md.com>; amd-gfx@lists.freedesktop.org
Cc: Wang, Daniel(Xiaowei) <daniel.wa...@amd.com>; Jiang, Jerry (SW)
<jerry.ji...@amd.com>
Subject: Re: [PATCH] drm/amdgpu:fix DMAframe for GFX8
> >>Briefly speaking: ring buffer should be fixed at some certain
> >>pattern so
ongs to P2, why we never meet vm fault ?
>
>
> Thanks for your time!
>
> BR Monk
>
>
>
>
> -----Original Message-
> From: Christian König [mailto:deathsim...@vodafone.de]
> Sent: Wednesday, August 24, 2016 4:16 PM
> To: Liu, Monk <monk@amd.com &l
md-gfx@lists.freedesktop.org
Cc: Wang, Daniel(Xiaowei) <daniel.wa...@amd.com>; Jiang, Jerry (SW)
<jerry.ji...@amd.com>
Subject: RE: [PATCH] drm/amdgpu:fix DMAframe for GFX8
Christian,
I'm not going to argue with you about policy or interface protocol , cuz those
things would block us from impleme
-P2 stage (P1/P2 means process1/process2), since this time vm page
> table of P2 yet not been binding to VMID3, if CE is already running in IB
> belongs to P2, why we never meet vm fault ?
>
>
> Thanks for your time!
>
> BR Monk
>
>
>
>
> -Original
age-
From: Christian König [mailto:deathsim...@vodafone.de]
Sent: Wednesday, August 24, 2016 4:16 PM
To: Liu, Monk <monk@amd.com>; amd-gfx@lists.freedesktop.org
Subject: Re: [PATCH] drm/amdgpu:fix DMAframe for GFX8
Am 24.08.2016 um 05:42 schrieb Monk Liu:
1,drop inserting double SWITCH_BUFF
, Monk <monk@amd.com>; amd-gfx@lists.freedesktop.org
Subject: Re: [PATCH] drm/amdgpu:fix DMAframe for GFX8
Am 24.08.2016 um 05:42 schrieb Monk Liu:
> 1,drop inserting double SWITCH_BUFFERS scheme, which impacts
> performance, because double SWITCH_BUFFER actaully doesn't switch
&g
Am 24.08.2016 um 05:42 schrieb Monk Liu:
1,drop inserting double SWITCH_BUFFERS scheme, which impacts
performance, because double SWITCH_BUFFER actaully doesn't
switch buffer, so the CE's ping-pong buffer is not used at
all. Now only insert one SWITCH_BUFFERS at the bottom of
each GFX dmaframe.
1,drop inserting double SWITCH_BUFFERS scheme, which impacts performance,
because double SWITCH_BUFFER actaully doesn't switch buffer, so the CE's
ping-pong buffer is not used at all. Now only insert one SWITCH_BUFFERS at the
bottom of each GFX dmaframe. And only inserting one SWITCH_BUFFER is