| Message ID | 20180328124129.6459-3-ard.biesheuvel@linaro.org (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On Wed, Mar 28, 2018 at 02:41:29PM +0200, Ard Biesheuvel wrote: > Add support macros to conditionally yield the NEON (and thus the CPU) > that may be called from the assembler code. > > In some cases, yielding the NEON involves saving and restoring a non > trivial amount of context (especially in the CRC folding algorithms), > and so the macro is split into three, and the code in between is only > executed when the yield path is taken, allowing the context to be preserved. > The third macro takes an optional label argument that marks the resume > path after a yield has been performed. Minor comments below, mostly just suggestions/observations. With the missing #include in asm-offsets.c fixed (if you think it's appropriate): Reviewed-by: Dave Martin <Dave.Martin@arm.com> > Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> > --- > arch/arm64/include/asm/assembler.h | 64 ++++++++++++++++++++ > arch/arm64/kernel/asm-offsets.c | 2 + > 2 files changed, 66 insertions(+) > > diff --git a/arch/arm64/include/asm/assembler.h b/arch/arm64/include/asm/assembler.h > index d354eb7f2f0c..fb11514273d9 100644 > --- a/arch/arm64/include/asm/assembler.h > +++ b/arch/arm64/include/asm/assembler.h > @@ -623,4 +623,68 @@ USER(\label, ic ivau, \tmp2) // invalidate I line PoU > .endif > .endm > > +/* > + * Check whether to yield to another runnable task from kernel mode NEON code > + * (which runs with preemption disabled). > + * > + * if_will_cond_yield_neon > + * // pre-yield patchup code > + * do_cond_yield_neon > + * // post-yield patchup code > + * endif_yield_neon <label> > + * > + * where <label> is optional, and marks the point where execution will resume > + * after a yield has been performed. If omitted, execution resumes right after > + * the endif_yield_neon invocation. Maybe add a comment describing cond_yield_neon, e.g.: * * As a convenience, in the case where no patchup code is required * the above sequence may be abbreviated to: * * cond_yield_neon <label> > + * > + * Note that the patchup code does not support assembler directives that change > + * the output section, any use of such directives is undefined. > + * > + * The yield itself consists of the following: > + * - Check whether the preempt count is exactly 1, in which case disabling > + * preemption once will make the task preemptible. If this is not the case, > + * yielding is pointless. > + * - Check whether TIF_NEED_RESCHED is set, and if so, disable and re-enable > + * kernel mode NEON (which will trigger a reschedule), and branch to the > + * yield fixup code. > + * > + * This macro sequence clobbers x0, x1 and the flags register unconditionally, > + * and may clobber x2 .. x18 if the yield path is taken. > + */ Does this mean that the pre-yield patchup code can safely refer to x2..x18, but the post-yield patchup code and the code at <label> (or otherwise immediately following endif_yield_neon) can't? > + > + .macro cond_yield_neon, lbl > + if_will_cond_yield_neon > + do_cond_yield_neon > + endif_yield_neon \lbl > + .endm > + > + .macro if_will_cond_yield_neon > +#ifdef CONFIG_PREEMPT > + get_thread_info x0 > + ldr w1, [x0, #TSK_TI_PREEMPT] > + ldr x0, [x0, #TSK_TI_FLAGS] > + cmp w1, #PREEMPT_DISABLE_OFFSET > + csel x0, x0, xzr, eq > + tbnz x0, #TIF_NEED_RESCHED, .Lyield_\@ // needs rescheduling? > +#endif > + /* fall through to endif_yield_neon */ > + .subsection 1 Can we junk the code in this case rather than including it in the kernel, like .section .discard.cond_yield_neon (this seems to conform to some notion of a standard discarded section name, see <asm-generic/vmlinux.lds.h>). This additionally discards the do_cond_yield_neon invocation (which I guess is what we'd expect for a non-preemptible kernel?) If doing that discard, a note could be added in the comment block to warn people not to assume that the patchup code and any labels defined in it will definitely end up in the kernel image. Since the patchup sequences aren't likely to be many or large, it's not the end of the world if we don't do this discarding though. > +.Lyield_\@ : > + .endm > + > + .macro do_cond_yield_neon > + bl kernel_neon_end > + bl kernel_neon_begin > + .endm > + > + .macro endif_yield_neon, lbl > + .ifnb \lbl > + b \lbl > + .else > + b .Lyield_out_\@ > + .endif Should you include .purgem do_cond_yield_neon .purgem endif_yield_neon here? Otherwise, I think you would get macro redefinition errors if if_will_cond_yield_neon is used more than once in a given file. You could maybe protect against nested and misordered macro uses by the following, though it feels a bit like overkill. Alternatively you could use a magic symbol