the PTB bo.
With this change, the vm->pt_freed list and work is not needed. Add
WARN_ON(unlocked) in amdgpu_vm_pt_free_dfs to catch if unmap to free the
PTB.
Signed-off-by: Philip Yang
---
drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c| 4 ---
drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h| 4 ---
d
On 2025-01-10 11:23, Chen, Xiaogang
wrote:
On 1/10/2025 8:37 AM, Philip Yang wrote:
On 2025-01-10 02:49, Emily Deng wrote:
For partial migrate from ram to vram,
the migrate->cpages
On 2025-01-09 12:14, Felix Kuehling
wrote:
On 2025-01-08 20:11, Philip Yang wrote:
On 2025-01-07 22:08, Deng, Emily wrote:
[AMD Official Use Only - AMD Internal
On 2025-01-10 09:25, Emily Deng wrote:
For partial migrate from ram to vram, the migrate->cpages is not
equal to migrate->npages, should use migrate->npages to check all needed
migrate pages which could be copied or not.
And only need to set those pages could be m
hat fixed, this patch
is
Reviewed-by: Philip Yang
Signed-off-by: Emily Deng
---
drivers/gpu/drm/amd/amdkfd/kfd_migrate.c | 17 ++---
1 file changed, 10 insertions(+), 7 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu
On 2025-01-07 22:08, Deng, Emily wrote:
[AMD Official Use Only - AMD Internal Distribution Only]
Hi
Philip,
It
still has the deadlock, maybe the best way is
On 2025-01-08 08:19, Emily Deng wrote:
For partial migrate from ram to vram, the migrate->cpages is not
equal to migrate->npages, should use migrate->npages to check all needed
migrate pages which could be copied or not.
And only need to set those pages could be m
On 2025-01-07 19:31, Deng, Emily wrote:
[AMD Official Use Only - AMD Internal Distribution Only]
From:
Yan
On 2025-01-07 10:50, Chen, Xiaogang
wrote:
On 1/6/2025 8:02 PM, Deng, Emily
wrote:
[AMD Official Use Only - AMD Internal
Distribution Only]
On 2025-01-07 07:30, Deng, Emily wrote:
[AMD Official Use Only - AMD Internal Distribution Only]
Hi Felix,
You are right, it is easily to hit deadlock, don't know why LOCKDEP doesn't catch this. Need to find another solution.
Hi Philip,
Do you have a sol
On 2025-01-06 21:31, Deng, Emily wrote:
[AMD Official Use Only - AMD Internal Distribution Only]
From:
Yang, Philip
On 2025-01-02 19:06, Emily Deng wrote:
For partial migrate from ram to vram, the migrate->cpages is not
equal to migrate->npages, should use migrate->npages to check all needed
migrate pages which could be copied or not.
And only need to set those pages could be m
*_ih.c except ASICs older than Vega which has only one ih ring.
Signed-off-by: Philip Yang
Reviewed-by: Christian König
Reviewed-by: Felix Kuehling
---
drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c | 6 ++
drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h | 1 +
drivers/gpu/drm/amd/amdgpu/navi10_ih.c
then driver process
the first event interrupt, set_event and event slot is auto-reset, then
for the second event interrupt, KFD goes to slow path as event is not
signaled, just drop the second event interrupt because the application
only need wakeup once.
Signed-off-by: Philip Yang
Reviewed-by:
handle the gfx v9 path, cover retry on/off and CAM filter
on/off cases.
Signed-off-by: Philip Yang
Reviewed-by: Felix Kuehling
---
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h | 10
drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 4 ++
drivers/gpu/drm/amd/amdkfd/kfd_device.c| 67
queue with number of workers equals to
number of partitions, let queue_work select the next CPU round robin
among the local CPUs of same NUMA.
