drm/xe/vf: Add drm_dev guards when detaching CCS read/write buffers

CCS read/write buffers are freed during BO destruction. In some cases,
BOs may be destroyed after the device is unbound but while the DRM
structure remains valid, leading to NULL pointer dereferences when
accessing device resources.

BUG: kernel NULL pointer dereference, address: 0000000000000000
PGD 0 P4D 0
Oops: Oops: 0000 [#1] SMP NOPTI
CPU: 0 UID: 0 PID: 9376 Comm: xe_pat Not tainted 7.2.0-rc2+ #1 PREEMPT(lazy)
RIP: 0010:xe_sriov_vf_ccs_rw_update_bb_addr+0x4d/0xa0 [xe]
RSP: 0018:ffffcf304110b9c8 EFLAGS: 00010246
RAX: ffff8a85c38a0a00 RBX: 00000000810ef000 RCX: 0000000000000000
RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffff8a85c39c1888
RBP: ffffcf304110b9e8 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000000 R12: ffff8a85c39c1888
R13: 0000000000000000 R14: ffff8a85c39b4f28 R15: ffff8a85c3885000
FS:  0000000000000000(0000) GS:ffff8a878b809000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000000000000000 CR3: 000000010314a002 CR4: 0000000000772ef0
PKRU: 55555554
Call Trace:
 <TASK>
 xe_migrate_ccs_rw_copy_clear+0x98/0x120 [xe]
 xe_sriov_vf_ccs_detach_bo+0x2c/0x60 [xe]
 xe_ttm_bo_delete_mem_notify+0xc8/0xe0 [xe]
 ttm_bo_cleanup_memtype_use+0x26/0x80 [ttm]
 ttm_bo_release+0x29e/0x2d0 [ttm]
 ttm_bo_fini+0x39/0x70 [ttm]
 xe_gem_object_free+0x1f/0x30 [xe]
 drm_gem_object_free+0x1d/0x40
 ttm_bo_vm_close+0x5f/0x90 [ttm]
 remove_vma+0x2c/0x70
 tear_down_vmas+0x63/0xf0
 exit_mmap+0x20d/0x3f0
 __mmput+0x45/0x170
 mmput+0x31/0x40
 do_exit+0x2ba/0xac0
 do_group_exit+0x2d/0xb0
 __x64_sys_exit_group+0x18/0x20
 x64_sys_call+0x14a0/0x2390
 do_syscall_64+0xdd/0x640
 ? count_memcg_events+0xea/0x240
 ? handle_mm_fault+0x1ec/0x2f0

Fixes: 864690cf4dd6 ("drm/xe/vf: Attach and detach CCS copy commands with BO")
Signed-off-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://patch.msgid.link/20260721052215.2267228-2-satyanarayana.k.v.p@intel.com
(cherry picked from commit 1ae415a6eefe5004954a1d352b1718faca8844ef)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>

diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c
index 7d28290e7d1ca38f092b8a2a29ab4852b0fe0e76..be787f331768d56eb4bbd44d06a31e2c3b79161f 100644 (file)
--- a/drivers/gpu/drm/xe/xe_migrate.c
+++ b/drivers/gpu/drm/xe/xe_migrate.c
@@ -1313,6 +1313,7 @@ int xe_migrate_ccs_rw_copy(struct xe_tile *tile, struct xe_exec_queue *q,
  * content.
  * @src_bo: The buffer object @src is currently bound to.
  * @read_write : Creates BB commands for CCS read/write.
+ * @bound: Device is bound
  *
  * Directly clearing the BB lacks atomicity and can lead to undefined
  * behavior if the vCPU is halted mid-operation during the clearing
@@ -1325,7 +1326,8 @@ int xe_migrate_ccs_rw_copy(struct xe_tile *tile, struct xe_exec_queue *q,
  * Returns: None.
  */
 void xe_migrate_ccs_rw_copy_clear(struct xe_bo *src_bo,
-                                 enum xe_sriov_vf_ccs_rw_ctxs read_write)
+                                 enum xe_sriov_vf_ccs_rw_ctxs read_write,
+                                 bool bound)
 {
        struct xe_mem_pool_node *bb = src_bo->bb_ccs[read_write];
        struct xe_device *xe = xe_bo_device(src_bo);
@@ -1339,13 +1341,15 @@ void xe_migrate_ccs_rw_copy_clear(struct xe_bo *src_bo,
        bb_pool = ctx->mem.ccs_bb_pool;
 
        scoped_guard(mutex, xe_mem_pool_bo_swap_guard(bb_pool)) {
-               xe_mem_pool_swap_shadow_locked(bb_pool);
+               if (bound) {
+                       xe_mem_pool_swap_shadow_locked(bb_pool);
 
