[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20200610021409.GH19604@bombadil.infradead.org>
Date: Tue, 9 Jun 2020 19:14:09 -0700
From: Matthew Wilcox <willy@...radead.org>
To: Andrew Morton <akpm@...ux-foundation.org>
Cc: Qian Cai <cai@....pw>, linux-kernel@...r.kernel.org,
dave@...olabs.net, manfred@...orfullife.com,
mm-commits@...r.kernel.org
Subject: Re: + ipc-convert-ipcs_idr-to-xarray-update.patch added to -mm tree
On Fri, Jun 05, 2020 at 02:20:39PM -0700, Andrew Morton wrote:
> On Fri, 5 Jun 2020 13:11:34 -0700 Matthew Wilcox <willy@...radead.org> wrote:
>
> > On Fri, Jun 05, 2020 at 03:58:48PM -0400, Qian Cai wrote:
> > > This will trigger,
> > >
> > > [ 8853.759549] LTP: starting semget05
> > > [ 8867.257088] BUG: sleeping function called from invalid context at mm/slab.h:567
> > > [ 8867.270259] in_atomic(): 1, irqs_disabled(): 0, non_block: 0, pid: 22556, name: semget05
> > > [ 8867.270309] 2 locks held by semget05/22556:
> > > [ 8867.270345] #0: 00000000512de7e0 (&ids->rwsem){++++}-{3:3}, at: ipcget+0x4e/0x230
> > > [ 8867.270426] #1: 00000000552b9018 (&new->lock){+.+.}-{2:2}, at: ipc_addid+0xf4/0xf50
> >
> > Did the fix for this not make it into -next?
>
> I don't think I've seen this fix before. And I'm unclear on how it
> pertains to the current patch(es) in -mm. And it has no changelog!
>
> Perhaps it would be best to do a formal send of the latest version?
In testing, I discovered a lockdep warning. I don't think it's readily
fixable, so please drop this patch for the moment. I'll be back with
a patch set that rejiggers the locking in ipc to be more normal.
Powered by blists - more mailing lists