Thread (34 messages) 34 messages, 2 authors, 8d ago

Re: [PATCH v6 20/24] dyndbg,module: make proper substructs in _ddebug_info

From: sashiko-bot@kernel.org
Date: 2026-07-08 02:30:14
Also in: dri-devel

Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
- [Medium] Built-in modules without class maps erroneously inherit all global class maps
- [Low] Build failure in DRM subsystem due to struct name typo
--
--- Patch [20]: [PATCH v6 20/24] dyndbg,module: make proper substructs in _ddebug_info ---
commit 715ff68bbb0cbb69dec753531f42001f8daeef9e
Author: Jim Cromie [off-list ref]

dyndbg,module: make proper substructs in _ddebug_info

recompose struct _ddebug_info, inserting proper sub-structs.

The struct _ddebug_info has 2 pairs of _vec, num_##_vec fields, for
descs and classes respectively.  for_subvec() makes walking these
vectors less cumbersome, now lets move those field pairs into their
own "vec" structs: _ddebug_descs & _ddebug_class_maps, and re-compose
struct _ddebug_info to contain them cleanly.  This also lets us get
rid of for_subvec()'s num_##_vec paste-up.
quoted hunk ↗ jump to hunk
diff --git a/drivers/gpu/drm/drm_print.c b/drivers/gpu/drm/drm_print.c
index 711ae6606c6ea..da2171186c6bf 100644
--- a/drivers/gpu/drm/drm_print.c
+++ b/drivers/gpu/drm/drm_print.c
@@ -69,7 +69,7 @@ DECLARE_DYNDBG_CLASSMAP(drm_debug_classes, DD_CLASS_TYPE_DISJOINT_BITS, 0,
 			"DRM_UT_DP",
 			"DRM_UT_DRMRES");
 
-static struct ddebug_class_param drm_debug_bitmap = {
+static struct _ddebug_class_param drm_debug_bitmap = {
[Severity: Low]
Will this cause a build failure? 

The struct defined in include/linux/dynamic_debug.h is named 
ddebug_class_param, without the leading underscore.

[ ... ]
quoted hunk ↗ jump to hunk
diff --git a/lib/dynamic_debug.c b/lib/dynamic_debug.c
index af05f4ae3b55e..5a884cbd6294a 100644
--- a/lib/dynamic_debug.c
+++ b/lib/dynamic_debug.c
[ ... ]
quoted hunk ↗ jump to hunk
@@ -1184,12 +1185,12 @@ static void ddebug_attach_module_classes(struct ddebug_table *dt, struct _ddebug
 	 * the builtin/modular classmap vector/section.  Save the start
 	 * and length of the subrange at its edges.
 	 */
-	for_subvec(i, cm, di, classes) {
+	for_subvec(i, cm, di, maps) {
[ ... ]
quoted hunk ↗ jump to hunk
 	if (nc) {
-		dt->num_classes = nc;
+		dt->info.maps.len = nc;
 		vpr_info("module:%s attached %d classes\n", dt->mod_name, nc);
 	}
 }
@@ -1211,10 +1212,10 @@ static int ddebug_add_module(struct _ddebug_info *di, const char *modname)
[ ... ]
 	dt->mod_name = modname;
-	dt->ddebugs = di->descs;
-	dt->num_ddebugs = di->num_descs;
+	dt->info = *di;
 
 	INIT_LIST_HEAD(&dt->link);
 
-	if (di->classes && di->num_classes)
+	if (di->maps.len)
 		ddebug_attach_module_classes(dt, di);
[Severity: Medium]
If a built-in module has no class maps, does it unintentionally inherit 
the global list of class maps here?
        
When dt->info is assigned *di, it copies the global maps array, including 
the global di->maps.len. 

If the module has no class maps, ddebug_attach_module_classes() will run, 
but nc remains 0. Since dt->info.maps.len is only updated when nc > 0:

	if (nc) {
		dt->info.maps.len = nc;
		vpr_info("module:%s attached %d classes\n", dt->mod_name, nc);
	}

It won't be cleared to 0, leaving the module with the full global list of 
class maps. Should dt->info.maps.len be explicitly cleared when no module
classes are found?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260707-dd-maint-2-v6-0-381f3edb0045@gmail.com?part=20
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help