to record the current state, similarly to frame_{push,pop}. .macro __if_will_cond_yield_neon .purgem if_will_cond_yield_neon //... .macro do_cond_yield_neon .purgem do_cond_yield_neon //... .macro endif_yield_neon .purgem endif_yield_neon //... .macro if_will_cond_yield_neon __if_will_cond_yield_neon .endm // if_will_cond_yield_neon .endm // endif_yield_neon .endm // do_cond_yield_neon .endm // __if_will_cond_yield_neon .macro if_will_cond_yield_neon __if_will_cond_yield_neon .endm > + .previous > +.Lyield_out_\@ : > + .endm > + > #endif /* __ASM_ASSEMBLER_H */ > diff --git a/arch/arm64/kernel/asm-offsets.c b/arch/arm64/kernel/asm-offsets.c > index 1303e04110cd..1e2ea2e51acb 100644 > --- a/arch/arm64/kernel/asm-offsets.c > +++ b/arch/arm64/kernel/asm-offsets.c > @@ -93,6 +93,8 @@ int main(void) > DEFINE(DMA_TO_DEVICE, DMA_TO_DEVICE); > DEFINE(DMA_FROM_DEVICE, DMA_FROM_DEVICE); > BLANK(); #include <linux/preempt.h> ? > + DEFINE(PREEMPT_DISABLE_OFFSET, PREEMPT_DISABLE_OFFSET); > + BLANK(); [...] Cheers ---Dave
On 28 March 2018 at 18:18, Dave Martin <Dave.Martin@arm.com> wrote: > On Wed, Mar 28, 2018 at 02:41:29PM +0200, Ard Biesheuvel wrote: >> Add support macros to conditionally yield the NEON (and thus the CPU) >> that may be called from the assembler code. >> >> In some cases, yielding the NEON involves saving and restoring a non >> trivial amount of context (especially in the CRC folding algorithms), >> and so the macro is split into three, and the code in between is only >> executed when the yield path is taken, allowing the context to be preserved. >> The third macro takes an optional label argument that marks the resume >> path after a yield has been performed. > > Minor comments below, mostly just suggestions/observations. > > With the missing #include in asm-offsets.c fixed (if you think it's > appropriate): > > Reviewed-by: Dave Martin <Dave.Martin@arm.com> > Thanks Dave Replies below >> Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> >> --- >> arch/arm64/include/asm/assembler.h | 64 ++++++++++++++++++++ >> arch/arm64/kernel/asm-offsets.c | 2 + >> 2 files changed, 66 insertions(+) >> >> diff --git a/arch/arm64/include/asm/assembler.h b/arch/arm64/include/asm/assembler.h >> index d354eb7f2f0c..fb11514273d9 100644 >> --- a/arch/arm64/include/asm/assembler.h >> +++ b/arch/arm64/include/asm/assembler.h >> @@ -623,4 +623,68 @@ USER(\label, ic ivau, \tmp2) // invalidate I line PoU >> .endif >> .endm >> >> +/* >> + * Check whether to yield to another runnable task from kernel mode NEON code >> + * (which runs with preemption disabled). >> + * >> + * if_will_cond_yield_neon >> + * // pre-yield patchup code >> + * do_cond_yield_neon >> + * // post-yield patchup code >> + * endif_yield_neon <label> >> + * >> + * where <label> is optional, and marks the point where execution will resume >> + * after a yield has been performed. If omitted, execution resumes right after >> + * the endif_yield_neon invocation. > > Maybe add a comment describing cond_yield_neon, e.g.: > > * > * As a convenience, in the case where no patchup code is required > * the above sequence may be abbreviated to: > * > * cond_yield_neon <label> > Makes sense. I will add that. >> + * >> + * Note that the patchup code does not support assembler directives that change >> + * the output section, any use of such directives is undefined. >> + * >> + * The yield itself consists of the following: >> + * - Check whether the preempt count is exactly 1, in which case disabling >> + * preemption once will make the task preemptible. If this is not the case, >> + * yielding is pointless. >> + * - Check whether TIF_NEED_RESCHED is set, and if so, disable and re-enable >> + * kernel mode NEON (which will trigger a reschedule), and branch to the >> + * yield fixup code. >> + * >> + * This macro sequence clobbers x0, x1 and the flags register unconditionally, >> + * and may clobber x2 .. x18 if the yield path is taken. >> + */ > > Does this mean that the pre-yield patchup code can safely refer to > x2..x18, but the post-yield patchup code and the code at <label> (or > otherwise immediately following endif_yield_neon) can't? > In theory, yes, but it doesn't really matter in practice. If you go down the yield path, you will always run the pre and post sequences, and the main code will need to keep state in x19 and up anyway if it wants it to be preserved. I should probably rephrase this to say that x0 .. x18 may be clobbered. >> + >> + .macro cond_yield_neon, lbl >> + if_will_cond_yield_neon >> + do_cond_yield_neon >> + endif_yield_neon \lbl >> + .endm >> + >> + .macro if_will_cond_yield_neon >> +#ifdef CONFIG_PREEMPT >> + get_thread_info x0 >> + ldr w1, [x0, #TSK_TI_PREEMPT] >> + ldr x0, [x0, #TSK_TI_FLAGS] >> + cmp