Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Subject: Re: [PATCH] dma-buf: Move BUG_ON from _add_shared_fence to
 _add_shared_inplace
To:     christian.koenig@amd.com, Sumit Semwal <sumit.semwal@linaro.org>
Cc:     linaro-mm-sig@lists.linaro.org, dri-devel@lists.freedesktop.org,
        linux-kernel@vger.kernel.org, amd-gfx@lists.freedesktop.org,
        linux-media@vger.kernel.org
References: <20180626143147.14296-1-michel@daenzer.net>
 <249b84ea-affe-2e27-abdd-81d61da9cce6@gmail.com>
From:   =?UTF-8?Q?Michel_D=c3=a4nzer?= <michel@daenzer.net>
Openpgp: preference=signencrypt
Autocrypt: addr=michel@daenzer.net; prefer-encrypt=mutual; keydata=
 xsDiBDsehS8RBACbsIQEX31aYSIuEKxEnEX82ezMR8z3LG8ktv1KjyNErUX9Pt7AUC7W3W0b
 LUhu8Le8S2va6hi7GfSAifl0ih3k6Bv1Itzgnd+7ZmSrvCN8yGJaHNQfAevAuEboIb+MaVHo
 9EMJj4ikOcRZCmQWw7evu/D9uQdtkCnRY9iJiAGxbwCguBHtpoGMxDOINCr5UU6qt+m4O+UD
 /355ohBBzzyh49lTj0kTFKr0Ozd20G2FbcqHgfFL1dc1MPyigej2gLga2osu2QY0ObvAGkOu
 WBi3LTY8Zs8uqFGDC4ZAwMPoFy3yzu3ne6T7d/68rJil0QcdQjzzHi6ekqHuhst4a+/+D23h
 Za8MJBEcdOhRhsaDVGAJSFEQB1qLBACOs0xN+XblejO35gsDSVVk8s+FUUw3TSWJBfZa3Imp
 V2U2tBO4qck+wqbHNfdnU/crrsHahjzBjvk8Up7VoY8oT+z03sal2vXEonS279xN2B92Tttr
 AgwosujguFO/7tvzymWC76rDEwue8TsADE11ErjwaBTs8ZXfnN/uAANgPM0jTWljaGVsIERh
 ZW56ZXIgPG1pY2hlbEBkYWVuemVyLm5ldD7CXgQTEQIAHgUCQFXxJgIbAwYLCQgHAwIDFQID
 AxYCAQIeAQIXgAAKCRBaga+OatuyAIrPAJ9ykonXI3oQcX83N2qzCEStLNW47gCeLWm/QiPY
 jqtGUnnSbyuTQfIySkLOwE0EOx6FRRAEAJZkcvklPwJCgNiw37p0GShKmFGGqf/a3xZZEpjI
 qNxzshFRFneZze4f5LhzbX1/vIm5+ZXsEWympJfZzyCmYPw86QcFxyZflkAxHx9LeD+89Elx
 bw6wT0CcLvSv8ROfU1m8YhGbV6g2zWyLD0/naQGVb8e4FhVKGNY2EEbHgFBrAAMGA/0VktFO
 CxFBdzLQ17RCTwCJ3xpyP4qsLJH0yCoA26rH2zE2RzByhrTFTYZzbFEid3ddGiHOBEL+bO+2
 GNtfiYKmbTkj1tMZJ8L6huKONaVrASFzLvZa2dlc2zja9ZSksKmge5BOTKWgbyepEc5qxSju
 YsYrX5xfLgTZC5abhhztpcJGBBgRAgAGBQI7HoVFAAoJEFqBr45q27IAlscAn2Ufk2d6/3p4
 Cuyz/NX7KpL2dQ8WAJ9UD5JEakhfofed8PSqOM7jOO3LCA==
Message-ID: <f6100513-d31b-b7a2-cc9f-104a60127277@daenzer.net>
Date:   Wed, 4 Jul 2018 11:09:58 +0200
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101
 Thunderbird/52.8.0
MIME-Version: 1.0
In-Reply-To: <249b84ea-affe-2e27-abdd-81d61da9cce6@gmail.com>
Content-Type: text/plain; charset=utf-8
Content-Language: en-CA
Content-Transfer-Encoding: 8bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On 2018-07-04 10:31 AM, Christian König wrote:
> Am 26.06.2018 um 16:31 schrieb Michel Dänzer:
>> From: Michel Dänzer <michel.daenzer@amd.com>
>>
>> Fixes the BUG_ON spuriously triggering under the following
>> circumstances:
>>
>> * ttm_eu_reserve_buffers processes a list containing multiple BOs using
>>    the same reservation object, so it calls
>>    reservation_object_reserve_shared with that reservation object once
>>    for each such BO.
>> * In reservation_object_reserve_shared, old->shared_count ==
>>    old->shared_max - 1, so obj->staged is freed in preparation of an
>>    in-place update.
>> * ttm_eu_fence_buffer_objects calls reservation_object_add_shared_fence
>>    once for each of the BOs above, always with the same fence.
>> * The first call adds the fence in the remaining free slot, after which
>>    old->shared_count == old->shared_max.
> 
> Well, the explanation here is not correct. For multiple BOs using the
> same reservation object we won't call
> reservation_object_add_shared_fence() multiple times because we move
> those to the duplicates list in ttm_eu_reserve_buffers().
> 
> But this bug can still happen because we call
> reservation_object_add_shared_fence() manually with fences for the same
> context in a couple of places.
> 
> One prominent case which comes to my mind are for the VM BOs during
> updates. Another possibility are VRAM BOs which need to be cleared.

Thanks. How about the following:

* ttm_eu_reserve_buffers calls reservation_object_reserve_shared.
* In reservation_object_reserve_shared, shared_count == shared_max - 1,
  so obj->staged is freed in preparation of an in-place update.
* ttm_eu_fence_buffer_objects calls reservation_object_add_shared_fence,
  after which shared_count == shared_max.
* The amdgpu driver also calls reservation_object_add_shared_fence for
  the same reservation object, and the BUG_ON triggers.

However, nothing bad would happen in
reservation_object_add_shared_inplace, since all fences use the same
context, so they can only occupy a single slot.

Prevent this by moving the BUG_ON to where an overflow would actually
happen (e.g. if a buggy caller didn't call
reservation_object_reserve_shared before).


Also, I'll add a reference to https://bugs.freedesktop.org/106418 in v2,
as I suspect this fix is necessary under the circumstances described
there as well.


-- 
Earthling Michel Dänzer               |               http://www.amd.com
Libre software enthusiast             |             Mesa and X developer