-               cs = xe_mem_pool_node_cpu_addr(bb);
-               memset(cs, MI_NOOP, bb->sa_node.size);
-               xe_sriov_vf_ccs_rw_update_bb_addr(ctx);
+                       cs = xe_mem_pool_node_cpu_addr(bb);
+                       memset(cs, MI_NOOP, bb->sa_node.size);
+                       xe_sriov_vf_ccs_rw_update_bb_addr(ctx);
 
-               xe_mem_pool_sync_shadow_locked(bb);
+                       xe_mem_pool_sync_shadow_locked(bb);
+               }
                xe_mem_pool_free_node(bb);
                src_bo->bb_ccs[read_write] = NULL;
        }

diff --git a/drivers/gpu/drm/xe/xe_migrate.h b/drivers/gpu/drm/xe/xe_migrate.h
index 78e5b63f3ebeafc95d59862972e6b5a566f0d34a..c3a268b01768ae6c45329864c83bc4013ca749e0 100644 (file)
--- a/drivers/gpu/drm/xe/xe_migrate.h
+++ b/drivers/gpu/drm/xe/xe_migrate.h
@@ -142,7 +142,8 @@ int xe_migrate_ccs_rw_copy(struct xe_tile *tile, struct xe_exec_queue *q,
                           enum xe_sriov_vf_ccs_rw_ctxs read_write);
 
 void xe_migrate_ccs_rw_copy_clear(struct xe_bo *src_bo,
-                                 enum xe_sriov_vf_ccs_rw_ctxs read_write);
+                                 enum xe_sriov_vf_ccs_rw_ctxs read_write,
+                                 bool bound);
 
 struct xe_lrc *xe_migrate_lrc(struct xe_migrate *migrate);
 struct xe_exec_queue *xe_migrate_exec_queue(struct xe_migrate *migrate);

diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
index 6787564629c65f8315e62d28eb153617f787b3a6..a8c831fbee3b5a241b2207e044484ea8843d90ac 100644 (file)
--- a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
+++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
@@ -3,6 +3,8 @@
  * Copyright © 2025 Intel Corporation
  */
 
+#include <drm/drm_drv.h>
+
 #include "instructions/xe_mi_commands.h"
 #include "instructions/xe_gpu_commands.h"
 #include "xe_bb.h"
@@ -446,7 +448,7 @@ err_unwind:
         */
        for_each_ccs_rw_ctx(ctx_id) {
                if (bo->bb_ccs[ctx_id])
-                       xe_migrate_ccs_rw_copy_clear(bo, ctx_id);
+                       xe_migrate_ccs_rw_copy_clear(bo, ctx_id, true);
        }
        return err;
 }
@@ -466,19 +468,27 @@ int xe_sriov_vf_ccs_detach_bo(struct xe_bo *bo)
        struct xe_device *xe = xe_bo_device(bo);
        enum xe_sriov_vf_ccs_rw_ctxs ctx_id;
        struct xe_mem_pool_node *bb;
+       bool bound;
+       int idx;
 
        xe_assert(xe, IS_VF_CCS_READY(xe));
 
        if (!xe_bo_has_valid_ccs_bb(bo))
                return 0;
 
+       bound = drm_dev_enter(&xe->drm, &idx);
+
        for_each_ccs_rw_ctx(ctx_id) {
                bb = bo->bb_ccs[ctx_id];
                if (!bb)
                        continue;
 
-               xe_migrate_ccs_rw_copy_clear(bo, ctx_id);
+               xe_migrate_ccs_rw_copy_clear(bo, ctx_id, bound);
        }
+
+       if (bound)
+               drm_dev_exit(idx);
+
        return 0;
 }