w1, #PREEMPT_DISABLE_OFFSET >> + csel x0, x0, xzr, eq >> + tbnz x0, #TIF_NEED_RESCHED, .Lyield_\@ // needs rescheduling? >> +#endif >> + /* fall through to endif_yield_neon */ >> + .subsection 1 > > Can we junk the code in this case rather than including it in the > kernel, like > > .section .discard.cond_yield_neon > > (this seems to conform to some notion of a standard discarded section > name, see <asm-generic/vmlinux.lds.h>). This additionally discards > the do_cond_yield_neon invocation (which I guess is what we'd expect > for a non-preemptible kernel?) > > If doing that discard, a note could be added in the comment block > to warn people not to assume that the patchup code and any labels > defined in it will definitely end up in the kernel image. > > Since the patchup sequences aren't likely to be many or large, it's > not the end of the world if we don't do this discarding though. > I chose not to bother. These are handcrafted assembly files that are usually kept in modules, which means the .text footprint is a 4k multiple anyway, and the code is complex enough as it is, so discarding ~10 instructions that have been moved out of the hot path already doesn't seem that useful to me. >> +.Lyield_\@ : >> + .endm >> + >> + .macro do_cond_yield_neon >> + bl kernel_neon_end >> + bl kernel_neon_begin >> + .endm >> + >> + .macro endif_yield_neon, lbl >> + .ifnb \lbl >> + b \lbl >> + .else >> + b .Lyield_out_\@ >> + .endif > > Should you include > > .purgem do_cond_yield_neon > .purgem endif_yield_neon > > here? > > Otherwise, I think you would get macro redefinition errors if > if_will_cond_yield_neon is used more than once in a given file. > if_will_cond_yield_neon does not define any macros itself, so this shouldn't be a problem. > You could maybe protect against nested and misordered macro uses by the > following, though it feels a bit like overkill. Alternatively you > could use a magic symbol to record the current state, similarly to > frame_{push,pop}. > > .macro __if_will_cond_yield_neon > .purgem if_will_cond_yield_neon > //... > > .macro do_cond_yield_neon > .purgem do_cond_yield_neon > //... > > .macro endif_yield_neon > .purgem endif_yield_neon > //... > > .macro if_will_cond_yield_neon > __if_will_cond_yield_neon > .endm // if_will_cond_yield_neon > .endm // endif_yield_neon > .endm // do_cond_yield_neon > .endm // __if_will_cond_yield_neon > > .macro if_will_cond_yield_neon > __if_will_cond_yield_neon > .endm > >> + .previous >> +.Lyield_out_\@ : >> + .endm >> + >> #endif /* __ASM_ASSEMBLER_H */ >> diff --git a/arch/arm64/kernel/asm-offsets.c b/arch/arm64/kernel/asm-offsets.c >> index 1303e04110cd..1e2ea2e51acb 100644 >> --- a/arch/arm64/kernel/asm-offsets.c >> +++ b/arch/arm64/kernel/asm-offsets.c >> @@ -93,6 +93,8 @@ int main(void) >> DEFINE(DMA_TO_DEVICE, DMA_TO_DEVICE); >> DEFINE(DMA_FROM_DEVICE, DMA_FROM_DEVICE); >> BLANK(); > > #include <linux/preempt.h> ? > Good point, will add. >> + DEFINE(PREEMPT_DISABLE_OFFSET, PREEMPT_DISABLE_OFFSET); >> + BLANK(); > > [...] > > Cheers > ---Dave
On Thu, Mar 29, 2018 at 10:02:18AM +0100, Ard Biesheuvel wrote: > On 28 March 2018 at 18:18, Dave Martin <Dave.Martin@arm.com> wrote: > > On Wed, Mar 28, 2018 at 02:41:29PM +0200, Ard Biesheuvel wrote: > >> Add support macros to conditionally yield the NEON (and thus the CPU) > >> that may be called from the assembler code. > >> > >> In some cases, yielding the NEON involves saving and restoring a non > >> trivial amount of context (especially in the CRC folding algorithms), > >> and so the macro is split into three, and the code in between is only > >> executed when the yield path is taken, allowing the context to be preserved. > >> The third macro takes an optional label argument that marks the resume > >> path after a yield has been performed. > > > > Minor comments below, mostly just suggestions/observations. > > > > With the missing #include in asm-offsets.c fixed (if you think it's > > appropriate): > > > > Reviewed-by: Dave Martin <Dave.Martin@arm.com> > > > > Thanks Dave > > Replies below > > >> Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> > >> --- > >> arch/arm64/include/asm/assembler.h | 64 ++++++++++++++++++++ > >> arch/arm64/kernel/asm-offsets.c | 2 + > >> 2 files changed, 66 insertions(+) > >> > >> diff --git a/arch/arm64/include/asm/assembler.h b/arch/arm64/include/asm/assembler.h > >> index d354eb7f2f0c..fb11514273d9 100644 > >> --- a/arch/arm64/include/asm/assembler.h > >> +++ b/arch/arm64/include/asm/assembler.h > >> @@ -623,4 +623,68 @@ USER(\label, ic ivau, \tmp2) // invalidate I line PoU > >> .endif > >> .endm > >> > >> +/* > >> + * Check whether to yield to another runnable task from kernel mode NEON code > >> + * (which runs with preemption