| Message ID | 3c1ebfa5-38d9-e6b2-fcf7-d8b17c0fa515@kernel.dk (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | arm: add support for TIF_NOTIFY_SIGNAL | expand |
On Thu, Oct 29, 2020 at 10:11:07AM -0600, Jens Axboe wrote: > Wire up TIF_NOTIFY_SIGNAL handling for arm. > > Cc: linux-arm-kernel@lists.infradead.org > Signed-off-by: Jens Axboe <axboe@kernel.dk> > --- > > 5.11 has support queued up for TIF_NOTIFY_SIGNAL, see this posting > for details: > > https://lore.kernel.org/io-uring/20201026203230.386348-1-axboe@kernel.dk/ > > As part of that work, I'm adding TIF_NOTIFY_SIGNAL support to all archs, > as that will enable a set of cleanups once all of them support it. > > This needs a bit of asm help, immediate doesn't like anything outside > of 1 byte, it seems. Any clues? Correct - immediates take an 8 bit value shifted by an even number of bits. I'm tempted to suggest that we simplify things by making TIF bits 0..15 invoke do_work_pending() no matter what - which will require a comment in thread_info.h that all those bits will have that effect. The resulting assembly would be: ldr r1, [tsk, #TI_FLAGS] @ re-check for syscall tracing - tst r1, #_TIF_SYSCALL_WORK | _TIF_WORK_MASK + movs r1, r1, lsl #16 which avoids the additional load caused by "ldr r2, =..." It's not like we're desperate for bits here. Further comments below if we don't decide on that approach... > > arch/arm/include/asm/thread_info.h | 5 ++++- > arch/arm/kernel/entry-common.S | 9 ++++++--- > arch/arm/kernel/signal.c | 2 +- > 3 files changed, 11 insertions(+), 5 deletions(-) > > diff --git a/arch/arm/include/asm/thread_info.h b/arch/arm/include/asm/thread_info.h > index 536b6b979f63..fec16d770180 100644 > --- a/arch/arm/include/asm/thread_info.h > +++ b/arch/arm/include/asm/thread_info.h > @@ -135,6 +135,7 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, > #define TIF_SYSCALL_AUDIT 5 /* syscall auditing active */ > #define TIF_SYSCALL_TRACEPOINT 6 /* syscall tracepoint instrumentation */ > #define TIF_SECCOMP 7 /* seccomp syscall filtering active */ > +#define TIF_NOTIFY_SIGNAL 8 /* signal notifications exist */ > > #define TIF_USING_IWMMXT 17 > #define TIF_MEMDIE 18 /* is terminating due to OOM killer */ > @@ -148,6 +149,7 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, > #define _TIF_SYSCALL_AUDIT (1 << TIF_SYSCALL_AUDIT) > #define _TIF_SYSCALL_TRACEPOINT (1 << TIF_SYSCALL_TRACEPOINT) > #define _TIF_SECCOMP (1 << TIF_SECCOMP) > +#define _TIF_NOTIFY_SIGNAL (1 << TIF_NOTIFY_SIGNAL) > #define _TIF_USING_IWMMXT (1 << TIF_USING_IWMMXT) > > /* Checks for any syscall work in entry-common.S */ > @@ -158,7 +160,8 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, > * Change these and you break ASM code in entry-common.S > */ > #define _TIF_WORK_MASK (_TIF_NEED_RESCHED | _TIF_SIGPENDING | \ > - _TIF_NOTIFY_RESUME | _TIF_UPROBE) > + _TIF_NOTIFY_RESUME | _TIF_UPROBE | \ > + _TIF_NOTIFY_SIGNAL) > > #endif /* __KERNEL__ */ > #endif /* __ASM_ARM_THREAD_INFO_H */ > diff --git a/arch/arm/kernel/entry-common.S b/arch/arm/kernel/entry-common.S > index 271cb8a1eba1..7485b58673b4 100644 > --- a/arch/arm/kernel/entry-common.S > +++ b/arch/arm/kernel/entry-common.S > @@ -53,7 +53,8 @@ __ret_fast_syscall: > cmp r2, #TASK_SIZE > blne addr_limit_check_failed > ldr r1, [tsk, #TI_FLAGS] @ re-check for syscall tracing > - tst r1, #_TIF_SYSCALL_WORK | _TIF_WORK_MASK > + ldr r2, =#_TIF_SYSCALL_WORK | _TIF_WORK_MASK No # required. > + tst r1, r2 > bne fast_work_pending > > > @@ -90,7 +91,8 @@ __ret_fast_syscall: > cmp r2, #TASK_SIZE > blne addr_limit_check_failed > ldr r1, [tsk, #TI_FLAGS] @ re-check for syscall tracing > - tst r1, #_TIF_SYSCALL_WORK | _TIF_WORK_MASK > + ldr r2, =#_TIF_SYSCALL_WORK | _TIF_WORK_MASK Ditto. > + tst r1, r2 > beq no_work_pending > UNWIND(.fnend ) > ENDPROC(ret_fast_syscall) > @@ -131,7 +133,8 @@ ENTRY(ret_to_user_from_irq) > cmp r2, #TASK_SIZE > blne addr_limit_check_failed > ldr r1, [tsk, #TI_FLAGS] > - tst r1, #_TIF_WORK_MASK > + ldr r2, =#_TIF_WORK_MASK Ditto.
On 10/29/20 10:29 AM, Russell King - ARM Linux admin wrote: > On Thu, Oct 29, 2020 at 10:11:07AM -0600, Jens Axboe wrote: >> Wire up TIF_NOTIFY_SIGNAL handling for arm. >> >> Cc: linux-arm-kernel@lists.infradead.org >> Signed-off-by: Jens Axboe <axboe@kernel.dk> >> --- >> >> 5.11 has support queued up for TIF_NOTIFY_SIGNAL, see this posting >> for details: >> >> https://lore.kernel.org/io-uring/20201026203230.386348-1-axboe@kernel.dk/ >> >> As part of that work, I'm adding TIF_NOTIFY_SIGNAL support to all archs, >> as that will enable a set of cleanups once all of them support it. >> >> This needs a bit of asm help, immediate doesn't like anything outside >> of 1 byte, it seems. Any clues? > > Correct - immediates take an 8 bit value shifted by an even number of > bits. > > I'm tempted to suggest that we simplify things by making TIF bits 0..15 > invoke do_work_pending() no matter what - which will require a comment > in thread_info.h that all those bits will have that effect. The > resulting assembly would be: > > ldr r1, [tsk, #TI_FLAGS] @ re-check for syscall tracing > - tst r1, #_TIF_SYSCALL_WORK | _TIF_WORK_MASK > + movs r1, r1, lsl #16 > > which avoids the additional load caused by "ldr r2, =..." > > It's not like we're desperate for bits here. So renumber TIF bits that don't need to worry about do_work_pending() > 15 then? I agree, there's plenty of bits available, so seems reasonable to me. But probably more work in terms of other bits being tested with tstcurrently - at least when I looked, I could not find any of them I could successfully remap from < 8 to > 8. Any chance I can talk you into hacking that up? > Further comments below if we don't decide on that approach... Totally up to you, of course. I'm fine with shifting the bits around.
On Thu, Oct 29, 2020 at 10:35:54AM -0600, Jens Axboe wrote: > On 10/29/20 10:29 AM, Russell King - ARM Linux admin wrote: > > On Thu, Oct 29, 2020 at 10:11:07AM -0600, Jens Axboe wrote: > >> Wire up TIF_NOTIFY_SIGNAL handling for arm. > >> > >> Cc: linux-arm-kernel@lists.infradead.org > >> Signed-off-by: Jens Axboe <axboe@kernel.dk> > >> --- > >> > >> 5.11 has support queued up for TIF_NOTIFY_SIGNAL, see this posting > >> for details: > >> > >> https://lore.kernel.org/io-uring/20201026203230.386348-1-axboe@kernel.dk/ > >> > >> As part of that work, I'm adding TIF_NOTIFY_SIGNAL support to all archs, > >> as that will enable a set of cleanups once all of them support it. > >> > >> This needs a bit of asm help, immediate doesn't like anything outside > >> of 1 byte, it seems. Any clues? > > > > Correct - immediates take an 8 bit value shifted by an even number of > > bits. > > > > I'm tempted to suggest that we simplify things by making TIF bits 0..15 > > invoke do_work_pending() no matter what - which will require a comment > > in thread_info.h that all those bits will have that effect. The > > resulting assembly would be: > > > > ldr r1, [tsk, #TI_FLAGS] @ re-check for syscall tracing > > - tst r1, #_TIF_SYSCALL_WORK | _TIF_WORK_MASK > > + movs r1, r1, lsl #16 > > > > which avoids the additional load caused by "ldr r2, =..." > > > > It's not like we're desperate for bits here. > > So renumber TIF bits that don't need to worry about do_work_pending() > > 15 then? I agree, there's plenty of bits available, so seems reasonable > to me. But probably more work in terms of other bits being tested with > tstcurrently - at least when I looked, I could not find any of them I > could successfully remap from < 8 to > 8. > > Any chance I can talk you into hacking that up? I don't believe that there's any need to do any renumbering. _TIF_WORK_MASK covers bits 0-3. _TIF_SYSCALL_WORK covers bits 4-7. Then we have 17, 18, and 20 which are not used to trigger the do_work_pending(). So, I think it's just a case of changing what you're doing in the assembly to my suggestion, and adding a comment to thread_info.h noting that bits 0-15 will trigger a call into do_work_pending() when returning to userspace.
