* [PATCH 0/6] Implement compression support on BMG
@ 2024-07-11 9:18 Akshata Jahagirdar
2024-07-11 9:19 ` Akshata Jahagirdar
2024-07-12 3:11 ` [PATCH 1/6] drm/xe/migrate: Handle clear ccs logic for xe2 dgfx Akshata Jahagirdar
0 siblings, 2 replies; 14+ messages in thread
From: Akshata Jahagirdar @ 2024-07-11 9:18 UTC (permalink / raw)
To: intel-xe
Cc: matthew.d.roper, matthew.auld, himal.prasad.ghimiray,
lucas.demarchi, Akshata Jahagirdar
Xe2+ has unified compression (exactly one compression mode/format),
where compression is now controlled via PAT at PTE level.
This simplifies KMD operations, as it can now decompress freely
without concern for the buffer's original compression format—unlike DG2,
which had multiple compression formats and thus required copying the
raw CCS state during VRAM eviction. In addition mixed VRAM and system
memory buffers were not supported with compression enabled.
On Xe2 dGPU compression is still only supported with VRAM, however we
can now support compression with VRAM and system memory buffers,
with GPU access being seamless underneath. So long as when doing
VRAM -> system memory the KMD uses compressed -> uncompressed,
to decompress it. This also allows CPU access to such buffers,
assuming that userspace first decompress the corresponding
pages being accessed.
If the pages are already in system memory then KMD would have already
decompressed them. When restoring such buffers with sysmem -> VRAM
the KMD can't easily know which pages were originally compressed,
so we always use uncompressed -> uncompressed here.
With this it also means we can drop all the raw CCS handling on such
platforms (including needing to allocate extra CCS storage).
In order to support this we now need to have two different identity
mappings for compressed and uncompressed VRAM.
In this patch, we set up the additional identity map for the VRAM with
compressed pat_index. We then select the appropriate mapping during
migration/clear.During eviction (vram->sysmem), we use the mapping from compressed -> uncompressed.
During restore (sysmem->vram), we need the mapping from uncompressed -> uncompressed.
Therefore, we need to have two different mappings for compressed and uncompressed
vram. We set up an additional identity map for the vram with compressed
pat_index. We then select the appropriate mapping during migration/clear.
Akshata Jahagirdar (6):
drm/xe/xe2: Introduce identity map for compressed pat for vram
drm/xe/migrate: Handle clear ccs logic for xe2 dgfx
drm/xe/migrate: Add kunit to test clear functionality
drm/xe/xe_migrate: Handle migration logic for xe2+ dgfx
drm/xe/migrate: Add kunit to test migration functionality for BMG
drm/xe/xe2: Do not run xe_bo_test for xe2+ dgfx
drivers/gpu/drm/xe/tests/xe_bo.c | 6 +
drivers/gpu/drm/xe/tests/xe_migrate.c | 389 +++++++++++++++++++++
drivers/gpu/drm/xe/tests/xe_migrate_test.c | 1 +
drivers/gpu/drm/xe/tests/xe_migrate_test.h | 1 +
drivers/gpu/drm/xe/xe_device.h | 5 +
drivers/gpu/drm/xe/xe_migrate.c | 66 +++-
6 files changed, 450 insertions(+), 18 deletions(-)
--
2.34.1
^ permalink raw reply [flat|nested] 14+ messages in thread
* [PATCH 0/6] Implement compression support on BMG
2024-07-11 9:18 [PATCH 0/6] Implement compression support on BMG Akshata Jahagirdar
@ 2024-07-11 9:19 ` Akshata Jahagirdar
2024-07-12 3:11 ` [PATCH 1/6] drm/xe/migrate: Handle clear ccs logic for xe2 dgfx Akshata Jahagirdar
1 sibling, 0 replies; 14+ messages in thread
From: Akshata Jahagirdar @ 2024-07-11 9:19 UTC (permalink / raw)
To: intel-xe; +Cc: akshatajahagirdar6, Akshata Jahagirdar
Xe2+ has unified compression (exactly one compression mode/format),
where compression is now controlled via PAT at PTE level.
This simplifies KMD operations, as it can now decompress freely
without concern for the buffer's original compression format—unlike DG2,
which had multiple compression formats and thus required copying the
raw CCS state during VRAM eviction. In addition mixed VRAM and system
memory buffers were not supported with compression enabled.
On Xe2 dGPU compression is still only supported with VRAM, however we
can now support compression with VRAM and system memory buffers,
with GPU access being seamless underneath. So long as when doing
VRAM -> system memory the KMD uses compressed -> uncompressed,
to decompress it. This also allows CPU access to such buffers,
assuming that userspace first decompress the corresponding
pages being accessed.
If the pages are already in system memory then KMD would have already
decompressed them. When restoring such buffers with sysmem -> VRAM
the KMD can't easily know which pages were originally compressed,
so we always use uncompressed -> uncompressed here.
