[PATCH 5/9] powerpc: Restore FPU/VEC/VSX if previously used

benchmarking it was determined that avoiding C in the common case is a performance benefit. The full check in asm greatly complicated that codepath for a negligible performance gain and the trade-off was deemed not worth it. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/i

[PATCH 0/9] FP/VEC/VSX switching optimisations

- Reordered some code in basic-asm.h and neatened some comments Cyril Bur (9): selftests/powerpc: Test the preservation of FPU and VMX regs across syscall selftests/powerpc: Test preservation of FPU and VMX regs across preemption selftests/powerpc: Test FPU and VMX regs in signal

[PATCH 2/9] selftests/powerpc: Test preservation of FPU and VMX regs across preemption

Loop in assembly checking the registers with many threads. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- tools/testing/selftests/powerpc/math/.gitignore| 2 + tools/testing/selftests/powerpc/math/Makefile | 5 +- tools/testing/selftests/powerpc/math/fpu_asm.S

[PATCH 3/9] selftests/powerpc: Test FPU and VMX regs in signal ucontext

Load up the non volatile FPU and VMX regs and ensure that they are the expected value in a signal handler Signed-off-by: Cyril Bur <cyril...@gmail.com> --- tools/testing/selftests/powerpc/math/.gitignore | 2 + tools/testing/selftests/powerpc/math/Makefile | 4 +- tools/t

[PATCH 7/9] powerpc: Add the ability to save FPU without giving it up

to userspace with hot registers avoiding a possibly pointless reload of FPU register state. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/include/asm/switch_to.h | 3 ++- arch/powerpc/kernel/fpu.S| 21 - arch/powerpc/kernel/process.c

Re: [PATCH v5 1/9] selftests/powerpc: Test the preservation of FPU and VMX regs across syscall

On Tue, 23 Feb 2016 14:38:14 +1100 Cyril Bur <cyril...@gmail.com> wrote: > Test that the non volatile floating point and Altivec registers get > correctly preserved across the fork() syscall. > > fork() works nicely for this purpose, the registers should be the same for >

Re: [PATCH 0/9] FP/VEC/VSX switching optimisations

On Mon, 29 Feb 2016 17:53:42 +1100 Cyril Bur <cyril...@gmail.com> wrote: Apologies this should have had a Version 6 prefix! > Cover-letter for V1 of the series is at > https://lists.ozlabs.org/pipermail/linuxppc-dev/2015-November/136350.html > > Cover-letter for V2 of the se

Re: selftests/powerpc: Remove -flto from common CFLAGS

On Mon, 29 Feb 2016 22:10:13 +1100 (AEDT) Michael Ellerman wrote: > Hi Suraj, > > On Mon, 2016-29-02 at 06:29:55 UTC, Suraj Jitindar Singh wrote: > > LTO can cause GCC to inline some functions which have attributes set. The > > You should define what LTO is the first

Re: [PATCH v5 1/9] selftests/powerpc: Test the preservation of FPU and VMX regs across syscall

On Thu, 25 Feb 2016 11:52:05 +0530 "Naveen N. Rao" <naveen.n@linux.vnet.ibm.com> wrote: > On 2016/02/25 10:44AM, Cyril Bur wrote: > > On Wed, 24 Feb 2016 19:57:38 +0530 > > "Naveen N. Rao" <naveen.n@linux.vnet.ibm.com> wrote: >

Re: [PATCH v5 1/9] selftests/powerpc: Test the preservation of FPU and VMX regs across syscall

On Wed, 24 Feb 2016 19:57:38 +0530 "Naveen N. Rao" <naveen.n@linux.vnet.ibm.com> wrote: > On 2016/02/23 02:38PM, Cyril Bur wrote: > > Test that the non volatile floating point and Altivec registers get > > correctly preserved across the fork() syscal

[PATCH v5 8/9] powerpc: Add the ability to save Altivec without giving it up

reload of VEC state. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/include/asm/switch_to.h | 3 ++- arch/powerpc/kernel/process.c| 12 +++- arch/powerpc/kernel/vector.S | 24 3 files changed, 17 insertions(+), 22 deletions(-)

[PATCH v5 6/9] powerpc: Prepare for splitting giveup_{fpu, altivec, vsx} in two

the saving of registers to thread structs and leave threads MSR with bits enabled. This patch introduces no functional change. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/include/asm/reg.h | 8 arch/powerpc/include/asm/switch_to.h | 7 +++ arch/powerpc/

[PATCH v5 1/9] selftests/powerpc: Test the preservation of FPU and VMX regs across syscall

