mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
To: Connor Abbott <cwabbott0@gmail.com>,
	Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Cc: Konrad Dybcio <konradybcio@kernel.org>,
	Bjorn Andersson <andersson@kernel.org>,
	Kees Cook <kees@kernel.org>,
	"Gustavo A. R. Silva" <gustavoars@kernel.org>,
	Rob Clark <robdclark@gmail.com>, Sean Paul <sean@poorly.run>,
	Abhinav Kumar <quic_abhinavk@quicinc.com>,
	Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>,
	David Airlie <airlied@gmail.com>, Simona Vetter <simona@ffwll.ch>,
	Dmitry Baryshkov <lumag@kernel.org>,
	Marijn Suijten <marijn.suijten@somainline.org>,
	linux-kernel@vger.kernel.org, linux-arm-msm@vger.kernel.org,
	linux-hardening@vger.kernel.org, dri-devel@lists.freedesktop.org,
	freedreno@lists.freedesktop.org
Subject: Re: [PATCH 3/4] drm/msm/a6xx: Get HBB dynamically, if available
Date: Wed, 9 Apr 2025 17:46:59 +0200	[thread overview]
Message-ID: <0fda3b72-8bff-4f3d-b9e2-cfe1ac1d3082@oss.qualcomm.com> (raw)
In-Reply-To: <CACu1E7Hto_Ms3RxQjXXYB+RB+FC5n1VFtohKbe4EzKWgEk9uiQ@mail.gmail.com>

On 4/9/25 5:44 PM, Connor Abbott wrote:
> On Wed, Apr 9, 2025 at 11:40 AM Konrad Dybcio
> <konrad.dybcio@oss.qualcomm.com> wrote:
>>
>> On 4/9/25 5:30 PM, Connor Abbott wrote:
>>> On Wed, Apr 9, 2025 at 11:22 AM Konrad Dybcio
>>> <konrad.dybcio@oss.qualcomm.com> wrote:
>>>>
>>>> On 4/9/25 5:12 PM, Connor Abbott wrote:
>>>>> On Wed, Apr 9, 2025 at 10:48 AM Konrad Dybcio <konradybcio@kernel.org> wrote:
>>>>>>
>>>>>> From: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
>>>>>>
>>>>>> The Highest Bank address Bit value can change based on memory type used.
>>>>>>
>>>>>> Attempt to retrieve it dynamically, and fall back to a reasonable
>>>>>> default (the one used prior to this change) on error.
>>>>>>
>>>>>> Signed-off-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
>>>>>> ---
>>>>>>  drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 22 ++++++++++++++++------
>>>>>>  1 file changed, 16 insertions(+), 6 deletions(-)
>>>>>>
>>>>>> diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
>>>>>> index 06465bc2d0b4b128cddfcfcaf1fe4252632b6777..0cc397378c99db35315209d0265ad9223e8b55c7 100644
>>>>>> --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
>>>>>> +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
>>>>>> @@ -13,6 +13,7 @@
>>>>>>  #include <linux/firmware/qcom/qcom_scm.h>
>>>>>>  #include <linux/pm_domain.h>
>>>>>>  #include <linux/soc/qcom/llcc-qcom.h>
>>>>>> +#include <linux/soc/qcom/smem.h>
>>>>>>
>>>>>>  #define GPU_PAS_ID 13
>>>>>>
>>>>>> @@ -669,17 +670,22 @@ static void a6xx_calc_ubwc_config(struct adreno_gpu *gpu)
>>>>>>  static void a6xx_set_ubwc_config(struct msm_gpu *gpu)
>>>>>>  {
>>>>>>         struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
>>>>>> +       u32 hbb = qcom_smem_dram_get_hbb();
>>>>>> +       u32 ubwc_mode = adreno_gpu->ubwc_config.ubwc_swizzle & 1;
>>>>>> +       u32 level2_swizzling_dis = !(adreno_gpu->ubwc_config.ubwc_swizzle & 2);
>>>>>> +       u32 hbb_hi, hbb_lo;
>>>>>> +
>>>>>>         /*
>>>>>>          * We subtract 13 from the highest bank bit (13 is the minimum value
>>>>>>          * allowed by hw) and write the lowest two bits of the remaining value
>>>>>>          * as hbb_lo and the one above it as hbb_hi to the hardware.
>>>>>>          */
>>>>>> -       BUG_ON(adreno_gpu->ubwc_config.highest_bank_bit < 13);
>>>>>> -       u32 hbb = adreno_gpu->ubwc_config.highest_bank_bit - 13;
>>>>>> -       u32 hbb_hi = hbb >> 2;
>>>>>> -       u32 hbb_lo = hbb & 3;
>>>>>> -       u32 ubwc_mode = adreno_gpu->ubwc_config.ubwc_swizzle & 1;
>>>>>> -       u32 level2_swizzling_dis = !(adreno_gpu->ubwc_config.ubwc_swizzle & 2);
>>>>>> +       if (hbb < 0)
>>>>>> +               hbb = adreno_gpu->ubwc_config.highest_bank_bit;
>>>>>
>>>>> No. The value we expose to userspace must match what we program.
>>>>> You'll break VK_EXT_host_image_copy otherwise.
>>>>
>>>> I didn't know that was exposed to userspace.
>>>>
>>>> The value must be altered either way - ultimately, the hardware must
>>>> receive the correct information. ubwc_config doesn't seem to be const,
>>>> so I can edit it there if you like it better.
>>>>
>>>> Konrad
>>>
>>> Yes, you should be calling qcom_smem_dram_get_hbb() in
>>> a6xx_calc_ubwc_config(). You can already see there's a TODO there to
>>> plug it in.
>>
>> Does this look good instead?
>>
>> diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
>> index 0cc397378c99..ae8dbc250e6a 100644
>> --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
>> +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
>> @@ -588,6 +588,8 @@ static void a6xx_set_cp_protect(struct msm_gpu *gpu)
>>
>>  static void a6xx_calc_ubwc_config(struct adreno_gpu *gpu)
>>  {
>> +       u8 hbb;
> 
> You can't make it u8 and then test for a negative value on error.