On 10/29/20 11:01 AM, Russell King - ARM Linux admin wrote: > On Thu, Oct 29, 2020 at 10:35:54AM -0600, Jens Axboe wrote: >> On 10/29/20 10:29 AM, Russell King - ARM Linux admin wrote: >>> On Thu, Oct 29, 2020 at 10:11:07AM -0600, Jens Axboe wrote: >>>> Wire up TIF_NOTIFY_SIGNAL handling for arm. >>>> >>>> Cc: linux-arm-kernel@lists.infradead.org >>>> Signed-off-by: Jens Axboe <axboe@kernel.dk> >>>> --- >>>> >>>> 5.11 has support queued up for TIF_NOTIFY_SIGNAL, see this posting >>>> for details: >>>> >>>> https://lore.kernel.org/io-uring/20201026203230.386348-1-axboe@kernel.dk/ >>>> >>>> As part of that work, I'm adding TIF_NOTIFY_SIGNAL support to all archs, >>>> as that will enable a set of cleanups once all of them support it. >>>> >>>> This needs a bit of asm help, immediate doesn't like anything outside >>>> of 1 byte, it seems. Any clues? >>> >>> Correct - immediates take an 8 bit value shifted by an even number of >>> bits. >>> >>> I'm tempted to suggest that we simplify things by making TIF bits 0..15 >>> invoke do_work_pending() no matter what - which will require a comment >>> in thread_info.h that all those bits will have that effect. The >>> resulting assembly would be: >>> >>> ldr r1, [tsk, #TI_FLAGS] @ re-check for syscall tracing >>> - tst r1, #_TIF_SYSCALL_WORK | _TIF_WORK_MASK >>> + movs r1, r1, lsl #16 >>> >>> which avoids the additional load caused by "ldr r2, =..." >>> >>> It's not like we're desperate for bits here. >> >> So renumber TIF bits that don't need to worry about do_work_pending() > >> 15 then? I agree, there's plenty of bits available, so seems reasonable >> to me. But probably more work in terms of other bits being tested with >> tstcurrently - at least when I looked, I could not find any of them I >> could successfully remap from < 8 to > 8. >> >> Any chance I can talk you into hacking that up? > > I don't believe that there's any need to do any renumbering. You are right! > _TIF_WORK_MASK covers bits 0-3. _TIF_SYSCALL_WORK covers bits 4-7. > Then we have 17, 18, and 20 which are not used to trigger the > do_work_pending(). > > So, I think it's just a case of changing what you're doing in the > assembly to my suggestion, and adding a comment to thread_info.h > noting that bits 0-15 will trigger a call into do_work_pending() > when returning to userspace. How about this? commit c03932936d8f99ff7c1c6c7d984e7a457284396c Author: Jens Axboe <axboe@kernel.dk> Date: Fri Oct 9 16:00:49 2020 -0600 arm: add support for TIF_NOTIFY_SIGNAL Wire up TIF_NOTIFY_SIGNAL handling for arm. Cc: linux-arm-kernel@lists.infradead.org Signed-off-by: Jens Axboe <axboe@kernel.dk> diff --git a/arch/arm/include/asm/thread_info.h b/arch/arm/include/asm/thread_info.h index 536b6b979f63..eb7ce2747eb0 100644 --- a/arch/arm/include/asm/thread_info.h +++ b/arch/arm/include/asm/thread_info.h @@ -126,6 +126,8 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, * thread information flags: * TIF_USEDFPU - FPU was used by this task this quantum (SMP) * TIF_POLLING_NRFLAG - true if poll_idle() is polling TIF_NEED_RESCHED + * + * Any bit in the range of 0..15 will cause do_work_pending() to be invoked. */ #define TIF_SIGPENDING 0 /* signal pending */ #define TIF_NEED_RESCHED 1 /* rescheduling necessary */ @@ -135,6 +137,7 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, #define TIF_SYSCALL_AUDIT 5 /* syscall auditing active */ #define TIF_SYSCALL_TRACEPOINT 6 /* syscall tracepoint instrumentation */ #define TIF_SECCOMP 7 /* seccomp syscall filtering active */ +#define TIF_NOTIFY_SIGNAL 8 /* signal notifications exist */ #define TIF_USING_IWMMXT 17 #define TIF_MEMDIE 18 /* is terminating due to OOM killer */ @@ -148,6 +151,7 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, #define _TIF_SYSCALL_AUDIT (1 << TIF_SYSCALL_AUDIT) #define _TIF_SYSCALL_TRACEPOINT (1 << TIF_SYSCALL_TRACEPOINT) #define _TIF_SECCOMP (1 << TIF_SECCOMP) +#define _TIF_NOTIFY_SIGNAL (1 << TIF_NOTIFY_SIGNAL) #define _TIF_USING_IWMMXT (1 << TIF_USING_IWMMXT) /* Checks for any syscall work in entry-common.S */ @@ -158,7 +162,8 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, * Change these and you break ASM code in entry-common.S */ #define _TIF_WORK_MASK (_TIF_NEED_RESCHED | _TIF_SIGPENDING | \ - _TIF_NOTIFY_RESUME | _TIF_UPROBE) + _TIF_NOTIFY_RESUME | _TIF_UPROBE | \ + _TIF_NOTIFY_SIGNAL) #endif /* __KERNEL__ */ #endif /* __ASM_ARM_THREAD_INFO_H */ diff --git a/arch/arm/kernel/entry-common.S b/arch/arm/kernel/entry-common.S index 271cb8a1eba1..77d16390a524 100644 --- a/arch/arm/kernel/entry-common.S +++ b/arch/arm/kernel/entry-common.S @@ -53,7 +53,7 @@ __ret_fast_syscall: cmp r2, #TASK_SIZE blne addr_limit_check_failed ldr r1, [tsk, #TI_FLAGS] @ re-check for syscall tracing - tst r1, #_TIF_SYSCALL_WORK | _TIF_WORK_MASK + movs r1, r1, lsl #16 bne fast_work_pending @@ -90,7 +90,7 @@ __ret_fast_syscall: cmp r2, #TASK_SIZE blne addr_limit_check_failed ldr r1, [tsk, #TI_FLAGS] @ re-check for syscall tracing - tst r1, #_TIF_SYSCALL_WORK | _TIF_WORK_MASK + movs r1, r1, lsl #16 beq no_work_pending UNWIND(.fnend ) ENDPROC(ret_fast_syscall) @@ -131,7 +131,7 @@ ENTRY(ret_to_user_from_irq) cmp r2, #TASK_SIZE blne addr_limit_check_failed ldr r1, [tsk, #TI_FLAGS] - tst r1, #_TIF_WORK_MASK + movs r1, r1, lsl #16 bne slow_work_pending no_work_pending: asm_trace_hardirqs_on save = 0 diff --git a/arch/arm/kernel/entry-v7m.S b/arch/arm/kernel/entry-v7m.S index de1f20624be1..d0e898608d30 100644 --- a/arch/arm/kernel/entry-v7m.S +++ b/arch/arm/kernel/entry-v7m.S @@ -59,7 +59,7 @@ __irq_entry: get_thread_info tsk ldr r2, [tsk, #TI_FLAGS] - tst r2, #_TIF_WORK_MASK + movs r2, r2, lsl #16 beq 2f @ no work pending mov r0, #V7M_SCB_ICSR_PENDSVSET str r0, [r1, V7M_SCB_ICSR] @ raise PendSV diff --git a/arch/arm/kernel/signal.c b/arch/arm/kernel/signal.c index 585edbfccf6d..9d2e916121be 100644 --- a/arch/arm/kernel/signal.c +++ b/arch/arm/kernel/signal.c @@ -655,7 +655,7 @@ do_work_pending(struct pt_regs *regs, unsigned int thread_flags, int syscall) if (unlikely(!user_mode(regs))) return 0; local_irq_enable(); - if (thread_flags & _TIF_SIGPENDING) { + if (thread_flags & (_TIF_SIGPENDING | _TIF_NOTIFY_SIGNAL)) { int restart = do_signal(regs, syscall); if (unlikely(restart)) { /*
On Thu, Oct 29, 2020 at 11:15:37AM -0600, Jens Axboe wrote: > How about this? > > commit c03932936d8f99ff7c1c6c7d984e7a457284396c > Author: Jens Axboe <axboe@kernel.dk> > Date: Fri Oct 9 16:00:49 2020 -0600 > > arm: add support for TIF_NOTIFY_SIGNAL > > Wire up TIF_NOTIFY_SIGNAL handling for arm. > > Cc: linux-arm-kernel@lists.infradead.org > Signed-off-by: Jens Axboe <axboe@kernel.dk> > > diff --git a/arch/arm/include/asm/thread_info.h b/arch/arm/include/asm/thread_info.h > index 536b6b979f63..eb7ce2747eb0 100644 > --- a/arch/arm/include/asm/thread_info.h > +++ b/arch/arm/include/asm/thread_info.h > @@ -126,6 +126,8 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, > * thread information flags: > * TIF_USEDFPU - FPU was used by this task this quantum (SMP) > * TIF_POLLING_NRFLAG - true if poll_idle() is polling TIF_NEED_RESCHED > + * > + * Any bit in the range of 0..15 will cause do_work_pending() to be invoked. > */ > #define TIF_SIGPENDING 0 /* signal pending */ > #define TIF_NEED_RESCHED 1 /* rescheduling necessary */ > @@ -135,6 +137,7 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, > #define TIF_SYSCALL_AUDIT 5 /* syscall auditing active */ > #define TIF_SYSCALL_TRACEPOINT 6 /* syscall tracepoint instrumentation */ > #define TIF_SECCOMP 7 /* seccomp syscall filtering active */ > +#define TIF_NOTIFY_SIGNAL 8 /* signal notifications exist */ > > #define TIF_USING_IWMMXT 17 > #define TIF_MEMDIE 18 /* is terminating due to OOM killer */ > @@ -148,6 +151,7 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, > #define _TIF_SYSCALL_AUDIT (1 << TIF_SYSCALL_AUDIT) > #define _TIF_SYSCALL_TRACEPOINT (1 << TIF_SYSCALL_TRACEPOINT) > #define _TIF_SECCOMP (1 << TIF_SECCOMP) > +#define _TIF_NOTIFY_SIGNAL (1 << TIF_NOTIFY_SIGNAL) > #define _TIF_USING_IWMMXT (1 << TIF_USING_IWMMXT) > > /* Checks for any syscall work in entry-common.S */ > @@ -158,7 +162,8 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, > * Change these and you break ASM code in entry-common.S > */ > #define _TIF_WORK_MASK (_TIF_NEED_RESCHED | _TIF_SIGPENDING | \ > - _TIF_NOTIFY_RESUME | _TIF_UPROBE) > + _TIF_NOTIFY_RESUME | _TIF_UPROBE | \ > + _TIF_NOTIFY_SIGNAL) > > #endif /* __KERNEL__ */ > #endif /* __ASM_ARM_THREAD_INFO_H */ > diff --git a/arch/arm/kernel/entry-common.S b/arch/arm/kernel/entry-common.S > index 271cb8a1eba1..77d16390a524 100644 > --- a/arch/arm/kernel/entry-common.S > +++ b/arch/arm/kernel/entry-common.S > @@ -53,7 +53,7 @@ __ret_fast_syscall: > cmp r2, #TASK_SIZE > blne addr_limit_check_failed > ldr r1, [tsk, #TI_FLAGS] @ re-check for syscall tracing > - tst r1, #_TIF_SYSCALL_WORK | _TIF_WORK_MASK > + movs r1, r1, lsl #16 > bne fast_work_pending > > > @@ -90,7 +90,7 @@ __ret_fast_syscall: > cmp r2, #TASK_SIZE > blne addr_limit_check_failed > ldr r1, [tsk, #TI_FLAGS] @ re-check for syscall tracing > - tst r1, #_TIF_SYSCALL_WORK | _TIF_WORK_MASK > + movs r1, r1, lsl #16 > beq no_work_pending > UNWIND(.fnend ) > ENDPROC(ret_fast_syscall) > @@ -131,7 +131,7 @@ ENTRY(ret_to_user_from_irq) > cmp r2, #TASK_SIZE > blne addr_limit_check_failed > ldr r1, [tsk, #TI_FLAGS] > - tst r1, #_TIF_WORK_MASK > + movs r1, r1, lsl #16 > bne slow_work_pending > no_work_pending: > asm_trace_hardirqs_on save = 0 > diff --git a/arch/arm/kernel/entry-v7m.S b/arch/arm/kernel/entry-v7m.S > index de1f20624be1..d0e898608d30 100644 > --- a/arch/arm/kernel/entry-v7m.S > +++ b/arch/arm/kernel/entry-v7m.S > @@ -59,7 +59,7 @@ __irq_entry: > > get_thread_info tsk > ldr r2, [tsk, #TI_FLAGS] > - tst r2, #_TIF_WORK_MASK > + movs r2, r2, lsl #16 > beq 2f @ no work pending > mov r0, #V7M_SCB_ICSR_PENDSVSET > str r0, [r1, V7M_SCB_ICSR] @ raise PendSV > diff --git a/arch/arm/kernel/signal.c b/arch/arm/kernel/signal.c > index 585edbfccf6d..9d2e916121be 100644 > --- a/arch/arm/kernel/signal.c > +++ b/arch/arm/kernel/signal.c > @@ -655,7 +655,7 @@ do_work_pending(struct pt_regs *regs, unsigned int thread_flags, int syscall) > if (unlikely(!user_mode(regs))) > return 0; > local_irq_enable(); > - if (thread_flags & _TIF_SIGPENDING) { > + if (thread_flags & (_TIF_SIGPENDING | _TIF_NOTIFY_SIGNAL)) { > int restart = do_signal(regs, syscall); > if (unlikely(restart)) { > /* > Looks perfect to me, thanks! I assume the pre-requisits for this are already in mainline or linux-next?