Signed-off-by: Philip Yang
Reviewed-by: Felix Kuehling
---
drivers/gpu/drm/amd/amdkfd/kfd_device.c| 25 --
drivers/gpu/drm/amd
-by: Philip Yang
Reviewed-by: Felix Kuehling
---
drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c | 8 +---
1 file changed, 5 insertions(+), 3 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
index 4c8308b2878b..56507ae919b0 100644
--- a
To handle 4 to 8 interrupts per second running CPX mode with 4
streams/queues per KFD node, KFD interrupt handler becomes the
performance bottleneck.
Remove the kfifo_out memcpy overhead by accessing ih_fifo data in-place
and updating rptr with kfifo_skip_count.
Signed-off-by: Philip
On 2024-10-21 04:12, Christian König
wrote:
Am
18.10.24 um 23:59 schrieb Philip Yang:
On 2024-10-18 14:28, Felix Kuehling wrote:
On 2024-10-17 04:34, Victor Zhao wrote:
make sure
On 2024-10-21 13:46, Alex Deucher
wrote:
This reverts commit a3ab2d45b9887ee609cd3bea39f668236935774c.
The userspace side for this code is not ready yet so revert
for now.
Signed-off-by: Alex Deucher
Cc: Philip Yang
Reviewed-by: Philip Yang
On 2024-10-18 14:28, Felix Kuehling
wrote:
On 2024-10-17 04:34, Victor Zhao wrote:
make sure KFD_FENCE_INIT write to
fence_addr before pm_send_query_status
called, to avoid qcm fence timeout caused by incorrect ord
It is safe to access dqm->sched status inside dqm_lock, no
race with gpu reset.
Reviewed-by: Philip Yang
On 2024-10-18 11:10, Shaoyun Liu wrote:
From: shaoyunl
Add back kfd queues in start scheduling that originally been
removed on stop schedul
ordering.
Signed-off-by: Victor Zhao
Reviewed-by: Philip Yang
---
drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c | 1 + drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h | 2 +-
2 files changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdkfd
On 2024-10-17 12:12, Shaoyun Liu wrote:
From: shaoyunl
Add back kfd queues in start scheduling that originally been
removed on stop scheduling.
Signed-off-by: Shaoyun Liu
---
.../drm/amd/amdkfd/kfd_device_queue_manager.c | 40 +--
1 file change
pe because it is updated outside process mutex
now.
Signed-off-by: Philip Yang
---
drivers/gpu/drm/amd/amdkfd/kfd_chardev.c | 6 +++---
drivers/gpu/drm/amd/amdkfd/kfd_priv.h| 2 +-
drivers/gpu/drm/amd/amdkfd/kfd_process.c | 4 ++--
drivers/gpu/drm/amd/amdkfd/kfd_svm.c | 26 +++
Drop this patch series, as Felix pointed out, the forked process
takes svm_bo device pages ref, svm_bo->pdd could refer to the
process that doesn't exist any more.
Regards,
Philip
On 2024-10-11 11:00, Philip Yang wrote:
KFD process device
c64_t because it is updated outside process mutex
now.
Signed-off-by: Philip Yang
---
drivers/gpu/drm/amd/amdkfd/kfd_chardev.c | 6 +++---
drivers/gpu/drm/amd/amdkfd/kfd_priv.h| 2 +-
drivers/gpu/drm/amd/amdkfd/kfd_process.c | 4 ++--
drivers/gpu/drm/amd/amdkfd/kfd_svm.c | 2 ++
4
KFD process device data pdd will be used for VRAM usage accounting, save
pdd to svm_bo to avoid searching pdd for every accounting, and get KFD
node from pdd->dev.
svm_bo->pdd will always be valid because KFD process release free all
svm_bo first, then destroy process pdds.
Signed-off-by:
On 2024-10-09 17:20, Felix Kuehling
wrote:
On 2024-10-04 16:28, Philip Yang wrote:
Per process device data pdd->vram_usage
is used by rocm-smi to report
VRAM usage, this is currently missing the svm_bo us
updated outside process mutex now.
