[PATCH 59/95] drm/amdgpu: Add debugfs to drop UALink protocol messages

From: Mukul Joshi <[email protected]>

Introduce a debugfs-controlled bitmap, amdgpu_ualink_drop_msg_bitmap,
where each bit corresponds to a value from
enum AMDGPU_UALINK_PROTOCOL_MESSAGES (e.g. NPA-REQ, NPA-RSP, NPA-FAIL,
NPA-REVOKE, NPA-RELEASE).

When a bit is set, the next incoming message of that type is dropped
in amdgpu_ualink_process_irq() and the bit is atomically cleared via
test_and_clear_bit(), so subsequent incoming messages are processed
normally. This allows exercising the UALink connection reset and
recovery paths (NPA-RSP timeout on the importer, retransmit/teardown
on the exporter, etc.) by injecting a single packet loss from
userspace, e.g.:

    # drop one NPA-REQ
    echo 0x08 > /sys/kernel/debug/dri/0/amdgpu_ualink_drop_msg_bitmap
    # drop one NPA-RSP
    echo 0x10 > /sys/kernel/debug/dri/0/amdgpu_ualink_drop_msg_bitmap

The bitmap lives in struct amdgpu_ualink_mgr and is exposed via
debugfs_create_ulong() next to the existing amdgpu_ualink_test entry.

Signed-off-by: Mukul Joshi <[email protected]>
Reviewed-by: Felix Kuehling <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c | 14 ++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_ualink.c  | 13 +++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_ualink.h  | 16 ++++++++++++++++
 3 files changed, 43 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
index 132d054900b5b..d80fe17556dab 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
@@ -2229,6 +2229,20 @@ int amdgpu_debugfs_init(struct amdgpu_device *adev)
 	debugfs_create_file("amdgpu_benchmark", 0200, root, adev,
 			    &amdgpu_benchmark_fops);
 
+	/* Debug-only: bitmap of incoming UALink protocol messages to drop.
+	 * Each bit position corresponds to an enum AMDGPU_UALINK_PROTOCOL_MESSAGES
+	 * value. Setting a bit causes exactly one matching incoming packet to be
+	 * dropped, after which the bit auto-clears and traffic resumes. Used to
+	 * exercise the connection reset paths.
+	 *
+	 * Examples (drop one NPA-REQ):
+	 *   echo 0x8  > /sys/kernel/debug/dri/0/amdgpu_ualink_drop_msg_bitmap
+	 * (drop one NPA-RSP):
+	 *   echo 0x10 > /sys/kernel/debug/dri/0/amdgpu_ualink_drop_msg_bitmap
+	 */
+	debugfs_create_ulong("amdgpu_ualink_drop_msg_bitmap", 0600, root,
+			     &adev->ualink.drop_msg_bitmap);
+
 	adev->debugfs_vbios_blob.data = adev->bios;
 	adev->debugfs_vbios_blob.size = adev->bios_size;
 	debugfs_create_blob("amdgpu_vbios", 0444, root,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ualink.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ualink.c
index cd9ee0484f9c2..1b8f0a6162b6b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ualink.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ualink.c
@@ -5344,6 +5344,19 @@ static int amdgpu_ualink_process_irq(struct amdgpu_device *adev,
 	dev_dbg(adev->dev, "Got MSG: remote acc_id %u msg_type %u\n",
 		src_acc_id, msg_type);
 
+	/* Debug hook: if the bit corresponding to this msg_type is set in
+	 * drop_msg_bitmap, drop this single packet and clear the bit so that
+	 * any subsequent incoming messages are processed normally. Used to
+	 * exercise the connection reset/recovery paths via debugfs.
+	 */
+	if (msg_type < BITS_PER_LONG &&
+	    test_and_clear_bit(msg_type, &adev->ualink.drop_msg_bitmap)) {
+		dev_warn(adev->dev,
+			 "DROP MSG (debugfs): src acc_id %u msg_type %u dw[0-3] 0x%x 0x%x 0x%x 0x%x\n",
+			 src_acc_id, msg_type, dw0, dw1, dw2, dw3);
+		return handled;
+	}
+
 	switch (msg_type) {
 	case AMDGPU_UALINK_HELLO_MSG:
 		receiver_acc_id = (dw0 >> AMDGPU_UALINK_HELLO_MSG_RECV_ACCID_SHIFT) &
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ualink.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ualink.h
index a1fd5fe953115..407220710d9b4 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ualink.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ualink.h
@@ -338,6 +338,22 @@ struct amdgpu_ualink_mgr {
 
 	/* Sequence number to track the need for TLB flushes */
 	atomic64_t last_flushed_tlb_seq;
+
+	/* Debug-only: bitmap of incoming UALink protocol messages to drop.
+	 *
+	 * Each bit position corresponds to a value from
+	 * enum AMDGPU_UALINK_PROTOCOL_MESSAGES:
+	 *   BIT(AMDGPU_UALINK_NPA_REQ_MSG) drops one incoming NPA-REQ.
+	 *   BIT(AMDGPU_UALINK_NPA_RSP_MSG) drops one incoming NPA-RSP.
+	 *
+	 *   On reception of a message whose corresponding bit is set, the bit
+	 *   is atomically cleared and the message is silently dropped. This
+	 *   means at most one packet per set bit is dropped; any further
+	 *   incoming packets of the same type are processed normally. This is
+	 *   intended to exercise the connection reset / recovery paths from a
+	 *   debugfs handle.
+	 */
+	unsigned long drop_msg_bitmap;
 };
 
 int amdgpu_ualink_init_interrupt(struct amdgpu_device *adev);
-- 
2.55.0