Re: [PATCH 3/4] x86_64 ia32 syscall audit fast-path

Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
From: Linus Torvalds
Date: Monday, July 7, 2008 - 5:37 pm

On Mon, 7 Jul 2008, Roland McGrath wrote:

On many microarchitectures, yes. Especially for a memory operand. 

That said, depending on the size of the constant, 'bt' may be _smaller_ 
than testl (8-bit constant vs 32-bit one). Which can make up for it.


I haven't checked recent CPU's, it may not matter much on ones that 
support 64-bit. But bt with a memop was traditionally quite a bit more 
expensive than 'test'. 

I too am too lazy to check. Once it's in the slow-path, it doesn't much 
matter. We're talking a few cycles here.

		Linus
--
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

Messages in current thread:
[PATCH 2/4] x86_64 syscall audit fast-path, Roland McGrath, (Sun Jul 6, 11:57 pm)
[PATCH 3/4] x86_64 ia32 syscall audit fast-path, Roland McGrath, (Sun Jul 6, 11:57 pm)
[PATCH 4/4] i386 syscall audit fast-path, Roland McGrath, (Sun Jul 6, 11:58 pm)
Re: [PATCH 3/4] x86_64 ia32 syscall audit fast-path, Linus Torvalds, (Mon Jul 7, 9:34 am)
Re: [PATCH 3/4] x86_64 ia32 syscall audit fast-path, Linus Torvalds, (Mon Jul 7, 9:39 am)
Re: [PATCH 3/4] x86_64 ia32 syscall audit fast-path, Roland McGrath, (Mon Jul 7, 5:02 pm)
Re: [PATCH 3/4] x86_64 ia32 syscall audit fast-path, Linus Torvalds, (Mon Jul 7, 5:37 pm)
Re: [PATCH 3/4] x86_64 ia32 syscall audit fast-path, Ingo Molnar, (Fri Jul 18, 4:36 am)
Re: [PATCH 3/4] x86_64 ia32 syscall audit fast-path, Roland McGrath, (Mon Jul 21, 4:54 pm)
Re: [PATCH 3/4] x86_64 ia32 syscall audit fast-path, Linus Torvalds, (Wed Jul 23, 5:22 pm)
Re: [PATCH 3/4] x86_64 ia32 syscall audit fast-path, Linus Torvalds, (Wed Jul 23, 5:40 pm)
Re: [PATCH 3/4] x86_64 ia32 syscall audit fast-path, Roland McGrath, (Wed Jul 23, 6:04 pm)