accel/amdxdna: Fix clflush buffer size

The firmware is told the buffer is req.buf_size bytes. It may read/write
the entire region. If the CPU only flushes a subset, the remaining cache
lines could contain stale data, causing the device to see garbage.

Fixes: 6e87001fe1 ("accel/amdxdna: Adjust size for copy_to_user()")
Reviewed-by: Mario Limonciello (AMD) <superm1@kernel.org>
Signed-off-by: Lizhi Hou <lizhi.hou@amd.com>
Link: https://patch.msgid.link/20260507040207.178111-1-lizhi.hou@amd.com
This commit is contained in:
Lizhi Hou 2026-05-06 21:02:07 -07:00
parent 55bd5e685b
commit ada61841ca

View File

@ -390,7 +390,7 @@ int aie2_query_status(struct amdxdna_dev_hdl *ndev, char __user *buf,
req.num_cols = hweight32(aie_bitmap);
req.aie_bitmap = aie_bitmap;
drm_clflush_virt_range(buff_addr, size); /* device can access */
drm_clflush_virt_range(buff_addr, req.dump_buff_size); /* device can access */
ret = aie_send_mgmt_msg_wait(&ndev->aie, &msg);
if (ret) {
XDNA_ERR(xdna, "Error during NPU query, status %d", ret);
@ -442,7 +442,7 @@ int aie2_query_telemetry(struct amdxdna_dev_hdl *ndev,
req.buf_size = buf_sz;
req.type = header->type;
drm_clflush_virt_range(addr, size); /* device can access */
drm_clflush_virt_range(addr, req.buf_size); /* device can access */
ret = aie_send_mgmt_msg_wait(&ndev->aie, &msg);
if (ret) {
XDNA_ERR(xdna, "Query telemetry failed, status %d", ret);
@ -1186,7 +1186,7 @@ int aie2_query_app_health(struct amdxdna_dev_hdl *ndev, u32 context_id,
req.context_id = context_id;
req.buf_size = buf_size;
drm_clflush_virt_range(buf, sizeof(*report));
drm_clflush_virt_range(buf, req.buf_size);
ret = aie_send_mgmt_msg_wait(&ndev->aie, &msg);
if (ret) {
XDNA_ERR(xdna, "Get app health failed, ret %d status 0x%x", ret, resp.status);