Re: TCP listenall

From: Matus Harvan <mharvan@xxxxxxxxxxx>
Date: Wed, 31 Oct 2007 01:53:06 +0100

On Sat, Oct 27, 2007 at 04:55:15AM +0100, Bruce M. Simpson wrote:

listenallr is static and is going to get trashed by concurrent threads,
unless there is a serialization with a lock, which I don't see.

I have used sys/netinet/ip_icmp.c:badport_bandlim() for
inspiration. There it did not seem to be protected by any locks
either. Would this be a bug in the ICMP code as well?

Matus

Attachment: pgpxQhCsEjlgG.pgp
Description: PGP signature

References:
- Re: TCP listenall
  - From: Matus Harvan
- Re: TCP listenall
  - From: Bruce M. Simpson

Prev by Date: Re: icmp echo_user
Next by Date: Re: UDP catchall
Previous by thread: Re: TCP listenall
Next by thread: Re: TCP listenall
Index(es):
- Date
- Thread

Relevant Pages

Re: Locking problem
... why they just remove the serialization ... This is a big argument to concurrency
(just ... If I create an index exactly to be a covering index to a ... Here's the
lock compatability matrix from ... (microsoft.public.sqlserver.programming)
Re: [PATCH] fix get_jiffies_64 to work on voyager
... I don't like very much this kind of construction, ... Since the "rmb" is as
expensive as the sequence lock, ... So converting that serialization to a "lockless"
algorithm that still ... (Linux-Kernel)
Re: page table lock patch V15 [0/7]: overview
... There's a nice standard way of doing that, namely sequence numbers. ... There
are two real costs in getting a lock: serialization and cache ... send the line "unsubscribe
linux-kernel" in ... (Linux-Kernel)
Re: [PATCH 1/5]: ufs: missed brelse and wrong baseblk
... they will be consequent because of synchronization. ... Now, which lock
would, in your opinion, provide serialization between these ... (Linux-Kernel)