| Message ID | 1398959381-8126-3-git-send-email-ard.biesheuvel@linaro.org (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On Thu, May 01, 2014 at 04:49:34PM +0100, Ard Biesheuvel wrote: > diff --git a/arch/arm64/kernel/fpsimd.c b/arch/arm64/kernel/fpsimd.c > index 4aef42a04bdc..86ac6a9bc86a 100644 > --- a/arch/arm64/kernel/fpsimd.c > +++ b/arch/arm64/kernel/fpsimd.c > @@ -87,6 +87,39 @@ void fpsimd_flush_thread(void) > preempt_enable(); > } > > +/* > + * Save the userland FPSIMD state of 'current' to memory > + */ > +void fpsimd_preserve_current_state(void) > +{ > + fpsimd_save_state(¤t->thread.fpsimd_state); > +} > + > +/* > + * Load the userland FPSIMD state of 'current' from memory > + */ > +void fpsimd_restore_current_state(void) > +{ > + fpsimd_load_state(¤t->thread.fpsimd_state); > +} > + > +/* > + * Load an updated userland FPSIMD state for 'current' from memory > + */ > +void fpsimd_update_current_state(struct fpsimd_state *state) > +{ > + preempt_disable(); > + fpsimd_load_state(state); > + preempt_enable(); > +} Minor - please update the comment above the functions to state that preemption needs to be disabled by the caller. > +/* > + * Invalidate live CPU copies of task t's FPSIMD state > + */ > +void fpsimd_flush_task_state(struct task_struct *t) > +{ > +} I guess this will be added in a subsequent patch. You could either move it there or add a comment in the commit log that it is a dummy function for now (I prefer the former).
On 6 May 2014 16:43, Catalin Marinas <catalin.marinas@arm.com> wrote: > On Thu, May 01, 2014 at 04:49:34PM +0100, Ard Biesheuvel wrote: >> diff --git a/arch/arm64/kernel/fpsimd.c b/arch/arm64/kernel/fpsimd.c >> index 4aef42a04bdc..86ac6a9bc86a 100644 >> --- a/arch/arm64/kernel/fpsimd.c >> +++ b/arch/arm64/kernel/fpsimd.c >> @@ -87,6 +87,39 @@ void fpsimd_flush_thread(void) >> preempt_enable(); >> } >> >> +/* >> + * Save the userland FPSIMD state of 'current' to memory >> + */ >> +void fpsimd_preserve_current_state(void) >> +{ >> + fpsimd_save_state(¤t->thread.fpsimd_state); >> +} >> + >> +/* >> + * Load the userland FPSIMD state of 'current' from memory >> + */ >> +void fpsimd_restore_current_state(void) >> +{ >> + fpsimd_load_state(¤t->thread.fpsimd_state); >> +} >> + >> +/* >> + * Load an updated userland FPSIMD state for 'current' from memory >> + */ >> +void fpsimd_update_current_state(struct fpsimd_state *state) >> +{ >> + preempt_disable(); >> + fpsimd_load_state(state); >> + preempt_enable(); >> +} > > Minor - please update the comment above the functions to state that > preemption needs to be disabled by the caller. > Do you mean in all three cases? And, by implication, that the preempt_disable()/enable() pair should be moved to the call site for fpsimd_update_current_state() ? >> +/* >> + * Invalidate live CPU copies of task t's FPSIMD state >> + */ >> +void fpsimd_flush_task_state(struct task_struct *t) >> +{ >> +} > > I guess this will be added in a subsequent patch. You could either move > it there or add a comment in the commit log that it is a dummy function > for now (I prefer the former). > OK
On Tue, May 06, 2014 at 03:48:08PM +0100, Ard Biesheuvel wrote: > On 6 May 2014 16:43, Catalin Marinas <catalin.marinas@arm.com> wrote: > > On Thu, May 01, 2014 at 04:49:34PM +0100, Ard Biesheuvel wrote: > >> diff --git a/arch/arm64/kernel/fpsimd.c b/arch/arm64/kernel/fpsimd.c > >> index 4aef42a04bdc..86ac6a9bc86a 100644 > >> --- a/arch/arm64/kernel/fpsimd.c > >> +++ b/arch/arm64/kernel/fpsimd.c > >> @@ -87,6 +87,39 @@ void fpsimd_flush_thread(void) > >> preempt_enable(); > >> } > >> > >> +/* > >> + * Save the userland FPSIMD state of 'current' to memory > >> + */ > >> +void fpsimd_preserve_current_state(void) > >> +{ > >> + fpsimd_save_state(¤t->thread.fpsimd_state); > >> +} > >> + > >> +/* > >> + * Load the userland FPSIMD state of 'current' from memory > >> + */ > >> +void fpsimd_restore_current_state(void) > >> +{ > >> + fpsimd_load_state(¤t->thread.fpsimd_state); > >> +} > >> + > >> +/* > >> + * Load an updated userland FPSIMD state for 'current' from memory > >> + */ > >> +void fpsimd_update_current_state(struct fpsimd_state *state) > >> +{ > >> + preempt_disable(); > >> + fpsimd_load_state(state); > >> + preempt_enable(); > >> +} > > > > Minor - please update the comment above the functions to state that > > preemption needs to be disabled by the caller. > > > > Do you mean in all three cases? And, by implication, that the > preempt_disable()/enable() pair should be moved to the call site for > fpsimd_update_current_state() ? No, just the comment for the first two functions updated.
