aboutsummaryrefslogtreecommitdiffstats
path: root/drivers/gpu/drm/amd/amdkfd
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/gpu/drm/amd/amdkfd')
-rw-r--r--drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler.h14
-rw-r--r--drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx10.asm2
-rw-r--r--drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx9.asm2
-rw-r--r--drivers/gpu/drm/amd/amdkfd/kfd_chardev.c4
-rw-r--r--drivers/gpu/drm/amd/amdkfd/kfd_migrate.c2
-rw-r--r--drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c4
-rw-r--r--drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c9
-rw-r--r--drivers/gpu/drm/amd/amdkfd/kfd_priv.h10
-rw-r--r--drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c4
-rw-r--r--drivers/gpu/drm/amd/amdkfd/kfd_topology.c10
10 files changed, 39 insertions, 22 deletions
diff --git a/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler.h b/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler.h
index df75863393fc..d1caaf0e6a7c 100644
--- a/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler.h
+++ b/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler.h
@@ -674,7 +674,7 @@ static const uint32_t cwsr_trap_gfx9_hex[] = {
0x86ea6a6a, 0x8f6e837a,
0xb96ee0c2, 0xbf800002,
0xb97a0002, 0xbf8a0000,
- 0xbe801f6c, 0xbf810000,
+ 0xbe801f6c, 0xbf9b0000,
};
static const uint32_t cwsr_trap_nv1x_hex[] = {
@@ -1091,7 +1091,7 @@ static const uint32_t cwsr_trap_nv1x_hex[] = {
0xb9eef807, 0x876dff6d,
0x0000ffff, 0x87fe7e7e,
0x87ea6a6a, 0xb9faf802,
- 0xbe80226c, 0xbf810000,
+ 0xbe80226c, 0xbf9b0000,
0xbf9f0000, 0xbf9f0000,
0xbf9f0000, 0xbf9f0000,
0xbf9f0000, 0x00000000,
@@ -1574,7 +1574,7 @@ static const uint32_t cwsr_trap_arcturus_hex[] = {
0x86ea6a6a, 0x8f6e837a,
0xb96ee0c2, 0xbf800002,
0xb97a0002, 0xbf8a0000,
- 0xbe801f6c, 0xbf810000,
+ 0xbe801f6c, 0xbf9b0000,
};
static const uint32_t cwsr_trap_aldebaran_hex[] = {
@@ -2065,7 +2065,7 @@ static const uint32_t cwsr_trap_aldebaran_hex[] = {
0x86ea6a6a, 0x8f6e837a,
0xb96ee0c2, 0xbf800002,
0xb97a0002, 0xbf8a0000,
- 0xbe801f6c, 0xbf810000,
+ 0xbe801f6c, 0xbf9b0000,
};
static const uint32_t cwsr_trap_gfx10_hex[] = {
@@ -2500,7 +2500,7 @@ static const uint32_t cwsr_trap_gfx10_hex[] = {
0x876dff6d, 0x0000ffff,
0x87fe7e7e, 0x87ea6a6a,
0xb9faf802, 0xbe80226c,
- 0xbf810000, 0xbf9f0000,
+ 0xbf9b0000, 0xbf9f0000,
0xbf9f0000, 0xbf9f0000,
0xbf9f0000, 0xbf9f0000,
};
@@ -2944,7 +2944,7 @@ static const uint32_t cwsr_trap_gfx11_hex[] = {
0xb8eef802, 0xbf0d866e,
0xbfa20002, 0xb97af802,
0xbe80486c, 0xb97af802,
- 0xbe804a6c, 0xbfb00000,
+ 0xbe804a6c, 0xbfb10000,
0xbf9f0000, 0xbf9f0000,
0xbf9f0000, 0xbf9f0000,
0xbf9f0000, 0x00000000,
@@ -3436,5 +3436,5 @@ static const uint32_t cwsr_trap_gfx9_4_3_hex[] = {
0x86ea6a6a, 0x8f6e837a,
0xb96ee0c2, 0xbf800002,
0xb97a0002, 0xbf8a0000,
- 0xbe801f6c, 0xbf810000,
+ 0xbe801f6c, 0xbf9b0000,
};
diff --git a/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx10.asm b/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx10.asm
index e0140df0b0ec..71b3dc0c7363 100644
--- a/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx10.asm
+++ b/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx10.asm
@@ -1104,7 +1104,7 @@ L_RETURN_WITHOUT_PRIV:
s_rfe_b64 s_restore_pc_lo //Return to the main shader program and resume execution
L_END_PGM:
- s_endpgm
+ s_endpgm_saved
end
function write_hwreg_to_mem(s, s_rsrc, s_mem_offset)
diff --git a/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx9.asm b/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx9.asm
index e506411ad28a..bb26338204f4 100644
--- a/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx9.asm
+++ b/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx9.asm
@@ -921,7 +921,7 @@ L_RESTORE:
/* the END */
/**************************************************************************/
L_END_PGM:
- s_endpgm
+ s_endpgm_saved
end
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index ce4c52ec34d8..80e90fdef291 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -1442,7 +1442,9 @@ static int kfd_ioctl_unmap_memory_from_gpu(struct file *filep,
kfd_flush_tlb(peer_pdd, TLB_FLUSH_HEAVYWEIGHT);
/* Remove dma mapping after tlb flush to avoid IO_PAGE_FAULT */
- amdgpu_amdkfd_gpuvm_dmaunmap_mem(mem, peer_pdd->drm_priv);
+ err = amdgpu_amdkfd_gpuvm_dmaunmap_mem(mem, peer_pdd->drm_priv);
+ if (err)
+ goto sync_memory_failed;
}
mutex_unlock(&p->mutex);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
index f856901055d3..bdc01ca9609a 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
@@ -574,7 +574,7 @@ svm_migrate_copy_to_ram(struct amdgpu_device *adev, struct svm_range *prange,
pr_debug("svms 0x%p [0x%lx 0x%lx]\n", prange->svms, prange->start,
prange->last);
- addr = prange->start << PAGE_SHIFT;
+ addr = migrate->start;
src = (uint64_t *)(scratch + npages);
dst = scratch;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c
index d722cbd31783..826bc4f6c8a7 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c
@@ -55,8 +55,8 @@ static void update_cu_mask(struct mqd_manager *mm, void *mqd,
m = get_mqd(mqd);
if (has_wa_flag) {
- uint32_t wa_mask = minfo->update_flag == UPDATE_FLAG_DBG_WA_ENABLE ?