disabled). > >> + * > >> + * if_will_cond_yield_neon > >> + * // pre-yield patchup code > >> + * do_cond_yield_neon > >> + * // post-yield patchup code > >> + * endif_yield_neon <label> > >> + * > >> + * where <label> is optional, and marks the point where execution will resume > >> + * after a yield has been performed. If omitted, execution resumes right after > >> + * the endif_yield_neon invocation. > > > > Maybe add a comment describing cond_yield_neon, e.g.: > > > > * > > * As a convenience, in the case where no patchup code is required > > * the above sequence may be abbreviated to: > > * > > * cond_yield_neon <label> > > > > Makes sense. I will add that. > > >> + * > >> + * Note that the patchup code does not support assembler directives that change > >> + * the output section, any use of such directives is undefined. > >> + * > >> + * The yield itself consists of the following: > >> + * - Check whether the preempt count is exactly 1, in which case disabling > >> + * preemption once will make the task preemptible. If this is not the case, > >> + * yielding is pointless. > >> + * - Check whether TIF_NEED_RESCHED is set, and if so, disable and re-enable > >> + * kernel mode NEON (which will trigger a reschedule), and branch to the > >> + * yield fixup code. > >> + * > >> + * This macro sequence clobbers x0, x1 and the flags register unconditionally, > >> + * and may clobber x2 .. x18 if the yield path is taken. > >> + */ > > > > Does this mean that the pre-yield patchup code can safely refer to > > x2..x18, but the post-yield patchup code and the code at <label> (or > > otherwise immediately following endif_yield_neon) can't? > > > > In theory, yes, but it doesn't really matter in practice. If you go > down the yield path, you will always run the pre and post sequences, > and the main code will need to keep state in x19 and up anyway if it > wants it to be preserved. True. > I should probably rephrase this to say that x0 .. x18 may be clobbered. Sure, that would be simpler. Or maybe just say that the set of clobbers is the same as for a function call -- this would cover NZCV for example. > >> + > >> + .macro cond_yield_neon, lbl > >> + if_will_cond_yield_neon > >> + do_cond_yield_neon > >> + endif_yield_neon \lbl > >> + .endm > >> + > >> + .macro if_will_cond_yield_neon > >> +#ifdef CONFIG_PREEMPT > >> + get_thread_info x0 > >> + ldr w1, [x0, #TSK_TI_PREEMPT] > >> + ldr x0, [x0, #TSK_TI_FLAGS] > >> + cmp w1, #PREEMPT_DISABLE_OFFSET > >> + csel x0, x0, xzr, eq > >> + tbnz x0, #TIF_NEED_RESCHED, .Lyield_\@ // needs rescheduling? > >> +#endif > >> + /* fall through to endif_yield_neon */ > >> + .subsection 1 > > > > Can we junk the code in this case rather than including it in the > > kernel, like > > > > .section .discard.cond_yield_neon > > > > (this seems to conform to some notion of a standard discarded section > > name, see <asm-generic/vmlinux.lds.h>). This additionally discards > > the do_cond_yield_neon invocation (which I guess is what we'd expect > > for a non-preemptible kernel?) > > > > If doing that discard, a note could be added in the comment block > > to warn people not to assume that the patchup code and any labels > > defined in it will definitely end up in the kernel image. > > > > Since the patchup sequences aren't likely to be many or large, it's > > not the end of the world if we don't do this discarding though. > > > > I chose not to bother. These are handcrafted assembly files that are > usually kept in modules, which means the .text footprint is a 4k > multiple anyway, and the code is complex enough as it is, so > discarding ~10 instructions that have been moved out of the hot path > already doesn't seem that useful to me. Agreed. Do you know who is building CONFIG_PREEMPT=n kernels these days? > >> +.Lyield_\@ : > >> + .endm > >> + > >> + .macro do_cond_yield_neon > >> + bl kernel_neon_end > >> + bl kernel_neon_begin > >> + .endm > >> + > >> + .macro endif_yield_neon, lbl > >> + .ifnb \lbl > >> + b \lbl > >> + .else > >> + b .Lyield_out_\@ > >> + .endif > > > > Should you include > > > > .purgem do_cond_yield_neon > > .purgem endif_yield_neon > > > > here? > > > > Otherwise, I think you would get macro redefinition errors if > > if_will_cond_yield_neon is used more than once in a given file. > > > > if_will_cond_yield_neon does not define any macros itself, so this > shouldn't be a problem. You're right. I skipped an .endm for some reason while reading and decided there were nested macros here. But there aren't. Protecting against misuse would be "nice", but people using them already need to know what they're doing, so it's low-priority and something that could be added in a later patch. So I agree that there's no need to add that here. [...] Cheers ---Dave
