From: "Christian König" <christian.koenig@amd.com>
To: "Matthew Brost" <matthew.brost@intel.com>,
"Christian König" <ckoenig.leichtzumerken@gmail.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>,
Huang Rui <ray.huang@amd.com>,
intel-xe@lists.freedesktop.org, dri-devel@lists.freedesktop.org,
matthew.auld@intel.com
Subject: Re: [PATCH v6 2/8] drm/ttm: Add ttm_bo_access
Date: Tue, 5 Nov 2024 08:41:07 +0100 [thread overview]
Message-ID: <c76830ae-c6f9-47a2-8087-32d9f7c3df2c@amd.com> (raw)
In-Reply-To: <ZylBWVjlW+GDYy5M@lstrano-desk.jf.intel.com>
[-- Attachment #1: Type: text/plain, Size: 8719 bytes --]
Am 04.11.24 um 22:49 schrieb Matthew Brost:
> On Mon, Nov 04, 2024 at 08:28:34PM +0100, Christian König wrote:
>> Am 04.11.24 um 18:34 schrieb Rodrigo Vivi:
>>> On Thu, Oct 31, 2024 at 04:43:19PM -0700, Matthew Brost wrote:
>>>> On Thu, Oct 31, 2024 at 11:10:42AM -0700, Matthew Brost wrote:
>>>>> Non-contiguous VRAM cannot easily be mapped in TTM nor can non-visible
>>>>> VRAM easily be accessed. Add ttm_bo_access, which is similar to
>>>>> ttm_bo_vm_access, to access such memory.
>>>>>
>>>>> v4:
>>>>> - Fix checkpatch warnings (CI)
>>>>> v5:
>>>>> - Fix checkpatch warnings (CI)
>>>>> v6:
>>>>> - Fix kernel doc (Auld)
>>>>>
>>>> Christian - Do you mind if I merge patch along with the rest of the
>>>> series to drm-xe-next?
>>> Ray, Christian,
>>>
>>> ack on getting this patch to drm-xe-next?
>> No, we actually spend quite some time removing the single page mapping
>> functionality for BOs.
>>
> I don't understand this statement. This patch just adds a TTM BO helper
> for access - it doesn't change anything wrt to single page mapping.
Well we spend quite some time removing single page mappings from device
drivers.
The only remaining use case of ttm_bo_kmap() with just one page is the
ttm_bo_vm_access_kmap() function and I was really hoping to make that
one TTM internal at some point.
>> You need a really good justification to bring that back.
>>
> The use case is EuDebugger requires essentially the same functionality
> as ptrace -> vm_access.
Then why don't you use ptrace in the first place?
> TTM mapping non-contiguous VRAM doesn't work unless I'm blind. User BOs
> which the EuDebugger accesses can be non-contiguous, hence the new
> helper.
Then why don't you handle that inside the driver in the first place
instead of going through a TTM midlayer?
Regards,
Christian.
>
> Matt
>
>> Regards,
>> Christian.
>>
>>>> Matt
>>>>
>>>>> Reported-by: Christoph Manszewski<christoph.manszewski@intel.com>
>>>>> Suggested-by: Thomas Hellström<thomas.hellstrom@linux.intel.com>
>>>>> Signed-off-by: Matthew Brost<matthew.brost@intel.com>
>>>>> Tested-by: Mika Kuoppala<mika.kuoppala@linux.intel.com>
>>>>> Reviewed-by: Matthew Auld<matthew.auld@intel.com>
>>>>> ---
>>>>> drivers/gpu/drm/ttm/ttm_bo_util.c | 86 +++++++++++++++++++++++++++++++
>>>>> drivers/gpu/drm/ttm/ttm_bo_vm.c | 65 +----------------------
>>>>> include/drm/ttm/ttm_bo.h | 2 +
>>>>> 3 files changed, 89 insertions(+), 64 deletions(-)
>>>>>
>>>>> diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c
