Date: Mon, 13 May 2019 20:24:37 -0400 From: Mark Johnston <markj@freebsd.org> To: Li-Wen Hsu <lwhsu@freebsd.org> Cc: "Andrey V. Elsukov" <ae@freebsd.org>, src-committers <src-committers@freebsd.org>, svn-src-all@freebsd.org, svn-src-head@freebsd.org, melifaro@freebsd.org Subject: Re: svn commit: r347526 - head/sys/net Message-ID: <20190514002437.GA51157@raichu> In-Reply-To: <CAKBkRUxDftmT-35P4uPxpF7QtDkLjMGyJewGR-pcV%2BfjdzJByw@mail.gmail.com> References: <201905131345.x4DDjTRn092669@repo.freebsd.org> <20190513193754.GA37655@raichu> <CAKBkRUxDftmT-35P4uPxpF7QtDkLjMGyJewGR-pcV%2BfjdzJByw@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, May 13, 2019 at 06:09:59PM -0400, Li-Wen Hsu wrote: > On Mon, May 13, 2019 at 3:38 PM Mark Johnston <markj@freebsd.org> wrote: > > > > On Mon, May 13, 2019 at 01:45:29PM +0000, Andrey V. Elsukov wrote: > > > Author: ae > > > Date: Mon May 13 13:45:28 2019 > > > New Revision: 347526 > > > URL: https://svnweb.freebsd.org/changeset/base/347526 > > > > > > Log: > > > Rework locking in BPF code to remove rwlock from fast path. > > > > > > On high packets rate the contention on rwlock in bpf_*tap*() functions > > > can lead to packets dropping. To avoid this, migrate this code to use > > > epoch(9) KPI and ConcurrencyKit's lists. > > > > > > * all lists changed to use CK_LIST; > > > * reference counting added to bpf_if and bpf_d; > > > * now bpf_if references ifnet and releases this reference on destroy; > > > * each bpf_d descriptor references bpf_if when it is attached; > > > * new struct bpf_program_buffer introduced to keep BPF filter programs; > > > * bpf_program_buffer, bpf_d and bpf_if structures are freed by > > > epoch_call(); > > > * bpf_freelist and ifnet_departure event are no longer needed, thus > > > both are removed; > > > > > > Reviewed by: melifaro > > > Sponsored by: Yandex LLC > > > Differential Revision: https://reviews.freebsd.org/D20224 > > > > This seems to introduce a possible recursion on the bpf cdev mutex. Now > > bpfwrite() holds the mutex over the if_output call, which may enter > > bpf_mtap() via bridge_transmit(), and bpf_mtap() may attempt to > > reacquire the mutex. > > It does, the sys.netinet.fibs_test.arpresolve_checks_interface_fib > case can trigger a panic: > > root@:/usr/tests/sys/netinet # kyua test > fibs_test:arpresolve_checks_interface_fib > fibs_test:arpresolve_checks_interface_fib -> panic: _mtx_lock_sleep: > recursed on non-recursive mutex bpf @ /usr/src/sys/net/bpf.c:2325 The problem should be fixed by r347549.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20190514002437.GA51157>