On Tue, May 06, 2014 at 04:12:55PM +0100, Catalin Marinas wrote: > On Tue, May 06, 2014 at 03:48:08PM +0100, Ard Biesheuvel wrote: > > On 6 May 2014 16:43, Catalin Marinas <catalin.marinas@arm.com> wrote: > > > On Thu, May 01, 2014 at 04:49:34PM +0100, Ard Biesheuvel wrote: > > >> diff --git a/arch/arm64/kernel/fpsimd.c b/arch/arm64/kernel/fpsimd.c > > >> index 4aef42a04bdc..86ac6a9bc86a 100644 > > >> --- a/arch/arm64/kernel/fpsimd.c > > >> +++ b/arch/arm64/kernel/fpsimd.c > > >> @@ -87,6 +87,39 @@ void fpsimd_flush_thread(void) > > >> preempt_enable(); > > >> } > > >> > > >> +/* > > >> + * Save the userland FPSIMD state of 'current' to memory > > >> + */ > > >> +void fpsimd_preserve_current_state(void) > > >> +{ > > >> + fpsimd_save_state(¤t->thread.fpsimd_state); > > >> +} > > >> + > > >> +/* > > >> + * Load the userland FPSIMD state of 'current' from memory > > >> + */ > > >> +void fpsimd_restore_current_state(void) > > >> +{ > > >> + fpsimd_load_state(¤t->thread.fpsimd_state); > > >> +} > > >> + > > >> +/* > > >> + * Load an updated userland FPSIMD state for 'current' from memory > > >> + */ > > >> +void fpsimd_update_current_state(struct fpsimd_state *state) > > >> +{ > > >> + preempt_disable(); > > >> + fpsimd_load_state(state); > > >> + preempt_enable(); > > >> +} > > > > > > Minor - please update the comment above the functions to state that > > > preemption needs to be disabled by the caller. > > > > > > > Do you mean in all three cases? And, by implication, that the > > preempt_disable()/enable() pair should be moved to the call site for > > fpsimd_update_current_state() ? > > No, just the comment for the first two functions updated. I noticed in a subsequent patch that you add preempt_disable/enable already in the first two functions. You could do it here as well to avoid confusion (and no need to update the comment).
diff --git a/arch/arm64/include/asm/fpsimd.h b/arch/arm64/include/asm/fpsimd.h index c43b4ac13008..9f9b8e438546 100644 --- a/arch/arm64/include/asm/fpsimd.h +++ b/arch/arm64/include/asm/fpsimd.h @@ -58,6 +58,12 @@ extern void fpsimd_load_state(struct fpsimd_state *state); extern void fpsimd_thread_switch(struct task_struct *next); extern void fpsimd_flush_thread(void); +extern void fpsimd_preserve_current_state(void); +extern void fpsimd_restore_current_state(void); +extern void fpsimd_update_current_state(struct fpsimd_state *state); + +extern void fpsimd_flush_task_state(struct task_struct *target); + #endif #endif diff --git a/arch/arm64/kernel/fpsimd.c b/arch/arm64/kernel/fpsimd.c index 4aef42a04bdc..86ac6a9bc86a 100644 --- a/arch/arm64/kernel/fpsimd.c +++ b/arch/arm64/kernel/fpsimd.c @@ -87,6 +87,39 @@ void fpsimd_flush_thread(void) preempt_enable(); } +/* + * Save the userland FPSIMD state of 'current' to memory + */ +void fpsimd_preserve_current_state(void) +{ + fpsimd_save_state(¤t->thread.fpsimd_state); +} + +/* + * Load the userland FPSIMD state of 'current' from memory + */ +void fpsimd_restore_current_state(void) +{ + fpsimd_load_state(¤t->thread.fpsimd_state); +} + +/* + * Load an updated userland FPSIMD state for 'current' from memory + */ +void fpsimd_update_current_state(struct fpsimd_state *state) +{ + preempt_disable(); + fpsimd_load_state(state); + preempt_enable(); +} + +/* + * Invalidate live CPU copies of task t's FPSIMD state + */ +void fpsimd_flush_task_state(struct task_struct *t) +{ +} + #ifdef CONFIG_KERNEL_MODE_NEON /* diff --git a/arch/arm64/kernel/process.c b/arch/arm64/kernel/process.c index 6391485f342d..c5693163408c 100644 --- a/arch/arm64/kernel/process.c +++ b/arch/arm64/kernel/process.c @@ -205,7 +205,7 @@ void release_thread(struct task_struct *dead_task) int arch_dup_task_struct(struct