With this it also means we can drop all the raw CCS handling on such
platforms (including needing to allocate extra CCS storage).
In order to support this we now need to have two different identity
mappings for compressed and uncompressed VRAM.
In this patch, we set up the additional identity map for the VRAM with
compressed pat_index. We then select the appropriate mapping during
migration/clear.During eviction (vram->sysmem), we use the mapping from compressed -> uncompressed.
During restore (sysmem->vram), we need the mapping from uncompressed -> uncompressed.
Therefore, we need to have two different mappings for compressed and uncompressed
vram. We set up an additional identity map for the vram with compressed
pat_index. We then select the appropriate mapping during migration/clear.
Akshata Jahagirdar (6):
drm/xe/xe2: Introduce identity map for compressed pat for vram
drm/xe/migrate: Handle clear ccs logic for xe2 dgfx
drm/xe/migrate: Add kunit to test clear functionality
drm/xe/xe_migrate: Handle migration logic for xe2+ dgfx
drm/xe/migrate: Add kunit to test migration functionality for BMG
drm/xe/xe2: Do not run xe_bo_test for xe2+ dgfx
drivers/gpu/drm/xe/tests/xe_bo.c | 6 +
drivers/gpu/drm/xe/tests/xe_migrate.c | 389 +++++++++++++++++++++
drivers/gpu/drm/xe/tests/xe_migrate_test.c | 1 +
drivers/gpu/drm/xe/tests/xe_migrate_test.h | 1 +
drivers/gpu/drm/xe/xe_device.h | 5 +
drivers/gpu/drm/xe/xe_migrate.c | 66 +++-
6 files changed, 450 insertions(+), 18 deletions(-)
--
2.34.1
^ permalink raw reply [flat|nested] 14+ messages in thread
* [PATCH 1/6] drm/xe/migrate: Handle clear ccs logic for xe2 dgfx
2024-07-11 9:18 [PATCH 0/6] Implement compression support on BMG Akshata Jahagirdar
2024-07-11 9:19 ` Akshata Jahagirdar
@ 2024-07-12 3:11 ` Akshata Jahagirdar
1 sibling, 0 replies; 14+ messages in thread
From: Akshata Jahagirdar @ 2024-07-12 3:11 UTC (permalink / raw)
To: intel-xe; +Cc: san0582, Jahagirdar, Akshata
From: "Jahagirdar, Akshata" <akshata.jahagirdar@intel.com>
For Xe2 dGPU, we clear the bo by modifying the VRAM using an
uncompressed pat index which then indirectly updates the
compression status as uncompressed i.e zeroed CCS.
So xe_migrate_clear() should be updated for BMG to not
emit CCS surf copy commands.
Signed-off-by: Akshata Jahagirdar <akshata.jahagirdar@intel.com>
---
drivers/gpu/drm/xe/xe_device.h | 5 +++++
drivers/gpu/drm/xe/xe_migrate.c | 6 +++---
2 files changed, 8 insertions(+), 3 deletions(-)
diff --git a/drivers/gpu/drm/xe/xe_device.h b/drivers/gpu/drm/xe/xe_device.h
index 0a2a3e7fd402..c3093506c28c 100644
--- a/drivers/gpu/drm/xe/xe_device.h
+++ b/drivers/gpu/drm/xe/xe_device.h
@@ -144,6 +144,11 @@ static inline bool xe_device_has_flat_ccs(struct xe_device *xe)
return xe->info.has_flat_ccs;
}
+static inline bool xe_device_needs_ccs_emit(struct xe_device *xe)
+{
+ return xe_device_has_flat_ccs(xe) && !(GRAPHICS_VER(xe) >= 20 && IS_DGFX(xe));
+}
+
static inline bool xe_device_has_sriov(struct xe_device *xe)
{
return xe->info.has_sriov;
diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c
index fa23a7e7ec43..2fc2cf375b1e 100644
--- a/drivers/gpu/drm/xe/xe_migrate.c
+++ b/drivers/gpu/drm/xe/xe_migrate.c
@@ -420,7 +420,7 @@ struct xe_migrate *xe_migrate_init(struct xe_tile *tile)
return ERR_PTR(err);
if (IS_DGFX(xe)) {
- if (xe_device_has_flat_ccs(xe))
+ if (xe_device_needs_ccs_emit(xe))
/* min chunk size corresponds to 4K of CCS Metadata */
m->min_chunk_size = SZ_4K * SZ_64K /
xe_device_ccs_bytes(xe, SZ_64K);
@@ -1034,7 +1034,7 @@ struct dma_fence *xe_migrate_clear(struct xe_migrate *m,
clear_system_ccs ? 0 : emit_clear_cmd_len(gt), 0,
avail_pts);
- if (xe_device_has_flat_ccs(xe))
+ if (xe_device_needs_ccs_emit(xe))
batch_size += EMIT_COPY_CCS_DW;
/* Clear commands */
@@ -1062,7 +1062,7 @@ struct dma_fence *xe_migrate_clear(struct xe_migrate *m,
if (!clear_system_ccs)
emit_clear(gt, bb, clear_L0_ofs, clear_L0, XE_PAGE_SIZE, clear_vram);
- if (xe_device_has_flat_ccs(xe)) {
+ if (xe_device_needs_ccs_emit(xe)) {
emit_copy_ccs(gt, bb, clear_L0_ofs, true,
m->cleared_mem_ofs, false, clear_L0);
flush_flags = MI_FLUSH_DW_CCS;
--
2.34.1
^ permalink raw reply related [flat|nested] 14+ messages in thread
* [PATCH 0/6] Implement compression support on BMG
@ 2024-07-12 6:39 Akshata Jahagirdar
2024-07-12 7:24 ` Akshata Jahagirdar
0 siblings, 1 reply; 14+ messages in thread
From: Akshata Jahagirdar @ 2024-07-12 6:39 UTC (permalink / raw)
To: intel-xe; +Cc: akshatajahagirdar6, Akshata Jahagirdar
Xe2+ has unified compression (exactly one compression mode/format),
where compression is now controlled via PAT at PTE level.
