Gitweb:     
http://git.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=c41bf8fa5e777b6a8a19cf2484937a7167eac77f
Commit:     c41bf8fa5e777b6a8a19cf2484937a7167eac77f
Parent:     02b64dab5675bc08048c7f70cbb0d8a417d20dbe
Author:     Jan Kiszka <[EMAIL PROTECTED]>
AuthorDate: Wed May 2 19:27:21 2007 +0200
Committer:  Andi Kleen <[EMAIL PROTECTED]>
CommitDate: Wed May 2 19:27:21 2007 +0200

    [PATCH] i386: avoid redundant preempt_disable in __unlazy_fpu
    
    There are two callers of __unlazy_fpu, unlazy_fpu and __switch_to, and
    none of them appear to require additional preempt_disable/enable here.
    Let's open-code save_init_fpu in __unlazy_fpu to save a few ops.
    
    Signed-off-by: Jan Kiszka <[EMAIL PROTECTED]>
    Signed-off-by: Andi Kleen <[EMAIL PROTECTED]>
---
 include/asm-i386/i387.h |   11 ++++++-----
 1 files changed, 6 insertions(+), 5 deletions(-)

diff --git a/include/asm-i386/i387.h b/include/asm-i386/i387.h
index 49dc8e1..cdd1e24 100644
--- a/include/asm-i386/i387.h
+++ b/include/asm-i386/i387.h
@@ -74,11 +74,12 @@ static inline void __save_init_fpu( struct task_struct *tsk 
)
        task_thread_info(tsk)->status &= ~TS_USEDFPU;
 }
 
-#define __unlazy_fpu( tsk ) do { \
-       if (task_thread_info(tsk)->status & TS_USEDFPU) \
-               save_init_fpu( tsk );                   \
-       else                                            \
-               tsk->fpu_counter = 0;                   \
+#define __unlazy_fpu( tsk ) do {                               \
+       if (task_thread_info(tsk)->status & TS_USEDFPU) {       \
+               __save_init_fpu(tsk);                           \
+               stts();                                         \
+       } else                                                  \
+               tsk->fpu_counter = 0;                           \
 } while (0)
 
 #define __clear_fpu( tsk )                                     \
-
To unsubscribe from this list: send the line "unsubscribe git-commits-head" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Reply via email to