>>>>> index d939925efa81..77e760ea7193 100644
>>>>> --- a/drivers/gpu/drm/ttm/ttm_bo_util.c
>>>>> +++ b/drivers/gpu/drm/ttm/ttm_bo_util.c
>>>>> @@ -919,3 +919,89 @@ s64 ttm_lru_walk_for_evict(struct ttm_lru_walk *walk, struct ttm_device *bdev,
>>>>> return progress;
>>>>> }
>>>>> +
>>>>> +static int ttm_bo_access_kmap(struct ttm_buffer_object *bo,
>>>>> + unsigned long offset,
>>>>> + void *buf, int len, int write)
>>>>> +{
>>>>> + unsigned long page = offset >> PAGE_SHIFT;
>>>>> + unsigned long bytes_left = len;
>>>>> + int ret;
>>>>> +
>>>>> + /* Copy a page at a time, that way no extra virtual address
>>>>> + * mapping is needed
>>>>> + */
>>>>> + offset -= page << PAGE_SHIFT;
>>>>> + do {
>>>>> + unsigned long bytes = min(bytes_left, PAGE_SIZE - offset);
>>>>> + struct ttm_bo_kmap_obj map;
>>>>> + void *ptr;
>>>>> + bool is_iomem;
>>>>> +
>>>>> + ret = ttm_bo_kmap(bo, page, 1, &map);
>>>>> + if (ret)
>>>>> + return ret;
>>>>> +
>>>>> + ptr = (void *)ttm_kmap_obj_virtual(&map, &is_iomem) + offset;
>>>>> + WARN_ON_ONCE(is_iomem);
>>>>> + if (write)
>>>>> + memcpy(ptr, buf, bytes);
>>>>> + else
>>>>> + memcpy(buf, ptr, bytes);
>>>>> + ttm_bo_kunmap(&map);
>>>>> +
>>>>> + page++;
>>>>> + buf += bytes;
>>>>> + bytes_left -= bytes;
>>>>> + offset = 0;
>>>>> + } while (bytes_left);
>>>>> +
>>>>> + return len;
>>>>> +}
>>>>> +
>>>>> +/**
>>>>> + * ttm_bo_access - Helper to access a buffer object
>>>>> + *
>>>>> + * @bo: ttm buffer object
>>>>> + * @offset: access offset into buffer object
>>>>> + * @buf: pointer to caller memory to read into or write from
>>>>> + * @len: length of access
>>>>> + * @write: write access
>>>>> + *
>>>>> + * Utility function to access a buffer object. Useful when buffer object cannot
>>>>> + * be easily mapped (non-contiguous, non-visible, etc...).
>>>>> + *
>>>>> + * Returns:
>>>>> + * @len if successful, negative error code on failure.
>>>>> + */
>>>>> +int ttm_bo_access(struct ttm_buffer_object *bo, unsigned long offset,
>>>>> + void *buf, int len, int write)
>>>>> +{
>>>>> + int ret;
>>>>> +
>>>>> + if (len < 1 || (offset + len) > bo->base.size)
>>>>> + return -EIO;
>>>>> +
>>>>> + ret = ttm_bo_reserve(bo, true, false, NULL);
>>>>> + if (ret)
>>>>> + return ret;
>>>>> +
>>>>> + switch (bo->resource->mem_type) {
>>>>> + case TTM_PL_SYSTEM:
>>>>> + fallthrough;
>>>>> + case TTM_PL_TT:
>>>>> + ret = ttm_bo_access_kmap(bo, offset, buf, len, write);
>>>>> + break;
>>>>> + default:
>>>>> + if (bo->bdev->funcs->access_memory)
>>>>> + ret = bo->bdev->funcs->access_memory
>>>>> + (bo, offset, buf, len, write);
>>>>> + else
>>>>> + ret = -EIO;
>>>>> + }
>>>>> +
>>>>> + ttm_bo_unreserve(bo);
>>>>> +
>>>>> + return ret;
>>>>> +}
>>>>> +EXPORT_SYMBOL(ttm_bo_access);
>>>>> diff --git a/drivers/gpu/drm/ttm/ttm_bo_vm.c b/drivers/gpu/drm/ttm/ttm_bo_vm.c
>>>>> index 2c699ed1963a..20b1e5f78684 100644
>>>>> --- a/drivers/gpu/drm/ttm/ttm_bo_vm.c
>>>>> +++ b/drivers/gpu/drm/ttm/ttm_bo_vm.c
>>>>> @@ -366,45 +366,6 @@ void ttm_bo_vm_close(struct vm_area_struct *vma)
>>>>> }
>>>>> EXPORT_SYMBOL(ttm_bo_vm_close);