This simplifies KMD operations, as it can now decompress freely
without concern for the buffer's original compression format—unlike DG2,
which had multiple compression formats and thus required copying the
raw CCS state during VRAM eviction. In addition mixed VRAM and system
memory buffers were not supported with compression enabled.
On Xe2 dGPU compression is still only supported with VRAM, however we
can now support compression with VRAM and system memory buffers,
with GPU access being seamless underneath. So long as when doing
VRAM -> system memory the KMD uses compressed -> uncompressed,
to decompress it. This also allows CPU access to such buffers,
assuming that userspace first decompress the corresponding
pages being accessed.
If the pages are already in system memory then KMD would have already
decompressed them. When restoring such buffers with sysmem -> VRAM
the KMD can't easily know which pages were originally compressed,
so we always use uncompressed -> uncompressed here.
With this it also means we can drop all the raw CCS handling on such
platforms (including needing to allocate extra CCS storage).
In order to support this we now need to have two different identity
mappings for compressed and uncompressed VRAM.
In this patch, we set up the additional identity map for the VRAM with
compressed pat_index. We then select the appropriate mapping during
migration/clear. During eviction (vram->sysmem), we use the mapping
from compressed -> uncompressed. During restore (sysmem->vram), we need
the mapping from uncompressed -> uncompressed.
Therefore, we need to have two different mappings for compressed and
uncompressed vram. We set up an additional identity map for the vram
with compressed pat_index.
We then select the appropriate mapping during migration/clear.
Akshata Jahagirdar (6):
drm/xe/xe2: Introduce identity map for compressed pat for vram
drm/xe/migrate: Handle clear ccs logic for xe2 dgfx
drm/xe/migrate: Add kunit to test clear functionality
drm/xe/xe_migrate: Handle migration logic for xe2+ dgfx
drm/xe/migrate: Add kunit to test migration functionality for BMG
drm/xe/xe2: Do not run xe_bo_test for xe2+ dgfx
drivers/gpu/drm/xe/tests/xe_bo.c | 6 +
drivers/gpu/drm/xe/tests/xe_migrate.c | 389 +++++++++++++++++++++
drivers/gpu/drm/xe/tests/xe_migrate_test.c | 1 +
drivers/gpu/drm/xe/tests/xe_migrate_test.h | 1 +
drivers/gpu/drm/xe/xe_migrate.c | 91 ++++-
5 files changed, 470 insertions(+), 18 deletions(-)
--
2.34.1
^ permalink raw reply [flat|nested] 14+ messages in thread
* [PATCH 0/6] Implement compression support on BMG
2024-07-12 6:39 [PATCH 0/6] Implement compression support on BMG Akshata Jahagirdar
@ 2024-07-12 7:24 ` Akshata Jahagirdar
0 siblings, 0 replies; 14+ messages in thread
From: Akshata Jahagirdar @ 2024-07-12 7:24 UTC (permalink / raw)
To: intel-xe; +Cc: akshatajahagirdar6, Akshata Jahagirdar
Xe2+ has unified compression (exactly one compression mode/format),
where compression is now controlled via PAT at PTE level.
This simplifies KMD operations, as it can now decompress freely
without concern for the buffer's original compression format—unlike DG2,
which had multiple compression formats and thus required copying the
raw CCS state during VRAM eviction. In addition mixed VRAM and system
memory buffers were not supported with compression enabled.
On Xe2 dGPU compression is still only supported with VRAM, however we
can now support compression with VRAM and system memory buffers,
with GPU access being seamless underneath. So long as when doing
VRAM -> system memory the KMD uses compressed -> uncompressed,
to decompress it. This also allows CPU access to such buffers,
assuming that userspace first decompress the corresponding
pages being accessed.
