From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <stable-owner@vger.kernel.org>
Received: from mail.fireflyinternet.com ([109.228.58.192]:52856 "EHLO
        fireflyinternet.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org
        with ESMTP id S1726547AbeJCRfz (ORCPT
        <rfc822;stable@vger.kernel.org>); Wed, 3 Oct 2018 13:35:55 -0400
Content-Type: text/plain; charset="utf-8"
MIME-Version: 1.0
Content-Transfer-Encoding: 8BIT
To: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>,
        intel-gfx@lists.freedesktop.org
From: Chris Wilson <chris@chris-wilson.co.uk>
In-Reply-To: <f75a2723-7957-5ddb-9d5d-ac79eddecfc0@linux.intel.com>
Cc: stable@vger.kernel.org
References: <20181001194447.29910-2-chris@chris-wilson.co.uk>
 <20181003082422.23214-1-chris@chris-wilson.co.uk>
 <f75a2723-7957-5ddb-9d5d-ac79eddecfc0@linux.intel.com>
Message-ID: <153856367929.17386.13325019531737250354@skylake-alporthouse-com>
Subject: Re: [Intel-gfx] [PATCH v3] drm/i915: Handle incomplete Z_FINISH for
 compressed error states
Date: Wed, 03 Oct 2018 11:47:59 +0100
Sender: stable-owner@vger.kernel.org
List-ID: <stable.vger.kernel.org>

Quoting Tvrtko Ursulin (2018-10-03 10:04:04)
> 
> On 03/10/2018 09:24, Chris Wilson wrote:
> > The final call to zlib_deflate(Z_FINISH) may require more output
> > space to be allocated and so needs to re-invoked. Failure to do so in
> > the current code leads to incomplete zlib streams (albeit intact due to
> > the use of Z_SYNC_FLUSH) resulting in the occasional short object
> > capture.
> > 
> > v2: Check against overrunning our pre-allocated page array
> > v3: Drop Z_SYNC_FLUSH entirely
> > 
> > Testcase: igt/i915-error-capture.js
> > Fixes: 0a97015d45ee ("drm/i915: Compress GPU objects in error state")
> > Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> > Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
> > Cc: <stable@vger.kernel.org> # v4.10+
> > Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>

> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>

Ta, thanks for the review. gem_exec_capture with several million
objects/pages was happy so pushed.

Now, we just need the O(N^2) fix so that gem_exec_capture runs in a few
minutes and not a few _decades_.
-Chris