>>>>> -static int ttm_bo_vm_access_kmap(struct ttm_buffer_object *bo,
>>>>> - unsigned long offset,
>>>>> - uint8_t *buf, int len, int write)
>>>>> -{
>>>>> - unsigned long page = offset >> PAGE_SHIFT;
>>>>> - unsigned long bytes_left = len;
>>>>> - int ret;
>>>>> -
>>>>> - /* Copy a page at a time, that way no extra virtual address
>>>>> - * mapping is needed
>>>>> - */
>>>>> - offset -= page << PAGE_SHIFT;
>>>>> - do {
>>>>> - unsigned long bytes = min(bytes_left, PAGE_SIZE - offset);
>>>>> - struct ttm_bo_kmap_obj map;
>>>>> - void *ptr;
>>>>> - bool is_iomem;
>>>>> -
>>>>> - ret = ttm_bo_kmap(bo, page, 1, &map);
>>>>> - if (ret)
>>>>> - return ret;
>>>>> -
>>>>> - ptr = (uint8_t *)ttm_kmap_obj_virtual(&map, &is_iomem) + offset;
>>>>> - WARN_ON_ONCE(is_iomem);
>>>>> - if (write)
>>>>> - memcpy(ptr, buf, bytes);
>>>>> - else
>>>>> - memcpy(buf, ptr, bytes);
>>>>> - ttm_bo_kunmap(&map);
>>>>> -
>>>>> - page++;
>>>>> - buf += bytes;
>>>>> - bytes_left -= bytes;
>>>>> - offset = 0;
>>>>> - } while (bytes_left);
>>>>> -
>>>>> - return len;
>>>>> -}
>>>>> -
>>>>> int ttm_bo_vm_access(struct vm_area_struct *vma, unsigned long addr,
>>>>> void *buf, int len, int write)
>>>>> {
>>>>> @@ -412,32 +373,8 @@ int ttm_bo_vm_access(struct vm_area_struct *vma, unsigned long addr,
>>>>> unsigned long offset = (addr) - vma->vm_start +
>>>>> ((vma->vm_pgoff - drm_vma_node_start(&bo->base.vma_node))
>>>>> << PAGE_SHIFT);
>>>>> - int ret;
>>>>> -
>>>>> - if (len < 1 || (offset + len) > bo->base.size)
>>>>> - return -EIO;
>>>>> - ret = ttm_bo_reserve(bo, true, false, NULL);
>>>>> - if (ret)
>>>>> - return ret;
>>>>> -
>>>>> - switch (bo->resource->mem_type) {
>>>>> - case TTM_PL_SYSTEM:
>>>>> - fallthrough;
>>>>> - case TTM_PL_TT:
>>>>> - ret = ttm_bo_vm_access_kmap(bo, offset, buf, len, write);
>>>>> - break;
>>>>> - default:
>>>>> - if (bo->bdev->funcs->access_memory)
>>>>> - ret = bo->bdev->funcs->access_memory(
>>>>> - bo, offset, buf, len, write);
>>>>> - else
>>>>> - ret = -EIO;
>>>>> - }
>>>>> -
>>>>> - ttm_bo_unreserve(bo);
>>>>> -
>>>>> - return ret;
>>>>> + return ttm_bo_access(bo, offset, buf, len, write);
>>>>> }
>>>>> EXPORT_SYMBOL(ttm_bo_vm_access);
>>>>> diff --git a/include/drm/ttm/ttm_bo.h b/include/drm/ttm/ttm_bo.h
>>>>> index 5804408815be..8ea11cd8df39 100644
>>>>> --- a/include/drm/ttm/ttm_bo.h
>>>>> +++ b/include/drm/ttm/ttm_bo.h
>>>>> @@ -421,6 +421,8 @@ void ttm_bo_unpin(struct ttm_buffer_object *bo);
>>>>> int ttm_bo_evict_first(struct ttm_device *bdev,
>>>>> struct ttm_resource_manager *man,
>>>>> struct ttm_operation_ctx *ctx);
>>>>> +int ttm_bo_access(struct ttm_buffer_object *bo, unsigned long offset,
>>>>> + void *buf, int len, int write);
>>>>> vm_fault_t ttm_bo_vm_reserve(struct ttm_buffer_object *bo,
>>>>> struct vm_fault *vmf);
>>>>> vm_fault_t ttm_bo_vm_fault_reserved(struct vm_fault *vmf,
>>>>> --
>>>>> 2.34.1
>>>>>
[-- Attachment #2: Type: text/html, Size: 10076 bytes --]
next prev parent reply other threads:[~2024-11-05 7:41 UTC|newest]
