed toolchains: - GCC is configured with default settings - All the libs built for many different CPU flavors I.e. that check gets in the way of usage of multilibbed toolchains. And even non-multilibbed toolchains are affected. OpenEmbedded also builds GCC without "--with-cpu" because each and every target component later is compiled with explicitly set "-mcpu=ZZZ". Acked-by: Rob Herring Signed-off-by: Alexey Brodkin Signed-off-by: Vineet Gupta --- arch/arc/Makefile | 14 -------------- 1 file changed, 14 deletions(-) diff --git a/arch/arc/Makefile b/arch/arc/Makefile index dcc0f0cb0ca5d..0dbce02356172 100644 --- a/arch/arc/Makefile +++ b/arch/arc/Makefile @@ -20,20 +20,6 @@ cflags-y += -fno-common -pipe -fno-builtin -mmedium-calls -D__linux__ cflags-$(CONFIG_ISA_ARCOMPACT) += -mA7 cflags-$(CONFIG_ISA_ARCV2) += -mcpu=hs38 -is_700 = $(shell $(CC) -dM -E - < /dev/null | grep -q "ARC700" && echo 1 || echo 0) - -ifdef CONFIG_ISA_ARCOMPACT -ifeq ($(is_700), 0) - $(error Toolchain not configured for ARCompact builds) -endif -endif - -ifdef CONFIG_ISA_ARCV2 -ifeq ($(is_700), 1) - $(error Toolchain not configured for ARCv2 builds) -endif -endif - ifdef CONFIG_ARC_CURR_IN_REG # For a global register defintion, make sure it gets passed to every file # We had a customer reported bug where some code built in kernel was NOT using From 7c2020c3022dfa6801d420b55ef0c507322d2c60 Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Fri, 14 Sep 2018 12:27:27 +0100 Subject: [PATCH 3/5] ARC: fix spelling mistake "entires" -> "entries" Trivial fix to spelling mistake in Kconfig Signed-off-by: Colin Ian King Signed-off-by: Vineet Gupta --- arch/arc/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arc/Kconfig b/arch/arc/Kconfig index b4441b0764d71..a045f30860477 100644 --- a/arch/arc/Kconfig +++ b/arch/arc/Kconfig @@ -149,7 +149,7 @@ config ARC_CPU_770 Support for ARC770 core introduced with Rel 4.10 (Summer 2011) This core has a bunch of cool new features: -MMU-v3: Variable Page Sz (4k, 8k, 16k), bigger J-TLB (128x4) - Shared Address Spaces (for sharing TLB entires in MMU) + Shared Address Spaces (for sharing TLB entries in MMU) -Caches: New Prog Model, Region Flush -Insns: endian swap, load-locked/store-conditional, time-stamp-ctr From 40660f1fcee8d524a60b5101538e42b1f39f106d Mon Sep 17 00:00:00 2001 From: Alexey Brodkin Date: Sun, 16 Sep 2018 23:47:57 +0300 Subject: [PATCH 4/5] ARC: build: Don't set CROSS_COMPILE in arch's Makefile There's not much sense in doing that because if user or his build-system didn't set CROSS_COMPILE we still may very well make incorrect guess. But as it turned out setting CROSS_COMPILE is not as harmless as one may think: with recent changes that implemented automatic discovery of __host__ gcc features unconditional setup of CROSS_COMPILE leads to failures on execution of "make xxx_defconfig" with absent cross-compiler, for more info see [1]. Set CROSS_COMPILE as well gets in the way if we want only to build .dtb's (again with absent cross-compiler which is not really needed for building .dtb's), see [2]. Note, we had to change LIBGCC assignment type from ":=" to "=" so that is is resolved on its usage, otherwise if it is resolved at declaration time with missing CROSS_COMPILE we're getting this error message from host GCC: | gcc: error: unrecognized command line option -mmedium-calls | gcc: error: unrecognized command line option -mno-sdata [1] http://lists.infradead.org/pipermail/linux-snps-arc/2018-September/004308.html [2] http://lists.infradead.org/pipermail/linux-snps-arc/2018-September/004320.html