If the pages are already in system memory then KMD would have already
decompressed them. When restoring such buffers with sysmem -> VRAM
the KMD can't easily know which pages were originally compressed,
so we always use uncompressed -> uncompressed here.
With this it also means we can drop all the raw CCS handling on such
platforms (including needing to allocate extra CCS storage).
In order to support this we now need to have two different identity
mappings for compressed and uncompressed VRAM.
In this patch, we set up the additional identity map for the VRAM with
compressed pat_index. We then select the appropriate mapping during
migration/clear. During eviction (vram->sysmem), we use the mapping
from compressed -> uncompressed. During restore (sysmem->vram), we need
the mapping from uncompressed -> uncompressed.
Therefore, we need to have two different mappings for compressed and
uncompressed vram. We set up an additional identity map for the vram
with compressed pat_index.
We then select the appropriate mapping during migration/clear.
Akshata Jahagirdar (6):
drm/xe/xe2: Introduce identity map for compressed pat for vram
drm/xe/migrate: Handle clear ccs logic for xe2 dgfx
drm/xe/migrate: Add kunit to test clear functionality
drm/xe/xe_migrate: Handle migration logic for xe2+ dgfx
drm/xe/migrate: Add kunit to test migration functionality for BMG
drm/xe/xe2: Do not run xe_bo_test for xe2+ dgfx
drivers/gpu/drm/xe/tests/xe_bo.c | 6 +
drivers/gpu/drm/xe/tests/xe_migrate.c | 389 +++++++++++++++++++++
drivers/gpu/drm/xe/tests/xe_migrate_test.c | 1 +
drivers/gpu/drm/xe/tests/xe_migrate_test.h | 1 +
drivers/gpu/drm/xe/xe_migrate.c | 91 ++++-
5 files changed, 470 insertions(+), 18 deletions(-)
--
2.34.1
^ permalink raw reply [flat|nested] 14+ messages in thread
* [PATCH 0/6] Implement compression support on BMG
@ 2024-07-10 7:53 Akshata Jahagirdar
2024-07-09 10:33 ` Matthew Auld
` (6 more replies)
0 siblings, 7 replies; 14+ messages in thread
From: Akshata Jahagirdar @ 2024-07-10 7:53 UTC (permalink / raw)
To: intel-xe; +Cc: lucas.demarchi, lucas.demarchi, Akshata Jahagirdar
According to the SAS for BMG compression, we need to decompress during eviction,
and not recompress on restore. Due to this, we need to introduce encoding pat_index
in case of vram too. This patch explores the solution of setting up an additional
identity map for the vram, this time at the end of previous mapping offset and
with compressed pat_index.
We then select the appropriate mapping during eviction/restore/clear.
Akshata Jahagirdar (6):
drm/xe/xe2: Introduce identity map for compressed pat for vram
drm/xe/migrate: Handle clear ccs logic for xe2 dgfx
drm/xe/migrate: Add kunit to test clear functionality
drm/xe/xe_migrate: Handle migration logic for xe2+ dgfx
drm/xe/migrate: Add kunit to test migration functionality for BMG
drm/xe/xe2: Do not run xe_bo_test for xe2+ dgfx
drivers/gpu/drm/xe/tests/xe_bo.c | 6 +
drivers/gpu/drm/xe/tests/xe_migrate.c | 388 +++++++++++++++++++++
drivers/gpu/drm/xe/tests/xe_migrate_test.c | 1 +
drivers/gpu/drm/xe/tests/xe_migrate_test.h | 1 +
drivers/gpu/drm/xe/xe_device.h | 5 +
drivers/gpu/drm/xe/xe_migrate.c | 66 +++-
6 files changed, 449 insertions(+), 18 deletions(-)
--
2.34.1
^ permalink raw reply [flat|nested] 14+ messages in thread* Re: [PATCH 0/6] Implement compression support on BMG
2024-07-10 7:53 Akshata Jahagirdar
@ 2024-07-09 10:33 ` Matthew Auld
2024-07-09 19:07 ` Jahagirdar, Akshata
2024-07-10 8:17 ` Akshata Jahagirdar
` (5 subsequent siblings)
6 siblings, 1 reply; 14+ messages in thread
From: Matthew Auld @ 2024-07-09 10:33 UTC (permalink / raw)
To: Akshata Jahagirdar, intel-xe
Cc: matthew.d.roper, himal.prasad.ghimiray, lucas.demarchi
On 09/07/2024 11:49, Akshata Jahagirdar wrote:
> According to the SAS for BMG compression, we need to decompress during eviction,
> and not recompress on restore. Due to this, we need to introduce encoding pat_index
> in case of vram too. This patch explores the solution of setting up an additional
> identity map for the vram, this time at the end of previous mapping offset and
> with compressed pat_index.