Signed-off-by: Philip Yang
---
drivers/gpu/drm/amd/amdkfd/kfd_chardev.c | 6 +++---
drivers/gpu/drm/amd/amdkfd/kfd_priv.h| 2 +-
drivers/gpu/drm/amd/amdkfd/kfd_process.c | 4 ++--
drivers/gpu/drm/amd/amdkfd/kfd_svm.c | 22 ++
4 f
get_wave_state is not defined for sdma queue, copy_context_work_handler
calls it for sdma queue will crash.
Signed-off-by: Philip Yang
---
drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdkfd
On 2024-09-11 02:54, Christian König
wrote:
Yeah, I completely agree with Xiaogang.
The PASID is an identifier of an address space. And the idea of
the KFD was that we can just use the same address space and with
it the page ta
On 2024-09-09 14:46, Christian König
wrote:
Am
09.09.24 um 18:02 schrieb Kim, Jonathan:
[Public]
-Original Message-
From: Christian König
Sent: Thursday, September 5, 202
ff-by: Ramesh Errabolu
With 2 below nitpicks fixed, this patch is
Reviewed-by: Philip Yang
change subject to "drm/amdkfd: Add svm_default_granularity module
parameter"
---
drivers/gpu/drm/amd/amdgpu/amdgpu.h | 1 +
drivers/gpu/drm/amd/amdgpu/am
leased before ttm_resource_manager is finilized, drain the workqueue in ttm_device.
v2: move drain_workqueue to amdgpu_ttm.c
Fixes:d99fbd9aab62 ("drm/ttm: Always take the bo delayed cleanup path for imported bos")
Suggested-by: Christian König
Signed-off-by: Asher Song
Acked-by: Ph
On 2024-09-02 05:06, Christian König
wrote:
Am
02.09.24 um 05:03 schrieb Lang Yu:
Fixes: 5a1c27951966 ("drm/amdgpu:
implement TLB flush fence")
Signed-off-by: Lang Yu
Ah yes, that exp
On 2024-08-29 18:31, Chen, Xiaogang
wrote:
On 8/29/2024 5:13 PM, Ramesh Errabolu wrote:
Caution: This message originated from an
External Source. Use proper caution when opening attachments,
clicking links, or respon
On 2024-08-29 17:15, Felix Kuehling
wrote:
On
2024-08-23 15:49, Philip Yang wrote:
If GPU reset kick in while KFD
restore_process_worker running, this may
causes different issues, for example below rcu stall warning
On 2024-08-28 18:01, Felix Kuehling
wrote:
On 2024-08-23 15:49, Philip Yang wrote:
If GPU reset kick in while KFD
restore_process_worker running, this may
causes different issues, for example below rcu stall
On 2024-08-26 15:34, Ramesh Errabolu
wrote:
Enables users to update the default size of buffer used
in migration either from Sysmem to VRAM or vice versa.
The param GOBM refers to granularity of buffer migration,
and is specified in terms of log(numPages(buff
.
v3:
Simplify event drop count handling (James Zhu)
Philip Yang (4):
drm/amdkfd: Document and define SVM events message macro
drm/amdkfd: Output migrate end event if migrate failed
drm/amdkfd: Increase SMI event fifo size
drm/amdkfd: SMI report dropped event count
drivers/gpu/drm/amd
and reset drop count to zero.
Signed-off-by: Philip Yang
---
drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c | 25 +
include/uapi/linux/kfd_ioctl.h | 6 +
2 files changed, 27 insertions(+), 4 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c
b
and reset drop count to zero.
Signed-off-by: Philip Yang
---
drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c | 25 +
include/uapi/linux/kfd_ioctl.h | 6 +
2 files changed, 27 insertions(+), 4 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c
b
future.
No functional changes.
Signed-off-by: Philip Yang
---
drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c | 45 +
include/uapi/linux/kfd_ioctl.h | 100 +---
2 files changed, 109 insertions(+), 36 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdkfd
prefix to the macro name.
Signed-off-by: Philip Yang
---
drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c
b/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c
index 1d94b445a060..9b8169761ec5
If page migration failed, also output migrate end event to match with
migrate start event, with failure error_code added to the end of the
migrate message macro. This will not break uAPI because application uses
old message macro sscanf drop and ignore the error_code.