On 10/29/20 11:17 AM, Russell King - ARM Linux admin wrote: > On Thu, Oct 29, 2020 at 11:15:37AM -0600, Jens Axboe wrote: >> How about this? >> >> commit c03932936d8f99ff7c1c6c7d984e7a457284396c >> Author: Jens Axboe <axboe@kernel.dk> >> Date: Fri Oct 9 16:00:49 2020 -0600 >> >> arm: add support for TIF_NOTIFY_SIGNAL >> >> Wire up TIF_NOTIFY_SIGNAL handling for arm. >> >> Cc: linux-arm-kernel@lists.infradead.org >> Signed-off-by: Jens Axboe <axboe@kernel.dk> >> >> diff --git a/arch/arm/include/asm/thread_info.h b/arch/arm/include/asm/thread_info.h >> index 536b6b979f63..eb7ce2747eb0 100644 >> --- a/arch/arm/include/asm/thread_info.h >> +++ b/arch/arm/include/asm/thread_info.h >> @@ -126,6 +126,8 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, >> * thread information flags: >> * TIF_USEDFPU - FPU was used by this task this quantum (SMP) >> * TIF_POLLING_NRFLAG - true if poll_idle() is polling TIF_NEED_RESCHED >> + * >> + * Any bit in the range of 0..15 will cause do_work_pending() to be invoked. >> */ >> #define TIF_SIGPENDING 0 /* signal pending */ >> #define TIF_NEED_RESCHED 1 /* rescheduling necessary */ >> @@ -135,6 +137,7 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, >> #define TIF_SYSCALL_AUDIT 5 /* syscall auditing active */ >> #define TIF_SYSCALL_TRACEPOINT 6 /* syscall tracepoint instrumentation */ >> #define TIF_SECCOMP 7 /* seccomp syscall filtering active */ >> +#define TIF_NOTIFY_SIGNAL 8 /* signal notifications exist */ >> >> #define TIF_USING_IWMMXT 17 >> #define TIF_MEMDIE 18 /* is terminating due to OOM killer */ >> @@ -148,6 +151,7 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, >> #define _TIF_SYSCALL_AUDIT (1 << TIF_SYSCALL_AUDIT) >> #define _TIF_SYSCALL_TRACEPOINT (1 << TIF_SYSCALL_TRACEPOINT) >> #define _TIF_SECCOMP (1 << TIF_SECCOMP) >> +#define _TIF_NOTIFY_SIGNAL (1 << TIF_NOTIFY_SIGNAL) >> #define _TIF_USING_IWMMXT (1 << TIF_USING_IWMMXT) >> >> /* Checks for any syscall work in entry-common.S */ >> @@ -158,7 +162,8 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, >> * Change these and you break ASM code in entry-common.S >> */ >> #define _TIF_WORK_MASK (_TIF_NEED_RESCHED | _TIF_SIGPENDING | \ >> - _TIF_NOTIFY_RESUME | _TIF_UPROBE) >> + _TIF_NOTIFY_RESUME | _TIF_UPROBE | \ >> + _TIF_NOTIFY_SIGNAL) >> >> #endif /* __KERNEL__ */ >> #endif /* __ASM_ARM_THREAD_INFO_H */ >> diff --git a/arch/arm/kernel/entry-common.S b/arch/arm/kernel/entry-common.S >> index 271cb8a1eba1..77d16390a524 100644 >> --- a/arch/arm/kernel/entry-common.S >> +++ b/arch/arm/kernel/entry-common.S >> @@ -53,7 +53,7 @@ __ret_fast_syscall: >> cmp r2, #TASK_SIZE >> blne addr_limit_check_failed >> ldr r1, [tsk, #TI_FLAGS] @ re-check for syscall tracing >> - tst r1, #_TIF_SYSCALL_WORK | _TIF_WORK_MASK >> + movs r1, r1, lsl #16 >> bne fast_work_pending >> >> >> @@ -90,7 +90,7 @@ __ret_fast_syscall: >> cmp r2, #TASK_SIZE >> blne addr_limit_check_failed >> ldr r1, [tsk, #TI_FLAGS] @ re-check for syscall tracing >> - tst r1, #_TIF_SYSCALL_WORK | _TIF_WORK_MASK >> + movs r1, r1, lsl #16 >> beq no_work_pending >> UNWIND(.fnend ) >> ENDPROC(ret_fast_syscall) >> @@ -131,7 +131,7 @@ ENTRY(ret_to_user_from_irq) >> cmp r2, #TASK_SIZE >> blne addr_limit_check_failed >> ldr r1, [tsk, #TI_FLAGS] >> - tst r1, #_TIF_WORK_MASK >> + movs r1, r1, lsl #16 >> bne slow_work_pending >> no_work_pending: >> asm_trace_hardirqs_on save = 0 >> diff --git a/arch/arm/kernel/entry-v7m.S b/arch/arm/kernel/entry-v7m.S >> index de1f20624be1..d0e898608d30 100644 >> --- a/arch/arm/kernel/entry-v7m.S >> +++ b/arch/arm/kernel/entry-v7m.S >> @@ -59,7 +59,7 @@ __irq_entry: >> >> get_thread_info tsk >> ldr r2, [tsk, #TI_FLAGS] >> - tst r2, #_TIF_WORK_MASK >> + movs r2, r2, lsl #16 >> beq 2f @ no work pending >> mov r0, #V7M_SCB_ICSR_PENDSVSET >> str r0, [r1, V7M_SCB_ICSR] @ raise PendSV >> diff --git a/arch/arm/kernel/signal.c b/arch/arm/kernel/signal.c >> index 585edbfccf6d..9d2e916121be 100644 >> --- a/arch/arm/kernel/signal.c >> +++ b/arch/arm/kernel/signal.c >> @@ -655,7 +655,7 @@ do_work_pending(struct pt_regs *regs, unsigned int thread_flags, int syscall) >> if (unlikely(!user_mode(regs))) >> return 0; >> local_irq_enable(); >> - if (thread_flags & _TIF_SIGPENDING) { >> + if (thread_flags & (_TIF_SIGPENDING | _TIF_NOTIFY_SIGNAL)) { >> int restart = do_signal(regs, syscall); >> if (unlikely(restart)) { >> /* >> > > Looks perfect to me, thanks! I assume the pre-requisits for this are > already in mainline or linux-next? Great! Thanks for your expedient attention and help. The bits using this are queued in tip for 5.11, so not in mainline yet, but should be in linux-next tomorrow I guess. But it was done such that arch patches could be queued up independently, so we didn't have weird cross dependencies.
On Thu, Oct 29, 2020 at 11:20:07AM -0600, Jens Axboe wrote: > On 10/29/20 11:17 AM, Russell King - ARM Linux admin wrote: > > On Thu, Oct 29, 2020 at 11:15:37AM -0600, Jens Axboe wrote: > >> How about this? > >> > >> commit c03932936d8f99ff7c1c6c7d984e7a457284396c > >> Author: Jens Axboe <axboe@kernel.dk> > >> Date: Fri Oct 9 16:00:49 2020 -0600 > >> > >> arm: add support for TIF_NOTIFY_SIGNAL > >> > >> Wire up TIF_NOTIFY_SIGNAL handling for arm. > >> > >> Cc: linux-arm-kernel@lists.infradead.org > >> Signed-off-by: Jens Axboe <axboe@kernel.dk> > >> > >> diff --git a/arch/arm/include/asm/thread_info.h b/arch/arm/include/asm/thread_info.h > >> index 536b6b979f63..eb7ce2747eb0 100644 > >> --- a/arch/arm/include/asm/thread_info.h > >> +++ b/arch/arm/include/asm/thread_info.h > >> @@ -126,6 +126,8 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, > >> * thread information flags: > >> * TIF_USEDFPU - FPU was used by this task this quantum (SMP) > >> * TIF_POLLING_NRFLAG - true if poll_idle() is polling TIF_NEED_RESCHED > >> + * > >> + * Any bit in the range of 0..15 will cause do_work_pending() to be invoked. > >> */ > >> #define TIF_SIGPENDING 0 /* signal pending */ > >> #define TIF_NEED_RESCHED 1 /* rescheduling necessary */ > >> @@ -135,6 +137,7 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, > >> #define TIF_SYSCALL_AUDIT 5 /* syscall auditing active */ > >> #define TIF_SYSCALL_TRACEPOINT 6 /* syscall tracepoint instrumentation */ > >> #define TIF_SECCOMP 7 /* seccomp syscall filtering active */ > >> +#define TIF_NOTIFY_SIGNAL 8 /* signal notifications exist */ > >> > >> #define TIF_USING_IWMMXT 17 > >> #define TIF_MEMDIE 18 /* is terminating due to OOM killer */ > >> @@ -148,6 +151,7 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, > >> #define _TIF_SYSCALL_AUDIT (1 << TIF_SYSCALL_AUDIT) > >> #define _TIF_SYSCALL_TRACEPOINT (1 << TIF_SYSCALL_TRACEPOINT) > >> #define _TIF_SECCOMP (1 << TIF_SECCOMP) > >> +#define _TIF_NOTIFY_SIGNAL (1 << TIF_NOTIFY_SIGNAL) > >> #define _TIF_USING_IWMMXT (1 << TIF_USING_IWMMXT) > >> > >> /* Checks for any syscall work in entry-common.S */ > >> @@ -158,7 +162,8 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, > >> * Change these and you break ASM code in entry-common.S > >> */ > >> #define _TIF_WORK_MASK (_TIF_NEED_RESCHED | _TIF_SIGPENDING | \ > >> - _TIF_NOTIFY_RESUME | _TIF_UPROBE) > >> + _TIF_NOTIFY_RESUME | _TIF_UPROBE | \ > >> + _TIF_NOTIFY_SIGNAL) > >> > >> #endif /* __KERNEL__ */ > >> #endif /* __ASM_ARM_THREAD_INFO_H */ > >> diff --git a/arch/arm/kernel/entry-common.S b/arch/arm/kernel/entry-common.S > >> index 271cb8a1eba1..77d16390a524 100644 > >> --- a/arch/arm/kernel/entry-common.S > >> +++ b/arch/arm/kernel/entry-common.S > >> @@ -53,7 +53,7 @@ __ret_fast_syscall: > >> cmp r2, #TASK_SIZE > >> blne addr_limit_check_failed > >> ldr r1, [tsk, #TI_FLAGS] @ re-check for syscall tracing > >> - tst r1, #_TIF_SYSCALL_WORK | _TIF_WORK_MASK > >> + movs r1, r1, lsl #16 > >> bne fast_work_pending > >> > >> > >> @@ -90,7 +90,7 @@ __ret_fast_syscall: > >> cmp r2, #TASK_SIZE > >> blne addr_limit_check_failed > >> ldr r1, [tsk, #TI_FLAGS] @ re-check for syscall tracing > >> - tst r1, #_TIF_SYSCALL_WORK | _TIF_WORK_MASK > >> + movs r1, r1, lsl #16 > >> beq no_work_pending > >> UNWIND(.fnend ) > >> ENDPROC(ret_fast_syscall) > >> @@ -131,7 +131,7 @@ ENTRY(ret_to_user_from_irq) > >> cmp r2, #TASK_SIZE > >> blne addr_limit_check_failed > >> ldr r1, [tsk, #TI_FLAGS] > >> - tst r1, #_TIF_WORK_MASK > >> + movs r1, r1, lsl #16 > >> bne slow_work_pending > >> no_work_pending: > >> asm_trace_hardirqs_on save = 0 > >> diff --git a/arch/arm/kernel/entry-v7m.S b/arch/arm/kernel/entry-v7m.S > >> index de1f20624be1..d0e898608d30 100644 > >> --- a/arch/arm/kernel/entry-v7m.S > >> +++ b/arch/arm/kernel/entry-v7m.S > >> @@ -59,7 +59,7 @@ __irq_entry: > >> > >> get_thread_info tsk > >> ldr r2, [tsk, #TI_FLAGS] > >> - tst r2, #_TIF_WORK_MASK > >> + movs r2, r2, lsl #16 > >> beq 2f @ no work pending > >> mov r0, #V7M_SCB_ICSR_PENDSVSET > >> str r0, [r1, V7M_SCB_ICSR] @ raise PendSV > >> diff --git a/arch/arm/kernel/signal.c b/arch/arm/kernel/signal.c > >> index 585edbfccf6d..9d2e916121be 100644 > >> --- a/arch/arm/kernel/signal.c > >> +++ b/arch/arm/kernel/signal.c > >> @@ -655,7 +655,7 @@ do_work_pending(struct pt_regs *regs, unsigned int thread_flags, int syscall) > >> if (unlikely(!user_mode(regs))) > >> return 0; > >> local_irq_enable(); > >> - if (thread_flags & _TIF_SIGPENDING) { > >> + if (thread_flags & (_TIF_SIGPENDING | _TIF_NOTIFY_SIGNAL)) { > >> int restart = do_signal(regs, syscall); > >> if (unlikely(restart)) { > >> /* > >> > > > > Looks perfect to me, thanks! I assume the pre-requisits for this are > > already in mainline or linux-next? > > Great! Thanks for your expedient attention and help. > > The bits using this are queued in tip for 5.11, so not in mainline yet, > but should be in linux-next tomorrow I guess. But it was done such that > arch patches could be queued up independently, so we didn't have weird > cross dependencies. Okay, I'll wait a few days and see about temporarily dumping it in my for-next branch so it gets a spin through kernelci next week. I'm not anticipating any breakage, so (if I remember) I'll give you a reviewed-by next week once it seems good. As I say, if I remember. Thanks.