> We then select the appropriate mapping during eviction/restore/clear.
Hey, I don't see this series on the ml. Are you subscribed?
>
> Akshata Jahagirdar (6):
> drm/xe/xe2: Introduce identity map for compressed pat for vram
> drm/xe/migrate: Handle clear ccs logic for xe2 dgfx
> drm/xe/migrate: Add kunit to test clear functionality
> drm/xe/xe_migrate: Handle migration logic for xe2+ dgfx
> drm/xe/migrate: Add kunit to test migration functionality for BMG
> drm/xe/xe2: Do not run xe_bo_test for xe2+ dgfx
>
> drivers/gpu/drm/xe/tests/xe_bo.c | 6 +
> drivers/gpu/drm/xe/tests/xe_migrate.c | 388 +++++++++++++++++++++
> drivers/gpu/drm/xe/tests/xe_migrate_test.c | 1 +
> drivers/gpu/drm/xe/tests/xe_migrate_test.h | 1 +
> drivers/gpu/drm/xe/xe_device.h | 5 +
> drivers/gpu/drm/xe/xe_migrate.c | 66 +++-
> 6 files changed, 449 insertions(+), 18 deletions(-)
>
^ permalink raw reply [flat|nested] 14+ messages in thread* Re: [PATCH 0/6] Implement compression support on BMG
2024-07-09 10:33 ` Matthew Auld
@ 2024-07-09 19:07 ` Jahagirdar, Akshata
0 siblings, 0 replies; 14+ messages in thread
From: Jahagirdar, Akshata @ 2024-07-09 19:07 UTC (permalink / raw)
To: Matthew Auld, intel-xe
Cc: matthew.d.roper, himal.prasad.ghimiray, lucas.demarchi
On 7/9/2024 3:33 AM, Matthew Auld wrote:
> On 09/07/2024 11:49, Akshata Jahagirdar wrote:
>> According to the SAS for BMG compression, we need to decompress
>> during eviction,
>> and not recompress on restore. Due to this, we need to introduce
>> encoding pat_index
>> in case of vram too. This patch explores the solution of setting up
>> an additional
>> identity map for the vram, this time at the end of previous mapping
>> offset and
>> with compressed pat_index.
>> We then select the appropriate mapping during eviction/restore/clear.
>
> Hey, I don't see this series on the ml. Are you subscribed?
Hey Matt,
Yes, I am.. seems like there is some issue with my mailing list
subscription,
so its not getting reflected on the website. I just resubscribed , would
send this out again.
>
>>
>> Akshata Jahagirdar (6):
>> drm/xe/xe2: Introduce identity map for compressed pat for vram
>> drm/xe/migrate: Handle clear ccs logic for xe2 dgfx
>> drm/xe/migrate: Add kunit to test clear functionality
>> drm/xe/xe_migrate: Handle migration logic for xe2+ dgfx
>> drm/xe/migrate: Add kunit to test migration functionality for BMG
>> drm/xe/xe2: Do not run xe_bo_test for xe2+ dgfx
>>
>> drivers/gpu/drm/xe/tests/xe_bo.c | 6 +
>> drivers/gpu/drm/xe/tests/xe_migrate.c | 388 +++++++++++++++++++++
>> drivers/gpu/drm/xe/tests/xe_migrate_test.c | 1 +
>> drivers/gpu/drm/xe/tests/xe_migrate_test.h | 1 +
>> drivers/gpu/drm/xe/xe_device.h | 5 +
>> drivers/gpu/drm/xe/xe_migrate.c | 66 +++-
>> 6 files changed, 449 insertions(+), 18 deletions(-)
>>
^ permalink raw reply [flat|nested] 14+ messages in thread
* [PATCH 0/6] Implement compression support on BMG
2024-07-10 7:53 Akshata Jahagirdar
2024-07-09 10:33 ` Matthew Auld
@ 2024-07-10 8:17 ` Akshata Jahagirdar
2024-07-11 5:54 ` Akshata Jahagirdar
` (4 subsequent siblings)
6 siblings, 0 replies; 14+ messages in thread
From: Akshata Jahagirdar @ 2024-07-10 8:17 UTC (permalink / raw)
To: intel-xe; +Cc: lucas.demarchi, lucas.demarchi, Akshata Jahagirdar
According to the SAS for BMG compression, we need to decompress during eviction,
and not recompress on restore. Due to this, we need to introduce encoding pat_index
in case of vram too. This patch explores the solution of setting up an additional
identity map for the vram, this time at the end of previous mapping offset and
with compressed pat_index.
We then select the appropriate mapping during eviction/restore/clear.
