On Thu, 22 Dec 2005, Eric Dumazet wrote: > Ingo Molnar a écrit : > > > > CLI/STI is extremely fast. (In fact in the -rt tree i'm using them within > > mutexes instead of preempt_enable()/preempt_disable(), because they are > > faster and generate less register side-effect.) > > > > Yes, but most of my machines have a ! CONFIG_PREEMPT kernel, so > preempt_enable()/preempt_disable() are empty, thus faster than CLI/STI for > sure :) > > Then, maybe the patch that moves 'current' in a dedicated x86_64 register may > help to lower the cost of preempt_enable()/preempt_disable() on a > CONFIG_PREEMPT kernel ? I'm not sure if it'll make much of a difference over; mov %gs:offset,%reg So 'current' already is fairly fast on x86_64.
- References:
- Re: 2.6.15-rc5-rt2 slowness
- From: Steven Rostedt <[email protected]>
- [PATCH RT 00/02] SLOB optimizations
- From: Steven Rostedt <[email protected]>
- Re: [PATCH RT 00/02] SLOB optimizations
- From: Matt Mackall <[email protected]>
- Re: [PATCH RT 00/02] SLOB optimizations
- From: Steven Rostedt <[email protected]>
- Re: [PATCH RT 00/02] SLOB optimizations
- From: Pekka Enberg <[email protected]>
- Re: [PATCH RT 00/02] SLOB optimizations
- From: Steven Rostedt <[email protected]>
- Re: [PATCH RT 00/02] SLOB optimizations
- From: Ingo Molnar <[email protected]>
- Re: [PATCH RT 00/02] SLOB optimizations
- From: Eric Dumazet <[email protected]>
- Re: [PATCH RT 00/02] SLOB optimizations
- From: Ingo Molnar <[email protected]>
- Re: [PATCH RT 00/02] SLOB optimizations
- From: Eric Dumazet <[email protected]>
- Re: [PATCH RT 00/02] SLOB optimizations
- From: Ingo Molnar <[email protected]>
- Re: [PATCH RT 00/02] SLOB optimizations
- From: Eric Dumazet <[email protected]>
- Re: 2.6.15-rc5-rt2 slowness
- Prev by Date: RE: [PATCH/RFT] tlclk: convert to the new platform device interface
- Next by Date: Re: [PATCH v2:3/3]Export cpu topology by sysfs
- Previous by thread: Re: [PATCH RT 00/02] SLOB optimizations
- Next by thread: Re: [PATCH RT 00/02] SLOB optimizations
- Index(es):
 |