[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <784a7813-b024-452e-8d7e-8cbaea761bcd@quicinc.com>
Date: Mon, 25 Nov 2024 18:03:52 -0800
From: Abhinav Kumar <quic_abhinavk@...cinc.com>
To: Dmitry Baryshkov <dmitry.baryshkov@...aro.org>,
Rob Clark
<robdclark@...il.com>, Sean Paul <sean@...rly.run>,
Marijn Suijten
<marijn.suijten@...ainline.org>,
Connor Abbott <cwabbott0@...il.com>, David
Airlie <airlied@...il.com>,
Simona Vetter <simona@...ll.ch>
CC: <linux-kernel@...r.kernel.org>, <linux-arm-msm@...r.kernel.org>,
<dri-devel@...ts.freedesktop.org>, <freedreno@...ts.freedesktop.org>
Subject: Re: [PATCH v2 1/3] drm/msm/mdss: define bitfields for the UBWC_STATIC
register
On 11/22/2024 9:44 PM, Dmitry Baryshkov wrote:
> Rather than hand-coding UBWC_STATIC value calculation, define
> corresponding bitfields and use them to setup the register value.
>
> Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@...aro.org>
> ---
> drivers/gpu/drm/msm/msm_mdss.c | 38 +++++++++++++++-----------
> drivers/gpu/drm/msm/msm_mdss.h | 3 +-
> drivers/gpu/drm/msm/registers/display/mdss.xml | 11 +++++++-
> 3 files changed, 34 insertions(+), 18 deletions(-)
>
<snip>
>
> diff --git a/drivers/gpu/drm/msm/registers/display/mdss.xml b/drivers/gpu/drm/msm/registers/display/mdss.xml
> index ac85caf1575c7908bcf68f0249da38dccf4f07b6..b6f93984928522a35a782cbad9de006eac225725 100644
> --- a/drivers/gpu/drm/msm/registers/display/mdss.xml
> +++ b/drivers/gpu/drm/msm/registers/display/mdss.xml
> @@ -21,7 +21,16 @@ xsi:schemaLocation="https://gitlab.freedesktop.org/freedreno/ rules-fd.xsd">
>
> <reg32 offset="0x00058" name="UBWC_DEC_HW_VERSION"/>
>
> - <reg32 offset="0x00144" name="UBWC_STATIC"/>
> + <reg32 offset="0x00144" name="UBWC_STATIC">
> + <bitfield name="UBWC_SWIZZLE" low="0" high="2"/>
> + <bitfield name="UBWC_BANK_SPREAD" pos="3"/>
> + <!-- high=5 for UBWC < 4.0 -->
> + <bitfield name="HIGHEST_BANK_BIT" low="4" high="6"/>
> + <bitfield name="UBWC_MIN_ACC_LEN" pos="8"/>
MIN_ACC_LEN OR MALSIZE has 2 bits , bits 8 and 9.
But bit 9 is unused today. Hence we were using it as a 1 or 0 today.
Its unused on all the chipsets I checked. Do you want to continue using
the same way or correct this?
Powered by blists - more mailing lists