Thread overview: 56+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-10-31 18:10 [PATCH v6 0/8] Fix non-contiguous VRAM BO access in Xe Matthew Brost
2024-10-31 18:10 ` [PATCH v6 1/8] drm/xe: Add xe_bo_vm_access Matthew Brost
2024-10-31 18:10 ` [PATCH v6 2/8] drm/ttm: Add ttm_bo_access Matthew Brost
2024-10-31 23:43 ` Matthew Brost
2024-11-04 17:34 ` Rodrigo Vivi
2024-11-04 19:28 ` Christian König
2024-11-04 21:49 ` Matthew Brost
2024-11-05 7:41 ` Christian König [this message]
2024-11-05 18:35 ` Matthew Brost
2024-11-06 9:48 ` Christian König
2024-11-06 15:25 ` Matthew Brost
2024-11-06 15:44 ` Christian König
2024-11-06 17:00 ` Matthew Brost
2024-11-07 9:44 ` Christian König
2024-11-11 8:00 ` Joonas Lahtinen
2024-11-11 10:10 ` Simona Vetter
2024-11-11 11:34 ` Christian König
2024-11-11 14:00 ` Joonas Lahtinen
2024-11-11 15:54 ` Christian König
2024-11-11 22:45 ` Matthew Brost
2024-11-12 9:23 ` Christian König
2024-11-12 13:41 ` Joonas Lahtinen
2024-11-12 16:22 ` Thomas Hellström
2024-11-12 16:25 ` Christian König
2024-11-12 16:33 ` Thomas Hellström
2024-11-13 8:37 ` Christian König
2024-11-13 10:44 ` Thomas Hellström
2024-11-13 11:42 ` Christian König
2024-11-15 18:27 ` Matthew Brost
2024-11-25 15:29 ` Matthew Brost
2024-11-25 16:19 ` Christian König
2024-11-25 17:27 ` Matthew Brost
2024-11-26 8:19 ` Christian König
2024-11-26 17:49 ` Matthew Brost
2024-11-27 13:21 ` Christian König
2024-11-12 8:28 ` Simona Vetter
2024-11-12 8:58 ` Christian König
2024-11-12 13:30 ` Joonas Lahtinen
2024-11-11 11:27 ` Christian König
2024-11-04 19:47 ` Christian König
2024-11-04 21:30 ` Matthew Brost
2024-11-04 22:26 ` Rodrigo Vivi
2024-10-31 18:10 ` [PATCH v6 3/8] drm/xe: Add xe_ttm_access_memory Matthew Brost
2024-10-31 18:10 ` [PATCH v6 4/8] drm/xe: Take PM ref in delayed snapshot capture worker Matthew Brost
2024-10-31 18:10 ` [PATCH v6 5/8] drm/xe/display: Update intel_bo_read_from_page to use ttm_bo_access Matthew Brost
2024-10-31 18:10 ` [PATCH v6 6/8] drm/xe: Use ttm_bo_access in xe_vm_snapshot_capture_delayed Matthew Brost
2024-10-31 18:10 ` [PATCH v6 7/8] drm/xe: Set XE_BO_FLAG_PINNED in migrate selftest BOs Matthew Brost
2024-10-31 18:10 ` [PATCH v6 8/8] drm/xe: Only allow contiguous BOs to use xe_bo_vmap Matthew Brost
2024-10-31 18:15 ` ✓ CI.Patch_applied: success for Fix non-contiguous VRAM BO access in Xe (rev6) Patchwork
2024-10-31 18:15 ` ✗ CI.checkpatch: warning " Patchwork
2024-10-31 18:17 ` ✓ CI.KUnit: success " Patchwork
2024-10-31 18:28 ` ✓ CI.Build: " Patchwork
2024-10-31 18:31 ` ✓ CI.Hooks: " Patchwork
2024-10-31 18:32 ` ✗ CI.checksparse: warning " Patchwork
2024-10-31 18:57 ` ✓ CI.BAT: success " Patchwork
2024-10-31 21:27 ` ✗ CI.FULL: failure " Patchwork
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=c76830ae-c6f9-47a2-8087-32d9f7c3df2c@amd.com \
--to=christian.koenig@amd.com \
--cc=ckoenig.leichtzumerken@gmail.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=intel-xe@lists.freedesktop.org \
--cc=matthew.auld@intel.com \
--cc=matthew.brost@intel.com \
--cc=ray.huang@amd.com \
--cc=rodrigo.vivi@intel.com \
/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