Akshata Jahagirdar (6):
drm/xe/xe2: Introduce identity map for compressed pat for vram
drm/xe/migrate: Handle clear ccs logic for xe2 dgfx
drm/xe/migrate: Add kunit to test clear functionality
drm/xe/xe_migrate: Handle migration logic for xe2+ dgfx
drm/xe/migrate: Add kunit to test migration functionality for BMG
drm/xe/xe2: Do not run xe_bo_test for xe2+ dgfx
drivers/gpu/drm/xe/tests/xe_bo.c | 6 +
drivers/gpu/drm/xe/tests/xe_migrate.c | 388 +++++++++++++++++++++
drivers/gpu/drm/xe/tests/xe_migrate_test.c | 1 +
drivers/gpu/drm/xe/tests/xe_migrate_test.h | 1 +
drivers/gpu/drm/xe/xe_device.h | 5 +
drivers/gpu/drm/xe/xe_migrate.c | 66 +++-
6 files changed, 449 insertions(+), 18 deletions(-)
--
2.34.1
^ permalink raw reply [flat|nested] 14+ messages in thread* [PATCH 0/6] Implement compression support on BMG
2024-07-10 7:53 Akshata Jahagirdar
2024-07-09 10:33 ` Matthew Auld
2024-07-10 8:17 ` Akshata Jahagirdar
@ 2024-07-11 5:54 ` Akshata Jahagirdar
2024-07-11 11:27 ` Akshata Jahagirdar
` (3 subsequent siblings)
6 siblings, 0 replies; 14+ messages in thread
From: Akshata Jahagirdar @ 2024-07-11 5:54 UTC (permalink / raw)
To: intel-xe; +Cc: akshatajahagirdar6, Akshata Jahagirdar
On Xe2 the compression has moved to a unified universal model
(exactly one compression mode/format), where compression is now
controlled via PAT on per-page basis. This now means KMD can
decompress freely. This was problematic on DG2 since we had
multiple compression formats, and the compression format used
on a particular buffer was unknown to the KMD, so instead the
raw CCS state needed to be copied around when evicting VRAM.
In addition mixed VRAM and system memory buffers were not
supported with compression enabled.
On Xe2 dGPU compression is still only supported with VRAM,
however we can now support compression with VRAM and system
memory buffers, with GPU access being seamless underneath.
So long as when doing VRAM -> sysmem the KMD does the move
using compressed -> uncompressed, to decompress it.
CPU access to such buffers is also possible, under the premise
that userspace first decompress the corresponding pages being
accessed. If the pages are already in system memory then KMD would
have already decompressed them. When restoring such buffers with
sysmem -> VRAM the KMD can't easily know which pages were originally
compressed, so we always use uncompressed -> uncompressed here.
With this it also means we can drop all the raw CCS handling
on such platforms (including needing to allocate extra CCS storage).
In order to support this we now need to have two different identity
mappings for compressed and uncompressed VRAM.
The additional identity map is the VRAM with compressed pat_index.
We then select the appropriate mapping during migration/clear.
Akshata Jahagirdar (6):
drm/xe/xe2: Introduce identity map for compressed pat for vram
drm/xe/migrate: Handle clear ccs logic for xe2 dgfx
drm/xe/migrate: Add kunit to test clear functionality
drm/xe/xe_migrate: Handle migration logic for xe2+ dgfx
drm/xe/migrate: Add kunit to test migration functionality for BMG
drm/xe/xe2: Do not run xe_bo_test for xe2+ dgfx
drivers/gpu/drm/xe/tests/xe_bo.c | 6 +
drivers/gpu/drm/xe/tests/xe_migrate.c | 388 +++++++++++++++++++++
drivers/gpu/drm/xe/tests/xe_migrate_test.c | 1 +
drivers/gpu/drm/xe/tests/xe_migrate_test.h | 1 +
drivers/gpu/drm/xe/xe_device.h | 5 +
drivers/gpu/drm/xe/xe_migrate.c | 66 +++-
6 files changed, 449 insertions(+), 18 deletions(-)
--
2.34.1
^ permalink raw reply [flat|nested] 14+ messages in thread* [PATCH 0/6] Implement compression support on BMG
2024-07-10 7:53 Akshata Jahagirdar
` (2 preceding siblings ...)
2024-07-11 5:54 ` Akshata Jahagirdar
@ 2024-07-11 11:27 ` Akshata Jahagirdar
2024-07-11 12:34 ` Ghimiray, Himal Prasad
` (2 subsequent siblings)
6 siblings, 0 replies; 14+ messages in thread
From: Akshata Jahagirdar @ 2024-07-11 11:27 UTC (permalink / raw)
To: intel-xe; +Cc: lucas.demarchi, lucas.demarchi, Akshata Jahagirdar
According to the SAS for BMG compression, we need to decompress during eviction,
and not recompress on restore. Due to this, we need to introduce encoding pat_index
in case of vram too. This patch explores the solution of setting up an additional
identity map for the vram, this time at the end of previous mapping offset and
with compressed pat_index.