task_struct *dst, struct task_struct *src) { - fpsimd_save_state(¤t->thread.fpsimd_state); + fpsimd_preserve_current_state(); *dst = *src; return 0; } diff --git a/arch/arm64/kernel/ptrace.c b/arch/arm64/kernel/ptrace.c index 6a8928bba03c..f8700eca24e7 100644 --- a/arch/arm64/kernel/ptrace.c +++ b/arch/arm64/kernel/ptrace.c @@ -517,6 +517,7 @@ static int fpr_set(struct task_struct *target, const struct user_regset *regset, return ret; target->thread.fpsimd_state.user_fpsimd = newstate; + fpsimd_flush_task_state(target); return ret; } @@ -764,6 +765,7 @@ static int compat_vfp_set(struct task_struct *target, uregs->fpcr = fpscr & VFP_FPSCR_CTRL_MASK; } + fpsimd_flush_task_state(target); return ret; } diff --git a/arch/arm64/kernel/signal.c b/arch/arm64/kernel/signal.c index 890a591f75dd..06448a77ff53 100644 --- a/arch/arm64/kernel/signal.c +++ b/arch/arm64/kernel/signal.c @@ -51,7 +51,7 @@ static int preserve_fpsimd_context(struct fpsimd_context __user *ctx) int err; /* dump the hardware registers to the fpsimd_state structure */ - fpsimd_save_state(fpsimd); + fpsimd_preserve_current_state(); /* copy the FP and status/control registers */ err = __copy_to_user(ctx->vregs, fpsimd->vregs, sizeof(fpsimd->vregs)); @@ -86,11 +86,8 @@ static int restore_fpsimd_context(struct fpsimd_context __user *ctx) __get_user_error(fpsimd.fpcr, &ctx->fpcr, err); /* load the hardware registers from the fpsimd_state structure */ - if (!err) { - preempt_disable(); - fpsimd_load_state(&fpsimd); - preempt_enable(); - } + if (!err) + fpsimd_update_current_state(&fpsimd); return err ? -EFAULT : 0; } diff --git a/arch/arm64/kernel/signal32.c b/arch/arm64/kernel/signal32.c index b3fc9f5ec6d3..ac7e237d0bda 100644 --- a/arch/arm64/kernel/signal32.c +++ b/arch/arm64/kernel/signal32.c @@ -219,7 +219,7 @@ static int compat_preserve_vfp_context(struct compat_vfp_sigframe __user *frame) * Note that this also saves V16-31, which aren't visible * in AArch32. */ - fpsimd_save_state(fpsimd); + fpsimd_preserve_current_state(); /* Place structure header on the stack */ __put_user_error(magic, &frame->magic, err); @@ -282,11 +282,8 @@ static int compat_restore_vfp_context(struct compat_vfp_sigframe __user *frame) * We don't need to touch the exception register, so * reload the hardware state. */ - if (!err) { - preempt_disable(); - fpsimd_load_state(&fpsimd); - preempt_enable(); - } + if (!err) + fpsimd_update_current_state(&fpsimd); return err ? -EFAULT : 0; }
There are two tacit assumptions in the FPSIMD handling code that will no longer hold after the next patch that optimizes away some FPSIMD state restores: . the FPSIMD registers of this CPU contain the userland FPSIMD state of task 'current'; . when switching to a task, its FPSIMD state will always be restored from memory. This patch adds the following functions to abstract away from straight FPSIMD register file saves and restores: - fpsimd_preserve_current_state -> ensure current's FPSIMD state is saved - fpsimd_restore_current_state -> ensure current's FPSIMD state is loaded - fpsimd_update_current_state -> replace current's FPSIMD state - fpsimd_flush_task_state -> invalidate live copies of a task's FPSIMD state Where necessary, the ptrace, signal handling and fork code are updated to use the above wrappers instead of poking into the FPSIMD registers directly. Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> --- arch/arm64/include/asm/fpsimd.h | 6 ++++++ arch/arm64/kernel/fpsimd.c | 33 +++++++++++++++++++++++++++++++++ arch/arm64/kernel/process.c | 2 +- arch/arm64/kernel/ptrace.c | 2 ++ arch/arm64/kernel/signal.c | 9 +++------ arch/arm64/kernel/signal32.c | 9 +++------ 6 files changed, 48 insertions(+), 13 deletions(-)