Signed-off-by: Philip Yang
]
Call Trace:
update_process_times+0x94/0xd0
RIP: 0010:amdgpu_vm_handle_moved+0x9a/0x210 [amdgpu]
amdgpu_amdkfd_gpuvm_restore_process_bos+0x3d6/0x7d0 [amdgpu]
restore_process_helper+0x27/0x80 [amdgpu]
Signed-off-by: Philip Yang
---
drivers/gpu/drm/amd/amdkfd/kfd_process.c | 56
On 2024-08-22 10:34, James Zhu wrote:
On 2024-07-30 16:15, Philip Yang wrote:
SMI event fifo size 1KB was enough to
report GPU vm fault or reset
[JZ] There is a typo here. it should be NOT enough
On 2024-08-22 10:32, James Zhu wrote:
On 2024-07-30 16:15, Philip Yang
wrote:
Document how to use SMI system management interface to enable and
receive SVM events. Document SVM event triggers.
Define SVM events message
page faults at deferred
work. So, the time period that kfd does not handle page faults is reduced
and can be controlled.
Signed-off-by: Xiaogang.Chen
Some nitpicks below.
This patch is Reviewed-by: Philip Yang
---
drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 4 +-
drivers/gpu
On 2024-08-21 19:22, Ramesh Errabolu
wrote:
KFD's design of unified memory (UM) does not allow users to
configure the size of buffer used in migrating buffer either
from Sysmem to VRAM or vice versa.
This is not true, app can change range granularit
, to handle error in case failed to
hold vm lock. Then hold dqm_lock to remove queue from queue list and
then release queue buffers.
Restore process worker restore queue hold dqm_lock, will always find
the queue with valid queue buffers.
Signed-off-by: Philip Yang
---
drivers/gpu/drm/amd/amdkfd
On 2024-07-31 04:10, Shikang Fan wrote:
Move kgd2kfd_init _zone_device() after release_full_gpu()
as it takes long time for asics with huge bar size and it could
potentially hit full access timeout for SRIOV during init.
Signed-off-by: Shikang Fan
---
drivers/gp
future.
No functional changes.
Signed-off-by: Philip Yang
---
drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c | 45 +
include/uapi/linux/kfd_ioctl.h | 100 +---
2 files changed, 109 insertions(+), 36 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdkfd
events dropped,
generate a dropped event record and add to kfifo.
Signed-off-by: Philip Yang
---
drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c | 41 ++---
include/uapi/linux/kfd_ioctl.h | 6 +++
2 files changed, 41 insertions(+), 6 deletions(-)
diff --git a/drivers/gpu/drm
SMI event fifo size 1KB was enough to report GPU vm fault or reset
event, increase it to 8KB to store about 100 migrate events, less chance
to drop the migrate events if lots of migration happened in the short
period of time. Add KFD prefix to the macro name.
Signed-off-by: Philip Yang
If page migration failed, also output migrate end event to match with
migrate start event, with failure error_code added to the end of the
migrate message macro. This will not break uAPI because application uses
old message macro sscanf drop and ignore the error_code.
Signed-off-by: Philip Yang
.
Philip Yang (4):
drm/amdkfd: Document and define SVM events message macro
drm/amdkfd: Output migrate end event if migrate failed
drm/amdkfd: Increase SMI event fifo size
drm/amdkfd: SMI report dropped event count
drivers/gpu/drm/amd/amdkfd/kfd_migrate.c| 14 ++-
drivers/gpu/drm/amd
The kfdtest user queue validation cases don't cover those error
condition path, thanks for catching it.