We then select the appropriate mapping during eviction/restore/clear.
Akshata Jahagirdar (6):
drm/xe/xe2: Introduce identity map for compressed pat for vram
drm/xe/migrate: Handle clear ccs logic for xe2 dgfx
drm/xe/migrate: Add kunit to test clear functionality
drm/xe/xe_migrate: Handle migration logic for xe2+ dgfx
drm/xe/migrate: Add kunit to test migration functionality for BMG
drm/xe/xe2: Do not run xe_bo_test for xe2+ dgfx
drivers/gpu/drm/xe/tests/xe_bo.c | 6 +
drivers/gpu/drm/xe/tests/xe_migrate.c | 388 +++++++++++++++++++++
drivers/gpu/drm/xe/tests/xe_migrate_test.c | 1 +
drivers/gpu/drm/xe/tests/xe_migrate_test.h | 1 +
drivers/gpu/drm/xe/xe_device.h | 5 +
drivers/gpu/drm/xe/xe_migrate.c | 66 +++-
6 files changed, 449 insertions(+), 18 deletions(-)
--
2.34.1
^ permalink raw reply [flat|nested] 14+ messages in thread* Re: [PATCH 0/6] Implement compression support on BMG
2024-07-10 7:53 Akshata Jahagirdar
` (3 preceding siblings ...)
2024-07-11 11:27 ` Akshata Jahagirdar
@ 2024-07-11 12:34 ` Ghimiray, Himal Prasad
2024-07-11 12:42 ` Akshata Jahagirdar
2024-07-11 13:07 ` Akshata Jahagirdar
6 siblings, 0 replies; 14+ messages in thread
From: Ghimiray, Himal Prasad @ 2024-07-11 12:34 UTC (permalink / raw)
To: Akshata Jahagirdar, intel-xe; +Cc: akshatajahagirdar6
On 11-07-2024 11:24, Akshata Jahagirdar wrote:
> On Xe2 the compression has moved to a unified universal model
> (exactly one compression mode/format), where compression is now
> controlled via PAT on per-page basis. This now means KMD can
> decompress freely. This was problematic on DG2 since we had
> multiple compression formats, and the compression format used
> on a particular buffer was unknown to the KMD, so instead the
> raw CCS state needed to be copied around when evicting VRAM.
> In addition mixed VRAM and system memory buffers were not
> supported with compression enabled.
>
> On Xe2 dGPU compression is still only supported with VRAM,
> however we can now support compression with VRAM and system
> memory buffers, with GPU access being seamless underneath.
> So long as when doing VRAM -> sysmem the KMD does the move
> using compressed -> uncompressed, to decompress it.
> CPU access to such buffers is also possible, under the premise
> that userspace first decompress the corresponding pages being
> accessed. If the pages are already in system memory then KMD would
> have already decompressed them. When restoring such buffers with
> sysmem -> VRAM the KMD can't easily know which pages were originally
> compressed, so we always use uncompressed -> uncompressed here.
> With this it also means we can drop all the raw CCS handling
> on such platforms (including needing to allocate extra CCS storage).
>
> In order to support this we now need to have two different identity
> mappings for compressed and uncompressed VRAM.
> The additional identity map is the VRAM with compressed pat_index.
> We then select the appropriate mapping during migration/clear.
>
Have gone through all the patches. Please address indentation and
checkpatch errors.
With above addressed all patches lgtm.
Reviewed-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
> Akshata Jahagirdar (6):
> drm/xe/xe2: Introduce identity map for compressed pat for vram
> drm/xe/migrate: Handle clear ccs logic for xe2 dgfx
> drm/xe/migrate: Add kunit to test clear functionality
> drm/xe/xe_migrate: Handle migration logic for xe2+ dgfx
> drm/xe/migrate: Add kunit to test migration functionality for BMG
> drm/xe/xe2: Do not run xe_bo_test for xe2+ dgfx
>
> drivers/gpu/drm/xe/tests/xe_bo.c | 6 +
> drivers/gpu/drm/xe/tests/xe_migrate.c | 388 +++++++++++++++++++++
> drivers/gpu/drm/xe/tests/xe_migrate_test.c | 1 +
> drivers/gpu/drm/xe/tests/xe_migrate_test.h | 1 +
> drivers/gpu/drm/xe/xe_device.h | 5 +
> drivers/gpu/drm/xe/xe_migrate.c | 66 +++-
> 6 files changed, 449 insertions(+), 18 deletions(-)
>
^ permalink raw reply [flat|nested] 14+ messages in thread* [PATCH 0/6] Implement compression support on BMG
2024-07-10 7:53 Akshata Jahagirdar
` (4 preceding siblings ...)