On 29 March 2018 at 10:36, Dave Martin <Dave.Martin@arm.com> wrote: > On Thu, Mar 29, 2018 at 10:02:18AM +0100, Ard Biesheuvel wrote: >> On 28 March 2018 at 18:18, Dave Martin <Dave.Martin@arm.com> wrote: >> > On Wed, Mar 28, 2018 at 02:41:29PM +0200, Ard Biesheuvel wrote: >> >> Add support macros to conditionally yield the NEON (and thus the CPU) >> >> that may be called from the assembler code. >> >> >> >> In some cases, yielding the NEON involves saving and restoring a non >> >> trivial amount of context (especially in the CRC folding algorithms), >> >> and so the macro is split into three, and the code in between is only >> >> executed when the yield path is taken, allowing the context to be preserved. >> >> The third macro takes an optional label argument that marks the resume >> >> path after a yield has been performed. >> > >> > Minor comments below, mostly just suggestions/observations. >> > >> > With the missing #include in asm-offsets.c fixed (if you think it's >> > appropriate): >> > >> > Reviewed-by: Dave Martin <Dave.Martin@arm.com> >> > >> >> Thanks Dave >> >> Replies below >> >> >> Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> >> >> --- >> >> arch/arm64/include/asm/assembler.h | 64 ++++++++++++++++++++ >> >> arch/arm64/kernel/asm-offsets.c | 2 + >> >> 2 files changed, 66 insertions(+) >> >> >> >> diff --git a/arch/arm64/include/asm/assembler.h b/arch/arm64/include/asm/assembler.h >> >> index d354eb7f2f0c..fb11514273d9 100644 >> >> --- a/arch/arm64/include/asm/assembler.h >> >> +++ b/arch/arm64/include/asm/assembler.h >> >> @@ -623,4 +623,68 @@ USER(\label, ic ivau, \tmp2) // invalidate I line PoU >> >> .endif >> >> .endm >> >> >> >> +/* >> >> + * Check whether to yield to another runnable task from kernel mode NEON code >> >> + * (which runs with preemption disabled). >> >> + * >> >> + * if_will_cond_yield_neon >> >> + * // pre-yield patchup code >> >> + * do_cond_yield_neon >> >> + * // post-yield patchup code >> >> + * endif_yield_neon <label> >> >> + * >> >> + * where <label> is optional, and marks the point where execution will resume >> >> + * after a yield has been performed. If omitted, execution resumes right after >> >> + * the endif_yield_neon invocation. >> > >> > Maybe add a comment describing cond_yield_neon, e.g.: >> > >> > * >> > * As a convenience, in the case where no patchup code is required >> > * the above sequence may be abbreviated to: >> > * >> > * cond_yield_neon <label> >> > >> >> Makes sense. I will add that. >> >> >> + * >> >> + * Note that the patchup code does not support assembler directives that change >> >> + * the output section, any use of such directives is undefined. >> >> + * >> >> + * The yield itself consists of the following: >> >> + * - Check whether the preempt count is exactly 1, in which case disabling >> >> + * preemption once will make the task preemptible. If this is not the case, >> >> + * yielding is pointless. >> >> + * - Check whether TIF_NEED_RESCHED is set, and if so, disable and re-enable >> >> + * kernel mode NEON (which will trigger a reschedule), and branch to the >> >> + * yield fixup code. >> >> + * >> >> + * This macro sequence clobbers x0, x1 and the flags register unconditionally, >> >> + * and may clobber x2 .. x18 if the yield path is taken. >> >> + */ >> > >> > Does this mean that the pre-yield patchup code can safely refer to >> > x2..x18, but the post-yield patchup code and the code at <label> (or >> > otherwise immediately following endif_yield_neon) can't? >> > >> >> In theory, yes, but it doesn't really matter in practice. If you go >> down the yield path, you will always run the pre and post sequences, >> and the main code will need to keep state in x19 and up anyway if it >> wants it to be preserved. > > True. > >> I should probably rephrase this to say that x0 .. x18 may be clobbered. > > Sure, that would be simpler. Or maybe just say that the set of clobbers > is the same as for a function call -- this would cover NZCV for example. > Even better. >> >> + >> >> + .macro cond_yield_neon, lbl >> >> + if_will_cond_yield_neon >> >> + do_cond_yield_neon >> >> + endif_yield_neon \lbl >> >> + .endm >> >> + >> >> + .macro if_will_cond_yield_neon >> >> +#ifdef CONFIG_PREEMPT >> >> + get_thread_info x0 >> >> + ldr w1, [x0, #TSK_TI_PREEMPT] >> >> + ldr x0, [x0, #TSK_TI_FLAGS] >> >> + cmp w1, #PREEMPT_DISABLE_OFFSET >> >> + csel x0, x0, xzr, eq >> >> + tbnz x0, #TIF_NEED_RESCHED, .Lyield_\@ // needs rescheduling? >> >> +#endif >> >> + /* fall through to endif_yield_neon */ >> >> + .subsection 1 >> > >> > Can we junk the code in this case rather than including it in the >> > kernel, like >> > >> > .section .discard.cond_yield_neon >> > >> > (this seems to conform to some notion of a standard discarded