This patch is
Reviewed-by: Philip Yang
On 2024-07-26 02:47, Srinivasan
Shanmugam wrote:
The fix involves setting 'err&#x
On 2024-07-25 14:19, Xiaogang.Chen
wrote:
From: Xiaogang Chen
When app unmap vm ranges(munmap) kfd/svm starts drain pending page fault and
not handle any incoming pages fault of this process until a deferred work item
got executed by default system wq. The
mory residency")
Reported-by: kernel test robot
Closes:
https://lore.kernel.org/oe-kbuild-all/202407252127.zvnxakra-...@intel.com/
Signed-off-by: Philip Yang
---
drivers/gpu/drm/amd/amdkfd/kfd_queue.c | 14 ++
1 file changed, 14 insertions(+)
diff --git a/drivers/gpu/drm/am
On 2024-07-24 09:58, Chen, Xiaogang
wrote:
On 7/23/2024 4:02 PM, Philip Yang
wrote:
On 2024-07-19 18:17, Xiaogang.Chen
wrote:
From: Xiaogang Chen
On 2024-07-19 18:17, Xiaogang.Chen
wrote:
From: Xiaogang Chen
When app unmap vm ranges(munmap) kfd/svm starts drain pending page fault and
not handle any incoming pages fault of this process until a deferred work item
got executed by default system wq. The
parameter from pqm_create_queue.
Rename structure queue wptr_bo_gart to hold wptr_bo reference for GART
mapping and umapping. Move MES wptr_bo_gart mapping to init_user_queue,
the same location with queue ctx_bo GART mapping.
Signed-off-by: Philip Yang
---
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
KFD node properties, to
remove the duplicate calculation code from Thunk.
Signed-off-by: Philip Yang
Reviewed-by: Felix Kuehling
---
drivers/gpu/drm/amd/amdkfd/kfd_priv.h | 1 +
drivers/gpu/drm/amd/amdkfd/kfd_queue.c| 75 +++
drivers/gpu/drm/amd/amdkfd/kfd_topology.c
Pass pointer reference to amdgpu_bo_unref to clear the correct pointer,
otherwise amdgpu_bo_unref clear the local variable, the original pointer
not set to NULL, this could cause use-after-free bug.
Signed-off-by: Philip Yang
Reviewed-by: Felix Kuehling
---
drivers/gpu/drm/amd/amdgpu
Find user queue rptr, ring buf, eop buffer and cwsr area BOs, and
check BOs are mapped on the GPU with correct size and take the BO
reference.
Signed-off-by: Philip Yang
Reviewed-by: Felix Kuehling
---
drivers/gpu/drm/amd/amdkfd/kfd_priv.h | 4 +++
drivers/gpu/drm/amd/amdkfd/kfd_queue.c | 38
.
Signed-off-by: Philip Yang
---
.../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c | 14 ++--
drivers/gpu/drm/amd/amdgpu/amdgpu_object.h| 6
drivers/gpu/drm/amd/amdkfd/kfd_chardev.c | 3 +-
drivers/gpu/drm/amd/amdkfd/kfd_priv.h | 1 +
drivers/gpu/drm/amd/amdkfd
When creating KFD user compute queue, check if queue eop buffer size,
cwsr area size, ctl stack size equal to the size of KFD node
properities.
Check the entire cwsr area which may split into multiple svm ranges
aligned to gramularity boundary.
Signed-off-by: Philip Yang
Reviewed-by: Felix
unmap the CWSR area while queue is active, pr_warn message
in dmesg log. To be safe, evict user queue.
Signed-off-by: Philip Yang
Reviewed-by: Felix Kuehling
---
drivers/gpu/drm/amd/amdkfd/kfd_queue.c | 110 -
drivers/gpu/drm/amd/amdkfd/kfd_svm.c | 12 +++
drivers/gpu/drm
Ensure update queue new ring buffer is mapped on GPU with correct size.
Decrease queue old ring_bo queue_refcount and increase new ring_bo
queue_refcount.
Signed-off-by: Philip Yang
Reviewed-by: Felix Kuehling
---
.../amd/amdkfd/kfd_process_queue_manager.c| 32 ++-
1 file
Change amdgpu_amdkfd_bo_mapped_to_dev to use drm_priv as parameter
instead of adev, to support spatial partition. This is only used by CRIU
checkpoint restore now. No functional change.
Signed-off-by: Philip Yang
Reviewed-by: Felix Kuehling
---
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
free and unmap them
when the qeueu is active, or evict the queue if SVM memory is unmapped and
freed from CPU.
