Mail Archive Search: <+subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*\[RFC\]\[PATCH\]\s+net\s+drivers\s+and\s+cache\s+alignment\s*$/>

Results:

References: [ +subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*\[RFC\]\[PATCH\]\s+net\s+drivers\s+and\s+cache\s+alignment\s*$/: 32 ]

Total 32 documents matching your query.

1. [RFC][PATCH] net drivers and cache alignment (score: 1): Author: xxxxxx>; Date: Sat, 07 Dec 2002 17:37:17 -0500; One of the [many] nice properties of the traditional Don Becker drivers has been that often the driver-private structures are arranged such that the structure is broken up on cacheline boundaries. Th; /archives/netdev/2002-12/msg00037.html (15,115 bytes)
2. Re: [RFC][PATCH] net drivers and cache alignment (score: 1): Author: xxxxxx>; Date: Sat, 07 Dec 2002 14:40:04 -0800 (PST); Can't the cacheline_aligned attribute be applied to individual struct members? I remember doing this for thread_struct on sparc ages ago.; /archives/netdev/2002-12/msg00038.html (8,619 bytes)
3. Re: [RFC][PATCH] net drivers and cache alignment (score: 1): Author: xxxxxx>; Date: Sat, 07 Dec 2002 17:46:32 -0500; David S. Miller wrote: Can't the cacheline_aligned attribute be applied to individual struct members? I remember doing this for thread_struct on sparc ages ago. I was hoping someone who knows gcc bet; /archives/netdev/2002-12/msg00039.html (9,378 bytes)
4. [RFC][PATCH] net drivers and cache alignment (score: 1): Author: xxxxxx>; Date: Sat, 07 Dec 2002 18:06:15 -0500; David S. Miller wrote: Can't the cacheline_aligned attribute be applied to individual struct members? I remember doing this for thread_struct on sparc ages ago. Looks like it from the 2.4 processor.h; /archives/netdev/2002-12/msg00040.html (15,662 bytes)
5. Re: [RFC][PATCH] net drivers and cache alignment (score: 1): Author: xxxxxx>; Date: Sat, 07 Dec 2002 15:29:16 -0800; It needs padding _only_ on SMP. ____cacheline_aligned_in_smp. struct foo { int a; int b __attribute__((__aligned__(1024))); int c; } foo; main() { printf("%d\n", sizeof(struct foo)); printf("%d\n", o; /archives/netdev/2002-12/msg00041.html (10,191 bytes)
6. Re: [RFC][PATCH] net drivers and cache alignment (score: 1): Author: xxxxxx>; Date: Sat, 07 Dec 2002 15:30:45 -0800 (PST); It needs padding _only_ on SMP. ____cacheline_aligned_in_smp. non-smp machines lack L2 caches? That's new to me :-) More seriously, there are real benefits on non-SMP systems.; /archives/netdev/2002-12/msg00042.html (9,765 bytes)
7. Re: [RFC][PATCH] net drivers and cache alignment (score: 1): Author: xxxxxx>; Date: Sat, 07 Dec 2002 18:36:48 -0500; Andrew Morton wrote: It needs padding _only_ on SMP. ____cacheline_aligned_in_smp. [...] So your patch will do what you want it to do. You should just tag the first member of a group with ____cacheli; /archives/netdev/2002-12/msg00043.html (10,699 bytes)
8. Re: [RFC][PATCH] net drivers and cache alignment (score: 1): Author: xxxxxx>; Date: Sun, 8 Dec 2002 00:37:46 +0100; What do you all think about this: include <stdio.h> struct S { __cacheline_start int x; __cacheline_start int y; int z; }; int main() { struct S s; printf("%d\n",sizeof(struct S)); printf("%d\n",offs; /archives/netdev/2002-12/msg00044.html (11,252 bytes)
9. Re: [RFC][PATCH] net drivers and cache alignment (score: 1): Author: xxxxxx>; Date: Sat, 07 Dec 2002 15:42:00 -0800; Then I am most confused. None of these fields will be put under busmastering or anything like that, so what advantage is there in spreading them out?; /archives/netdev/2002-12/msg00045.html (9,784 bytes)
10. Re: [RFC][PATCH] net drivers and cache alignment (score: 1): Author: xxxxxx>; Date: Sat, 07 Dec 2002 18:42:57 -0500; J.A. Magallon wrote: On 2002.12.08 Andrew Morton wrote: Jeff Garzik wrote: David S. Miller wrote: Can't the cacheline_aligned attribute be applied to individual struct members? I remember doing this; /archives/netdev/2002-12/msg00046.html (11,308 bytes)
