fix i386 mutex fastpath on FRAME_POINTER && !DEBUG_MUTEXES
In-Reply-To: <[email protected]>
Quoted 12 lines · Ingo Molnar
Quoted 10 lines · Chuck Ebbert
LOCK " decl (%%eax) \n" \ " jns 1f \n" \ " call "#fail_fn" \n" \ "1: \n" \ \ :"=a" (dummy) \ : "a" (count) \
Quoted 4 lines
Quoted 2 lines
Will the extra taken forward conditional jump in the fastpath cause much of a slowdown?
yeah - the fastpath is much more common than the slowpath.
But that's how the spinlock code does it. Should that be changed to put the spinloops in .text.lock and make the fastpaths a fall-through?
Signature · 2 lines
Chuck Currently reading: _Olympos_ by Dan Simmons