Patch 1-2 is prepration work and general fix.
v2:
- patch 3/9, keep wptr_bo_gart in struct queue
Philip Yang (9):
drm/amdkfd: kfd_bo_mapped_dev support partition
drm/amdkfd
See inline.
kfd_queue_release_buffers return value is handled in queue destroy
path, to return -ERESTARTSYS if fail to hold vm lock to release
buffers because signal is received. See inline.
On 2024-07-15 08:34, Philip Yang wrote:
Add helpe
On 2024-07-17 16:10, Felix Kuehling
wrote:
@@ -603,8 +606,6
@@ struct queue {
void *gang_ctx_bo;
uint64_t gang_ctx_gpu_addr;
void *gang_ctx_cpu_ptr;
-
When creating KFD user compute queue, check if queue eop buffer size,
cwsr area size, ctl stack size equal to the size of KFD node
properities.
Check the entire cwsr area which may split into multiple svm ranges
aligned to gramularity boundary.
Signed-off-by: Philip Yang
Reviewed-by: Felix
Find user queue rptr, ring buf, eop buffer and cwsr area BOs, and
check BOs are mapped on the GPU with correct size and take the BO
reference.
Signed-off-by: Philip Yang
Reviewed-by: Felix Kuehling
---
drivers/gpu/drm/amd/amdkfd/kfd_priv.h | 4 +++
drivers/gpu/drm/amd/amdkfd/kfd_queue.c | 38
.
Signed-off-by: Philip Yang
---
.../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c | 14 ++--
drivers/gpu/drm/amd/amdgpu/amdgpu_object.h| 6
drivers/gpu/drm/amd/amdkfd/kfd_chardev.c | 3 +-
drivers/gpu/drm/amd/amdkfd/kfd_priv.h | 1 +
drivers/gpu/drm/amd/amdkfd
KFD node properties, to
remove the duplicate calculation code from Thunk.
Signed-off-by: Philip Yang
Reviewed-by: Felix Kuehling
---
drivers/gpu/drm/amd/amdkfd/kfd_priv.h | 1 +
drivers/gpu/drm/amd/amdkfd/kfd_queue.c| 75 +++
drivers/gpu/drm/amd/amdkfd/kfd_topology.c
Ensure update queue new ring buffer is mapped on GPU with correct size.
Decrease queue old ring_bo queue_refcount and increase new ring_bo
queue_refcount.
Signed-off-by: Philip Yang
Reviewed-by: Felix Kuehling
---
.../amd/amdkfd/kfd_process_queue_manager.c| 32 ++-
1 file
, the same location with
queue ctx_bo GART mapping.
Signed-off-by: Philip Yang
---
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h| 2 +-
.../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c | 5 +-
drivers/gpu/drm/amd/amdkfd/kfd_chardev.c | 56 +++---
.../drm/amd/amdkfd
Change amdgpu_amdkfd_bo_mapped_to_dev to use drm_priv as parameter
instead of adev, to support spatial partition. This is only used by CRIU
checkpoint restore now. No functional change.
Signed-off-by: Philip Yang
Reviewed-by: Felix Kuehling
---
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
unmap the CWSR area while queue is active, pr_warn message
in dmesg log. To be safe, evict user queue.
Signed-off-by: Philip Yang
---
drivers/gpu/drm/amd/amdkfd/kfd_queue.c | 110 -
drivers/gpu/drm/amd/amdkfd/kfd_svm.c | 12 +++
drivers/gpu/drm/amd/amdkfd/kfd_svm.h | 1
free and unmap them
when the qeueu is active, or evict the queue if SVM memory is unmapped and
freed from CPU.
Patch 1-2 is prepration work and general fix.
Philip Yang (9):
drm/amdkfd: kfd_bo_mapped_dev support partition
drm/amdkfd: amdkfd_free_gtt_mem clear the correct pointer
drm/amdkfd
Pass pointer reference to amdgpu_bo_unref to clear the correct pointer,
otherwise amdgpu_bo_unref clear the local variable, the original pointer
not set to NULL, this could cause use-after-free bug.
