Mail Archive Search: <+subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*MSEC_TO

Results:

References: [ +subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*MSEC_TO_JIFFIES\s+is\s+messed\s+up\.\.\.\s*$/: 92 ]

Total 92 documents matching your query.

1. Re: MSEC_TO_JIFFIES is messed up... (score: 1): Author: Valdis.Kletnieks@xxxxxx; Date: Wed, 12 May 2004 15:47:18 -0400; If the kernel jiffie is anything other than exactly 1 msec, you're screwed... [/usr/src/linux-2.6.6-mm1]2 find include -name '*.h' | xargs egrep '#define HZ ' include/asm-ppc64/param.h:#define HZ 100; /archives/netdev/2004-05/msg00079.html (11,518 bytes)
2. Re: MSEC_TO_JIFFIES is messed up... (score: 1): Author: Davide Libenzi <davidel@xxxxxxxxxxxxxxx>; Date: Wed, 12 May 2004 12:49:38 -0700 (PDT); For HZ == 1000 it's fine, even if it'd better to explicitly make it HZ dependent and let the compiler to discard them. - Davide; /archives/netdev/2004-05/msg00080.html (10,814 bytes)
3. Re: MSEC_TO_JIFFIES is messed up... (score: 1): Author: Davide Libenzi <davidel@xxxxxxxxxxxxxxx>; Date: Wed, 12 May 2004 12:56:04 -0700 (PDT); I believe they were talking about include/asm-i386/param.h ^^^^^^^^ - Davide; /archives/netdev/2004-05/msg00081.html (10,741 bytes)
4. Re: MSEC_TO_JIFFIES is messed up... (score: 1): Author: Valdis.Kletnieks@xxxxxx; Date: Wed, 12 May 2004 16:07:16 -0400; .. True. The problem is that even for the i386 family, there's no inherent reason why the value of HZ is nailed to 1000 - it was changed from the default 100 in the 2.4 kernel for reasons that apply; /archives/netdev/2004-05/msg00083.html (11,282 bytes)
5. Re: MSEC_TO_JIFFIES is messed up... (score: 1): Author: Valdis.Kletnieks@xxxxxx; Date: Wed, 12 May 2004 16:18:17 -0400; That should read we "know" that HZ is 1000.... How about this instead? That will DTRT if somebody changes HZ for their build, and still allow us to avoud the *1000/1000 conversion for most builds....; /archives/netdev/2004-05/msg00084.html (11,318 bytes)
6. Re: MSEC_TO_JIFFIES is messed up... (score: 1): Author: Andrew Morton <akpm@xxxxxxxx>; Date: Wed, 12 May 2004 13:20:50 -0700; Yes, that's a correct optimisation. This is simply a namespace clash. How about we do: in some kernel-wide header then kill off all the private implementations?; /archives/netdev/2004-05/msg00085.html (11,552 bytes)
7. Re: MSEC_TO_JIFFIES is messed up... (score: 1): Author: Greg KH <greg@xxxxxxxxx>; Date: Wed, 12 May 2004 13:32:51 -0700; Looks good to me. thanks, greg k-h; /archives/netdev/2004-05/msg00086.html (11,086 bytes)
8. Re: MSEC_TO_JIFFIES is messed up... (score: 1): Author: Andrew Morton <akpm@xxxxxxxx>; Date: Wed, 12 May 2004 13:35:20 -0700; OK, I'll do it.; /archives/netdev/2004-05/msg00087.html (11,226 bytes)
9. Re: MSEC_TO_JIFFIES is messed up... (score: 1): Author: William Lee Irwin III <wli@xxxxxxxxxxxxxx>; Date: Wed, 12 May 2004 13:38:29 -0700; How about this? if HZ <= 1000 && !(1000 % HZ) -- wli; /archives/netdev/2004-05/msg00088.html (12,258 bytes)
10. RE: MSEC_TO_JIFFIES is messed up... (score: 1): Author: Jan Olderdissen <jan@xxxxxxxxxxx>; Date: Wed, 12 May 2004 13:40:18 -0700; Couple nitpicks: Jan; /archives/netdev/2004-05/msg00089.html (8,638 bytes)
11. Re: MSEC_TO_JIFFIES is messed up... (score: 1): Author: Ingo Molnar <mingo@xxxxxxx>; Date: Wed, 12 May 2004 22:35:00 +0200; new patch attached - this adopts the overflow-safe variant from sctp.h, removes it from sctp.h and moves it into a generic include file and also does the HZ=1000 simplification. Ingo Attachment: hz-c; /archives/netdev/2004-05/msg00090.html (11,206 bytes)
