This is a note to let you know that I've just added the patch titled
drm/amdkfd: range check cp bad op exception interrupts
to the 6.6-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary
The filename of the patch is:
drm-amdkfd-range-check-cp-bad-op-exception-interrupt.patch
and it can be found in the queue-6.6 subdirectory.
If you, or anyone else, feels it should not be added to the stable tree,
please let <stable@xxxxxxxxxxxxxxx> know about it.
commit ef1d808a900f2b8686bd05e994c9db3be1ed9316
Author: Jonathan Kim <Jonathan.Kim@xxxxxxx>
Date: Fri Mar 1 12:25:17 2024 -0500
drm/amdkfd: range check cp bad op exception interrupts
[ Upstream commit 0cac183b98d8a8c692c98e8dba37df15a9e9210d ]
Due to a CP interrupt bug, bad packet garbage exception codes are raised.
Do a range check so that the debugger and runtime do not receive garbage
codes.
Update the user api to guard exception code type checking as well.
Signed-off-by: Jonathan Kim <jonathan.kim@xxxxxxx>
Tested-by: Jesse Zhang <jesse.zhang@xxxxxxx>
Reviewed-by: Felix Kuehling <felix.kuehling@xxxxxxx>
Signed-off-by: Alex Deucher <alexander.deucher@xxxxxxx>
Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx>
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v10.c b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v10.c
index a7697ec8188e0..f85ca6cb90f56 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v10.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v10.c
@@ -336,7 +336,8 @@ static void event_interrupt_wq_v10(struct kfd_node *dev,
break;
}
kfd_signal_event_interrupt(pasid, context_id0 & 0x7fffff, 23);
- } else if (source_id == SOC15_INTSRC_CP_BAD_OPCODE) {
+ } else if (source_id == SOC15_INTSRC_CP_BAD_OPCODE &&
+ KFD_DBG_EC_TYPE_IS_PACKET(KFD_DEBUG_CP_BAD_OP_ECODE(context_id0))) {
kfd_set_dbg_ev_from_interrupt(dev, pasid,
KFD_DEBUG_DOORBELL_ID(context_id0),
KFD_EC_MASK(KFD_DEBUG_CP_BAD_OP_ECODE(context_id0)),
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c
index 2a65792fd1162..3ca9c160da7c2 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c
@@ -325,7 +325,8 @@ static void event_interrupt_wq_v11(struct kfd_node *dev,
/* CP */
if (source_id == SOC15_INTSRC_CP_END_OF_PIPE)
kfd_signal_event_interrupt(pasid, context_id0, 32);
- else if (source_id == SOC15_INTSRC_CP_BAD_OPCODE)
+ else if (source_id == SOC15_INTSRC_CP_BAD_OPCODE &&
+ KFD_DBG_EC_TYPE_IS_PACKET(KFD_CTXID0_CP_BAD_OP_ECODE(context_id0)))
kfd_set_dbg_ev_from_interrupt(dev, pasid,
KFD_CTXID0_DOORBELL_ID(context_id0),
KFD_EC_MASK(KFD_CTXID0_CP_BAD_OP_ECODE(context_id0)),
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c
index 27cdaea405017..8a6729939ae55 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c
@@ -385,7 +385,8 @@ static void event_interrupt_wq_v9(struct kfd_node *dev,
break;
}
kfd_signal_event_interrupt(pasid, sq_int_data, 24);
- } else if (source_id == SOC15_INTSRC_CP_BAD_OPCODE) {
+ } else if (source_id == SOC15_INTSRC_CP_BAD_OPCODE &&
+ KFD_DBG_EC_TYPE_IS_PACKET(KFD_DEBUG_CP_BAD_OP_ECODE(context_id0))) {
kfd_set_dbg_ev_from_interrupt(dev, pasid,
KFD_DEBUG_DOORBELL_ID(context_id0),
KFD_EC_MASK(KFD_DEBUG_CP_BAD_OP_ECODE(context_id0)),
diff --git a/include/uapi/linux/kfd_ioctl.h b/include/uapi/linux/kfd_ioctl.h
index eeb2fdcbdcb70..cd924c959d732 100644
--- a/include/uapi/linux/kfd_ioctl.h
+++ b/include/uapi/linux/kfd_ioctl.h
@@ -909,14 +909,25 @@ enum kfd_dbg_trap_exception_code {
KFD_EC_MASK(EC_DEVICE_NEW))
#define KFD_EC_MASK_PROCESS (KFD_EC_MASK(EC_PROCESS_RUNTIME) | \
KFD_EC_MASK(EC_PROCESS_DEVICE_REMOVE))
+#define KFD_EC_MASK_PACKET (KFD_EC_MASK(EC_QUEUE_PACKET_DISPATCH_DIM_INVALID) | \
+ KFD_EC_MASK(EC_QUEUE_PACKET_DISPATCH_GROUP_SEGMENT_SIZE_INVALID) | \
+ KFD_EC_MASK(EC_QUEUE_PACKET_DISPATCH_CODE_INVALID) | \
+ KFD_EC_MASK(EC_QUEUE_PACKET_RESERVED) | \
+ KFD_EC_MASK(EC_QUEUE_PACKET_UNSUPPORTED) | \
+ KFD_EC_MASK(EC_QUEUE_PACKET_DISPATCH_WORK_GROUP_SIZE_INVALID) | \
+ KFD_EC_MASK(EC_QUEUE_PACKET_DISPATCH_REGISTER_INVALID) | \
+ KFD_EC_MASK(EC_QUEUE_PACKET_VENDOR_UNSUPPORTED))
/* Checks for exception code types for KFD search */
+#define KFD_DBG_EC_IS_VALID(ecode) (ecode > EC_NONE && ecode < EC_MAX)
#define KFD_DBG_EC_TYPE_IS_QUEUE(ecode) \
- (!!(KFD_EC_MASK(ecode) & KFD_EC_MASK_QUEUE))
+ (KFD_DBG_EC_IS_VALID(ecode) && !!(KFD_EC_MASK(ecode) & KFD_EC_MASK_QUEUE))
#define KFD_DBG_EC_TYPE_IS_DEVICE(ecode) \
- (!!(KFD_EC_MASK(ecode) & KFD_EC_MASK_DEVICE))
+ (KFD_DBG_EC_IS_VALID(ecode) && !!(KFD_EC_MASK(ecode) & KFD_EC_MASK_DEVICE))
#define KFD_DBG_EC_TYPE_IS_PROCESS(ecode) \
- (!!(KFD_EC_MASK(ecode) & KFD_EC_MASK_PROCESS))
+ (KFD_DBG_EC_IS_VALID(ecode) && !!(KFD_EC_MASK(ecode) & KFD_EC_MASK_PROCESS))
+#define KFD_DBG_EC_TYPE_IS_PACKET(ecode) \
+ (KFD_DBG_EC_IS_VALID(ecode) && !!(KFD_EC_MASK(ecode) & KFD_EC_MASK_PACKET))
/* Runtime enable states */
