Messages in this thread | | | Date | Tue, 11 Mar 2008 08:07:34 +0300 | From | Alexey Dobriyan <> | Subject | Re: [patch 1/2] x86, fpu: split FPU state from task struct - v5 |
| |
On Mon, Mar 10, 2008 at 03:28:04PM -0700, Suresh Siddha wrote: > Split the FPU save area from the task struct. This allows easy migration > of FPU context, and it's generally cleaner. It also allows the following > two optimizations: > > 1) only allocate when the application actually uses FPU, so in the first > lazy FPU trap. This could save memory for non-fpu using apps. Next patch > does this lazy allocation. > > 2) allocate the right size for the actual cpu rather than 512 bytes always. > Patches enabling xsave/xrstor support (coming shortly) will take advantage > of this.
Ugh, not seeing patch, but judging from description it will make "choose wrong CONFIG_M* and fxsave will corrupt random FPU state" situation likely?
> --- linux-2.6-x86.orig/arch/x86/kernel/process_64.c > +++ linux-2.6-x86/arch/x86/kernel/process_64.c > @@ -634,7 +634,7 @@ > > /* we're going to use this soon, after a few expensive things */ > if (next_p->fpu_counter>5) > - prefetch(&next->i387.fxsave); > + prefetch(next->xstate);
Can we please give it better name, like fpu_state? It's a member of task_struct after all.
> --- linux-2.6-x86.orig/arch/x86/kernel/traps_64.c > +++ linux-2.6-x86/arch/x86/kernel/traps_64.c > @@ -1157,6 +1157,10 @@ > #endif > > /* > + * initialize the per thread extended state: > + */ > + init_thread_xstate();
Useless comment after xstate renaming :)
> --- linux-2.6-x86.orig/kernel/fork.c > +++ linux-2.6-x86/kernel/fork.c > @@ -144,6 +148,9 @@ > ARCH_MIN_TASKALIGN, SLAB_PANIC | SLAB_NOTRACK, NULL); > #endif > > + /* do the arch specific task caches init */ > + arch_task_cache_init();
Useless comment.
> --- linux-2.6-x86.orig/arch/x86/kernel/i387.c 2008-03-07 10:24:09.000000000 -0800 > +++ linux-2.6-x86/arch/x86/kernel/i387.c 2008-03-10 10:42:04.000000000 -0700 > @@ -61,10 +74,6 @@ > void __cpuinit fpu_init(void) > { > unsigned long oldcr0 = read_cr0(); > - extern void __bad_fxsave_alignment(void); > - > - if (offsetof(struct task_struct, thread.i387.fxsave) & 15) > - __bad_fxsave_alignment();
I think removal of such checks needs giving necessary alignment to cache. Previously it worked because of __aligned((16)) and L1_CACHE_SHIFT combo.
> Index: linux-2.6-x86/include/asm-x86/i387.h > =================================================================== > --- linux-2.6-x86.orig/include/asm-x86/i387.h 2008-03-07 10:24:11.000000000 -0800 > +++ linux-2.6-x86/include/asm-x86/i387.h 2008-03-10 10:42:04.000000000 -0700 > @@ -324,25 +323,25 @@ > static inline unsigned short get_fpu_cwd(struct task_struct *tsk) > { > if (cpu_has_fxsr) { > - return tsk->thread.i387.fxsave.cwd; > + return tsk->thread.xstate->fxsave.cwd; > } else { > - return (unsigned short)tsk->thread.i387.fsave.cwd; > + return (unsigned short) tsk->thread.xstate->fsave.cwd; ^^^ > } > } > > static inline unsigned short get_fpu_swd(struct task_struct *tsk) > { > if (cpu_has_fxsr) { > - return tsk->thread.i387.fxsave.swd; > + return tsk->thread.xstate->fxsave.swd; > } else { > - return (unsigned short)tsk->thread.i387.fsave.swd; > + return (unsigned short) tsk->thread.xstate->fsave.swd; ^^^
> --- linux-2.6-x86.orig/include/asm-x86/processor.h 2008-03-07 10:24:11.000000000 -0800 > +++ linux-2.6-x86/include/asm-x86/processor.h 2008-03-10 10:42:04.000000000 -0700 > @@ -355,7 +355,7 @@ > u32 entry_eip; > }; > > -union i387_union { > +union thread_xstate {
thread_fpu_state.
> Index: linux-2.6-x86/arch/x86/kernel/process.c > =================================================================== > --- /dev/null 1970-01-01 00:00:00.000000000 +0000 > +++ linux-2.6-x86/arch/x86/kernel/process.c 2008-03-10 10:42:04.000000000 -0700 > +void free_thread_info(struct thread_info *ti) > +{ > + kmem_cache_free(task_xstate_cachep, ti->task->thread.xstate); > + ti->task->thread.xstate = NULL; > + > + free_pages((unsigned long)(ti), get_order(THREAD_SIZE));
Uselesss () ^ ^
| |