12. Re: MSEC_TO_JIFFIES is messed up... (score: 1): Author: Andrew Morton <akpm@xxxxxxxx>; Date: Wed, 12 May 2004 13:47:18 -0700; You promise it's correct and generates good code? -- 25-akpm/drivers/net/irda/act200l-sir.c | 2 +- 25-akpm/drivers/net/irda/act200l.c | 10 +++++-- 25-akpm/drivers/net/irda/actisys.c | 2 +- 25-akpm/dr; /archives/netdev/2004-05/msg00091.html (48,485 bytes)
13. Re: MSEC_TO_JIFFIES is messed up... (score: 1): Author: "Bill Rugolsky Jr." <brugolsky@xxxxxxxxxxxxxxxxxxxxxxxxx>; Date: Wed, 12 May 2004 16:54:07 -0400; Yes, that was Sridhar Samudrala <sri@xxxxxxxxxx>. I've done some work to create an accurate implementation of msecs_to_jiffies() and jiffies_to_msecs(), which I was going to submit later this week a; /archives/netdev/2004-05/msg00092.html (14,000 bytes)
14. Re: MSEC_TO_JIFFIES is messed up... (score: 1): Author: Ingo Molnar <mingo@xxxxxxx>; Date: Wed, 12 May 2004 22:28:07 +0200; The code is 100% correct, but not 100% clean, and definitely not documented. The reason for the change was to avoid unnecessary integer multiplications and divisions in sched.c. the attached patch (a; /archives/netdev/2004-05/msg00093.html (11,099 bytes)
15. Re: MSEC_TO_JIFFIES is messed up... (score: 1): Author: Ingo Molnar <mingo@xxxxxxx>; Date: Wed, 12 May 2004 22:50:28 +0200; yet another patch - this time it's: complete, covers irda, accelerates HZ=100, unifies the slightly differing namespaces and compiles/boots as well. Ingo Attachment: hz-cleanup-2.6.6-A2 Description:; /archives/netdev/2004-05/msg00094.html (11,499 bytes)
16. Re: MSEC_TO_JIFFIES is messed up... (score: 1): Author: Ingo Molnar <mingo@xxxxxxx>; Date: Wed, 12 May 2004 21:33:49 +0200; why is it wrong? Ingo; /archives/netdev/2004-05/msg00095.html (9,469 bytes)
17. Re: MSEC_TO_JIFFIES is messed up... (score: 1): Author: William Lee Irwin III <wli@xxxxxxxxxxxxxx>; Date: Wed, 12 May 2004 13:59:15 -0700; Yes. Changelog: Optimize the cases where HZ is a divisor of 1000 or vice-versa in JIFFIES_TO_MSEC() and MSEC_TO_JIFFIES() by allowing the nonvanishing(!) integral ratios to appear as a parenthesized; /archives/netdev/2004-05/msg00096.html (12,482 bytes)
18. Re: MSEC_TO_JIFFIES is messed up... (score: 1): Author: Davide Libenzi <davidel@xxxxxxxxxxxxxxx>; Date: Wed, 12 May 2004 14:01:58 -0700 (PDT); $ cat foo.c int foo(int i) { return i * 1000 / 1000; } $ gcc -S -c foo.c $ cat foo.s .file "foo.c" .text .globl foo .type foo, @function foo: pushl %ebp movl %esp, %ebp movl 8(%ebp), %eax leave ret .; /archives/netdev/2004-05/msg00097.html (11,685 bytes)
19. Re: MSEC_TO_JIFFIES is messed up... (score: 1): Author: Valdis.Kletnieks@xxxxxx; Date: Wed, 12 May 2004 17:03:10 -0400; 1) - 1) Looks good to me.. :) Attachment: pgptvREZ6XExz.pgp Description: PGP signature; /archives/netdev/2004-05/msg00098.html (12,113 bytes)
20. Re: MSEC_TO_JIFFIES is messed up... (score: 1): Author: Sridhar Samudrala <sri@xxxxxxxxxx>; Date: Wed, 12 May 2004 14:03:24 -0700 (PDT); I have submitted a patch just to do this some time back on netdev mailing list. You can find it at http://marc.theaimsgroup.com/?l=linux-netdev&m=108024598716537&w=2 static inline unsigned long msecs; /archives/netdev/2004-05/msg00099.html (13,141 bytes)

Current List: 1 - 20
Page: [1] [2] [3] [4] [5]