From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ville =?iso-8859-1?Q?Syrj=E4l=E4?= Date: Fri, 07 Dec 2012 14:45:02 +0000 Subject: Re: [PATCH 2/5] OMAPFB: simplify locking Message-Id: <20121207144502.GG32230@intel.com> List-Id: References: <1354881309-17625-1-git-send-email-tomi.valkeinen@ti.com> <1354881309-17625-2-git-send-email-tomi.valkeinen@ti.com> <20121207125350.GE32230@intel.com> <50C1F232.80508@ti.com> In-Reply-To: <50C1F232.80508@ti.com> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable To: Tomi Valkeinen Cc: Archit Taneja , linux-omap@vger.kernel.org, linux-fbdev@vger.kernel.org On Fri, Dec 07, 2012 at 03:42:10PM +0200, Tomi Valkeinen wrote: > On 2012-12-07 14:53, Ville Syrj=E4l=E4 wrote: > > On Fri, Dec 07, 2012 at 01:55:06PM +0200, Tomi Valkeinen wrote: > >> Kernel lock verification code has lately detected possible circular > >> locking in omapfb. The exact problem is unclear, but omapfb's current > >> locking seems to be overly complex. > >> > >> This patch simplifies the locking in the following ways: > >> > >> - Remove explicit omapfb mem region locking. I couldn't figure out the > >> need for this, as long as we take care to take omapfb lock. > >=20 > > I suppose the idea with that was that you wouldn't need the global > > omapfb lock, and also it was an rwsem so it allowed parallel access to > > the mem regions, unless the region size was being changed, in which > > case it took the write lock. I can't really remember what the reason > > for using an rwsem was, but I suppose there was one at the time. >=20 > Right. Yes, I have no recollection either of the possible reason for it > =3D). Did we have multiple concurrerent users for the fbs? It still sounds > like a useless optimization, as all the region locks were only held for > a short time, as far as I saw. >=20 > It could also be that we're missing something from the mainline kernel, > which we had in the Nokia kernel, and which would explain the need for > region locks. Yeah, perhaps there was some use for it at the time. Possibly the zero copy video playback benefited from it in the beginning. But IIRC almost everything was handled via the X server in the end, so I don't think there's any compelling reason for keeping the rwsem. --=20 Ville Syrj=E4l=E4 Intel OTC