Signed-off-by: Philip Yang
---
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c | 14
On 2024-06-21 13:28, Xiaogang.Chen
wrote:
From: Xiaogang Chen
When user adds new vm range that has overlapping with existing svm pranges
current kfd clones new prange and remove existing pranges including all data
associate with it. It is not necessary. We
amdgpu_vm_handle_fault and then to gmc interrupt handler
to show vm fault message.
Signed-off-by: Philip Yang
---
drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 5 +++--
drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h | 2 +-
drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c | 2 +-
drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 7
On 2024-05-02 08:42, James Zhu wrote:
On 2024-05-01 18:56, Philip Yang wrote:
On system with khugepaged enabled and user
cases with THP buffer, the
hmm_range_fault may takes > 15 seconds to return -EBUSY,
On 2024-05-02 00:09, Chen, Xiaogang
wrote:
On 5/1/2024 5:56 PM, Philip Yang wrote:
Caution: This message originated from an
External Source. Use proper caution when opening attachments,
clicking links, or responding
urn EBUSY, then userspace libdrm and Thunk will call
ioctl again.
Change EAGAIN to debug message as this is not error.
Signed-off-by: Philip Yang
---
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c | 5 -
drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.c | 12 +++-
drivers/gpu/drm/
: Ramesh Errabolu
Reviewed-by: Philip Yang
---
drivers/gpu/drm/amd/amdkfd/kfd_svm.c | 6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index 386875e6eb96..dcb1d5d3f860 100644
TTM_PL_FLAG_CONTIFUOUS flag, and ask
VRAM buddy allocator to get contiguous VRAM.
Signed-off-by: Philip Yang
---
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c | 4
include/uapi/linux/kfd_ioctl.h | 1 +
2 files changed, 5 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu
Bump the kfd ioctl minor version to delcare the contiguous VRAM
allocation flag support.
Signed-off-by: Philip Yang
---
include/uapi/linux/kfd_ioctl.h | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/include/uapi/linux/kfd_ioctl.h b/include/uapi/linux/kfd_ioctl.h
index
If the BO pages pinned for RDMA is not contiguous on VRAM, evict it to
system memory first to free the VRAM space, then allocate contiguous
VRAM space, and then move it from system memory back to VRAM.
v6: user context should use interruptible call (Felix)
Signed-off-by: Philip Yang
prefix to the macro
name.
v6: use shorter flag name, use interruptible wait ctx, drop patch 5/6 (Felix)
Philip Yang (5):
drm/amdgpu: Support contiguous VRAM allocation
drm/amdgpu: Handle sg size limit for contiguous allocation
drm/amdgpu: Evict BOs from same process for contiguous alloca
from the same process, this will evict the user queues
first, and restore the queues later after contiguous VRAM allocation.
Signed-off-by: Philip Yang
Reviewed-by: Felix Kuehling
---
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a
guous VRAM memory. To workaround the sg table segment
size limit, allocate multiple segments if contiguous size is bigger than
AMDGPU_MAX_SG_SEGMENT_SIZE.
Signed-off-by: Philip Yang
Reviewed-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c | 12 ++--
1 file chang
On 2024-04-23 18:17, Felix Kuehling
wrote:
On 2024-04-23 11:28, Philip Yang wrote:
RDMA device with limited scatter-gather
ability requires contiguous VRAM
buffer allocation for RDMA peer direct support
On 2024-04-23 18:15, Felix Kuehling
wrote:
On
2024-04-23 11:28, Philip Yang wrote:
If the BO pages pinned for RDMA is not
contiguous on VRAM, evict it to
system memory first to free the VRAM space, then allocate
Bump the kfd ioctl minor version to delcare the contiguous VRAM
allocation flag support.
Signed-off-by: Philip Yang
---
include/uapi/linux/kfd_ioctl.h | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/include/uapi/linux/kfd_ioctl.h b/include/uapi/linux/kfd_ioctl.h
index
1 - 100 of 723 matches
Mail list logo