On 10/29/20 11:42 AM, Russell King - ARM Linux admin wrote: > On Thu, Oct 29, 2020 at 11:20:07AM -0600, Jens Axboe wrote: >> On 10/29/20 11:17 AM, Russell King - ARM Linux admin wrote: >>> On Thu, Oct 29, 2020 at 11:15:37AM -0600, Jens Axboe wrote: >>>> How about this? >>>> >>>> commit c03932936d8f99ff7c1c6c7d984e7a457284396c >>>> Author: Jens Axboe <axboe@kernel.dk> >>>> Date: Fri Oct 9 16:00:49 2020 -0600 >>>> >>>> arm: add support for TIF_NOTIFY_SIGNAL >>>> >>>> Wire up TIF_NOTIFY_SIGNAL handling for arm. >>>> >>>> Cc: linux-arm-kernel@lists.infradead.org >>>> Signed-off-by: Jens Axboe <axboe@kernel.dk> >>>> >>>> diff --git a/arch/arm/include/asm/thread_info.h b/arch/arm/include/asm/thread_info.h >>>> index 536b6b979f63..eb7ce2747eb0 100644 >>>> --- a/arch/arm/include/asm/thread_info.h >>>> +++ b/arch/arm/include/asm/thread_info.h >>>> @@ -126,6 +126,8 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, >>>> * thread information flags: >>>> * TIF_USEDFPU - FPU was used by this task this quantum (SMP) >>>> * TIF_POLLING_NRFLAG - true if poll_idle() is polling TIF_NEED_RESCHED >>>> + * >>>> + * Any bit in the range of 0..15 will cause do_work_pending() to be invoked. >>>> */ >>>> #define TIF_SIGPENDING 0 /* signal pending */ >>>> #define TIF_NEED_RESCHED 1 /* rescheduling necessary */ >>>> @@ -135,6 +137,7 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, >>>> #define TIF_SYSCALL_AUDIT 5 /* syscall auditing active */ >>>> #define TIF_SYSCALL_TRACEPOINT 6 /* syscall tracepoint instrumentation */ >>>> #define TIF_SECCOMP 7 /* seccomp syscall filtering active */ >>>> +#define TIF_NOTIFY_SIGNAL 8 /* signal notifications exist */ >>>> >>>> #define TIF_USING_IWMMXT 17 >>>> #define TIF_MEMDIE 18 /* is terminating due to OOM killer */ >>>> @@ -148,6 +151,7 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, >>>> #define _TIF_SYSCALL_AUDIT (1 << TIF_SYSCALL_AUDIT) >>>> #define _TIF_SYSCALL_TRACEPOINT (1 << TIF_SYSCALL_TRACEPOINT) >>>> #define _TIF_SECCOMP (1 << TIF_SECCOMP) >>>> +#define _TIF_NOTIFY_SIGNAL (1 << TIF_NOTIFY_SIGNAL) >>>> #define _TIF_USING_IWMMXT (1 << TIF_USING_IWMMXT) >>>> >>>> /* Checks for any syscall work in entry-common.S */ >>>> @@ -158,7 +162,8 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, >>>> * Change these and you break ASM code in entry-common.S >>>> */ >>>> #define _TIF_WORK_MASK (_TIF_NEED_RESCHED | _TIF_SIGPENDING | \ >>>> - _TIF_NOTIFY_RESUME | _TIF_UPROBE) >>>> + _TIF_NOTIFY_RESUME | _TIF_UPROBE | \ >>>> + _TIF_NOTIFY_SIGNAL) >>>> >>>> #endif /* __KERNEL__ */ >>>> #endif /* __ASM_ARM_THREAD_INFO_H */ >>>> diff --git a/arch/arm/kernel/entry-common.S b/arch/arm/kernel/entry-common.S >>>> index 271cb8a1eba1..77d16390a524 100644 >>>> --- a/arch/arm/kernel/entry-common.S >>>> +++ b/arch/arm/kernel/entry-common.S >>>> @@ -53,7 +53,7 @@ __ret_fast_syscall: >>>> cmp r2, #TASK_SIZE >>>> blne addr_limit_check_failed >>>> ldr r1, [tsk, #TI_FLAGS] @ re-check for syscall tracing >>>> - tst r1, #_TIF_SYSCALL_WORK | _TIF_WORK_MASK >>>> + movs r1, r1, lsl #16 >>>> bne fast_work_pending >>>> >>>> >>>> @@ -90,7 +90,7 @@ __ret_fast_syscall: >>>> cmp r2, #TASK_SIZE >>>> blne addr_limit_check_failed >>>> ldr r1, [tsk, #TI_FLAGS] @ re-check for syscall tracing >>>> - tst r1, #_TIF_SYSCALL_WORK | _TIF_WORK_MASK >>>> + movs r1, r1, lsl #16 >>>> beq no_work_pending >>>> UNWIND(.fnend ) >>>> ENDPROC(ret_fast_syscall) >>>> @@ -131,7 +131,7 @@ ENTRY(ret_to_user_from_irq) >>>> cmp r2, #TASK_SIZE >>>> blne addr_limit_check_failed >>>> ldr r1, [tsk, #TI_FLAGS] >>>> - tst r1, #_TIF_WORK_MASK >>>> + movs r1, r1, lsl #16 >>>> bne slow_work_pending >>>> no_work_pending: >>>> asm_trace_hardirqs_on save = 0 >>>> diff --git a/arch/arm/kernel/entry-v7m.S b/arch/arm/kernel/entry-v7m.S >>>> index de1f20624be1..d0e898608d30 100644 >>>> --- a/arch/arm/kernel/entry-v7m.S >>>> +++ b/arch/arm/kernel/entry-v7m.S >>>> @@ -59,7 +59,7 @@ __irq_entry: >>>> >>>> get_thread_info tsk >>>> ldr r2, [tsk, #TI_FLAGS] >>>> - tst r2, #_TIF_WORK_MASK >>>> + movs r2, r2, lsl #16 >>>> beq 2f @ no work pending >>>> mov r0, #V7M_SCB_ICSR_PENDSVSET >>>> str r0, [r1, V7M_SCB_ICSR] @ raise PendSV >>>> diff --git a/arch/arm/kernel/signal.c b/arch/arm/kernel/signal.c >>>> index 585edbfccf6d..9d2e916121be 100644 >>>> --- a/arch/arm/kernel/signal.c >>>> +++ b/arch/arm/kernel/signal.c >>>> @@ -655,7 +655,7 @@ do_work_pending(struct pt_regs *regs, unsigned int thread_flags, int syscall) >>>> if (unlikely(!user_mode(regs))) >>>> return 0; >>>> local_irq_enable(); >>>> - if (thread_flags & _TIF_SIGPENDING) { >>>> + if (thread_flags & (_TIF_SIGPENDING | _TIF_NOTIFY_SIGNAL)) { >>>> int restart = do_signal(regs, syscall); >>>> if (unlikely(restart)) { >>>> /* >>>> >>> >>> Looks perfect to me, thanks! I assume the pre-requisits for this are >>> already in mainline or linux-next? >> >> Great! Thanks for your expedient attention and help. >> >> The bits using this are queued in tip for 5.11, so not in mainline yet, >> but should be in linux-next tomorrow I guess. But it was done such that >> arch patches could be queued up independently, so we didn't have weird >> cross dependencies. > > Okay, I'll wait a few days and see about temporarily dumping it in > my for-next branch so it gets a spin through kernelci next week. > I'm not anticipating any breakage, so (if I remember) I'll give you > a reviewed-by next week once it seems good. As I say, if I remember. Thanks! I sent out 23 of these today, I'll do some followup next week if need be :-)
On 10/29/20 11:42 AM, Russell King - ARM Linux admin wrote: > On Thu, Oct 29, 2020 at 11:20:07AM -0600, Jens Axboe wrote: >> On 10/29/20 11:17 AM, Russell King - ARM Linux admin wrote: >>> On Thu, Oct 29, 2020 at 11:15:37AM -0600, Jens Axboe wrote: >>>> How about this? >>>> >>>> commit c03932936d8f99ff7c1c6c7d984e7a457284396c >>>> Author: Jens Axboe <axboe@kernel.dk> >>>> Date: Fri Oct 9 16:00:49 2020 -0600 >>>> >>>> arm: add support for TIF_NOTIFY_SIGNAL >>>> >>>> Wire up TIF_NOTIFY_SIGNAL handling for arm. >>>> >>>> Cc: linux-arm-kernel@lists.infradead.org >>>> Signed-off-by: Jens Axboe <axboe@kernel.dk> >>>> >>>> diff --git a/arch/arm/include/asm/thread_info.h b/arch/arm/include/asm/thread_info.h >>>> index 536b6b979f63..eb7ce2747eb0 100644 >>>> --- a/arch/arm/include/asm/thread_info.h >>>> +++ b/arch/arm/include/asm/thread_info.h >>>> @@ -126,6 +126,8 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, >>>> * thread information flags: >>>> * TIF_USEDFPU - FPU was used by this task this quantum (SMP) >>>> * TIF_POLLING_NRFLAG - true if poll_idle() is polling TIF_NEED_RESCHED >>>> + * >>>> + * Any bit in the range of 0..15 will cause do_work_pending() to be invoked. >>>> */ >>>> #define TIF_SIGPENDING 0 /* signal pending */ >>>> #define TIF_NEED_RESCHED 1 /* rescheduling necessary */ >>>> @@ -135,6 +137,7 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, >>>> #define TIF_SYSCALL_AUDIT 5 /* syscall auditing active */ >>>> #define TIF_SYSCALL_TRACEPOINT 6 /* syscall tracepoint instrumentation */ >>>> #define TIF_SECCOMP 7 /* seccomp syscall filtering active */ >>>> +#define TIF_NOTIFY_SIGNAL 8 /* signal notifications exist */ >>>> >>>> #define TIF_USING_IWMMXT 17 >>>> #define TIF_MEMDIE 18 /* is terminating due to OOM killer */ >>>> @@ -148,6 +151,7 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, >>>> #define _TIF_SYSCALL_AUDIT (1 << TIF_SYSCALL_AUDIT) >>>> #define _TIF_SYSCALL_TRACEPOINT (1 << TIF_SYSCALL_TRACEPOINT) >>>> #define _TIF_SECCOMP (1 << TIF_SECCOMP) >>>> +#define _TIF_NOTIFY_SIGNAL (1 << TIF_NOTIFY_SIGNAL) >>>> #define _TIF_USING_IWMMXT (1 << TIF_USING_IWMMXT) >>>> >>>> /* Checks for any syscall work in entry-common.S */ >>>> @@ -158,7 +162,8 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, >>>> * Change these and you break ASM code in entry-common.S >>>> */ >>>> #define _TIF_WORK_MASK (_TIF_NEED_RESCHED | _TIF_SIGPENDING | \ >>>> - _TIF_NOTIFY_RESUME | _TIF_UPROBE) >>>> + _TIF_NOTIFY_RESUME | _TIF_UPROBE | \ >>>> + _TIF_NOTIFY_SIGNAL) >>>> >>>> #endif /* __KERNEL__ */ >>>> #endif /* __ASM_ARM_THREAD_INFO_H */ >>>> diff --git a/arch/arm/kernel/entry-common.S b/arch/arm/kernel/entry-common.S >>>> index 271cb8a1eba1..77d16390a524 100644 >>>> --- a/arch/arm/kernel/entry-common.S >>>> +++ b/arch/arm/kernel/entry-common.S >>>> @@ -53,7 +53,7 @@ __ret_fast_syscall: >>>> cmp r2, #TASK_SIZE >>>> blne addr_limit_check_failed >>>> ldr r1, [tsk, #TI_FLAGS] @ re-check for syscall tracing >>>> - tst r1, #_TIF_SYSCALL_WORK | _TIF_WORK_MASK >>>> + movs r1, r1, lsl #16 >>>> bne fast_work_pending >>>> >>>> >>>> @@ -90,7 +90,7 @@ __ret_fast_syscall: >>>> cmp r2, #TASK_SIZE >>>> blne addr_limit_check_failed >>>> ldr r1, [tsk, #TI_FLAGS] @ re-check for syscall tracing >>>> - tst r1, #_TIF_SYSCALL_WORK | _TIF_WORK_MASK >>>> + movs r1, r1, lsl #16 >>>> beq no_work_pending >>>> UNWIND(.fnend ) >>>> ENDPROC(ret_fast_syscall) >>>> @@ -131,7 +131,7 @@ ENTRY(ret_to_user_from_irq) >>>> cmp r2, #TASK_SIZE >>>> blne addr_limit_check_failed >>>> ldr r1, [tsk, #TI_FLAGS] >>>> - tst r1, #_TIF_WORK_MASK >>>> + movs r1, r1, lsl #16 >>>> bne slow_work_pending >>>> no_work_pending: >>>> asm_trace_hardirqs_on save = 0 >>>> diff --git a/arch/arm/kernel/entry-v7m.S b/arch/arm/kernel/entry-v7m.S >>>> index de1f20624be1..d0e898608d30 100644 >>>> --- a/arch/arm/kernel/entry-v7m.S >>>> +++ b/arch/arm/kernel/entry-v7m.S >>>> @@ -59,7 +59,7 @@ __irq_entry: >>>> >>>> get_thread_info tsk >>>> ldr r2, [tsk, #TI_FLAGS] >>>> - tst r2, #_TIF_WORK_MASK >>>> + movs r2, r2, lsl #16 >>>> beq 2f @ no work pending >>>> mov r0, #V7M_SCB_ICSR_PENDSVSET >>>> str r0, [r1, V7M_SCB_ICSR] @ raise PendSV >>>> diff --git a/arch/arm/kernel/signal.c b/arch/arm/kernel/signal.c >>>> index 585edbfccf6d..9d2e916121be 100644 >>>> --- a/arch/arm/kernel/signal.c >>>> +++ b/arch/arm/kernel/signal.c >>>> @@ -655,7 +655,7 @@ do_work_pending(struct pt_regs *regs, unsigned int thread_flags, int syscall) >>>> if (unlikely(!user_mode(regs))) >>>> return 0; >>>> local_irq_enable(); >>>> - if (thread_flags & _TIF_SIGPENDING) { >>>> + if (thread_flags & (_TIF_SIGPENDING | _TIF_NOTIFY_SIGNAL)) { >>>> int restart = do_signal(regs, syscall); >>>> if (unlikely(restart)) { >>>> /* >>>> >>> >>> Looks perfect to me, thanks! I assume the pre-requisits for this are >>> already in mainline or linux-next? >> >> Great! Thanks for your expedient attention and help. >> >> The bits using this are queued in tip for 5.11, so not in mainline yet, >> but should be in linux-next tomorrow I guess. But it was done such that >> arch patches could be queued up independently, so we didn't have weird >> cross dependencies. > > Okay, I'll wait a few days and see about temporarily dumping it in > my for-next branch so it gets a spin through kernelci next week. > I'm not anticipating any breakage, so (if I remember) I'll give you > a reviewed-by next week once it seems good. As I say, if I remember. Russell, did you have a chance to run it through the machinery?
On Tue, Nov 03, 2020 at 08:11:49AM -0700, Jens Axboe wrote: > On 10/29/20 11:42 AM, Russell King - ARM Linux admin wrote: > > On Thu, Oct 29, 2020 at 11:20:07AM -0600, Jens Axboe wrote: > >> On 10/29/20 11:17 AM, Russell King - ARM Linux admin wrote: > >>> On Thu, Oct 29, 2020 at 11:15:37AM -0600, Jens Axboe wrote: > >>>> How about this? > >>>> > >>>> commit c03932936d8f99ff7c1c6c7d984e7a457284396c > >>>> Author: Jens Axboe <axboe@kernel.dk> > >>>> Date: Fri Oct 9 16:00:49 2020 -0600 > >>>> > >>>> arm: add support for TIF_NOTIFY_SIGNAL > >>>> > >>>> Wire up TIF_NOTIFY_SIGNAL handling for arm. > >>>> > >>>> Cc: linux-arm-kernel@lists.infradead.org > >>>> Signed-off-by: Jens Axboe <axboe@kernel.dk> > >>>> > >>>> diff --git a/arch/arm/include/asm/thread_info.h b/arch/arm/include/asm/thread_info.h > >>>> index 536b6b979f63..eb7ce2747eb0 100644 > >>>> --- a/arch/arm/include/asm/thread_info.h > >>>> +++ b/arch/arm/include/asm/thread_info.h > >>>> @@ -126,6 +126,8 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, > >>>> * thread information flags: > >>>> * TIF_USEDFPU - FPU was used by this task this quantum (SMP) > >>>> * TIF_POLLING_NRFLAG - true if poll_idle() is polling TIF_NEED_RESCHED > >>>> + * > >>>> + * Any bit in the range of 0..15 will cause do_work_pending() to be invoked. > >>>> */ > >>>> #define TIF_SIGPENDING 0 /* signal pending */ > >>>> #define TIF_NEED_RESCHED 1 /* rescheduling necessary */ > >>>> @@ -135,6 +137,7 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, > >>>> #define TIF_SYSCALL_AUDIT 5 /* syscall auditing active */ > >>>> #define TIF_SYSCALL_TRACEPOINT 6 /* syscall tracepoint instrumentation */ > >>>> #define TIF_SECCOMP 7 /* seccomp syscall filtering active */ > >>>> +#define TIF_NOTIFY_SIGNAL 8 /* signal notifications exist */ > >>>> > >>>> #define TIF_USING_IWMMXT 17 > >>>> #define TIF_MEMDIE 18 /* is terminating due to OOM killer */ > >>>> @@ -148,6 +151,7 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, > >>>> #define _TIF_SYSCALL_AUDIT (1 << TIF_SYSCALL_AUDIT) > >>>> #define _TIF_SYSCALL_TRACEPOINT (1 << TIF_SYSCALL_TRACEPOINT) > >>>> #define _TIF_SECCOMP (1 << TIF_SECCOMP) > >>>> +#define _TIF_NOTIFY_SIGNAL (1 << TIF_NOTIFY_SIGNAL) > >>>> #define _TIF_USING_IWMMXT (1 << TIF_USING_IWMMXT) > >>>> > >>>> /* Checks for any syscall work in entry-common.S */ > >>>> @@ -158,7 +162,8 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, > >>>> * Change these and you break ASM code in entry-common.S > >>>> */ > >>>> #define _TIF_WORK_MASK (_TIF_NEED_RESCHED | _TIF_SIGPENDING | \ > >>>> - _TIF_NOTIFY_RESUME | _TIF_UPROBE) > >>>> + _TIF_NOTIFY_RESUME | _TIF_UPROBE | \ > >>>> + _TIF_NOTIFY_SIGNAL) > >>>> > >>>> #endif /* __KERNEL__ */ > >>>> #endif /* __ASM_ARM_THREAD_INFO_H */ > >>>> diff --git a/arch/arm/kernel/entry-common.S b/arch/arm/kernel/entry-common.S > >>>> index 271cb8a1eba1..77d16390a524 100644 > >>>> --- a/arch/arm/kernel/entry-common.S > >>>> +++ b/arch/arm/kernel/entry-common.S > >>>> @@ -53,7 +53,7 @@ __ret_fast_syscall: > >>>> cmp r2, #TASK_SIZE > >>>> blne addr_limit_check_failed > >>>> ldr r1, [tsk, #TI_FLAGS] @ re-check for syscall tracing > >>>> - tst r1, #_TIF_SYSCALL_WORK | _TIF_WORK_MASK > >>>> + movs r1, r1, lsl #16 > >>>> bne fast_work_pending > >>>> > >>>> > >>>> @@ -90,7 +90,7 @@ __ret_fast_syscall: > >>>> cmp r2, #TASK_SIZE > >>>> blne addr_limit_check_failed > >>>> ldr r1, [tsk, #TI_FLAGS] @ re-check for syscall tracing > >>>> - tst r1, #_TIF_SYSCALL_WORK | _TIF_WORK_MASK > >>>> + movs r1, r1, lsl #16 > >>>> beq no_work_pending > >>>> UNWIND(.fnend ) > >>>> ENDPROC(ret_fast_syscall) > >>>> @@ -131,7 +131,7 @@ ENTRY(ret_to_user_from_irq) > >>>> cmp r2, #TASK_SIZE > >>>> blne addr_limit_check_failed > >>>> ldr r1, [tsk, #TI_FLAGS] > >>>> - tst r1, #_TIF_WORK_MASK > >>>> + movs r1, r1, lsl #16 > >>>> bne slow_work_pending > >>>> no_work_pending: > >>>> asm_trace_hardirqs_on save = 0 > >>>> diff --git a/arch/arm/kernel/entry-v7m.S b/arch/arm/kernel/entry-v7m.S > >>>> index de1f20624be1..d0e898608d30 100644 > >>>> --- a/arch/arm/kernel/entry-v7m.S > >>>> +++ b/arch/arm/kernel/entry-v7m.S > >>>> @@ -59,7 +59,7 @@ __irq_entry: > >>>> > >>>> get_thread_info tsk > >>>> ldr r2, [tsk, #TI_FLAGS] > >>>> - tst r2, #_TIF_WORK_MASK > >>>> + movs r2, r2, lsl #16 > >>>> beq 2f @ no work pending > >>>> mov r0, #V7M_SCB_ICSR_PENDSVSET > >>>> str r0, [r1, V7M_SCB_ICSR] @ raise PendSV > >>>> diff --git a/arch/arm/kernel/signal.c b/arch/arm/kernel/signal.c > >>>> index 585edbfccf6d..9d2e916121be 100644 > >>>> --- a/arch/arm/kernel/signal.c > >>>> +++ b/arch/arm/kernel/signal.c > >>>> @@ -655,7 +655,7 @@ do_work_pending(struct pt_regs *regs, unsigned int thread_flags, int syscall) > >>>> if (unlikely(!user_mode(regs))) > >>>> return 0; > >>>> local_irq_enable(); > >>>> - if (thread_flags & _TIF_SIGPENDING) { > >>>> + if (thread_flags & (_TIF_SIGPENDING | _TIF_NOTIFY_SIGNAL)) { > >>>> int restart = do_signal(regs, syscall); > >>>> if (unlikely(restart)) { > >>>> /* > >>>> > >>> > >>> Looks perfect to me, thanks! I assume the pre-requisits for this are > >>> already in mainline or linux-next? > >> > >> Great! Thanks for your expedient attention and help. > >> > >> The bits using this are queued in tip for 5.11, so not in mainline yet, > >> but should be in linux-next tomorrow I guess. But it was done such that > >> arch patches could be queued up independently, so we didn't have weird > >> cross dependencies. > > > > Okay, I'll wait a few days and see about temporarily dumping it in > > my for-next branch so it gets a spin through kernelci next week. > > I'm not anticipating any breakage, so (if I remember) I'll give you > > a reviewed-by next week once it seems good. As I say, if I remember. > > Russell, did you have a chance to run it through the machinery? I threw the patch on top of my for-next branch, and let the various autobuilders chew on it for a few days. I haven't had any reports back, not even of any breakage through me adding it to my tree. I guess that's a positive indication.