Test that the non volatile floating point and Altivec registers get correctly preserved across the fork() syscall. fork() works nicely for this purpose, the registers should be the same for both parent and child Signed-off-by: Cyril Bur <cyril...@gmail.com> --- tools/testing/selftests/p

[PATCH v5 0/9] FP/VEC/VSX switching optimisations

with the custom ASM. - Added some extra debugging output to the vmx_signal testcase - Fixed comments in testing code - Updated VSX test code to use GCC Altivec macros Cyril Bur (9): selftests/powerpc: Test the preservation of FPU and VMX regs across syscall selftests/powerpc: Test preservation

[PATCH v5 3/9] selftests/powerpc: Test FPU and VMX regs in signal ucontext

Load up the non volatile FPU and VMX regs and ensure that they are the expected value in a signal handler Signed-off-by: Cyril Bur <cyril...@gmail.com> --- tools/testing/selftests/powerpc/math/.gitignore | 2 + tools/testing/selftests/powerpc/math/Makefile | 4 +- tools/t

[PATCH v5 4/9] powerpc: Explicitly disable math features when copying thread

are hot while not having MSR_{FP,VEC,VSX} means that the registers must be loaded. This allows for a smarter return to userspace. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/kernel/process.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/powerpc/kernel/process.c

[PATCH v5 5/9] powerpc: Restore FPU/VEC/VSX if previously used

benchmarking it was determined that avoiding C in the common case is a performance benefit. The full check in asm greatly complicated that codepath for a negligible performance gain and the trade-off was deemed not worth it. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/i

[PATCH v5 9/9] powerpc: Add the ability to save VSX without giving it up

pointless reload of VSX state. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/include/asm/switch_to.h | 4 arch/powerpc/kernel/ppc_ksyms.c | 4 arch/powerpc/kernel/process.c| 42 +--- arch/powerpc/kernel/vector.S

[PATCH v5 7/9] powerpc: Add the ability to save FPU without giving it up

to userspace with hot registers avoiding a possibly pointless reload of FPU register state. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/include/asm/switch_to.h | 3 ++- arch/powerpc/kernel/fpu.S| 21 - arch/powerpc/kernel/process.c

[PATCH v5 2/9] selftests/powerpc: Test preservation of FPU and VMX regs across preemption

Loop in assembly checking the registers with many threads. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- tools/testing/selftests/powerpc/math/.gitignore| 2 + tools/testing/selftests/powerpc/math/Makefile | 5 +- tools/testing/selftests/powerpc/math/fpu_asm.S

[PATCH v3 2/9] selftests/powerpc: Test preservation of FPU and VMX regs across preemption

Loop in assembly checking the registers with many threads. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- tools/testing/selftests/powerpc/math/.gitignore| 2 + tools/testing/selftests/powerpc/math/Makefile | 5 +- tools/testing/selftests/powerpc/math/fpu_asm.S

[PATCH v3 4/9] powerpc: Explicitly disable math features when copying thread

are hot while not having MSR_{FP,VEC,VSX} means that the registers must be loaded. This allows for a smarter return to userspace. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/kernel/process.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/powerpc/kernel/process.c

[PATCH v3 1/9] selftests/powerpc: Test the preservation of FPU and VMX regs across syscall

Test that the non volatile floating point and Altivec registers get correctly preserved across the fork() syscall. fork() works nicely for this purpose, the registers should be the same for both parent and child Signed-off-by: Cyril Bur <cyril...@gmail.com> --- tools/testing/selftests/p

[PATCH v3 5/9] powerpc: Restore FPU/VEC/VSX if previously used

benchmarking it was determined that avoiding C in the common case is a performance benefit. The full check in asm greatly complicated that codepath for a negligible performance gain and the trade-off was deemed not worth it. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/i

[PATCH v3 9/9] powerpc: Add the ability to save VSX without giving it up

pointless reload of VSX state. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/include/asm/switch_to.h | 4 arch/powerpc/kernel/ppc_ksyms.c | 4 arch/powerpc/kernel/process.c| 42 +--- arch/powerpc/kernel/vector.S

[PATCH v3 8/9] powerpc: Add the ability to save Altivec without giving it up

reload of VEC state. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/include/asm/switch_to.h | 3 ++- arch/powerpc/kernel/process.c| 12 +++- arch/powerpc/kernel/vector.S | 24 3 files changed, 17 insertions(+), 22 deletions(-)

[PATCH v3 7/9] powerpc: Add the ability to save FPU without giving it up

to userspace with hot registers avoiding a possibly pointless reload of FPU register state. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/include/asm/switch_to.h | 3 ++- arch/powerpc/kernel/fpu.S| 21 - arch/powerpc/kernel/process.c

[PATCH v3 6/9] powerpc: Prepare for splitting giveup_{fpu, altivec, vsx} in two

the saving of registers to thread structs and leave threads MSR with bits enabled. This patch introduces no functional change. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/include/asm/switch_to.h | 7 +++ arch/powerpc/kernel/process.c

[PATCH v3 3/9] selftests/powerpc: Test FPU and VMX regs in signal ucontext

Load up the non volatile FPU and VMX regs and ensure that they are the expected value in a signal handler Signed-off-by: Cyril Bur <cyril...@gmail.com> --- tools/testing/selftests/powerpc/math/.gitignore | 2 + tools/testing/selftests/powerpc/math/Makefile | 4 +- tools/t

[PATCH v3 0/9] FP/VEC/VSX switching optimisations

commit message in 4/9 better reflect the patch - Removed overuse of #ifdef blocks and redundant condition in 5/9 - Split 6/8 in two to better prepare for 7,8,9 - Removed #ifdefs in 6/9 Cyril Bur (9): selftests/powerpc: Test the preservation of FPU and VMX regs across syscall selftests

Re: [PATCH v3 4/9] powerpc: Explicitly disable math features when copying thread

On Mon, 25 Jan 2016 11:04:23 +1100 Balbir Singh <bsinghar...@gmail.com> wrote: > On Thu, 21 Jan 2016 11:55:44 +1100 > Cyril Bur <cyril...@gmail.com> wrote: > > > Currently when threads get scheduled off they always giveup the FPU, > > Altivec (VMX) and Vector (V

[PATCH v4 0/9] FP/VEC/VSX switching optimisations

#ifdefs in V3 (6/9) - Reordered some conditions in if statements Cyril Bur (9): selftests/powerpc: Test the preservation of FPU and VMX regs across syscall selftests/powerpc: Test preservation of FPU and VMX regs across preemption selftests/powerpc: Test FPU and VMX regs in signal ucontext

[PATCH v4 2/9] selftests/powerpc: Test preservation of FPU and VMX regs across preemption

Loop in assembly checking the registers with many threads. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- tools/testing/selftests/powerpc/math/.gitignore| 2 + tools/testing/selftests/powerpc/math/Makefile | 5 +- tools/testing/selftests/powerpc/math/fpu_asm.S

[PATCH v4 3/9] selftests/powerpc: Test FPU and VMX regs in signal ucontext

Load up the non volatile FPU and VMX regs and ensure that they are the expected value in a signal handler Signed-off-by: Cyril Bur <cyril...@gmail.com> --- tools/testing/selftests/powerpc/math/.gitignore | 2 + tools/testing/selftests/powerpc/math/Makefile | 4 +- tools/t

[PATCH v4 1/9] selftests/powerpc: Test the preservation of FPU and VMX regs across syscall

Test that the non volatile floating point and Altivec registers get correctly preserved across the fork() syscall. fork() works nicely for this purpose, the registers should be the same for both parent and child Signed-off-by: Cyril Bur <cyril...@gmail.com> --- tools/testing/selftests/p

[PATCH v4 4/9] powerpc: Explicitly disable math features when copying thread

are hot while not having MSR_{FP,VEC,VSX} means that the registers must be loaded. This allows for a smarter return to userspace. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/kernel/process.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/powerpc/kernel/process.c

[PATCH v4 5/9] powerpc: Restore FPU/VEC/VSX if previously used

benchmarking it was determined that avoiding C in the common case is a performance benefit. The full check in asm greatly complicated that codepath for a negligible performance gain and the trade-off was deemed not worth it. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/i

[PATCH v4 6/9] powerpc: Prepare for splitting giveup_{fpu, altivec, vsx} in two

the saving of registers to thread structs and leave threads MSR with bits enabled. This patch introduces no functional change. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/include/asm/reg.h | 8 arch/powerpc/include/asm/switch_to.h | 7 +++ arch/powerpc/

[PATCH v4 8/9] powerpc: Add the ability to save Altivec without giving it up

reload of VEC state. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/include/asm/switch_to.h | 3 ++- arch/powerpc/kernel/process.c| 12 +++- arch/powerpc/kernel/vector.S | 24 3 files changed, 17 insertions(+), 22 deletions(-)

[PATCH v4 7/9] powerpc: Add the ability to save FPU without giving it up

to userspace with hot registers avoiding a possibly pointless reload of FPU register state. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/include/asm/switch_to.h | 3 ++- arch/powerpc/kernel/fpu.S| 21 - arch/powerpc/kernel/process.c

[PATCH v4 9/9] powerpc: Add the ability to save VSX without giving it up

pointless reload of VSX state. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/include/asm/switch_to.h | 4 arch/powerpc/kernel/ppc_ksyms.c | 4 arch/powerpc/kernel/process.c| 42 +--- arch/powerpc/kernel/vector.S

Re: [PATCH v4 1/9] selftests/powerpc: Test the preservation of FPU and VMX regs across syscall

On Mon, 15 Feb 2016 22:29:17 +0530 "Naveen N. Rao" <naveen.n@linux.vnet.ibm.com> wrote: > On 2016/02/15 04:07PM, Cyril Bur wrote: > > Test that the non volatile floating point and Altivec registers get > > correctly preserved across the fork() syscal

[PATCH] selftests/powerpc: PMU libs: use signed long to read perf_event_paranoid

. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- tools/testing/selftests/powerpc/pmu/lib.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/tools/testing/selftests/powerpc/pmu/lib.c b/tools/testing/selftests/powerpc/pmu/lib.c index a361ad3..8b992fa 100644 --- a/tools/t

Re: selftests/powerpc: Remove -flto from common CFLAGS

IL *.out cpu_event_pinned_vs_ebb_test.out:[FAIL] Test FAILED on line 87 ipc_unmuxed.out:[FAIL] Test FAILED on line 38 per_event_excludes.out:[FAIL] Test FAILED on line 95 There were no matches for grep SKIP *.out for any of the four. This patch appears to not have affected any of the tests. Reviewed-

[PATCH 1/2] powerpc: Batch up loads/stores on saving and restoring Altivec

instruction. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- These patches need to be applied on to of my rework of FPU/VMX/VSX switching: https://patchwork.ozlabs.org/patch/589703/ I left in some of my comments indicating if functions are called from C or not. Looking at them now, they might be

[PATCH 2/2] powerpc: Batch up loads/stores on saving and restoring VSX

to the overlapping layout of FPU registers and VSX registers on POWER chips, this patch also benefits FPU loads and stores when VSX is compiled in and the CPU is VSX capable. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/include/asm/ppc_asm.

Re: [PATCH V10 02/28] powerpc, process: Add the function flush_tmregs_to_thread

On Tue, 16 Feb 2016 14:29:32 +0530 Anshuman Khandual wrote: > This patch creates a function flush_tmregs_to_thread which > will then be used by subsequent patches in this series. The > function checks for self tracing ptrace interface attempts > while in the TM

Re: [PATCH V10 18/28] selftests, powerpc: Add ptrace tests for GPR/FPR registers

On Tue, 16 Feb 2016 14:29:48 +0530 Anshuman Khandual wrote: > This patch adds ptrace interface test for GPR/FPR registers. > This adds ptrace interface based helper functions related to > GPR/FPR access and some assembly helper functions related to > GPR/FPR

Re: [PATCH V10 17/28] selftests, powerpc: Add ptrace tests for EBB

On Tue, 16 Feb 2016 14:29:47 +0530 Anshuman Khandual wrote: > This patch adds ptrace interface test for EBB specific > registers. This also adds some generic ptrace interface > based helper functions to be used by other patches later > on in the series. > >

Re: [PATCH 1/2] powerpc: Batch up loads/stores on saving and restoring Altivec

On Thu, 10 Mar 2016 10:01:07 +1100 Michael Neuling <mi...@neuling.org> wrote: > On Tue, 2016-03-01 at 16:55 +1100, Cyril Bur wrote: > > > Currently the assembly to save and restore Altivec registers boils down to > > a load immediate of the offset of the specific Al

[PATCH] powerpc: Fix possible unrecoverable exception caused by '70fe3d9'

a better fix in future. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/kernel/entry_64.S | 9 + 1 file changed, 9 insertions(+) diff --git a/arch/powerpc/kernel/entry_64.S b/arch/powerpc/kernel/entry_64.S index 038e0a1..f3aa4b4 100644 --- a/arch/powerpc/kernel/en

Re: [PATCH 1/2] powerpc: Batch up loads/stores on saving and restoring Altivec

On Thu, 10 Mar 2016 16:37:47 +1100 Cyril Bur <cyril...@gmail.com> wrote: > On Thu, 10 Mar 2016 10:01:07 +1100 > Michael Neuling <mi...@neuling.org> wrote: > > > On Tue, 2016-03-01 at 16:55 +1100, Cyril Bur wrote: > > > > > Currently the assembly t

