Thread (15 messages) 15 messages, 4 authors, 2021-07-01

Re: [PATCH] tcp: Do not reset the icsk_ca_initialized in tcp_init_transfer.

flat view

From: Neal Cardwell <ncardwell@google.com>
Date: 2021-06-28 16:24:46
Also in: bpf, linux-kernel-mentees, lkml

On Mon, Jun 28, 2021 at 12:18 PM Phi Nguyen [off-list ref] wrote:
On 6/28/2021 10:52 PM, Eric Dumazet wrote:
quoted
Unfortunately this patch might break things.

We keep changing this CC switching, with eBPF being mixed in the equation.

I would suggest you find a Fixes: tag first, so that we can continue
the discussion.

Thank you.
Thank for your feedback. I will resubmit it with a Fixes tag.

Regard.
Thanks.

Can you also please provide a summary of the event sequence that
triggers the bug? Based on your Reported-by tag, I guess this is based
on the syzbot reproducer:

 https://groups.google.com/g/syzkaller-bugs/c/VbHoSsBz0hk/m/cOxOoTgPCAAJ

but perhaps you can give a summary of the event sequence that causes
the bug? Is it that the call:

setsockopt$inet_tcp_TCP_CONGESTION(r0, 0x6, 0xd,
&(0x7f0000000000)='cdg\x00', 0x4)

initializes the CC and happens before the connection is established,
and then when the connection is established, the line that sets:
  icsk->icsk_ca_initialized = 0;
is incorrect, causing the CC to be initialized again without first
calling the cleanup code that deallocates the CDG-allocated memory?

thanks,
neal
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help