Mail Archive Search: <+subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*\[timers\]\s+net\/ipv4\s*$/>

Results:

References: [ +subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*\[timers\]\s+net\/ipv4\s*$/: 12 ]

Total 12 documents matching your query.

1. [timers] net/ipv4 (score: 1): Author: Andrew Morton <andrewm@xxxxxxxxxx>; Date: Tue, 30 May 2000 22:14:42 +1000; You're the first cab off the rank :) Here are the results of reviewing net/ipv4/* for timer deletion safety. I believe there are some races in there. I have marked these with the string REVIEWME. Ad; /archives/netdev/2000-05/msg00140.html (16,039 bytes)
2. Re: [timers] net/ipv4 (score: 1): Author: kuznet@xxxxxxxxxxxxx; Date: Tue, 30 May 2000 20:59:28 +0400 (MSK DST); [ Andi, please, look at the end ] ipmr_expire_timer is static stateless timer. There are no problems with it. Even not counting that del_timer_sync is deadlocky, it should not be ever used for stati; /archives/netdev/2000-05/msg00142.html (11,728 bytes)
3. Re: [timers] net/ipv4 (score: 1): Author: Andrew Morton <andrewm@xxxxxxxxxx>; Date: Wed, 31 May 2000 02:28:39 +0000; Alexey, I'm obviously missing something. If you want to say "Andrew, TCP is safe" then that's cool. I'll just buzz off and stare unhappily at the SCSI code. But I just need one more shot at it. :) st; /archives/netdev/2000-05/msg00149.html (8,554 bytes)
4. Re: [timers] net/ipv4 (score: 1): Author: "David S. Miller" <davem@xxxxxxxxxx>; Date: Tue, 30 May 2000 21:47:03 -0700; Bear with me Alexey, I'm Australian. Where are the refcounts held? For TCP, they are held in the socket. See all the sock_get/sock_put calls around adding/removing the TCP retransmit timer. Only the; /archives/netdev/2000-05/msg00150.html (8,148 bytes)
5. Re: [timers] net/ipv4 (score: 1): Author: Andrew Morton <andrewm@xxxxxxxxxx>; Date: Thu, 01 Jun 2000 01:07:23 +1000; I've revisited net/ipv4/* and I agree it's safe. I was assuming that the network rx path was called from bh_action() and hence that timers are serialised wrt network rx, But not process context. I se; /archives/netdev/2000-05/msg00153.html (7,467 bytes)
6. Re: [timers] net/ipv4 (score: 1): Author: kuznet@xxxxxxxxxxxxx; Date: Wed, 31 May 2000 21:34:22 +0400 (MSK DST); It is supposed to be safe. Please, audit all, but do not spin too long on one place. 8) You have already found bugs in ipv{4,6} defragmenters, big dusty hole in net/sched and bug in igmp (see below); /archives/netdev/2000-05/msg00156.html (9,572 bytes)
7. [timers] net/ipv4 (score: 1): Author: Andrew Morton <andrewm@xxxxxxxxxx>; Date: Tue, 30 May 2000 22:14:42 +1000; Hi, Alexey. You're the first cab off the rank :) Here are the results of reviewing net/ipv4/* for timer deletion safety. I believe there are some races in there. I have marked these with the string R; /archives/netdev/2000-05/msg00311.html (16,039 bytes)
8. Re: [timers] net/ipv4 (score: 1): Author: kuznet@xxxxxxxxxxxxx; Date: Tue, 30 May 2000 20:59:28 +0400 (MSK DST); Hello! [ Andi, please, look at the end ] ipmr_expire_timer is static stateless timer. There are no problems with it. Even not counting that del_timer_sync is deadlocky, it should not be ever used for; /archives/netdev/2000-05/msg00313.html (11,753 bytes)
9. Re: [timers] net/ipv4 (score: 1): Author: Andrew Morton <andrewm@xxxxxxxxxx>; Date: Wed, 31 May 2000 02:28:39 +0000; Alexey, I'm obviously missing something. If you want to say "Andrew, TCP is safe" then that's cool. I'll just buzz off and stare unhappily at the SCSI code. But I just need one more shot at it. :) st; /archives/netdev/2000-05/msg00320.html (8,608 bytes)
10. Re: [timers] net/ipv4 (score: 1): Author: "David S. Miller" <davem@xxxxxxxxxx>; Date: Tue, 30 May 2000 21:47:03 -0700; Bear with me Alexey, I'm Australian. Where are the refcounts held? For TCP, they are held in the socket. See all the sock_get/sock_put calls around adding/removing the TCP retransmit timer. Only the; /archives/netdev/2000-05/msg00321.html (8,252 bytes)
11. Re: [timers] net/ipv4 (score: 1): Author: Andrew Morton <andrewm@xxxxxxxxxx>; Date: Thu, 01 Jun 2000 01:07:23 +1000; I've revisited net/ipv4/* and I agree it's safe. I was assuming that the network rx path was called from bh_action() and hence that timers are serialised wrt network rx, But not process context. I se; /archives/netdev/2000-05/msg00324.html (7,521 bytes)
12. Re: [timers] net/ipv4 (score: 1): Author: kuznet@xxxxxxxxxxxxx; Date: Wed, 31 May 2000 21:34:22 +0400 (MSK DST); Hello! It is supposed to be safe. Please, audit all, but do not spin too long on one place. 8) You have already found bugs in ipv{4,6} defragmenters, big dusty hole in net/sched and bug in igmp (see; /archives/netdev/2000-05/msg00327.html (9,597 bytes)

Current List: 1 - 12
Page: [1]