section >> > name, see <asm-generic/vmlinux.lds.h>). This additionally discards >> > the do_cond_yield_neon invocation (which I guess is what we'd expect >> > for a non-preemptible kernel?) >> > >> > If doing that discard, a note could be added in the comment block >> > to warn people not to assume that the patchup code and any labels >> > defined in it will definitely end up in the kernel image. >> > >> > Since the patchup sequences aren't likely to be many or large, it's >> > not the end of the world if we don't do this discarding though. >> > >> >> I chose not to bother. These are handcrafted assembly files that are >> usually kept in modules, which means the .text footprint is a 4k >> multiple anyway, and the code is complex enough as it is, so >> discarding ~10 instructions that have been moved out of the hot path >> already doesn't seem that useful to me. > > Agreed. Do you know who is building CONFIG_PREEMPT=n kernels these > days? > AFAIK most distro kernels use voluntary preemption, so they'd still benefit from this. >> >> +.Lyield_\@ : >> >> + .endm >> >> + >> >> + .macro do_cond_yield_neon >> >> + bl kernel_neon_end >> >> + bl kernel_neon_begin >> >> + .endm >> >> + >> >> + .macro endif_yield_neon, lbl >> >> + .ifnb \lbl >> >> + b \lbl >> >> + .else >> >> + b .Lyield_out_\@ >> >> + .endif >> > >> > Should you include >> > >> > .purgem do_cond_yield_neon >> > .purgem endif_yield_neon >> > >> > here? >> > >> > Otherwise, I think you would get macro redefinition errors if >> > if_will_cond_yield_neon is used more than once in a given file. >> > >> >> if_will_cond_yield_neon does not define any macros itself, so this >> shouldn't be a problem. > > You're right. I skipped an .endm for some reason while reading and > decided there were nested macros here. But there aren't. > > Protecting against misuse would be "nice", but people using them already > need to know what they're doing, so it's low-priority and something that > could be added in a later patch. So I agree that there's no need to add > that here. > OK. I will respin with the minor issues addressed and your R-b added, and repost before the end of the day. Will, hopefully you're still ok with picking this up for v4.17? I'd hate to postpone the crypto pieces that depend on it to v4.19
On Thu, Mar 29, 2018 at 10:59:28AM +0100, Ard Biesheuvel wrote: > On 29 March 2018 at 10:36, Dave Martin <Dave.Martin@arm.com> wrote: > > On Thu, Mar 29, 2018 at 10:02:18AM +0100, Ard Biesheuvel wrote: > >> On 28 March 2018 at 18:18, Dave Martin <Dave.Martin@arm.com> wrote: [...] > >> I should probably rephrase this to say that x0 .. x18 may be clobbered. > > > > Sure, that would be simpler. Or maybe just say that the set of clobbers > > is the same as for a function call -- this would cover NZCV for example. > > > > Even better. [...] > >> > Since the patchup sequences aren't likely to be many or large, it's > >> > not the end of the world if we don't do this discarding though. > >> > > >> > >> I chose not to bother. These are handcrafted assembly files that are > >> usually kept in modules, which means the .text footprint is a 4k > >> multiple anyway, and the code is complex enough as it is, so > >> discarding ~10 instructions that have been moved out of the hot path > >> already doesn't seem that useful to me. > > > > Agreed. Do you know who is building CONFIG_PREEMPT=n kernels these > > days? > > > > AFAIK most distro kernels use voluntary preemption, so they'd still > benefit from this. OK, and given the size of the typical distro kernel, I doubt anyone will lose sleep over a couple of hundred extra bytes. I might try to hack it up later just for fun, just to see whether it works. [...] > >> > Should you include > >> > > >> > .purgem do_cond_yield_neon > >> > .purgem endif_yield_neon > >> > > >> > here? > >> > > >> > Otherwise, I think you would get macro redefinition errors if > >> > if_will_cond_yield_neon is used more than once in a given file. > >> > > >> > >> if_will_cond_yield_neon does not define any macros itself, so this > >> shouldn't be a problem. > > > > You're right. I skipped an .endm for some reason while reading and > > decided there were nested macros here. But there aren't. > > > > Protecting against misuse would be "nice", but people using them already > > need to know what they're doing, so it's low-priority and something that > > could be added in a later patch. So I agree that there's no need to add > > that here. > > > > OK. > > I will respin with the minor issues addressed and your R-b added, and > repost before the end of the day. Sounds good to me. Cheers ---Dave > Will, hopefully you're still ok with picking this up for v4.17? I'd > hate to postpone the crypto pieces that depend on it to v4.19 [...]
