From 55ccf3fe3f9a3441731aa79cf42a628fc4ecace9 Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Wed, 16 May 2012 15:03:51 -0700 Subject: fork: move the real prepare_to_copy() users to arch_dup_task_struct() Historical prepare_to_copy() is mostly a no-op, duplicated for majority of the architectures and the rest following the x86 model of flushing the extended register state like fpu there. Remove it and use the arch_dup_task_struct() instead. Suggested-by: Oleg Nesterov Suggested-by: Linus Torvalds Signed-off-by: Suresh Siddha Link: http://lkml.kernel.org/r/1336692811-30576-1-git-send-email-suresh.b.siddha@intel.com Acked-by: Benjamin Herrenschmidt Cc: David Howells Cc: Koichi Yasutake Cc: Paul Mackerras Cc: Paul Mundt Cc: Chris Zankel Cc: Richard Henderson Cc: Russell King Cc: Haavard Skinnemoen Cc: Mike Frysinger Cc: Mark Salter Cc: Aurelien Jacquiot Cc: Mikael Starvik Cc: Yoshinori Sato Cc: Richard Kuo Cc: Tony Luck Cc: Michal Simek Cc: Ralf Baechle Cc: Jonas Bonn Cc: James E.J. Bottomley Cc: Helge Deller Cc: Martin Schwidefsky Cc: Heiko Carstens Cc: Chen Liqin Cc: Lennox Wu Cc: David S. Miller Cc: Chris Metcalf Cc: Jeff Dike Cc: Richard Weinberger Cc: Guan Xuetao Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/processor.h | 3 --- arch/x86/kernel/process.c | 6 ++++++ arch/x86/kernel/process_32.c | 9 --------- arch/x86/kernel/process_64.c | 9 --------- 4 files changed, 6 insertions(+), 21 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h index 4fa7dcceb6c..97fe04318e9 100644 --- a/arch/x86/include/asm/processor.h +++ b/arch/x86/include/asm/processor.h @@ -579,9 +579,6 @@ extern int kernel_thread(int (*fn)(void *), void *arg, unsigned long flags); /* Free all resources held by a thread. */ extern void release_thread(struct task_struct *); -/* Prepare to copy thread state - unlazy all lazy state */ -extern void prepare_to_copy(struct task_struct *tsk); - unsigned long get_wchan(struct task_struct *p); /* diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index 1d92a5ab6e8..b7e1e0e5398 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -47,10 +47,16 @@ EXPORT_SYMBOL_GPL(idle_notifier_unregister); struct kmem_cache *task_xstate_cachep; EXPORT_SYMBOL_GPL(task_xstate_cachep); +/* + * this gets called so that we can store lazy state into memory and copy the + * current task into the new thread. + */ int arch_dup_task_struct(struct task_struct *dst, struct task_struct *src) { int ret; + unlazy_fpu(src); + *dst = *src; if (fpu_allocated(&src->thread.fpu)) { memset(&dst->thread.fpu, 0, sizeof(dst->thread.fpu)); diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c index ae6847303e2..2aa57dc909d 100644 --- a/arch/x86/kernel/process_32.c +++ b/arch/x86/kernel/process_32.c @@ -126,15 +126,6 @@ void release_thread(struct task_struct *dead_task) release_vm86_irqs(dead_task); } -/* - * This gets called before we allocate a new thread and copy - * the current task into it. - */ -void prepare_to_copy(struct task_struct *tsk) -{ - unlazy_fpu(tsk); -} - int copy_thread(unsigned long clone_flags, unsigned long sp, unsigned long unused, struct task_struct *p, struct pt_regs *regs) diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index 43d8b48b23e..c4c0645a401 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -145,15 +145,6 @@ static inline u32 read_32bit_tls(struct task_struct *t, int tls) return get_desc_base(&t->thread.tls_array[tls]); } -/* - * This gets called before we allocate a new thread and copy - * the current task into it. - */ -void prepare_to_copy(struct task_struct *tsk) -{ - unlazy_fpu(tsk); -} - int copy_thread(unsigned long clone_flags, unsigned long sp, unsigned long unused, struct task_struct *p, struct pt_regs *regs) -- cgit v1.2.3-70-g09d2 From d75f1b391f5ef73016d14bc6f7e4725820ebaa5b Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Wed, 16 May 2012 15:03:53 -0700 Subject: x86, xsave: remove thread_has_fpu() bug check in __sanitize_i387_state() Code paths like fork(), exit() and signal handling flush the fpu state explicitly to the structures in memory. BUG_ON() in __sanitize_i387_state() is checking that the fpu state is not live any more. But for preempt kernels, task can be scheduled out and in at any place and the preload_fpu logic during context switch can make the fpu registers live again. For example, consider a 64-bit Task which uses fpu frequently and as such you will find its fpu_counter mostly non-zero. During its time slice, kernel used fpu by doing kernel_fpu_begin/kernel_fpu_end(). After this, in the same scheduling slice, task-A got a signal to handle. Then during the signal setup path we got preempted when we are just before the sanitize_i387_state() in arch/x86/kernel/xsave.c:save_i387_xstate(). And when we come back we will have the fpu registers live that can hit the bug_on. Similarly during core dump, other threads can context-switch in and out (because of spurious wakeups while waiting for the coredump to finish in kernel/exit.c:exit_mm()) and the main thread dumping core can run into this bug when it finds some other thread with its fpu registers live on some other cpu. So remove the paranoid check for now, even though it caught a bug in the multi-threaded core dump case (fixed in the previous patch). Signed-off-by: Suresh Siddha Link: http://lkml.kernel.org/r/1336692811-30576-3-git-send-email-suresh.b.siddha@intel.com Cc: Oleg Nesterov Cc: Linus Torvalds Signed-off-by: H. Peter Anvin --- arch/x86/kernel/xsave.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/xsave.c b/arch/x86/kernel/xsave.c index e62728e30b0..bd18149b2b0 100644 --- a/arch/x86/kernel/xsave.c +++ b/arch/x86/kernel/xsave.c @@ -48,8 +48,6 @@ void __sanitize_i387_state(struct task_struct *tsk) if (!fx) return; - BUG_ON(__thread_has_fpu(tsk)); - xstate_bv = tsk->thread.fpu.state->xsave.xsave_hdr.xstate_bv; /* -- cgit v1.2.3-70-g09d2 From 1dcc8d7ba235a316a056f993e88f0d18b92c60d9 Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Wed, 16 May 2012 15:03:54 -0700 Subject: x86, fpu: drop the fpu state during thread exit There is no need to save any active fpu state to the task structure memory if the task is dead. Just drop the state instead. For example, this saved some 1770 xsave's during the system boot of a two socket Xeon system. Suggested-by: Linus Torvalds Signed-off-by: Suresh Siddha Link: http://lkml.kernel.org/r/1336692811-30576-4-git-send-email-suresh.b.siddha@intel.com Cc: Oleg Nesterov Signed-off-by: H. Peter Anvin --- arch/x86/kernel/process.c | 19 +++++++++++++------ 1 file changed, 13 insertions(+), 6 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index b7e1e0e5398..1219fe2be8f 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -87,6 +87,16 @@ void arch_task_cache_init(void) SLAB_PANIC | SLAB_NOTRACK, NULL); } +static inline void drop_fpu(struct task_struct *tsk) +{ + /* + * Forget coprocessor state.. + */ + tsk->fpu_counter = 0; + clear_fpu(tsk); + clear_used_math(); +} + /* * Free current thread data structures etc.. */ @@ -109,6 +119,8 @@ void exit_thread(void) put_cpu(); kfree(bp); } + + drop_fpu(me); } void show_regs(struct pt_regs *regs) @@ -149,12 +161,7 @@ void flush_thread(void) flush_ptrace_hw_breakpoint(tsk); memset(tsk->thread.tls_array, 0, sizeof(tsk->thread.tls_array)); - /* - * Forget coprocessor state.. - */ - tsk->fpu_counter = 0; - clear_fpu(tsk); - clear_used_math(); + drop_fpu(tsk); } static void hard_disable_TSC(void) -- cgit v1.2.3-70-g09d2