From mboxrd@z Thu Jan 1 00:00:00 1970 From: bugzilla-daemon@freedesktop.org Subject: [Bug 93460] [amdgpu] Ooops during shutdown - amdgpu_vm_grab_id Date: Tue, 22 Dec 2015 02:49:27 +0000 Message-ID: References: Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="===============1138684514==" Return-path: Received: from culpepper.freedesktop.org (unknown [131.252.210.165]) by gabe.freedesktop.org (Postfix) with ESMTP id BF6796E14B for ; Mon, 21 Dec 2015 18:49:27 -0800 (PST) In-Reply-To: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dri-devel-bounces@lists.freedesktop.org Sender: "dri-devel" To: dri-devel@lists.freedesktop.org List-Id: dri-devel@lists.freedesktop.org --===============1138684514== Content-Type: multipart/alternative; boundary="1450752567.DcA711.2033"; charset="UTF-8" --1450752567.DcA711.2033 Date: Tue, 22 Dec 2015 02:49:27 +0000 MIME-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable https://bugs.freedesktop.org/show_bug.cgi?id=3D93460 --- Comment #8 from david1.zhou@amd.com --- (In reply to Christian K=C3=B6nig from comment #7) > (In reply to david1.zhou@amd.com from comment #6) > > Maybe we shall avoid to use fence for vmid, instead using LRU list. >=20 > Yeah, thought about that as well. The problem is that we used to have an = LRU > list and I switched to fences because they had less overhead. >=20 > We still need to keep the fences around for synchronization, so I'm not s= ure > if that would really help. >=20 > The real price question is what is going wrong here? yes, we need to identify why the contexts of two fences are different, where two fences come from, what the kind of two fences are, which ring two fences belong. --=20 You are receiving this mail because: You are the assignee for the bug. --1450752567.DcA711.2033 Date: Tue, 22 Dec 2015 02:49:27 +0000 MIME-Version: 1.0 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable

Comment= # 8 on bug 93460<= /a> from david1.zhou@amd.com 
(In reply to Christian K=C3=B6nig from comment #7)
> (In reply to david1.zhou@amd.com from comment #6)
> > Maybe we shall avoid to use fence for vmid, instead using LRU lis=
t.
>=20
> Yeah, thought about that as well. The problem is that we used to have =
an LRU
> list and I switched to fences because they had less overhead.
>=20
> We still need to keep the fences around for synchronization, so I'm no=
t sure
> if that would really help.
>=20
> The real price question is what is going wrong here?

yes, we need to identify why the contexts of two fences are different, where
two fences come from, what the kind of two fences are, which ring two fences
belong.

You are receiving this mail because: =20=20=20=20=20=20
  • You are the assignee for the bug.
--1450752567.DcA711.2033-- --===============1138684514== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: base64 Content-Disposition: inline X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX18KZHJpLWRldmVs IG1haWxpbmcgbGlzdApkcmktZGV2ZWxAbGlzdHMuZnJlZWRlc2t0b3Aub3JnCmh0dHA6Ly9saXN0 cy5mcmVlZGVza3RvcC5vcmcvbWFpbG1hbi9saXN0aW5mby9kcmktZGV2ZWwK --===============1138684514==--