Mail Archive Search: <+subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*\[patch\]\s+do\s+not\s+readlock\s+all\s+buckets\s+in\s+\/proc\/net\/tcp\s*$/>

Results:

References: [ +subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*\[patch\]\s+do\s+not\s+readlock\s+all\s+buckets\s+in\s+\/proc\/net\/tcp\s*$/: 14 ]

Total 14 documents matching your query.

1. [patch] do not readlock all buckets in /proc/net/tcp (score: 1): Author: ert@xxxxxxxxxxxxxxxxxxx>; Date: Mon, 5 Jul 2004 13:09:49 +0200; This patch makes the files /proc/net/tcp and /proc/net/tcp6 not acquire the readlock for every bucket. On ppc64 and ia64 the readlocks are so expensive, that reading /proc/net/tcp takes 0.25 seconds; /archives/netdev/2004-07/msg00088.html (8,259 bytes)
2. Re: [patch] do not readlock all buckets in /proc/net/tcp (score: 1): Author: Hellwig <hch@xxxxxx>; Date: Mon, 05 Jul 2004 21:27:54 +1000; Why not use NETLINK+TCP_DIAG instead? It's much faster. -- Visit Openswan at http://www.openswan.org/ Email: Herbert Xu ~{PmV>HI~} <herbert@xxxxxxxxxxxxxxxxxxx> Home Page: http://gondor.apana.org.au/; /archives/netdev/2004-07/msg00089.html (9,082 bytes)
3. Re: [patch] do not readlock all buckets in /proc/net/tcp (score: 1): Author: ssner@xxxxxxx>; Date: Mon, 5 Jul 2004 22:06:10 +1000; Then you wouldn't mind adding this optimisation for tcp_diag as well, right? :) Cheers, -- Visit Openswan at http://www.openswan.org/ Email: Herbert Xu ~{PmV>HI~} <herbert@xxxxxxxxxxxxxxxxxxx> Home P; /archives/netdev/2004-07/msg00090.html (9,145 bytes)
4. Re: [patch] do not readlock all buckets in /proc/net/tcp (score: 1): Author: xxxxxxxxxxxxx>; Date: Mon, 5 Jul 2004 13:35:55 +0200; Not sure if you want / can fix all proprietary software. Oh, and NETLINK+TCP_DIAG seems to have the same readlock contention problem, see tcpdiag_dump(). Ciao, Marcus; /archives/netdev/2004-07/msg00091.html (9,146 bytes)
5. Re: [patch] do not readlock all buckets in /proc/net/tcp (score: 1): Author: xxxxxxxxxxxxxxxxxxx>; Date: Mon, 05 Jul 2004 21:25:22 +0900 (JST); here it is. :-) Signed-off-by: Hideaki YOSHIFUJI <yoshfuji@xxxxxxxxxxxxxx> == net/ipv4/tcp_diag.c 1.15 vs edited == -- 1.15/net/ipv4/tcp_diag.c 2004-06-08 07:27:58 +09:00 +++ edited/net/ipv4/tcp_diag; /archives/netdev/2004-07/msg00092.html (10,215 bytes)
6. Re: [patch] do not readlock all buckets in /proc/net/tcp (score: 1): Author: xxx>; Date: Mon, 05 Jul 2004 22:25:13 +0900 (JST); Oops... Signed-off-by: Hideaki YOSHIFUJI <yoshfuji@xxxxxxxxxxxxxx> == net/ipv4/tcp_diag.c 1.15 vs edited == -- 1.15/net/ipv4/tcp_diag.c 2004-06-08 07:27:58 +09:00 +++ edited/net/ipv4/tcp_diag.c 2004-; /archives/netdev/2004-07/msg00093.html (9,938 bytes)
7. Re: [patch] do not readlock all buckets in /proc/net/tcp (score: 1): Author: xxxxxxxxxxx>; Date: Mon, 5 Jul 2004 14:45:11 +0200; On Mon, Jul 05, 2004 at 09:25:22PM +0900, YOSHIFUJI Hideaki / موون The second hlist_empty is bad, you should be checking &tcp_ehash[i + tcp_ehash_size].chain ((head+tcp_ehash_size) I think). Ciao, Ma; /archives/netdev/2004-07/msg00094.html (10,380 bytes)