Re: [PATCH 2/2] powerpc: Batch up loads/stores on saving and restoring VSX

On Thu, 10 Mar 2016 11:09:32 +1100 Michael Neuling <mi...@neuling.org> wrote: > On Tue, 2016-03-01 at 16:55 +1100, Cyril Bur wrote: > > > Currently the assembly to save and restore VSX registers boils down to a > > load immediate of the offset of the specific VSX registe

Re: [PATCH 1/2] powerpc: Batch up loads/stores on saving and restoring Altivec

On Thu, 10 Mar 2016 10:01:07 +1100 Michael Neuling <mi...@neuling.org> wrote: > On Tue, 2016-03-01 at 16:55 +1100, Cyril Bur wrote: > > > Currently the assembly to save and restore Altivec registers boils down to > > a load immediate of the offset of the specific Al

Re: [PATCH] powerpc: Fix possible unrecoverable exception caused by '70fe3d9'

On Wed, 16 Mar 2016 22:45:20 +1100 Benjamin Herrenschmidt <b...@au1.ibm.com> wrote: > On Wed, 2016-03-16 at 13:29 +1100, Cyril Bur wrote: > > +#ifdef CONFIG_PPC_BOOK3S > > +   ld  r10,PACAKMSR(r13) > > +   li  r9,MSR_RI > > +   

[PATCH 5/5] powerpc: Preserve the SPR values across fork() syscalls

Currently copy_thread() doesn't flush SPRs to the parent thread struct. Currently this only affects the TAR register as perf takes care of some of the others and the remaining ones are all Event Based Branch (EBB) registers which are cleared across fork(). Signed-off-by: Cyril Bur <cy

[PATCH 3/5] powerpc: Fix TAR leak across exec() syscalls

as the Event Based Branch (EBB) registers are left alone by design as these fall into the same category as leaving file descriptors open across exec(), it is up the parent thread to sanitise what it deems necessary. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/kernel/pro

[PATCH 1/5] selftests/powerpc: Add exec() test to check for spr sanitisation

Signed-off-by: Cyril Bur <cyril...@gmail.com> --- tools/testing/selftests/powerpc/syscalls/Makefile | 3 +- .../testing/selftests/powerpc/syscalls/spr_exec.c | 78 ++ 2 files changed, 80 insertions(+), 1 deletion(-) create mode 100644 tools/testing/selftests/p

[PATCH 2/5] selftests/powerpc: Add fork() test to check for spr being preserved

Signed-off-by: Cyril Bur <cyril...@gmail.com> --- tools/testing/selftests/powerpc/syscalls/Makefile | 3 +- .../testing/selftests/powerpc/syscalls/spr_fork.c | 78 ++ 2 files changed, 80 insertions(+), 1 deletion(-) create mode 100644 tools/testing/selftests/p

[PATCH 4/5] powerpc: Move flush_all_to_thread() below save_sprs()

Signed-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/kernel/process.c | 34 +- 1 file changed, 17 insertions(+), 17 deletions(-) diff --git a/arch/powerpc/kernel/process.c b/arch/powerpc/kernel/process.c index 56444a6..7625976 100644 --- a/arch/p

[PATCH 1/2] selftests/powerpc: benchmarks/context_switch.c improve usage message

Signed-off-by: Cyril Bur <cyril...@gmail.com> --- tools/testing/selftests/powerpc/benchmarks/context_switch.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/tools/testing/selftests/powerpc/benchmarks/context_switch.c b/tools/testing/selftests/powerpc/benc

[PATCH 2/2] selftests/powerpc: benchmarks/context_switch.c use vector instructions/types

play well with __attribute__((__target__("no-vsx")), LTO can cause GCC to forget about the attribute and compile with VSX instructions regardless. Be weary when enabling -flfo for this test. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- tools/testing/selftests/powerpc/benchmarks/Makefile

Re: [PATCH V10 02/28] powerpc, process: Add the function flush_tmregs_to_thread

On Wed, 02 Mar 2016 09:59:06 +0530 Anshuman Khandual <khand...@linux.vnet.ibm.com> wrote: > On 03/02/2016 05:45 AM, Cyril Bur wrote: > > On Tue, 16 Feb 2016 14:29:32 +0530 > > Anshuman Khandual <khand...@linux.vnet.ibm.com> wrote: > > > >> This patch

[PATCH V2 1/5] selftests/powerpc: Add exec() test to check for spr sanitisation

Signed-off-by: Cyril Bur <cyril...@gmail.com> --- tools/testing/selftests/powerpc/syscalls/Makefile | 3 +- .../testing/selftests/powerpc/syscalls/spr_exec.c | 78 ++ 2 files changed, 80 insertions(+), 1 deletion(-) create mode 100644 tools/testing/selftests/p