> On 29 Mar 2018, at 13:12, Dave Martin <Dave.Martin@arm.com> wrote: > >> On Thu, Mar 29, 2018 at 10:59:28AM +0100, Ard Biesheuvel wrote: >>> On 29 March 2018 at 10:36, Dave Martin <Dave.Martin@arm.com> wrote: >>>> On Thu, Mar 29, 2018 at 10:02:18AM +0100, Ard Biesheuvel wrote: >>>> On 28 March 2018 at 18:18, Dave Martin <Dave.Martin@arm.com> wrote: > > [...] > >>>> I should probably rephrase this to say that x0 .. x18 may be clobbered. >>> >>> Sure, that would be simpler. Or maybe just say that the set of clobbers >>> is the same as for a function call -- this would cover NZCV for example. >>> >> >> Even better. > > [...] > >>>>> Since the patchup sequences aren't likely to be many or large, it's >>>>> not the end of the world if we don't do this discarding though. >>>>> >>>> >>>> I chose not to bother. These are handcrafted assembly files that are >>>> usually kept in modules, which means the .text footprint is a 4k >>>> multiple anyway, and the code is complex enough as it is, so >>>> discarding ~10 instructions that have been moved out of the hot path >>>> already doesn't seem that useful to me. >>> >>> Agreed. Do you know who is building CONFIG_PREEMPT=n kernels these >>> days? >>> >> >> AFAIK most distro kernels use voluntary preemption, so they'd still >> benefit from this. > > OK, and given the size of the typical distro kernel, I doubt anyone will > lose sleep over a couple of hundred extra bytes. > My point was that this is /not/ dead code on typical distro kernels given that this approach should work equally under voluntary preemption. > I might try to hack it up later just for fun, just to see whether it > works. > > [...] > >>>>> Should you include >>>>> >>>>> .purgem do_cond_yield_neon >>>>> .purgem endif_yield_neon >>>>> >>>>> here? >>>>> >>>>> Otherwise, I think you would get macro redefinition errors if >>>>> if_will_cond_yield_neon is used more than once in a given file. >>>>> >>>> >>>> if_will_cond_yield_neon does not define any macros itself, so this >>>> shouldn't be a problem. >>> >>> You're right. I skipped an .endm for some reason while reading and >>> decided there were nested macros here. But there aren't. >>> >>> Protecting against misuse would be "nice", but people using them already >>> need to know what they're doing, so it's low-priority and something that >>> could be added in a later patch. So I agree that there's no need to add >>> that here. >>> >> >> OK. >> >> I will respin with the minor issues addressed and your R-b added, and >> repost before the end of the day. > > Sounds good to me. > > Cheers > ---Dave > >> Will, hopefully you're still ok with picking this up for v4.17? I'd >> hate to postpone the crypto pieces that depend on it to v4.19 > > [...]
