login
Header Space

 
 

Re: pthread_create() slow for many threads; also time to revisit 64b context switch optimization?

Score:
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
To: Ingo Molnar <mingo@...>
Cc: Ulrich Drepper <drepper@...>, Arjan van de Ven <arjan@...>, <akpm@...>, <hugh@...>, <linux-mm@...>, <linux-kernel@...>, <briangrant@...>, <cgd@...>, <mbligh@...>, Linus Torvalds <torvalds@...>, Thomas Gleixner <tglx@...>, H. Peter Anvin <hpa@...>
Date: Friday, August 15, 2008 - 11:54 am

Ingo Molnar wrote:


I think the flag makes sense but it's name is confusing - 64BIT for a
flag which means "maybe request 32-bit stack"!  Suggest:

+#define MAP_STACK       0x20000         /* 31bit or 64bit address for stack, */
+                                        /* whichever is faster on this CPU */

Also, is this _only_ useful for thread stacks, or are there other
memory allocations where 31-bitness affects execution speed on old P4s?

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

Messages in current thread:
Re: pthread_create() slow for many threads; also time to rev..., Jamie Lokier, (Fri Aug 15, 11:54 am)
speck-geostationary