From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Benjamin Marzinski" Subject: Re: [PATCH v3 02/19] libmultipath: fix tur checker double locking Date: Thu, 4 Oct 2018 11:47:20 -0500 Message-ID: <20181004164720.GC3172@octiron.msp.redhat.com> References: <1537571127-10143-1-git-send-email-bmarzins@redhat.com> <1537571127-10143-3-git-send-email-bmarzins@redhat.com> <3bad257069d68444b0b2743d4b38ba6b6d15501d.camel@suse.com> <8d1b4d256211d4ee6222bdc24563b09695bc883a.camel@suse.de> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Content-Disposition: inline In-Reply-To: <8d1b4d256211d4ee6222bdc24563b09695bc883a.camel@suse.de> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: dm-devel-bounces@redhat.com Errors-To: dm-devel-bounces@redhat.com To: Martin Wilck Cc: Bart Van Assche , dm-devel@redhat.com List-Id: dm-devel.ids On Mon, Oct 01, 2018 at 10:44:06PM +0200, Martin Wilck wrote: > On Mon, 2018-10-01 at 22:09 +0200, Martin Wilck wrote: > > > > I like the lock removal a lot, but not so much the conversion into a > > string. Why not keep the dev_t? > > > > Or maybe even easier, the other way around: why don't we make it a > > char* and simply set checker->dev_t = &pp->dev_t? > > OK, that 2nd one won't work because pp may be destroyed before the > checker terminates. Got it. Still I'd prefer keeping a dev_t in the > checker structure rather than a char[32]. Why? We never do anything with it as a dev_t except change it into a string. If you feel strongly about this, I don't really care, and I'll change it. I just don't understand your objection. -Ben > > Martin >