8. [patch] do not readlock all buckets in /proc/net/tcp (score: 1): Author: Marcus Meissner <meissner@xxxxxxx>; Date: Mon, 5 Jul 2004 13:09:49 +0200; Hi, This patch makes the files /proc/net/tcp and /proc/net/tcp6 not acquire the readlock for every bucket. On ppc64 and ia64 the readlocks are so expensive, that reading /proc/net/tcp takes 0.25 seco; /archives/netdev/2004-07/msg00904.html (8,259 bytes)
9. Re: [patch] do not readlock all buckets in /proc/net/tcp (score: 1): Author: Herbert Xu <herbert@xxxxxxxxxxxxxxxxxxx>; Date: Mon, 05 Jul 2004 21:27:54 +1000; Why not use NETLINK+TCP_DIAG instead? It's much faster. -- Visit Openswan at http://www.openswan.org/ Email: Herbert Xu ~{PmV>HI~} <herbert@xxxxxxxxxxxxxxxxxxx> Home Page: http://gondor.apana.org.au/; /archives/netdev/2004-07/msg00905.html (9,156 bytes)
10. Re: [patch] do not readlock all buckets in /proc/net/tcp (score: 1): Author: Herbert Xu <herbert@xxxxxxxxxxxxxxxxxxx>; Date: Mon, 5 Jul 2004 22:06:10 +1000; Then you wouldn't mind adding this optimisation for tcp_diag as well, right? :) Cheers, -- Visit Openswan at http://www.openswan.org/ Email: Herbert Xu ~{PmV>HI~} <herbert@xxxxxxxxxxxxxxxxxxx> Home P; /archives/netdev/2004-07/msg00906.html (9,304 bytes)
11. Re: [patch] do not readlock all buckets in /proc/net/tcp (score: 1): Author: Marcus Meissner <meissner@xxxxxxx>; Date: Mon, 5 Jul 2004 13:35:55 +0200; Not sure if you want / can fix all proprietary software. Oh, and NETLINK+TCP_DIAG seems to have the same readlock contention problem, see tcpdiag_dump(). Ciao, Marcus; /archives/netdev/2004-07/msg00907.html (9,225 bytes)
12. Re: [patch] do not readlock all buckets in /proc/net/tcp (score: 1): Author: YOSHIFUJI Hideaki / <yoshfuji@xxxxxxxxxxxxxx>; Date: Mon, 05 Jul 2004 21:25:22 +0900 (JST); here it is. :-) Signed-off-by: Hideaki YOSHIFUJI <yoshfuji@xxxxxxxxxxxxxx> == net/ipv4/tcp_diag.c 1.15 vs edited == -- 1.15/net/ipv4/tcp_diag.c 2004-06-08 07:27:58 +09:00 +++ edited/net/ipv4/tcp_diag; /archives/netdev/2004-07/msg00908.html (10,328 bytes)
13. Re: [patch] do not readlock all buckets in /proc/net/tcp (score: 1): Author: YOSHIFUJI Hideaki / <yoshfuji@xxxxxxxxxxxxxx>; Date: Mon, 05 Jul 2004 22:25:13 +0900 (JST); Oops... Signed-off-by: Hideaki YOSHIFUJI <yoshfuji@xxxxxxxxxxxxxx> == net/ipv4/tcp_diag.c 1.15 vs edited == -- 1.15/net/ipv4/tcp_diag.c 2004-06-08 07:27:58 +09:00 +++ edited/net/ipv4/tcp_diag.c 2004-; /archives/netdev/2004-07/msg00909.html (10,069 bytes)
14. Re: [patch] do not readlock all buckets in /proc/net/tcp (score: 1): Author: Marcus Meissner <meissner@xxxxxxx>; Date: Mon, 5 Jul 2004 14:45:11 +0200; The second hlist_empty is bad, you should be checking &tcp_ehash[i + tcp_ehash_size].chain ((head+tcp_ehash_size) I think). Ciao, Marcus; /archives/netdev/2004-07/msg00910.html (10,563 bytes)

Current List: 1 - 14
Page: [1]