On 11/12/20 8:32 AM, Russell King - ARM Linux admin wrote: > On Tue, Nov 03, 2020 at 08:11:49AM -0700, Jens Axboe wrote: >> On 10/29/20 11:42 AM, Russell King - ARM Linux admin wrote: >>> On Thu, Oct 29, 2020 at 11:20:07AM -0600, Jens Axboe wrote: >>>> On 10/29/20 11:17 AM, Russell King - ARM Linux admin wrote: >>>>> On Thu, Oct 29, 2020 at 11:15:37AM -0600, Jens Axboe wrote: >>>>>> How about this? >>>>>> >>>>>> commit c03932936d8f99ff7c1c6c7d984e7a457284396c >>>>>> Author: Jens Axboe <axboe@kernel.dk> >>>>>> Date: Fri Oct 9 16:00:49 2020 -0600 >>>>>> >>>>>> arm: add support for TIF_NOTIFY_SIGNAL >>>>>> >>>>>> Wire up TIF_NOTIFY_SIGNAL handling for arm. >>>>>> >>>>>> Cc: linux-arm-kernel@lists.infradead.org >>>>>> Signed-off-by: Jens Axboe <axboe@kernel.dk> >>>>>> >>>>>> diff --git a/arch/arm/include/asm/thread_info.h b/arch/arm/include/asm/thread_info.h >>>>>> index 536b6b979f63..eb7ce2747eb0 100644 >>>>>> --- a/arch/arm/include/asm/thread_info.h >>>>>> +++ b/arch/arm/include/asm/thread_info.h >>>>>> @@ -126,6 +126,8 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, >>>>>> * thread information flags: >>>>>> * TIF_USEDFPU - FPU was used by this task this quantum (SMP) >>>>>> * TIF_POLLING_NRFLAG - true if poll_idle() is polling TIF_NEED_RESCHED >>>>>> + * >>>>>> + * Any bit in the range of 0..15 will cause do_work_pending() to be invoked. >>>>>> */ >>>>>> #define TIF_SIGPENDING 0 /* signal pending */ >>>>>> #define TIF_NEED_RESCHED 1 /* rescheduling necessary */ >>>>>> @@ -135,6 +137,7 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, >>>>>> #define TIF_SYSCALL_AUDIT 5 /* syscall auditing active */ >>>>>> #define TIF_SYSCALL_TRACEPOINT 6 /* syscall tracepoint instrumentation */ >>>>>> #define TIF_SECCOMP 7 /* seccomp syscall filtering active */ >>>>>> +#define TIF_NOTIFY_SIGNAL 8 /* signal notifications exist */ >>>>>> >>>>>> #define TIF_USING_IWMMXT 17 >>>>>> #define TIF_MEMDIE 18 /* is terminating due to OOM killer */ >>>>>> @@ -148,6 +151,7 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, >>>>>> #define _TIF_SYSCALL_AUDIT (1 << TIF_SYSCALL_AUDIT) >>>>>> #define _TIF_SYSCALL_TRACEPOINT (1 << TIF_SYSCALL_TRACEPOINT) >>>>>> #define _TIF_SECCOMP (1 << TIF_SECCOMP) >>>>>> +#define _TIF_NOTIFY_SIGNAL (1 << TIF_NOTIFY_SIGNAL) >>>>>> #define _TIF_USING_IWMMXT (1 << TIF_USING_IWMMXT) >>>>>> >>>>>> /* Checks for any syscall work in entry-common.S */ >>>>>> @@ -158,7 +162,8 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, >>>>>> * Change these and you break ASM code in entry-common.S >>>>>> */ >>>>>> #define _TIF_WORK_MASK (_TIF_NEED_RESCHED | _TIF_SIGPENDING | \ >>>>>> - _TIF_NOTIFY_RESUME | _TIF_UPROBE) >>>>>> + _TIF_NOTIFY_RESUME | _TIF_UPROBE | \ >>>>>> + _TIF_NOTIFY_SIGNAL) >>>>>> >>>>>> #endif /* __KERNEL__ */ >>>>>> #endif /* __ASM_ARM_THREAD_INFO_H */ >>>>>> diff --git a/arch/arm/kernel/entry-common.S b/arch/arm/kernel/entry-common.S >>>>>> index 271cb8a1eba1..77d16390a524 100644 >>>>>> --- a/arch/arm/kernel/entry-common.S >>>>>> +++ b/arch/arm/kernel/entry-common.S >>>>>> @@ -53,7 +53,7 @@ __ret_fast_syscall: >>>>>> cmp r2, #TASK_SIZE >>>>>> blne addr_limit_check_failed >>>>>> ldr r1, [tsk, #TI_FLAGS] @ re-check for syscall tracing >>>>>> - tst r1, #_TIF_SYSCALL_WORK | _TIF_WORK_MASK >>>>>> + movs r1, r1, lsl #16 >>>>>> bne fast_work_pending >>>>>> >>>>>> >>>>>> @@ -90,7 +90,7 @@ __ret_fast_syscall: >>>>>> cmp r2, #TASK_SIZE >>>>>> blne addr_limit_check_failed >>>>>> ldr r1, [tsk, #TI_FLAGS] @ re-check for syscall tracing >>>>>> - tst r1, #_TIF_SYSCALL_WORK | _TIF_WORK_MASK >>>>>> + movs r1, r1, lsl #16 >>>>>> beq no_work_pending >>>>>> UNWIND(.fnend ) >>>>>> ENDPROC(ret_fast_syscall) >>>>>> @@ -131,7 +131,7 @@ ENTRY(ret_to_user_from_irq) >>>>>> cmp r2, #TASK_SIZE >>>>>> blne addr_limit_check_failed >>>>>> ldr r1, [tsk, #TI_FLAGS] >>>>>> - tst r1, #_TIF_WORK_MASK >>>>>> + movs r1, r1, lsl #16 >>>>>> bne slow_work_pending >>>>>> no_work_pending: >>>>>> asm_trace_hardirqs_on save = 0 >>>>>> diff --git a/arch/arm/kernel/entry-v7m.S b/arch/arm/kernel/entry-v7m.S >>>>>> index de1f20624be1..d0e898608d30 100644 >>>>>> --- a/arch/arm/kernel/entry-v7m.S >>>>>> +++ b/arch/arm/kernel/entry-v7m.S >>>>>> @@ -59,7 +59,7 @@ __irq_entry: >>>>>> >>>>>> get_thread_info tsk >>>>>> ldr r2, [tsk, #TI_FLAGS] >>>>>> - tst r2, #_TIF_WORK_MASK >>>>>> + movs r2, r2, lsl #16 >>>>>> beq 2f @ no work pending >>>>>> mov r0, #V7M_SCB_ICSR_PENDSVSET >>>>>> str r0, [r1, V7M_SCB_ICSR] @ raise PendSV >>>>>> diff --git a/arch/arm/kernel/signal.c b/arch/arm/kernel/signal.c >>>>>> index 585edbfccf6d..9d2e916121be 100644 >>>>>> --- a/arch/arm/kernel/signal.c >>>>>> +++ b/arch/arm/kernel/signal.c >>>>>> @@ -655,7 +655,7 @@ do_work_pending(struct pt_regs *regs, unsigned int thread_flags, int syscall) >>>>>> if (unlikely(!user_mode(regs))) >>>>>> return 0; >>>>>> local_irq_enable(); >>>>>> - if (thread_flags & _TIF_SIGPENDING) { >>>>>> + if (thread_flags & (_TIF_SIGPENDING | _TIF_NOTIFY_SIGNAL)) { >>>>>> int restart = do_signal(regs, syscall); >>>>>> if (unlikely(restart)) { >>>>>> /* >>>>>> >>>>> >>>>> Looks perfect to me, thanks! I assume the pre-requisits for this are >>>>> already in mainline or linux-next? >>>> >>>> Great! Thanks for your expedient attention and help. >>>> >>>> The bits using this are queued in tip for 5.11, so not in mainline yet, >>>> but should be in linux-next tomorrow I guess. But it was done such that >>>> arch patches could be queued up independently, so we didn't have weird >>>> cross dependencies. >>> >>> Okay, I'll wait a few days and see about temporarily dumping it in >>> my for-next branch so it gets a spin through kernelci next week. >>> I'm not anticipating any breakage, so (if I remember) I'll give you >>> a reviewed-by next week once it seems good. As I say, if I remember. >> >> Russell, did you have a chance to run it through the machinery? > > I threw the patch on top of my for-next branch, and let the various > autobuilders chew on it for a few days. I haven't had any reports > back, not even of any breakage through me adding it to my tree. > > I guess that's a positive indication. Good enough to add your acked-by?