[PATCH V2 4/5] powerpc: Move flush_all_to_thread() below save_sprs()

Signed-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/kernel/process.c | 34 +- 1 file changed, 17 insertions(+), 17 deletions(-) diff --git a/arch/powerpc/kernel/process.c b/arch/powerpc/kernel/process.c index 56444a6..7625976 100644 --- a/arch/p

[PATCH V2 5/5] powerpc: Preserve the SPR values across fork() syscalls

Currently copy_thread() doesn't flush SPRs to the parent thread struct. Currently this only affects the TAR register as perf takes care of some of the others and the remaining ones are all Event Based Branch (EBB) registers which are cleared across fork(). Signed-off-by: Cyril Bur <cy

[PATCH V2 3/5] powerpc: Fix SPR leak across exec() syscalls

thread to sanitise what it deems necessary. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/kernel/process.c | 14 ++ 1 file changed, 14 insertions(+) diff --git a/arch/powerpc/kernel/process.c b/arch/powerpc/kernel/process.c index d7a9df5..56444a6 100644 ---

[PATCH V2 2/5] selftests/powerpc: Add fork() test to check for spr being preserved

Signed-off-by: Cyril Bur <cyril...@gmail.com> --- tools/testing/selftests/powerpc/syscalls/Makefile | 3 +- .../testing/selftests/powerpc/syscalls/spr_fork.c | 78 ++ 2 files changed, 80 insertions(+), 1 deletion(-) create mode 100644 tools/testing/selftests/p

Re: [PATCH] powerpc: Add support for userspace P9 copy paste

ted > processor is detected. > > NOTE: this is for userspace only, not in kernel, and does not deal > with KVM guests. > > Patch created with much assistance from Michael Neuling > <mi...@neuling.org> > Hi Chris, Patch looks good. Looks like you've put 8 spaces (instead o

Re: [PATCH] powerpc: Improve comment explaining why we modify VRSAVE

than zero, it might be worth noting that all 1's is 'best' in case anyone does ever follow that note. Perhaps ending with "... Set it to all 1s as a best effort to adhere to the programming note in '5.3.3 VR Save Register' of the ISA" Having said all that, a reminder that glibc does look at

Re: [PATCH] selftests/powerpc: Test cp_abort during context switch

Hi Chris, I must admit I didn't run on it on real hardware ;). Looks good. > Patch created with much assistance from Michael Neuling > <mi...@neuling.org> > > Signed-off-by: Chris Smart <ch...@distroguy.com> Reviewed-by: Cyril Bur <cyril...@gmail.com> > --- &

[PATCH v2 00/20] Consistent TM structures

As ptrace code was merged between v1 and v2, work was needed there to make it fit in with the final three patches. The overall aim of this series may have gotten lost here. The final three patches are the goal here. Cyril Bur (20): selftests/powerpc: Compile selftests against headers without

[PATCH v2 09/20] selftests/powerpc: Introduce GPR asm helper header file

Signed-off-by: Cyril Bur <cyril...@gmail.com> --- tools/testing/selftests/powerpc/gpr_asm.h | 96 +++ 1 file changed, 96 insertions(+) create mode 100644 tools/testing/selftests/powerpc/gpr_asm.h diff --git a/tools/testing/selftests/powerpc/gpr_asm.h b/tools/t

[PATCH v2 06/20] selftests/powerpc: Check for VSX preservation across userspace preemption

Ensure the kernel correctly switches VSX registers correctly. VSX registers are all volatile, and despite the kernel preserving VSX across syscalls, it doesn't have to. Test that during interrupts and timeslices ending the VSX regs remain the same. Signed-off-by: Cyril Bur <cyril...@gmail.

[PATCH v2 02/20] powerpc: Always restore FPU/VEC/VSX if hardware transactional memory in use

. Fixes: 70fe3d9 ("powerpc: Restore FPU/VEC/VSX if previously used") Signed-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/kernel/process.c | 8 +--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/arch/powerpc/kernel/process.c b/arch/powerpc/kernel/process

[PATCH v2 03/20] powerpc: Add check_if_tm_restore_required() to giveup_all()

giveup_all() causes FPU/VMX/VSX facilitities to be disabled in a threads MSR. If this thread was transactional this should be recorded as reclaiming/recheckpointing code will need to know. Fixes: c208505 ("powerpc: create giveup_all()") Signed-off-by: Cyril Bur <cyril...@gmail.c

