linux-kernel - Re: [PATCH v2 08/11] drm/msm: adreno: request for maximum bus bandwidth usage

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <1965cd01-7b31-4f16-82b2-27fd56fcb77e@linaro.org>
Date: Mon, 25 Nov 2024 09:14:22 +0100
From: Neil Armstrong <neil.armstrong@...aro.org>
To: Akhil P Oommen <quic_akhilpo@...cinc.com>
Cc: Viresh Kumar <vireshk@...nel.org>, Nishanth Menon <nm@...com>,
 Stephen Boyd <sboyd@...nel.org>, "Rafael J. Wysocki" <rafael@...nel.org>,
 Rob Clark <robdclark@...il.com>, Sean Paul <sean@...rly.run>,
 Konrad Dybcio <konradybcio@...nel.org>,
 Abhinav Kumar <quic_abhinavk@...cinc.com>,
 Dmitry Baryshkov <dmitry.baryshkov@...aro.org>,
 Marijn Suijten <marijn.suijten@...ainline.org>,
 David Airlie <airlied@...il.com>, Simona Vetter <simona@...ll.ch>,
 Bjorn Andersson <andersson@...nel.org>, Rob Herring <robh@...nel.org>,
 Krzysztof Kozlowski <krzk+dt@...nel.org>, Conor Dooley
 <conor+dt@...nel.org>, Connor Abbott <cwabbott0@...il.com>,
 linux-pm@...r.kernel.org, linux-kernel@...r.kernel.org,
 linux-arm-msm@...r.kernel.org, dri-devel@...ts.freedesktop.org,
 freedreno@...ts.freedesktop.org, devicetree@...r.kernel.org
Subject: Re: [PATCH v2 08/11] drm/msm: adreno: request for maximum bus
 bandwidth usage

On 23/11/2024 23:59, Akhil P Oommen wrote:
> On Tue, Nov 19, 2024 at 06:56:43PM +0100, Neil Armstrong wrote:
>> When requesting a DDR bandwidth level along a GPU frequency
>> level via the GMU, we can also specify the bus bandwidth usage in a 16bit
>> quantitized value.
>>
>> For now simply request the maximum bus usage.
> 
> Why? You don't care about power efficiency?
> Lets drop this patch. We don't care about AB vote yet.

I care about functionality, without this AB vote the spillall use
case that fails on SM8650 HDK fails on the SM8650 QRD.

AB is a quantitized value of the BW voted, so yes I expect we can have
100% of the BW voted, but since we scale the BW it's perfectly fine.

Neil

> 
> -Akhil
> 
>>
>> Signed-off-by: Neil Armstrong <neil.armstrong@...aro.org>
>> ---
>>   drivers/gpu/drm/msm/adreno/a6xx_gmu.c | 11 +++++++++++
>>   drivers/gpu/drm/msm/adreno/a6xx_hfi.h |  5 +++++
>>   2 files changed, 16 insertions(+)
>>
>> diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
>> index dc2d0035544e7848e5c4ea27f1ea9a191f9c4991..36c0f67fd8e109aabf09a0804bacbed3593c39d7 100644
>> --- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
>> +++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
>> @@ -134,6 +134,17 @@ void a6xx_gmu_set_freq(struct msm_gpu *gpu, struct dev_pm_opp *opp,
>>   			if (bw == gmu->gpu_bw_table[bw_index])
>>   				break;
>>   		}
>> +
>> +		if (bw_index) {
>> +			/*
>> +			 * Append AB vote to the maximum bus usage.
>> +			 * AB represents a quantitized 16bit value of the
>> +			 * max ddr bandwidth we could use, let's simply
>> +			 * request the maximum for now.
>> +			 */
>> +			bw_index |= AB_VOTE(MAX_AB_VOTE);
>> +			bw_index |= AB_VOTE_ENABLE;
>> +		}
>>   	}
>>   
>>   	gmu->current_perf_index = perf_index;
>> diff --git a/drivers/gpu/drm/msm/adreno/a6xx_hfi.h b/drivers/gpu/drm/msm/adreno/a6xx_hfi.h
>> index 528110169398f69f16443a29a1594d19c36fb595..52ba4a07d7b9a709289acd244a751ace9bdaab5d 100644
>> --- a/drivers/gpu/drm/msm/adreno/a6xx_hfi.h
>> +++ b/drivers/gpu/drm/msm/adreno/a6xx_hfi.h
>> @@ -173,6 +173,11 @@ struct a6xx_hfi_gx_bw_perf_vote_cmd {
>>   	u32 bw;
>>   };
>>   
>> +#define AB_VOTE_MASK		GENMASK(31, 16)
>> +#define MAX_AB_VOTE		(FIELD_MAX(AB_VOTE_MASK) - 1)
>> +#define AB_VOTE(vote)		FIELD_PREP(AB_VOTE_MASK, (vote))
>> +#define AB_VOTE_ENABLE		BIT(8)
>> +
>>   #define HFI_H2F_MSG_PREPARE_SLUMBER 33
>>   
>>   struct a6xx_hfi_prep_slumber_cmd {
>>
>> -- 
>> 2.34.1
>>