[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <1721b4cb-adab-4e5b-070f-c26d0e2c6bbf@amd.com>
Date: Wed, 30 Mar 2022 08:41:30 -0400
From: Rodrigo Siqueira Jordao <rjordrigo@....com>
To: Melissa Wen <mwen@...lia.com>, amd-gfx@...ts.freedesktop.org,
airlied@...ux.ie, alexander.deucher@....com,
christian.koenig@....com, daniel@...ll.ch, harry.wentland@....com,
Rodrigo.Siqueira@....com, sunpeng.li@....com, Xinhui.Pan@....com
Cc: Qingqing Zhuo <qingqing.zhuo@....com>,
Dmytro Laktyushkin <Dmytro.Laktyushkin@....com>,
Jasdeep Dhillon <jdhillon@....com>,
dri-devel@...ts.freedesktop.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 0/2] remove DC_FP_* wrappers in dml files
On 2022-03-26 16:24, Melissa Wen wrote:
> From FPU documentation, developers must not use DC_FP_START/END in dml
> files, but invoke it when calling FPU-associated functions (isolated in
> dml folder). Therefore, the first patch renames dcn10_validate_bandwidth
> in dml/calcs to dcn_ for generalization, declares dcn10_validate_bandwidth
> in dcn10 - that calls dcn_validate_bandwidth and wraps with DC_FP_*
> accordingly. The second patch removes invocations of DC_FP_* from dml
> files and properly wraps FPU functions in dc code outside dml folder.
>
> Melissa Wen (2):
> drm/amd/display: detach fpu operations from dcn10_validate_bandwidth
> in calcs
> drm/amd/display: remove DC_FP_* wrapper from dml folder
>
> .../amd/display/dc/dcn10/dcn10_hw_sequencer.c | 10 ++++++++--
> .../drm/amd/display/dc/dcn10/dcn10_resource.c | 16 ++++++++++++++++
> .../drm/amd/display/dc/dml/calcs/dcn_calcs.c | 19 +------------------
> .../drm/amd/display/dc/dml/dcn20/dcn20_fpu.c | 2 --
> .../gpu/drm/amd/display/dc/inc/dcn_calcs.h | 2 +-
> 5 files changed, 26 insertions(+), 23 deletions(-)
>
Hi,
Thanks a lot for your patch!
I reviewed and tested this series and lgtm. Applied to amd-staging-drm-next.
Btw, I agree with Christian. Can you try to find a way to add a
compilation error or warning if the developer tries to add DC_FP_*
inside DML?
Also, about the question of recursive calling to DC_FP_*, it should be
safe if using DC_FP_*.
Thanks
Siqueira
Powered by blists - more mailing lists