On Thu, Nov 12, 2020 at 08:42:55AM -0700, Jens Axboe wrote: > On 11/12/20 8:32 AM, Russell King - ARM Linux admin wrote: > > On Tue, Nov 03, 2020 at 08:11:49AM -0700, Jens Axboe wrote: > >> On 10/29/20 11:42 AM, Russell King - ARM Linux admin wrote: > >>> On Thu, Oct 29, 2020 at 11:20:07AM -0600, Jens Axboe wrote: > >>>> On 10/29/20 11:17 AM, Russell King - ARM Linux admin wrote: > >>>>> On Thu, Oct 29, 2020 at 11:15:37AM -0600, Jens Axboe wrote: > >>>>>> How about this? > >>>>>> > >>>>>> commit c03932936d8f99ff7c1c6c7d984e7a457284396c > >>>>>> Author: Jens Axboe <axboe@kernel.dk> > >>>>>> Date: Fri Oct 9 16:00:49 2020 -0600 > >>>>>> > >>>>>> arm: add support for TIF_NOTIFY_SIGNAL > >>>>>> > >>>>>> Wire up TIF_NOTIFY_SIGNAL handling for arm. > >>>>>> > >>>>>> Cc: linux-arm-kernel@lists.infradead.org > >>>>>> Signed-off-by: Jens Axboe <axboe@kernel.dk> > >>>>>> > >>>>>> diff --git a/arch/arm/include/asm/thread_info.h b/arch/arm/include/asm/thread_info.h > >>>>>> index 536b6b979f63..eb7ce2747eb0 100644 > >>>>>> --- a/arch/arm/include/asm/thread_info.h > >>>>>> +++ b/arch/arm/include/asm/thread_info.h > >>>>>> @@ -126,6 +126,8 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, > >>>>>> * thread information flags: > >>>>>> * TIF_USEDFPU - FPU was used by this task this quantum (SMP) > >>>>>> * TIF_POLLING_NRFLAG - true if poll_idle() is polling TIF_NEED_RESCHED > >>>>>> + * > >>>>>> + * Any bit in the range of 0..15 will cause do_work_pending() to be invoked. > >>>>>> */ > >>>>>> #define TIF_SIGPENDING 0 /* signal pending */ > >>>>>> #define TIF_NEED_RESCHED 1 /* rescheduling necessary */ > >>>>>> @@ -135,6 +137,7 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, > >>>>>> #define TIF_SYSCALL_AUDIT 5 /* syscall auditing active */ > >>>>>> #define TIF_SYSCALL_TRACEPOINT 6 /* syscall tracepoint instrumentation */ > >>>>>> #define TIF_SECCOMP 7 /* seccomp syscall filtering active */ > >>>>>> +#define TIF_NOTIFY_SIGNAL 8 /* signal notifications exist */ > >>>>>> > >>>>>> #define TIF_USING_IWMMXT 17 > >>>>>> #define TIF_MEMDIE 18 /* is terminating due to OOM killer */ > >>>>>> @@ -148,6 +151,7 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, > >>>>>> #define _TIF_SYSCALL_AUDIT (1 << TIF_SYSCALL_AUDIT) > >>>>>> #define _TIF_SYSCALL_TRACEPOINT (1 << TIF_SYSCALL_TRACEPOINT) > >>>>>> #define _TIF_SECCOMP (1 << TIF_SECCOMP) > >>>>>> +#define _TIF_NOTIFY_SIGNAL (1 << TIF_NOTIFY_SIGNAL) > >>>>>> #define _TIF_USING_IWMMXT (1 << TIF_USING_IWMMXT) > >>>>>> > >>>>>> /* Checks for any syscall work in entry-common.S */ > >>>>>> @@ -158,7 +162,8 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, > >>>>>> * Change these and you break ASM code in entry-common.S > >>>>>> */ > >>>>>> #define _TIF_WORK_MASK (_TIF_NEED_RESCHED | _TIF_SIGPENDING | \ > >>>>>> - _TIF_NOTIFY_RESUME | _TIF_UPROBE) > >>>>>> + _TIF_NOTIFY_RESUME | _TIF_UPROBE | \ > >>>>>> + _TIF_NOTIFY_SIGNAL) > >>>>>> > >>>>>> #endif /* __KERNEL__ */ > >>>>>> #endif /* __ASM_ARM_THREAD_INFO_H */ > >>>>>> diff --git a/arch/arm/kernel/entry-common.S b/arch/arm/kernel/entry-common.S > >>>>>> index 271cb8a1eba1..77d16390a524 100644 > >>>>>> --- a/arch/arm/kernel/entry-common.S > >>>>>> +++ b/arch/arm/kernel/entry-common.S > >>>>>> @@ -53,7 +53,7 @@ __ret_fast_syscall: > >>>>>> cmp r2, #TASK_SIZE > >>>>>> blne addr_limit_check_failed > >>>>>> ldr r1, [tsk, #TI_FLAGS] @ re-check for syscall tracing > >>>>>> - tst r1, #_TIF_SYSCALL_WORK | _TIF_WORK_MASK > >>>>>> + movs r1, r1, lsl #16 > >>>>>> bne fast_work_pending > >>>>>> > >>>>>> > >>>>>> @@ -90,7 +90,7 @@ __ret_fast_syscall: > >>>>>> cmp r2, #TASK_SIZE > >>>>>> blne addr_limit_check_failed > >>>>>> ldr r1, [tsk, #TI_FLAGS] @ re-check for syscall tracing > >>>>>> - tst r1, #_TIF_SYSCALL_WORK | _TIF_WORK_MASK > >>>>>> + movs r1, r1, lsl #16 > >>>>>> beq no_work_pending > >>>>>> UNWIND(.fnend ) > >>>>>> ENDPROC(ret_fast_syscall) > >>>>>> @@ -131,7 +131,7 @@ ENTRY(ret_to_user_from_irq) > >>>>>> cmp r2, #TASK_SIZE > >>>>>> blne addr_limit_check_failed > >>>>>> ldr r1, [tsk, #TI_FLAGS] > >>>>>> - tst r1, #_TIF_WORK_MASK > >>>>>> + movs r1, r1, lsl #16 > >>>>>> bne slow_work_pending > >>>>>> no_work_pending: > >>>>>> asm_trace_hardirqs_on save = 0 > >>>>>> diff --git a/arch/arm/kernel/entry-v7m.S b/arch/arm/kernel/entry-v7m.S > >>>>>> index de1f20624be1..d0e898608d30 100644 > >>>>>> --- a/arch/arm/kernel/entry-v7m.S > >>>>>> +++ b/arch/arm/kernel/entry-v7m.S > >>>>>> @@ -59,7 +59,7 @@ __irq_entry: > >>>>>> > >>>>>> get_thread_info tsk > >>>>>> ldr r2, [tsk, #TI_FLAGS] > >>>>>> - tst r2, #_TIF_WORK_MASK > >>>>>> + movs r2, r2, lsl #16 > >>>>>> beq 2f @ no work pending > >>>>>> mov r0, #V7M_SCB_ICSR_PENDSVSET > >>>>>> str r0, [r1, V7M_SCB_ICSR] @ raise PendSV > >>>>>> diff --git a/arch/arm/kernel/signal.c b/arch/arm/kernel/signal.c > >>>>>> index 585edbfccf6d..9d2e916121be 100644 > >>>>>> --- a/arch/arm/kernel/signal.c > >>>>>> +++ b/arch/arm/kernel/signal.c > >>>>>> @@ -655,7 +655,7 @@ do_work_pending(struct pt_regs *regs, unsigned int thread_flags, int syscall) > >>>>>> if (unlikely(!user_mode(regs))) > >>>>>> return 0; > >>>>>> local_irq_enable(); > >>>>>> - if (thread_flags & _TIF_SIGPENDING) { > >>>>>> + if (thread_flags & (_TIF_SIGPENDING | _TIF_NOTIFY_SIGNAL)) { > >>>>>> int restart = do_signal(regs, syscall); > >>>>>> if (unlikely(restart)) { > >>>>>> /* > >>>>>> > >>>>> > >>>>> Looks perfect to me, thanks! I assume the pre-requisits for this are > >>>>> already in mainline or linux-next? > >>>> > >>>> Great! Thanks for your expedient attention and help. > >>>> > >>>> The bits using this are queued in tip for 5.11, so not in mainline yet, > >>>> but should be in linux-next tomorrow I guess. But it was done such that > >>>> arch patches could be queued up independently, so we didn't have weird > >>>> cross dependencies. > >>> > >>> Okay, I'll wait a few days and see about temporarily dumping it in > >>> my for-next branch so it gets a spin through kernelci next week. > >>> I'm not anticipating any breakage, so (if I remember) I'll give you > >>> a reviewed-by next week once it seems good. As I say, if I remember. > >> > >> Russell, did you have a chance to run it through the machinery? > > > > I threw the patch on top of my for-next branch, and let the various > > autobuilders chew on it for a few days. I haven't had any reports > > back, not even of any breakage through me adding it to my tree. > > > > I guess that's a positive indication. > > Good enough to add your acked-by? I guess. Acked-by: Russell King <rmk+kernel@armlinux.org.uk> Thanks.