[PATCH v2 07/20] selftests/powerpc: Rework FPU stack placement macros and move to header file

in an different area of the powerpc selftests Signed-off-by: Cyril Bur <cyril...@gmail.com> --- tools/testing/selftests/powerpc/fpu_asm.h | 81 ++ tools/testing/selftests/powerpc/math/fpu_asm.S | 73 ++- 2 files changed, 86 insertions(+), 68 del

[PATCH v2 10/20] selftests/powerpc: Add transactional memory defines

Signed-off-by: Cyril Bur <cyril...@gmail.com> --- tools/testing/selftests/powerpc/basic_asm.h | 4 1 file changed, 4 insertions(+) diff --git a/tools/testing/selftests/powerpc/basic_asm.h b/tools/testing/selftests/powerpc/basic_asm.h index 3349a07..5131059 100644 --- a/tools/t

[PATCH v2 15/20] selftests/powerpc: Add checks for transactional FPUs in signal contexts

If a thread receives a signal while transactional the kernel creates a second context to show the transactional state of the process. This test loads some known values and waits for a signal and confirms that the expected values are in the signal context. Signed-off-by: Cyril Bur <cy

[PATCH v2 18/20] powerpc: tm: Always use fp_state and vr_state to store live registers

. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/include/asm/processor.h | 7 +- arch/powerpc/kernel/process.c| 63 +++- arch/powerpc/kernel/ptrace.c | 278 +-- arch/powerpc/kernel/signal_32.c | 50 +++ arch/p

[PATCH v2 20/20] powerpc: Remove do_load_up_transact_{fpu,altivec}

Previous rework of TM code leaves these functions unused Signed-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/include/asm/tm.h | 5 - arch/powerpc/kernel/fpu.S | 26 -- arch/powerpc/kernel/vector.S | 25 - 3 files chang

[PATCH] powerpc: ptrace: tm_reclaim_current() in flush_tmregs_to_thread()

best to flush even for current. This patch also moves flush_tmregs_to_thread() into ptrace.c as it is only called from that file. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/include/asm/switch_to.h | 8 arch/powerpc/kernel/process.c

Re: [PATCH v2] powernv: Simplify searching for compatible device nodes

nd prd similar to how opal-i2c nodes are found. > > Signed-off-by: Jack Miller <j...@codezen.org> Using a version of the related skiboot patch that may not be the final one: Tested-by: Cyril Bur <cyril...@gmail.com> > --- > arch/powerpc/platforms/powernv/opal.c | 24

Re: [PATCH v2 18/20] powerpc: tm: Always use fp_state and vr_state to store live registers

On Sun, 2016-08-14 at 09:56 +0800, Simon Guo wrote: > On Fri, Aug 12, 2016 at 09:28:17AM +1000, Cyril Bur wrote: > > > > @@ -846,7 +834,9 @@ static void tm_reclaim_thread(struct > > thread_struct *thr, > >   if (!MSR_TM_SUSPENDED(mfmsr())) > >   ret

[PATCH v2 04/20] powerpc: Return the new MSR from msr_check_and_set()

mfmsr() is a fairly expensive call and callers of msr_check_and_set() may want to make decisions bits in the MSR that it did not change but may not know the value of. This patch would avoid a two calls to mfmsr(). Signed-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/include/asm

[PATCH v2 11/20] selftests/powerpc: Allow tests to extend their kill timeout

Signed-off-by: Cyril Bur <cyril...@gmail.com> --- tools/testing/selftests/powerpc/harness.c | 9 +++-- tools/testing/selftests/powerpc/utils.h | 2 +- 2 files changed, 8 insertions(+), 3 deletions(-) diff --git a/tools/testing/selftests/powerpc/harness.c b/tools/testing/selftests/p

[PATCH v2 08/20] selftests/powerpc: Move VMX stack frame macros to header file

Signed-off-by: Cyril Bur <cyril...@gmail.com> --- tools/testing/selftests/powerpc/math/vmx_asm.S | 85 +- tools/testing/selftests/powerpc/vmx_asm.h | 98 ++ 2 files changed, 99 insertions(+), 84 deletions(-) create mode 100644 tools/t

[PATCH v2 14/20] selftests/powerpc: Add checks for transactional GPRs in signal contexts

If a thread receives a signal while transactional the kernel creates a second context to show the transactional state of the process. This test loads some known values and waits for a signal and confirms that the expected values are in the signal context. Signed-off-by: Cyril Bur <cy

[PATCH v2 17/20] selftests/powerpc: Add checks for transactional VSXs in signal contexts

