All of lore.kernel.org
 help / color / mirror / Atom feed
From: Sai Prakash Ranjan <saiprakash.ranjan@codeaurora.org>
To: Rob Clark <robdclark@gmail.com>
Cc: "Isaac J. Manjarres" <isaacm@codeaurora.org>,
	Will Deacon <will@kernel.org>, Joerg Roedel <joro@8bytes.org>,
	Akhil P Oommen <akhilpo@codeaurora.org>,
	"\" <iommu@lists.linux-foundation.org>,
	<linux-arm-kernel@lists.infradead.org>,
	Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
	linux-arm-msm <linux-arm-msm@vger.kernel.org>,
	freedreno <freedreno@lists.freedesktop.org>,
	Kristian H Kristensen <hoegsberg@google.com>,
	Sean Paul <sean@poorly.run>, David Airlie <airlied@linux.ie>,
	Daniel Vetter <daniel@ffwll.ch>,
	dri-devel" <dri-devel@lists.freedesktop.org>,
	Jordan Crouse <jordan@cosmicpenguin.net>,
	Robin Murphy <robin.murphy@arm.com>
Subject: Re: [PATCH 2/3] iommu/io-pgtable-arm: Add IOMMU_LLC page protection flag
Date: Wed, 17 Mar 2021 15:03:16 +0530	[thread overview]
Message-ID: <d44fc38c038be1165aa8f4212bd9c91f@codeaurora.org> (raw)
In-Reply-To: <CAF6AEGuc5i9hMtfU3HSpLVWi_e=emJTPLqntzJfAH69dO_gagA@mail.gmail.com>

Hi Rob,

On 2021-03-16 22:46, Rob Clark wrote:

<snip>...

>> > >
>> > > When the GPU has a buffer mapped with IOMMU_LLC, is the buffer also mapped
>> > > into the CPU and with what attributes? Rob said "writecombine for
>> > > everything" -- does that mean ioremap_wc() / MEMREMAP_WC?
>> >
>> > Currently userspace asks for everything WC, so pgprot_writecombine()
>> >
>> > The kernel doesn't enforce this, but so far provides no UAPI to do
>> > anything useful with non-coherent cached mappings (although there is
>> > interest to support this)
>> >
>> 
>> btw, I'm looking at a benchmark (gl_driver2_off) where (after some
>> other in-flight optimizations land) we end up bottlenecked on writing
>> to WC cmdstream buffers.  I assume in the current state, WC goes all
>> the way to main memory rather than just to system cache?
>> 
> 
> oh, I guess this (mentioned earlier in thread) is what I really want
> for this benchmark:
> 
> https://android-review.googlesource.com/c/kernel/common/+/1549097/3
> 

You can also check if the system cache lines are allocated for GPU
or not with patch in https://crrev.com/c/2766723

With the above patch applied,
cat /sys/kernel/debug/llcc_stats/llcc_scid_status

The SCIDs for GPU are listed in include/linux/soc/qcom/llcc-qcom.h

Thanks,
Sai

-- 
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a 
member
of Code Aurora Forum, hosted by The Linux Foundation
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

  reply	other threads:[~2021-03-17  9:33 UTC|newest]

