Date: Tue, 9 Oct 2018 11:57:53 -0700 From: John Baldwin <jhb@FreeBSD.org> To: "Jonathan T. Looney" <jtl@freebsd.org> Cc: src-committers <src-committers@freebsd.org>, svn-src-all@freebsd.org, svn-src-head@freebsd.org Subject: Re: svn commit: r339251 - in head/sys: net netinet netinet6 Message-ID: <d4887528-dd25-7500-9afc-d79af38781ee@FreeBSD.org> In-Reply-To: <CADrOrmviC-%2BVz3PMM0yA1=VRwhKnY6Kkv8aG=nFE=VYPL8mZyQ@mail.gmail.com> References: <201810091326.w99DQ7MN083891@repo.freebsd.org> <bb0f4bca-8cca-f2c2-f8d8-5a8ee548d3ef@FreeBSD.org> <CADrOrmviC-%2BVz3PMM0yA1=VRwhKnY6Kkv8aG=nFE=VYPL8mZyQ@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On 10/9/18 8:36 AM, Jonathan T. Looney wrote: > On Tue, Oct 9, 2018 at 11:02 AM John Baldwin <jhb@freebsd.org <mailto:jhb@freebsd.org>> wrote: > > Do we have some sort of simple per-thread epoch counter similar to > td->td_locks that we could assert on in userret() and in the ithread loop > when ithreads go to sleep to help catch leaked locks? > > > Yes: td->td_epochnest. > > There are already INVARIANTS checks in malloc (for M_WAITOK), userret(), and _sleep(). I think that covers the cases you mentioned. However, I am very much in favor of adding any additional checks which make sense. The one that would matter in this case would be idle ithreads. We have some existing checks (WITNESS_WARN) in ithread_loop() already. We should add a td_epochnest assertion next to that line I think. -- John Baldwin
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?d4887528-dd25-7500-9afc-d79af38781ee>