Fair. I think it was u8 in a pre-release version of the patchset and it
stuck in my mind.. though I'dve expected clang to warn me here..

> Other than that, looks good.

Thanks, I'll change it in v2.

Konrad

  reply	other threads:[~2025-04-09 15:47 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-04-09 14:47 [PATCH 0/4] Retrieve information about DDR from SMEM Konrad Dybcio
2025-04-09 14:47 ` [PATCH 1/4] soc: qcom: Expose DDR data " Konrad Dybcio
2025-04-10  2:21   ` Bjorn Andersson
2025-04-09 14:47 ` [PATCH 2/4] drm/msm/a5xx: Get HBB dynamically, if available Konrad Dybcio
2025-04-09 14:47 ` [PATCH 3/4] drm/msm/a6xx: " Konrad Dybcio
2025-04-09 15:12   ` Connor Abbott
2025-04-09 15:22     ` Konrad Dybcio
2025-04-09 15:30       ` Connor Abbott
2025-04-09 15:40         ` Konrad Dybcio
2025-04-09 15:44           ` Connor Abbott
2025-04-09 15:46             ` Konrad Dybcio [this message]
2025-04-09 14:47 ` [PATCH 4/4] drm/msm/mdss: " Konrad Dybcio
2025-04-09 15:44 ` [PATCH 0/4] Retrieve information about DDR from SMEM Dmitry Baryshkov
2025-04-09 15:49   ` Konrad Dybcio
2025-04-11  9:48     ` Konrad Dybcio
2025-04-11  9:57       ` Dmitry Baryshkov
2025-04-11 10:03         ` Konrad Dybcio
2025-04-11 10:50           ` Dmitry Baryshkov
2025-04-11 10:52             ` Konrad Dybcio
2025-04-14 12:28               ` Dmitry Baryshkov
2025-04-14 13:06                 ` Konrad Dybcio

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=0fda3b72-8bff-4f3d-b9e2-cfe1ac1d3082@oss.qualcomm.com \
    --to=konrad.dybcio@oss.qualcomm.com \
    --cc=airlied@gmail.com \
    --cc=andersson@kernel.org \
    --cc=cwabbott0@gmail.com \
    --cc=dmitry.baryshkov@oss.qualcomm.com \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=freedreno@lists.freedesktop.org \
    --cc=gustavoars@kernel.org \
    --cc=kees@kernel.org \
    --cc=konradybcio@kernel.org \
    --cc=linux-arm-msm@vger.kernel.org \
    --cc=linux-hardening@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=lumag@kernel.org \
    --cc=marijn.suijten@somainline.org \
    --cc=quic_abhinavk@quicinc.com \
    --cc=robdclark@gmail.com \
    --cc=sean@poorly.run \
    --cc=simona@ffwll.ch \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®