Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1204735

Re: [PATCH] net: Unbreak resetting default values for tcp_wmem/udp_wmem_min

From David Miller <davem@davemloft.net>
Newsgroups linux.kernel
Subject Re: [PATCH] net: Unbreak resetting default values for tcp_wmem/udp_wmem_min
Date 2015-08-11 05:50 +0200
Message-ID <pW8JB-1BB-33@gated-at.bofh.it> (permalink)
References <pUdu2-3Ci-9@gated-at.bofh.it> <pVO89-4Pi-3@gated-at.bofh.it> <pW8zT-1qd-5@gated-at.bofh.it>
Organization linux.* mail to news gateway

Show all headers | View raw


From: Calvin Owens <calvinowens@fb.com>
Date: Mon, 10 Aug 2015 20:34:06 -0700

> I'm really questioning the limitation itself: why enforce a minimum of
> SOCK_MIN_SNDBUF here? Why not SK_MEM_QUANTUM?
> 
> Commit 8133534c760d4083 referred to b1cb59cf2efe7971, which choose to
> use the SOCK_MIN constants as the lower limits to avoid nasty bugs. But
> AFAICS, a limit of SOCK_MIN_SNDBUF isn't necessary to do that: the
> BUG_ON cited in the commit message for b1cb59cf2efe7971 seems to have
> happened because unix_stream_sendmsg() expects a minimum of a full page
> (ie SK_MEM_QUANTUM) and the math broke, not because it had less than
> SOCK_MIN_SNDBUF allocated.
> 
> Nothing seems to assume that it has at least SOCK_MIN_SNDBUF to play
> with, so my argument is that enforcing a minimum of SK_MEM_QUANTUM
> avoids the sort of bugs commit 8133534c760d4083 was trying to avoid, and
> it does so without breaking anybody's sysctl configurations. What do you
> think?

The author of said commit argues that too small values lead to really
bad performance, but I guess he should have adjusted the default if he
cared about it so much.

Ok, can you respin your patch with some added details in the commit
message like what you said above?

Thanks.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Back to linux.kernel | Previous | NextPrevious in thread | Next in thread | Find similar | Unroll thread


Thread

[PATCH] net: Unbreak resetting default values for tcp_wmem/udp_wmem_min Calvin Owens <calvinowens@fb.com> - 2015-08-05 22:30 +0200
  Re: [PATCH] net: Unbreak resetting default values for  tcp_wmem/udp_wmem_min David Miller <davem@davemloft.net> - 2015-08-10 07:50 +0200
    Re: [PATCH] net: Unbreak resetting default values for  tcp_wmem/udp_wmem_min Calvin Owens <calvinowens@fb.com> - 2015-08-11 05:40 +0200
      Re: [PATCH] net: Unbreak resetting default values for  tcp_wmem/udp_wmem_min David Miller <davem@davemloft.net> - 2015-08-11 05:50 +0200
        [PATCH] net: Use SK_MEM_QUANTUM as minimum for tcp/udp rmem/wmem Calvin Owens <calvinowens@fb.com> - 2015-08-12 07:00 +0200
          Re: [PATCH] net: Use SK_MEM_QUANTUM as minimum for tcp/udp rmem/wmem Eric Dumazet <eric.dumazet@gmail.com> - 2015-08-12 16:30 +0200
            Re: [PATCH] net: Use SK_MEM_QUANTUM as minimum for tcp/udp rmem/wmem Sorin Dumitru <sorin@returnze.ro> - 2015-08-12 19:10 +0200
              Re: [PATCH] net: Use SK_MEM_QUANTUM as minimum for tcp/udp rmem/wmem Eric Dumazet <eric.dumazet@gmail.com> - 2015-08-12 19:50 +0200
                Re: [PATCH] net: Use SK_MEM_QUANTUM as minimum for tcp/udp rmem/wmem Calvin Owens <calvinowens@fb.com> - 2015-08-13 23:10 +0200
          [PATCH] Revert "net: limit tcp/udp rmem/wmem to  SOCK_{RCV,SND}BUF_MIN" Calvin Owens <calvinowens@fb.com> - 2015-08-13 23:30 +0200
            Re: [PATCH] Revert "net: limit tcp/udp rmem/wmem to  SOCK_{RCV,SND}BUF_MIN" Eric Dumazet <eric.dumazet@gmail.com> - 2015-08-14 01:00 +0200
            Re: [PATCH] Revert "net: limit tcp/udp rmem/wmem to  SOCK_{RCV,SND}BUF_MIN" David Miller <davem@davemloft.net> - 2015-08-17 21:20 +0200

csiph-web