Signed-off-by: Alexey Brodkin Cc: Masahiro Yamada Cc: Rob Herring Signed-off-by: Vineet Gupta --- arch/arc/Makefile | 10 +--------- 1 file changed, 1 insertion(+), 9 deletions(-) diff --git a/arch/arc/Makefile b/arch/arc/Makefile index 0dbce02356172..644815c0516e7 100644 --- a/arch/arc/Makefile +++ b/arch/arc/Makefile @@ -6,14 +6,6 @@ # published by the Free Software Foundation. # -ifeq ($(CROSS_COMPILE),) -ifndef CONFIG_CPU_BIG_ENDIAN -CROSS_COMPILE := arc-linux- -else -CROSS_COMPILE := arceb-linux- -endif -endif - KBUILD_DEFCONFIG := nsim_700_defconfig cflags-y += -fno-common -pipe -fno-builtin -mmedium-calls -D__linux__ @@ -65,7 +57,7 @@ cflags-$(disable_small_data) += -mno-sdata -fcall-used-gp cflags-$(CONFIG_CPU_BIG_ENDIAN) += -mbig-endian ldflags-$(CONFIG_CPU_BIG_ENDIAN) += -EB -LIBGCC := $(shell $(CC) $(cflags-y) --print-libgcc-file-name) +LIBGCC = $(shell $(CC) $(cflags-y) --print-libgcc-file-name) # Modules with short calls might break for calls into builtin-kernel KBUILD_CFLAGS_MODULE += -mlong-calls -mno-millicode From c58a584f05e35d1d4342923cd7aac07d9c3d3d16 Mon Sep 17 00:00:00 2001 From: Vineet Gupta Date: Fri, 5 Oct 2018 12:48:48 -0700 Subject: [PATCH 5/5] ARC: clone syscall to setp r25 as thread pointer Per ARC TLS ABI, r25 is designated TP (thread pointer register). However so far kernel didn't do any special treatment, like setting up usermode r25, even for CLONE_SETTLS. We instead relied on libc runtime to do this, in say clone libc wrapper [1]. This was deliberate to keep kernel ABI agnostic (userspace could potentially change TP, specially for different ARC ISA say ARCompact vs. ARCv2 with different spare registers etc) However userspace setting up r25, after clone syscall opens a race, if child is not scheduled and gets a signal instead. It starts off in userspace not in clone but in a signal handler and anything TP sepcific there such as pthread_self() fails which showed up with uClibc testsuite nptl/tst-kill6 [2] Fix this by having kernel populate r25 to TP value. So this locks in ABI, but it was not going to change anyways, and fwiw is same for both ARCompact (arc700 core) and ARCvs (HS3x cores) [1] https://cgit.uclibc-ng.org/cgi/cgit/uclibc-ng.git/tree/libc/sysdeps/linux/arc/clone.S [2] https://github.com/wbx-github/uclibc-ng-test/blob/master/test/nptl/tst-kill6.c Fixes: ARC STAR 9001378481 Cc: stable@vger.kernel.org Reported-by: Nikita Sobolev Signed-off-by: Vineet Gupta --- arch/arc/kernel/process.c | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) diff --git a/arch/arc/kernel/process.c b/arch/arc/kernel/process.c index 4674541eba3fd..8ce6e72359155 100644 --- a/arch/arc/kernel/process.c +++ b/arch/arc/kernel/process.c @@ -241,6 +241,26 @@ int copy_thread(unsigned long clone_flags, task_thread_info(current)->thr_ptr; } + + /* + * setup usermode thread pointer #1: + * when child is picked by scheduler, __switch_to() uses @c_callee to + * populate usermode callee regs: this works (despite being in a kernel + * function) since special return path for child @ret_from_fork() + * ensures those regs are not clobbered all the way to RTIE to usermode + */ + c_callee->r25 = task_thread_info(p)->thr_ptr; + +#ifdef CONFIG_ARC_CURR_IN_REG + /* + * setup usermode thread pointer #2: + * however for this special use of r25 in kernel, __switch_to() sets + * r25 for kernel needs and only in the final return path is usermode + * r25 setup, from pt_regs->user_r25. So set that up as well + */ + c_regs->user_r25 = c_callee->r25; +#endif + return 0; }