- 0xffff : 0xffffffff;
+ uint32_t wa_mask =
+ (minfo->update_flag & UPDATE_FLAG_DBG_WA_ENABLE) ? 0xffff : 0xffffffff;
m->compute_static_thread_mgmt_se0 = wa_mask;
m->compute_static_thread_mgmt_se1 = wa_mask;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
index 42d881809dc7..697b6d530d12 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
@@ -303,6 +303,15 @@ static void update_mqd(struct mqd_manager *mm, void *mqd,
update_cu_mask(mm, mqd, minfo, 0);
set_priority(m, q);
+ if (minfo && KFD_GC_VERSION(mm->dev) >= IP_VERSION(9, 4, 2)) {
+ if (minfo->update_flag & UPDATE_FLAG_IS_GWS)
+ m->compute_resource_limits |=
+ COMPUTE_RESOURCE_LIMITS__FORCE_SIMD_DIST_MASK;
+ else
+ m->compute_resource_limits &=
+ ~COMPUTE_RESOURCE_LIMITS__FORCE_SIMD_DIST_MASK;
+ }
+
q->is_active = QUEUE_IS_ACTIVE(*q);
}
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index 17fbedbf3651..80320b8603fc 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -532,6 +532,7 @@ struct queue_properties {
enum mqd_update_flag {
UPDATE_FLAG_DBG_WA_ENABLE = 1,
UPDATE_FLAG_DBG_WA_DISABLE = 2,
+ UPDATE_FLAG_IS_GWS = 4, /* quirk for gfx9 IP */
};
struct mqd_update_info {
@@ -1488,10 +1489,15 @@ void kfd_dec_compute_active(struct kfd_node *dev);
/* Cgroup Support */
/* Check with device cgroup if @kfd device is accessible */
-static inline int kfd_devcgroup_check_permission(struct kfd_node *kfd)
+static inline int kfd_devcgroup_check_permission(struct kfd_node *node)
{
#if defined(CONFIG_CGROUP_DEVICE) || defined(CONFIG_CGROUP_BPF)
- struct drm_device *ddev = adev_to_drm(kfd->adev);
+ struct drm_device *ddev;
+
+ if (node->xcp)
+ ddev = node->xcp->ddev;
+ else
+ ddev = adev_to_drm(node->adev);
return devcgroup_check_permission(DEVCG_DEV_CHAR, DRM_MAJOR,
ddev->render->index,
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
index 43eff221eae5..4858112f9a53 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
@@ -95,6 +95,7 @@ void kfd_process_dequeue_from_device(struct kfd_process_device *pdd)
int pqm_set_gws(struct process_queue_manager *pqm, unsigned int qid,
void *gws)
{
+ struct mqd_update_info minfo = {0};
struct kfd_node *dev = NULL;
struct process_queue_node *pqn;
struct kfd_process_device *pdd;
@@ -146,9 +147,10 @@ int pqm_set_gws(struct process_queue_manager *pqm, unsigned int qid,
}
pdd->qpd.num_gws = gws ? dev->adev->gds.gws_size : 0;
+ minfo.update_flag = gws ? UPDATE_FLAG_IS_GWS : 0;
return pqn->q->device->dqm->ops.update_queue(pqn->q->device->dqm,
- pqn->q, NULL);
+ pqn->q, &minfo);
}
void kfd_process_dequeue_from_all_devices(struct kfd_process *p)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
index e5f7c92eebcb..6ed2ec381aaa 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
@@ -1638,12 +1638,10 @@ static int fill_in_l2_l3_pcache(struct kfd_cache_properties **props_ext,
else
mode = UNKNOWN_MEMORY_PARTITION_MODE;
- if (pcache->cache_level == 2)
- pcache->cache_size = pcache_info[cache_type].cache_size * num_xcc;
- else if (mode)
- pcache->cache_size = pcache_info[cache_type].cache_size / mode;
- else
- pcache->cache_size = pcache_info[cache_type].cache_size;
+ pcache->cache_size = pcache_info[cache_type].cache_size;
+ /* Partition mode only affects L3 cache size */
+ if (mode && pcache->cache_level == 3)
+ pcache->cache_size /= mode;
if (pcache_info[cache_type].flags & CRAT_CACHE_FLAGS_DATA_CACHE)
pcache->cache_type |= HSA_CACHE_TYPE_DATA;
generated by cgit v1.2.3 (git 2.39.1) at 2025-12-26 03:59:51 +0000