If a thread receives a signal while transactional the kernel creates a second context to show the transactional state of the process. This test loads some known values and waits for a signal and confirms that the expected values are in the signal context. Signed-off-by: Cyril Bur <cy

[PATCH v2 19/20] powerpc: tm: Rename transct_(*) to ck(\1)_state

Make the structures being used for checkpointed state named consistently with the pt_regs/ckpt_regs. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/include/asm/processor.h | 8 ++--- arch/powerpc/kernel/asm-offsets.c| 12 arch/powerpc/kernel/fpu.S

[PATCH v2 05/20] powerpc: Never giveup a reclaimed thread when enabling kernel {fp, altivec, vsx}

of the checkpointed state to the live structures within the thread struct thus destroying the true live state for that thread. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/kernel/process.c | 39 --- 1 file changed, 36 insertions

[PATCH v2 13/20] selftests/powerpc: Check that signals always get delivered

Signed-off-by: Cyril Bur <cyril...@gmail.com> --- tools/testing/selftests/powerpc/Makefile | 1 + tools/testing/selftests/powerpc/signal/Makefile| 12 +++ tools/testing/selftests/powerpc/signal/signal.S| 50 ++ tools/testing/selftests/powerpc/signal/signal.c

[PATCH v2 12/20] selftests/powerpc: Add TM tcheck helpers in C

Signed-off-by: Cyril Bur <cyril...@gmail.com> --- tools/testing/selftests/powerpc/tm/tm.h | 27 +++ 1 file changed, 27 insertions(+) diff --git a/tools/testing/selftests/powerpc/tm/tm.h b/tools/testing/selftests/powerpc/tm/tm.h index 60318ba..2c8da74 100644 --- a

[PATCH v2 16/20] selftests/powerpc: Add checks for transactional VMXs in signal contexts

If a thread receives a signal while transactional the kernel creates a second context to show the transactional state of the process. This test loads some known values and waits for a signal and confirms that the expected values are in the signal context. Signed-off-by: Cyril Bur <cy

[PATCH v2 01/20] selftests/powerpc: Compile selftests against headers without AT_HWCAP2

It might be nice to compile selftests against older kernels and headers but which may not have HWCAP2. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- tools/testing/selftests/powerpc/utils.h | 7 +++ 1 file changed, 7 insertions(+) diff --git a/tools/testing/selftests/powerpc/util

Re: [PATCH 3/5] powerpc: tm: Always use fp_state and vr_state to store live registers

On Sun, 17 Jul 2016 11:25:43 +0800 Simon Guo <simon...@linux.vnet.ibm.com> wrote: > Hi Cyril, > On Wed, Jun 08, 2016 at 02:00:34PM +1000, Cyril Bur wrote: > > @@ -917,24 +907,10 @@ static inline void tm_recheckpoint_new_task(struct > > task_struct *new) > >

[RFC 1/3] selftests/powerpc: Add test to check TM ucontext creation

Signed-off-by: Cyril Bur <cyril...@gmail.com> --- .../selftests/powerpc/tm/tm-signal-context-chk.c | 102 + 1 file changed, 102 insertions(+) create mode 100644 tools/testing/selftests/powerpc/tm/tm-signal-context-chk.c diff --git a/tools/testing/selftests/powerpc

[RFC 3/3] powerpc: tm: Enable transactional memory (TM) lazily for userspace

to avoid the expensive swap of the TM registers. A TM unavailable exception will occur if a thread does use TM and the kernel will enable MSR_TM and leave it so for some time afterwards. Signed-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/include/asm/processor.h | 1 + arch/powerpc/

[RFC 2/3] powerpc: tm: Add TM Unavailable Exception

-off-by: Cyril Bur <cyril...@gmail.com> --- arch/powerpc/kernel/traps.c | 25 + 1 file changed, 25 insertions(+) diff --git a/arch/powerpc/kernel/traps.c b/arch/powerpc/kernel/traps.c index 3e4c84d..29260ee 100644 --- a/arch/powerpc/kernel/traps.c +++ b/arch/powerpc/

Re: [PATCH 3/5] powerpc: tm: Always use fp_state and vr_state to store live registers

On Tue, 28 Jun 2016 11:53:13 +0800 Simon Guo <simon...@linux.vnet.ibm.com> wrote: > hi Cyril, > > On Wed, Jun 08, 2016 at 02:00:34PM +1000, Cyril Bur wrote: > > @@ -1108,11 +1084,11 @@ struct task_struct *__switch_to(struct task_struct > > *prev, > >

<    1   2   3   4   5   >