11. Re: [RFC][PATCH] net drivers and cache alignment (score: 1): Author: xxxxxx>; Date: Sat, 07 Dec 2002 15:45:21 -0800; That will generate a warning on faster^Wolder versions of gcc. mnm:/home/akpm> gcc t2.c t2.c:11: warning: unnamed struct/union that defines no instances t2.c:15: warning: unnamed struct/union that de; /archives/netdev/2002-12/msg00047.html (9,666 bytes)
12. Re: [RFC][PATCH] net drivers and cache alignment (score: 1): Author: xxxxxx>; Date: Sat, 07 Dec 2002 15:51:36 -0800; Oh I see what you want - to be able to pick up all the operating fields in a single fetch. That will increase the overall cache footprint though. I wonder if it's really a net win, over just keeping; /archives/netdev/2002-12/msg00048.html (10,166 bytes)
13. Re: [RFC][PATCH] net drivers and cache alignment (score: 1): Author: xxxxxx>; Date: Sun, 8 Dec 2002 00:52:55 +0100; And how 'bout this (do not have any gcc oldie available to test): It passes gcc-3.2 -Wall... I think it's nicer to insert __c_s than to go field by field marking them... -- J.A. Magallon <jamagallon@; /archives/netdev/2002-12/msg00049.html (10,734 bytes)
14. Re: [RFC][PATCH] net drivers and cache alignment (score: 1): Author: xxxxxx>; Date: Sat, 7 Dec 2002 20:14:03 -0500; The structure contains an __aligned__(1024) item. Think about an array of 'struct foo' items. They have to be 2048 bytes or you won't align correctly. C allows for empty space in structure padding, b; /archives/netdev/2002-12/msg00050.html (11,644 bytes)
15. Re: [RFC][PATCH] net drivers and cache alignment (score: 1): Author: xxxxxx>; Date: Sun, 08 Dec 2002 12:00:44 -0800 (PST); Then I am most confused. None of these fields will be put under busmastering or anything like that, so what advantage is there in spreading them out? When you are in the "tx path" you'll take one L2; /archives/netdev/2002-12/msg00051.html (10,196 bytes)
16. Re: [RFC][PATCH] net drivers and cache alignment (score: 1): Author: xxxxxx>; Date: 15 Dec 2002 13:31:15 -0500; There's another case where it matters, I guess one could look at it as similar to the SMP case, but between CPU and device. Some devices have producer indices in host memory which they update whenev; /archives/netdev/2002-12/msg00091.html (10,447 bytes)
17. [RFC][PATCH] net drivers and cache alignment (score: 1): Author: Jeff Garzik <jgarzik@xxxxxxxxx>; Date: Sat, 07 Dec 2002 17:37:17 -0500; One of the [many] nice properties of the traditional Don Becker drivers has been that often the driver-private structures are arranged such that the structure is broken up on cacheline boundaries. Th; /archives/netdev/2002-12/msg00182.html (15,264 bytes)
18. Re: [RFC][PATCH] net drivers and cache alignment (score: 1): Author: "David S. Miller" <davem@xxxxxxxxxx>; Date: Sat, 07 Dec 2002 14:40:04 -0800 (PST); Can't the cacheline_aligned attribute be applied to individual struct members? I remember doing this for thread_struct on sparc ages ago.; /archives/netdev/2002-12/msg00183.html (8,667 bytes)
19. Re: [RFC][PATCH] net drivers and cache alignment (score: 1): Author: Jeff Garzik <jgarzik@xxxxxxxxx>; Date: Sat, 07 Dec 2002 17:46:32 -0500; I was hoping someone who knows gcc better than me knew that, and would speak up ;-); /archives/netdev/2002-12/msg00184.html (9,492 bytes)
20. [RFC][PATCH] net drivers and cache alignment (score: 1): Author: Jeff Garzik <jgarzik@xxxxxxxxx>; Date: Sat, 07 Dec 2002 18:06:15 -0500; Looks like it from the 2.4 processor.h code. Attached is cut #2. Thanks for all the near-instant feedback so far :) Andrew, does the attached still need padding on SMP? == drivers/net/tg3.c 1.41 vs e; /archives/netdev/2002-12/msg00185.html (15,755 bytes)

Current List: 1 - 20
Page: [1] [2]