2024-07-11 12:34 ` Ghimiray, Himal Prasad
@ 2024-07-11 12:42 ` Akshata Jahagirdar
2024-07-11 13:07 ` Akshata Jahagirdar
6 siblings, 0 replies; 14+ messages in thread
From: Akshata Jahagirdar @ 2024-07-11 12:42 UTC (permalink / raw)
To: intel-xe; +Cc: akshatajahagirdar6, Akshata Jahagirdar
According to the SAS for BMG compression, we need to decompress during eviction,
and not recompress on restore. Due to this, we need to introduce encoding pat_index
in case of vram too. This patch explores the solution of setting up an additional
identity map for the vram, this time at the end of previous mapping offset and
with compressed pat_index.
We then select the appropriate mapping during eviction/restore/clear.
Akshata Jahagirdar (6):
drm/xe/xe2: Introduce identity map for compressed pat for vram
drm/xe/migrate: Handle clear ccs logic for xe2 dgfx
drm/xe/migrate: Add kunit to test clear functionality
drm/xe/xe_migrate: Handle migration logic for xe2+ dgfx
drm/xe/migrate: Add kunit to test migration functionality for BMG
drm/xe/xe2: Do not run xe_bo_test for xe2+ dgfx
drivers/gpu/drm/xe/tests/xe_bo.c | 6 +
drivers/gpu/drm/xe/tests/xe_migrate.c | 388 +++++++++++++++++++++
drivers/gpu/drm/xe/tests/xe_migrate_test.c | 1 +
drivers/gpu/drm/xe/tests/xe_migrate_test.h | 1 +
drivers/gpu/drm/xe/xe_device.h | 5 +
drivers/gpu/drm/xe/xe_migrate.c | 66 +++-
6 files changed, 449 insertions(+), 18 deletions(-)
--
2.34.1
^ permalink raw reply [flat|nested] 14+ messages in thread* [PATCH 0/6] Implement compression support on BMG
2024-07-10 7:53 Akshata Jahagirdar
` (5 preceding siblings ...)
2024-07-11 12:42 ` Akshata Jahagirdar
@ 2024-07-11 13:07 ` Akshata Jahagirdar
6 siblings, 0 replies; 14+ messages in thread
From: Akshata Jahagirdar @ 2024-07-11 13:07 UTC (permalink / raw)
To: intel-xe; +Cc: akshatajahagirdar6, Akshata Jahagirdar
According to the SAS for BMG compression, we need to decompress during eviction,
and not recompress on restore. Due to this, we need to introduce encoding pat_index
in case of vram too. This patch explores the solution of setting up an additional
identity map for the vram, this time at the end of previous mapping offset and
with compressed pat_index.
We then select the appropriate mapping during eviction/restore/clear.
Akshata Jahagirdar (6):
drm/xe/xe2: Introduce identity map for compressed pat for vram
drm/xe/migrate: Handle clear ccs logic for xe2 dgfx
drm/xe/migrate: Add kunit to test clear functionality
drm/xe/xe_migrate: Handle migration logic for xe2+ dgfx
drm/xe/migrate: Add kunit to test migration functionality for BMG
drm/xe/xe2: Do not run xe_bo_test for xe2+ dgfx
drivers/gpu/drm/xe/tests/xe_bo.c | 6 +
drivers/gpu/drm/xe/tests/xe_migrate.c | 388 +++++++++++++++++++++
drivers/gpu/drm/xe/tests/xe_migrate_test.c | 1 +
drivers/gpu/drm/xe/tests/xe_migrate_test.h | 1 +
drivers/gpu/drm/xe/xe_device.h | 5 +
drivers/gpu/drm/xe/xe_migrate.c | 66 +++-
6 files changed, 449 insertions(+), 18 deletions(-)
--
2.34.1
^ permalink raw reply [flat|nested] 14+ messages in thread
end of thread, other threads:[~2024-07-12 7:24 UTC | newest]
Thread overview: 14+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-07-11 9:18 [PATCH 0/6] Implement compression support on BMG Akshata Jahagirdar
2024-07-11 9:19 ` Akshata Jahagirdar
2024-07-12 3:11 ` [PATCH 1/6] drm/xe/migrate: Handle clear ccs logic for xe2 dgfx Akshata Jahagirdar
-- strict thread matches above, loose matches on Subject: below --
2024-07-12 6:39 [PATCH 0/6] Implement compression support on BMG Akshata Jahagirdar
2024-07-12 7:24 ` Akshata Jahagirdar
2024-07-10 7:53 Akshata Jahagirdar
2024-07-09 10:33 ` Matthew Auld
2024-07-09 19:07 ` Jahagirdar, Akshata
2024-07-10 8:17 ` Akshata Jahagirdar
2024-07-11 5:54 ` Akshata Jahagirdar
2024-07-11 11:27 ` Akshata Jahagirdar
2024-07-11 12:34 ` Ghimiray, Himal Prasad
2024-07-11 12:42 ` Akshata Jahagirdar
2024-07-11 13:07 ` Akshata Jahagirdar
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox