From f1853505d9ca1c3ea27c29cf83c24661531c527b Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Tue, 2 Sep 2014 19:57:23 +0200 Subject: [PATCH] x86, fpu: copy_process: Avoid fpu_alloc/copy if !used_math() arch_dup_task_struct() copies thread.fpu if fpu_allocated(), this looks suboptimal and misleading. Say, a forking process could use FPU only once in a signal handler but now tsk_used_math(src) == F, in this case the child gets a copy of fpu->state for no reason. The child won't use the saved registers anyway even if it starts to use FPU, this can only avoid fpu_alloc() in do_device_not_available(). Change this code to check tsk_used_math(current) instead. We still need to clear fpu->has_fpu/state, we could do this memset(0) under fpu_allocated() check but I think this doesn't make sense. See also the next change. use_eager_fpu() assumes that fpu_allocated() is always true, but a forking task (and thus its child) must always have PF_USED_MATH set, otherwise the child can either use FPU without used_math() (note that switch_fpu_prepare() doesn't do stts() in this case), or it will be killed by do_device_not_available()->BUG_ON(use_eager_fpu). Signed-off-by: Oleg Nesterov Link: http://lkml.kernel.org/r/20140902175723.GA21659@redhat.com Reviewed-by: Suresh Siddha Signed-off-by: H. Peter Anvin --- arch/x86/kernel/process.c | 13 ++++++------- 1 file changed, 6 insertions(+), 7 deletions(-) diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index f804dc935d2a..b9ba9d52020e 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -64,14 +64,13 @@ EXPORT_SYMBOL_GPL(task_xstate_cachep); */ int arch_dup_task_struct(struct task_struct *dst, struct task_struct *src) { - int ret; - *dst = *src; - if (fpu_allocated(&src->thread.fpu)) { - memset(&dst->thread.fpu, 0, sizeof(dst->thread.fpu)); - ret = fpu_alloc(&dst->thread.fpu); - if (ret) - return ret; + + memset(&dst->thread.fpu, 0, sizeof(dst->thread.fpu)); + if (tsk_used_math(src)) { + int err = fpu_alloc(&dst->thread.fpu); + if (err) + return err; fpu_copy(dst, src); } return 0; -- 2.39.2