lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <ZqMYLWFWEC0OWjrl@mini-arch>
Date: Thu, 25 Jul 2024 20:29:49 -0700
From: Stanislav Fomichev <sdf@...ichev.me>
To: Rao Shoaib <rao.shoaib@...cle.com>
Cc: sdf@...gle.com, Eric Dumazet <edumazet@...gle.com>,
	priyarjha@...gle.com, ycheng@...gle.com, soheil@...gle.com,
	daniel@...earbox.net, netdev@...r.kernel.org, bpf@...r.kernel.org
Subject: Re: bpf: add BPF_CGROUP_SOCK_OPS callback that is executed on every
 RTT (commit: 23729ff23186424)

On 07/25, Rao Shoaib wrote:
> Hi Stanislav,
> 
> I have a question about the placement of tcp_bpf_rtt() call in
> tcp_rtt_estimator(). Why is the call made before the assignment
> 
> tp->srtt_us = max(1U, srtt);
> 
> How is the attached eBPF program suppose the get the new value?

Take a look at the way tcp_bpf_rtt is invoked. It gets mrtt_us
and srtt arguments. Those are passed via bpf_sock_ops args field.
See tools/testing/selftests/bpf/progs/tcp_rtt.c. Hope that helps!

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