Thread overview: 121+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-01-11 14:15 [PATCH 0/3] iommu/drm/msm: Allow non-coherent masters to use system cache Sai Prakash Ranjan
2021-01-11 14:15 ` Sai Prakash Ranjan
2021-01-11 14:15 ` Sai Prakash Ranjan
2021-01-11 14:15 ` [PATCH 1/3] iommu/io-pgtable: Rename last-level cache quirk to IO_PGTABLE_QUIRK_PTW_LLC Sai Prakash Ranjan
2021-01-11 14:15   ` Sai Prakash Ranjan
2021-01-11 14:15   ` Sai Prakash Ranjan
2021-01-11 14:15 ` [PATCH 2/3] iommu/io-pgtable-arm: Add IOMMU_LLC page protection flag Sai Prakash Ranjan
2021-01-11 14:15   ` Sai Prakash Ranjan
2021-01-11 14:15   ` Sai Prakash Ranjan
2021-01-29  9:05   ` Will Deacon
2021-01-29  9:05     ` Will Deacon
2021-01-29  9:05     ` Will Deacon
2021-01-29  9:42     ` Sai Prakash Ranjan
2021-01-29  9:42       ` Sai Prakash Ranjan
2021-02-01 11:15       ` Will Deacon
2021-02-01 11:15         ` Will Deacon
2021-02-01 11:15         ` Will Deacon
2021-02-01 11:15         ` Will Deacon
2021-02-01 16:20         ` Rob Clark
2021-02-01 16:20           ` Rob Clark
2021-02-01 16:20           ` Rob Clark
2021-02-01 16:20           ` Rob Clark
2021-02-01 18:20           ` Jordan Crouse
2021-02-01 18:20             ` Jordan Crouse
2021-02-01 18:20             ` Jordan Crouse
2021-02-02  6:26             ` Sai Prakash Ranjan
2021-02-02  6:26               ` Sai Prakash Ranjan
2021-02-02  6:26               ` Sai Prakash Ranjan
2021-02-03 21:46               ` Will Deacon
2021-02-03 21:46                 ` Will Deacon
2021-02-03 21:46                 ` Will Deacon
2021-02-03 21:46                 ` Will Deacon
2021-02-03 22:14                 ` Rob Clark
2021-02-03 22:14                   ` Rob Clark
2021-02-03 22:14                   ` Rob Clark
2021-02-03 22:14                   ` Rob Clark
2021-03-16 17:04                   ` Rob Clark
2021-03-16 17:04                     ` Rob Clark
2021-03-16 17:04                     ` Rob Clark
2021-03-16 17:04                     ` Rob Clark
2021-03-16 17:16                     ` Rob Clark
2021-03-16 17:16                       ` Rob Clark
2021-03-16 17:16                       ` Rob Clark
2021-03-16 17:16                       ` Rob Clark
2021-03-17  9:33                       ` Sai Prakash Ranjan [this message]
2021-03-25 18:36                         ` Rob Clark
2021-02-05 12:08                 ` Sai Prakash Ranjan
2021-02-05 12:08                   ` Sai Prakash Ranjan
2021-02-05 12:08                   ` Sai Prakash Ranjan
2021-03-09  6:40                   ` Sai Prakash Ranjan
2021-03-09  6:40                     ` Sai Prakash Ranjan
2021-03-09  6:40                     ` Sai Prakash Ranjan
2021-03-25 17:33                     ` Will Deacon
2021-03-25 17:33                       ` Will Deacon
2021-03-25 17:33                       ` Will Deacon
2021-03-25 17:33                       ` Will Deacon
2021-06-30 10:07                       ` Sai Prakash Ranjan
2021-06-30 10:07                         ` Sai Prakash Ranjan
2021-06-30 10:07                         ` Sai Prakash Ranjan
2021-02-02  6:28             ` Sai Prakash Ranjan
2021-02-02  6:28               ` Sai Prakash Ranjan
2021-02-02  6:28               ` Sai Prakash Ranjan
2021-01-11 14:15 ` [PATCH 3/3] drm/msm: Use IOMMU_LLC page protection flag to map gpu buffers Sai Prakash Ranjan
2021-01-11 14:15   ` Sai Prakash Ranjan
2021-01-11 14:15   ` Sai Prakash Ranjan
2021-01-20  5:18 ` [PATCH 0/3] iommu/drm/msm: Allow non-coherent masters to use system cache Sai Prakash Ranjan
2021-01-20  5:18   ` Sai Prakash Ranjan
2021-01-20  5:18   ` Sai Prakash Ranjan
2021-01-29  8:53   ` Sai Prakash Ranjan
2021-01-29  8:53     ` Sai Prakash Ranjan
2021-07-28 14:00 ` Georgi Djakov
2021-07-28 14:00   ` Georgi Djakov
2021-07-28 14:00   ` Georgi Djakov
2021-07-28 14:00   ` Georgi Djakov
2021-07-29  4:38   ` Sai Prakash Ranjan
2021-07-29  4:38     ` Sai Prakash Ranjan
2021-07-29  4:38     ` Sai Prakash Ranjan
2021-08-02 10:55     ` Will Deacon
2021-08-02 10:55       ` Will Deacon
2021-08-02 10:55       ` Will Deacon
2021-08-02 15:08       ` [Freedreno] " Rob Clark
2021-08-02 15:08         ` Rob Clark
2021-08-02 15:08         ` Rob Clark
2021-08-02 15:14         ` Will Deacon
2021-08-02 15:14           ` Will Deacon
2021-08-02 15:14           ` Will Deacon
2021-08-03  1:36           ` Rob Clark
2021-08-03  1:36             ` Rob Clark
2021-08-03  1:36             ` Rob Clark
2021-08-09 14:56             ` Will Deacon
2021-08-09 14:56               ` Will Deacon
2021-08-09 14:56               ` Will Deacon
2021-08-09 16:57               ` Rob Clark
2021-08-09 16:57                 ` Rob Clark
2021-08-09 16:57                 ` Rob Clark
2021-08-09 17:05                 ` Will Deacon
2021-08-09 17:05                   ` Will Deacon
2021-08-09 17:05                   ` Will Deacon
2021-08-09 17:18                   ` Rob Clark
2021-08-09 17:18                     ` Rob Clark
2021-08-09 17:18                     ` Rob Clark
2021-08-09 17:40                     ` Will Deacon
2021-08-09 17:40                       ` Will Deacon
2021-08-09 17:40                       ` Will Deacon
2021-08-09 17:47                       ` Sai Prakash Ranjan
2021-08-09 17:47                         ` Sai Prakash Ranjan
2021-08-09 18:07                         ` Rob Clark
2021-08-09 18:07                           ` Rob Clark
2021-08-09 18:07                           ` Rob Clark
2021-08-09 18:10                           ` Sai Prakash Ranjan
2021-08-09 18:10                             ` Sai Prakash Ranjan
2021-08-09 18:30                             ` Rob Clark
2021-08-09 18:30                               ` Rob Clark
2021-08-09 18:30                               ` Rob Clark
2021-08-09 18:32                               ` Sai Prakash Ranjan
2021-08-09 18:32                                 ` Sai Prakash Ranjan
2021-08-10  9:16                         ` Will Deacon
2021-08-10  9:16                           ` Will Deacon
2021-08-10  9:16                           ` Will Deacon
2021-08-10  9:54                           ` Sai Prakash Ranjan
2021-08-10  9:54                             ` Sai Prakash Ranjan

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=d44fc38c038be1165aa8f4212bd9c91f@codeaurora.org \
    --to=saiprakash.ranjan@codeaurora.org \
    --cc=akhilpo@codeaurora.org \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=isaacm@codeaurora.org \
    --cc=jordan@cosmicpenguin.net \
    --cc=joro@8bytes.org \
    --cc=robdclark@gmail.com \
    --cc=robin.murphy@arm.com \
    --cc=will@kernel.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.