On 11/12/20 8:44 AM, Russell King - ARM Linux admin wrote: > On Thu, Nov 12, 2020 at 08:42:55AM -0700, Jens Axboe wrote: >> On 11/12/20 8:32 AM, Russell King - ARM Linux admin wrote: >>> On Tue, Nov 03, 2020 at 08:11:49AM -0700, Jens Axboe wrote: >>>> On 10/29/20 11:42 AM, Russell King - ARM Linux admin wrote: >>>>> On Thu, Oct 29, 2020 at 11:20:07AM -0600, Jens Axboe wrote: >>>>>> On 10/29/20 11:17 AM, Russell King - ARM Linux admin wrote: >>>>>>> On Thu, Oct 29, 2020 at 11:15:37AM -0600, Jens Axboe wrote: >>>>>>>> How about this? >>>>>>>> >>>>>>>> commit c03932936d8f99ff7c1c6c7d984e7a457284396c >>>>>>>> Author: Jens Axboe <axboe@kernel.dk> >>>>>>>> Date: Fri Oct 9 16:00:49 2020 -0600 >>>>>>>> >>>>>>>> arm: add support for TIF_NOTIFY_SIGNAL >>>>>>>> >>>>>>>> Wire up TIF_NOTIFY_SIGNAL handling for arm. >>>>>>>> >>>>>>>> Cc: linux-arm-kernel@lists.infradead.org >>>>>>>> Signed-off-by: Jens Axboe <axboe@kernel.dk> >>>>>>>> >>>>>>>> diff --git a/arch/arm/include/asm/thread_info.h b/arch/arm/include/asm/thread_info.h >>>>>>>> index 536b6b979f63..eb7ce2747eb0 100644 >>>>>>>> --- a/arch/arm/include/asm/thread_info.h >>>>>>>> +++ b/arch/arm/include/asm/thread_info.h >>>>>>>> @@ -126,6 +126,8 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, >>>>>>>> * thread information flags: >>>>>>>> * TIF_USEDFPU - FPU was used by this task this quantum (SMP) >>>>>>>> * TIF_POLLING_NRFLAG - true if poll_idle() is polling TIF_NEED_RESCHED >>>>>>>> + * >>>>>>>> + * Any bit in the range of 0..15 will cause do_work_pending() to be invoked. >>>>>>>> */ >>>>>>>> #define TIF_SIGPENDING 0 /* signal pending */ >>>>>>>> #define TIF_NEED_RESCHED 1 /* rescheduling necessary */ >>>>>>>> @@ -135,6 +137,7 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, >>>>>>>> #define TIF_SYSCALL_AUDIT 5 /* syscall auditing active */ >>>>>>>> #define TIF_SYSCALL_TRACEPOINT 6 /* syscall tracepoint instrumentation */ >>>>>>>> #define TIF_SECCOMP 7 /* seccomp syscall filtering active */ >>>>>>>> +#define TIF_NOTIFY_SIGNAL 8 /* signal notifications exist */ >>>>>>>> >>>>>>>> #define TIF_USING_IWMMXT 17 >>>>>>>> #define TIF_MEMDIE 18 /* is terminating due to OOM killer */ >>>>>>>> @@ -148,6 +151,7 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, >>>>>>>> #define _TIF_SYSCALL_AUDIT (1 << TIF_SYSCALL_AUDIT) >>>>>>>> #define _TIF_SYSCALL_TRACEPOINT (1 << TIF_SYSCALL_TRACEPOINT) >>>>>>>> #define _TIF_SECCOMP (1 << TIF_SECCOMP) >>>>>>>> +#define _TIF_NOTIFY_SIGNAL (1 << TIF_NOTIFY_SIGNAL) >>>>>>>> #define _TIF_USING_IWMMXT (1 << TIF_USING_IWMMXT) >>>>>>>> >>>>>>>> /* Checks for any syscall work in entry-common.S */ >>>>>>>> @@ -158,7 +162,8 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, >>>>>>>> * Change these and you break ASM code in entry-common.S >>>>>>>> */ >>>>>>>> #define _TIF_WORK_MASK (_TIF_NEED_RESCHED | _TIF_SIGPENDING | \ >>>>>>>> - _TIF_NOTIFY_RESUME | _TIF_UPROBE) >>>>>>>> + _TIF_NOTIFY_RESUME | _TIF_UPROBE | \ >>>>>>>> + _TIF_NOTIFY_SIGNAL) >>>>>>>> >>>>>>>> #endif /* __KERNEL__ */ >>>>>>>> #endif /* __ASM_ARM_THREAD_INFO_H */ >>>>>>>> diff --git a/arch/arm/kernel/entry-common.S b/arch/arm/kernel/entry-common.S >>>>>>>> index 271cb8a1eba1..77d16390a524 100644 >>>>>>>> --- a/arch/arm/kernel/entry-common.S >>>>>>>> +++ b/arch/arm/kernel/entry-common.S >>>>>>>> @@ -53,7 +53,7 @@ __ret_fast_syscall: >>>>>>>> cmp r2, #TASK_SIZE >>>>>>>> blne addr_limit_check_failed >>>>>>>> ldr r1, [tsk, #TI_FLAGS] @ re-check for syscall tracing >>>>>>>> - tst r1, #_TIF_SYSCALL_WORK | _TIF_WORK_MASK >>>>>>>> + movs r1, r1, lsl #16 >>>>>>>> bne fast_work_pending >>>>>>>> >>>>>>>> >>>>>>>> @@ -90,7 +90,7 @@ __ret_fast_syscall: >>>>>>>> cmp r2, #TASK_SIZE >>>>>>>> blne addr_limit_check_failed >>>>>>>> ldr r1, [tsk, #TI_FLAGS] @ re-check for syscall tracing >>>>>>>> - tst r1, #_TIF_SYSCALL_WORK | _TIF_WORK_MASK >>>>>>>> + movs r1, r1, lsl #16 >>>>>>>> beq no_work_pending >>>>>>>> UNWIND(.fnend ) >>>>>>>> ENDPROC(ret_fast_syscall) >>>>>>>> @@ -131,7 +131,7 @@ ENTRY(ret_to_user_from_irq) >>>>>>>> cmp r2, #TASK_SIZE >>>>>>>> blne addr_limit_check_failed >>>>>>>> ldr r1, [tsk, #TI_FLAGS] >>>>>>>> - tst r1, #_TIF_WORK_MASK >>>>>>>> + movs r1, r1, lsl #16 >>>>>>>> bne slow_work_pending >>>>>>>> no_work_pending: >>>>>>>> asm_trace_hardirqs_on save = 0 >>>>>>>> diff --git a/arch/arm/kernel/entry-v7m.S b/arch/arm/kernel/entry-v7m.S >>>>>>>> index de1f20624be1..d0e898608d30 100644 >>>>>>>> --- a/arch/arm/kernel/entry-v7m.S >>>>>>>> +++ b/arch/arm/kernel/entry-v7m.S >>>>>>>> @@ -59,7 +59,7 @@ __irq_entry: >>>>>>>> >>>>>>>> get_thread_info tsk >>>>>>>> ldr r2, [tsk, #TI_FLAGS] >>>>>>>> - tst r2, #_TIF_WORK_MASK >>>>>>>> + movs r2, r2, lsl #16 >>>>>>>> beq 2f @ no work pending >>>>>>>> mov r0, #V7M_SCB_ICSR_PENDSVSET >>>>>>>> str r0, [r1, V7M_SCB_ICSR] @ raise PendSV >>>>>>>> diff --git a/arch/arm/kernel/signal.c b/arch/arm/kernel/signal.c >>>>>>>> index 585edbfccf6d..9d2e916121be 100644 >>>>>>>> --- a/arch/arm/kernel/signal.c >>>>>>>> +++ b/arch/arm/kernel/signal.c >>>>>>>> @@ -655,7 +655,7 @@ do_work_pending(struct pt_regs *regs, unsigned int thread_flags, int syscall) >>>>>>>> if (unlikely(!user_mode(regs))) >>>>>>>> return 0; >>>>>>>> local_irq_enable(); >>>>>>>> - if (thread_flags & _TIF_SIGPENDING) { >>>>>>>> + if (thread_flags & (_TIF_SIGPENDING | _TIF_NOTIFY_SIGNAL)) { >>>>>>>> int restart = do_signal(regs, syscall); >>>>>>>> if (unlikely(restart)) { >>>>>>>> /* >>>>>>>> >>>>>>> >>>>>>> Looks perfect to me, thanks! I assume the pre-requisits for this are >>>>>>> already in mainline or linux-next? >>>>>> >>>>>> Great! Thanks for your expedient attention and help. >>>>>> >>>>>> The bits using this are queued in tip for 5.11, so not in mainline yet, >>>>>> but should be in linux-next tomorrow I guess. But it was done such that >>>>>> arch patches could be queued up independently, so we didn't have weird >>>>>> cross dependencies. >>>>> >>>>> Okay, I'll wait a few days and see about temporarily dumping it in >>>>> my for-next branch so it gets a spin through kernelci next week. >>>>> I'm not anticipating any breakage, so (if I remember) I'll give you >>>>> a reviewed-by next week once it seems good. As I say, if I remember. >>>> >>>> Russell, did you have a chance to run it through the machinery? >>> >>> I threw the patch on top of my for-next branch, and let the various >>> autobuilders chew on it for a few days. I haven't had any reports >>> back, not even of any breakage through me adding it to my tree. >>> >>> I guess that's a positive indication. >> >> Good enough to add your acked-by? > > I guess. > > Acked-by: Russell King <rmk+kernel@armlinux.org.uk> Thanks Russell, added.
diff --git a/arch/arm/include/asm/thread_info.h b/arch/arm/include/asm/thread_info.h index 536b6b979f63..fec16d770180 100644 --- a/arch/arm/include/asm/thread_info.h +++ b/arch/arm/include/asm/thread_info.h @@ -135,6 +135,7 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, #define TIF_SYSCALL_AUDIT 5 /* syscall auditing active */ #define TIF_SYSCALL_TRACEPOINT 6 /* syscall tracepoint instrumentation */ #define TIF_SECCOMP 7 /* seccomp syscall filtering active */ +#define TIF_NOTIFY_SIGNAL 8 /* signal notifications exist */ #define TIF_USING_IWMMXT 17 #define TIF_MEMDIE 18 /* is terminating due to OOM killer */ @@ -148,6 +149,7 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, #define _TIF_SYSCALL_AUDIT (1 << TIF_SYSCALL_AUDIT) #define _TIF_SYSCALL_TRACEPOINT (1 << TIF_SYSCALL_TRACEPOINT) #define _TIF_SECCOMP (1 << TIF_SECCOMP) +#define _TIF_NOTIFY_SIGNAL (1 << TIF_NOTIFY_SIGNAL) #define _TIF_USING_IWMMXT (1 << TIF_USING_IWMMXT) /* Checks for any syscall work in entry-common.S */ @@ -158,7 +160,8 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, * Change these and you break ASM code in entry-common.S */ #define _TIF_WORK_MASK (_TIF_NEED_RESCHED | _TIF_SIGPENDING | \ - _TIF_NOTIFY_RESUME | _TIF_UPROBE) + _TIF_NOTIFY_RESUME | _TIF_UPROBE | \ + _TIF_NOTIFY_SIGNAL) #endif /* __KERNEL__ */ #endif /* __ASM_ARM_THREAD_INFO_H */ diff --git a/arch/arm/kernel/entry-common.S b/arch/arm/kernel/entry-common.S index 271cb8a1eba1..7485b58673b4 100644 --- a/arch/arm/kernel/entry-common.S +++ b/arch/arm/kernel/entry-common.S @@ -53,7 +53,8 @@ __ret_fast_syscall: cmp r2, #TASK_SIZE blne addr_limit_check_failed ldr r1, [tsk, #TI_FLAGS] @ re-check for syscall tracing - tst r1, #_TIF_SYSCALL_WORK | _TIF_WORK_MASK + ldr r2, =#_TIF_SYSCALL_WORK | _TIF_WORK_MASK + tst r1, r2 bne fast_work_pending @@ -90,7 +91,8 @@ __ret_fast_syscall: cmp r2, #TASK_SIZE blne addr_limit_check_failed ldr r1, [tsk, #TI_FLAGS] @ re-check for syscall tracing - tst r1, #_TIF_SYSCALL_WORK | _TIF_WORK_MASK + ldr r2, =#_TIF_SYSCALL_WORK | _TIF_WORK_MASK + tst r1, r2 beq no_work_pending UNWIND(.fnend ) ENDPROC(ret_fast_syscall) @@ -131,7 +133,8 @@ ENTRY(ret_to_user_from_irq) cmp r2, #TASK_SIZE blne addr_limit_check_failed ldr r1, [tsk, #TI_FLAGS] - tst r1, #_TIF_WORK_MASK + ldr r2, =#_TIF_WORK_MASK + tst r1, r2 bne slow_work_pending no_work_pending: asm_trace_hardirqs_on save = 0 diff --git a/arch/arm/kernel/signal.c b/arch/arm/kernel/signal.c index 585edbfccf6d..9d2e916121be 100644 --- a/arch/arm/kernel/signal.c +++ b/arch/arm/kernel/signal.c @@ -655,7 +655,7 @@ do_work_pending(struct pt_regs *regs, unsigned int thread_flags, int syscall) if (unlikely(!user_mode(regs))) return 0; local_irq_enable(); - if (thread_flags & _TIF_SIGPENDING) { + if (thread_flags & (_TIF_SIGPENDING | _TIF_NOTIFY_SIGNAL)) { int restart = do_signal(regs, syscall); if (unlikely(restart)) { /*
Wire up TIF_NOTIFY_SIGNAL handling for arm. Cc: linux-arm-kernel@lists.infradead.org Signed-off-by: Jens Axboe <axboe@kernel.dk> --- 5.11 has support queued up for TIF_NOTIFY_SIGNAL, see this posting for details: https://lore.kernel.org/io-uring/20201026203230.386348-1-axboe@kernel.dk/ As part of that work, I'm adding TIF_NOTIFY_SIGNAL support to all archs, as that will enable a set of cleanups once all of them support it. This needs a bit of asm help, immediate doesn't like anything outside of 1 byte, it seems. Any clues? arch/arm/include/asm/thread_info.h | 5 ++++- arch/arm/kernel/entry-common.S | 9 ++++++--- arch/arm/kernel/signal.c | 2 +- 3 files changed, 11 insertions(+), 5 deletions(-)