On Thu, Mar 29, 2018 at 01:36:44PM +0200, Ard Biesheuvel wrote: > > > On 29 Mar 2018, at 13:12, Dave Martin <Dave.Martin@arm.com> wrote: > > > >> On Thu, Mar 29, 2018 at 10:59:28AM +0100, Ard Biesheuvel wrote: > >>> On 29 March 2018 at 10:36, Dave Martin <Dave.Martin@arm.com> wrote: > >>>> On Thu, Mar 29, 2018 at 10:02:18AM +0100, Ard Biesheuvel wrote: > >>>> On 28 March 2018 at 18:18, Dave Martin <Dave.Martin@arm.com> wrote: [...] > >>>>> Since the patchup sequences aren't likely to be many or large, it's > >>>>> not the end of the world if we don't do this discarding though. > >>>>> > >>>> > >>>> I chose not to bother. These are handcrafted assembly files that are > >>>> usually kept in modules, which means the .text footprint is a 4k > >>>> multiple anyway, and the code is complex enough as it is, so > >>>> discarding ~10 instructions that have been moved out of the hot path > >>>> already doesn't seem that useful to me. > >>> > >>> Agreed. Do you know who is building CONFIG_PREEMPT=n kernels these > >>> days? > >>> > >> > >> AFAIK most distro kernels use voluntary preemption, so they'd still > >> benefit from this. > > > > OK, and given the size of the typical distro kernel, I doubt anyone will > > lose sleep over a couple of hundred extra bytes. > > > > My point was that this is /not/ dead code on typical distro kernels > given that this approach should work equally under voluntary preemption. I think CONFIG_PREEMPT and CONFIG_PREEMPT_VOLUNTARY are mutually exclusive, so in the PREEMPT_VOLUNTARY case the yield path code will get compiled out here. But that's probably the right thing to do IIUC: unless we introduce an explicit preemption point into do_cond_yield_neon, voluntary preemption won't occur anyway. And the crypto API probably doesn't expect us to do that... especially if we're in a softirq. [...] Cheers ---Dave
diff --git a/arch/arm64/include/asm/assembler.h b/arch/arm64/include/asm/assembler.h index d354eb7f2f0c..fb11514273d9 100644 --- a/arch/arm64/include/asm/assembler.h +++ b/arch/arm64/include/asm/assembler.h @@ -623,4 +623,68 @@ USER(\label, ic ivau, \tmp2) // invalidate I line PoU .endif .endm +/* + * Check whether to yield to another runnable task from kernel mode NEON code + * (which runs with preemption disabled). + * + * if_will_cond_yield_neon + * // pre-yield patchup code + * do_cond_yield_neon + * // post-yield patchup code + * endif_yield_neon <label> + * + * where <label> is optional, and marks the point where execution will resume + * after a yield has been performed. If omitted, execution resumes right after + * the endif_yield_neon invocation. + * + * Note that the patchup code does not support assembler directives that change + * the output section, any use of such directives is undefined. + * + * The yield itself consists of the following: + * - Check whether the preempt count is exactly 1, in which case disabling + * preemption once will make the task preemptible. If this is not the case, + * yielding is pointless. + * - Check whether TIF_NEED_RESCHED is set, and if so, disable and re-enable + * kernel mode NEON (which will trigger a reschedule), and branch to the + * yield fixup code. + * + * This macro sequence clobbers x0, x1 and the flags register unconditionally, + * and may clobber x2 .. x18 if the yield path is taken. + */ + + .macro cond_yield_neon, lbl + if_will_cond_yield_neon + do_cond_yield_neon + endif_yield_neon \lbl + .endm + + .macro if_will_cond_yield_neon +#ifdef CONFIG_PREEMPT + get_thread_info x0 + ldr w1, [x0, #TSK_TI_PREEMPT] + ldr x0, [x0, #TSK_TI_FLAGS] + cmp w1, #PREEMPT_DISABLE_OFFSET + csel x0, x0, xzr, eq + tbnz x0, #TIF_NEED_RESCHED, .Lyield_\@ // needs rescheduling? +#endif + /* fall through to endif_yield_neon */ + .subsection 1 +.Lyield_\@ : + .endm + + .macro do_cond_yield_neon + bl kernel_neon_end + bl kernel_neon_begin + .endm + + .macro endif_yield_neon, lbl + .ifnb \lbl + b \lbl + .else + b .Lyield_out_\@ + .endif + .previous +.Lyield_out_\@ : + .endm + #endif /* __ASM_ASSEMBLER_H */ diff --git a/arch/arm64/kernel/asm-offsets.c b/arch/arm64/kernel/asm-offsets.c index 1303e04110cd..1e2ea2e51acb 100644 --- a/arch/arm64/kernel/asm-offsets.c +++ b/arch/arm64/kernel/asm-offsets.c @@ -93,6 +93,8 @@ int main(void) DEFINE(DMA_TO_DEVICE, DMA_TO_DEVICE); DEFINE(DMA_FROM_DEVICE, DMA_FROM_DEVICE); BLANK(); + DEFINE(PREEMPT_DISABLE_OFFSET, PREEMPT_DISABLE_OFFSET); + BLANK(); DEFINE(CLOCK_REALTIME, CLOCK_REALTIME); DEFINE(CLOCK_MONOTONIC, CLOCK_MONOTONIC); DEFINE(CLOCK_MONOTONIC_RAW, CLOCK_MONOTONIC_RAW);
Add support macros to conditionally yield the NEON (and thus the CPU) that may be called from the assembler code. In some cases, yielding the NEON involves saving and restoring a non trivial amount of context (especially in the CRC folding algorithms), and so the macro is split into three, and the code in between is only executed when the yield path is taken, allowing the context to be preserved. The third macro takes an optional label argument that marks the resume path after a yield has been performed. Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> --- arch/arm64/include/asm/assembler.h | 64 ++++++++++++++++++++ arch/arm64/kernel/asm-offsets.c | 2 + 2 files changed, 66 insertions(+)