public inbox for drm-ai-reviews@public-inbox.freedesktop.org
 help / color / mirror / Atom feed
From: Jim Cromie <jim.cromie@gmail.com>
To: airlied@gmail.com, simona@ffwll.ch, jbaron@akamai.com,
	gregkh@linuxfoundation.org, Jim Cromie <jim.cromie@gmail.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	linux-kernel@vger.kernel.org
Cc: mripard@kernel.org, tzimmermann@suse.de,
	maarten.lankhorst@linux.intel.com, jani.nikula@intel.com,
	ville.syrjala@linux.intel.com, christian.koenig@amd.com,
	matthew.auld@intel.com, arunpravin.paneerselvam@amd.com,
	louis.chauvet@bootlin.com, skhan@linuxfoundation.org,
	pmladek@suse.com, ukaszb@chromium.org,
	dri-devel@lists.freedesktop.org, intel-gfx@lists.freedesktop.org,
	amd-gfx@lists.freedesktop.org
Subject: [PATCH v11 32/65] dyndbg: Harden classmap and callsite validation
Date: Fri, 13 Mar 2026 07:19:57 -0600	[thread overview]
Message-ID: <20260313132103.2529746-33-jim.cromie@gmail.com> (raw)
In-Reply-To: <20260313132103.2529746-1-jim.cromie@gmail.com>

Dynamic debug classmaps allow modules to _DEFINE or _USE classmaps,
which map classnames to a range of class-IDs (0..62) that they are
using.  This names => IDs mapping allows many modules to use 0..N
independently, and even allows (carefully) shared classnames.

Previously, class validation only caught some violations in _DEFINEd
classmaps; 0..62 range checks, overlaps among _DEFINEd classmaps, but
left blind spots in _USEs, for out of bounds manual offsets,
overlapping _USEd classmaps, unknown map types, and orphaned callsite
IDs. These oversights could lead to ambiguous routing or spammy
warnings during control query execution.

This commit hardens classmap and descriptor validation:

- Mirror the compile-time limits of _DEFINE by adding a static_assert
  to validate the _offset value passed to DYNAMIC_DEBUG_CLASSMAP_USE_.
- Add run-time overlap checks for _USEd classmaps in ddebug_add_module()
  to prevent collisions between private maps and imported APIs.
- Validate that module classmaps use known enum map types.
- Scan module descriptors at load time to print a single warning per
  missing class_id, rather than waiting for a user query to trip over it.
- Downgrade the global WARN_ONCE in ddebug_match_desc() to a
  pr_warn_ratelimited, since orphaned class IDs are now tracked and
  warned about early at module load.

Signed-off-by: Jim Cromie <jim.cromie@gmail.com>
---
 include/linux/dynamic_debug.h |  2 ++
 lib/dynamic_debug.c           | 52 +++++++++++++++++++++++++++++++----
 2 files changed, 48 insertions(+), 6 deletions(-)

diff --git a/include/linux/dynamic_debug.h b/include/linux/dynamic_debug.h
index 0d1245aefc69..28684cfc0937 100644
--- a/include/linux/dynamic_debug.h
+++ b/include/linux/dynamic_debug.h
@@ -262,6 +262,8 @@ struct _ddebug_class_param {
 
 #define __DYNAMIC_DEBUG_CLASSMAP_USE(_var, _offset, _uname)		\
 	extern struct _ddebug_class_map _var;				\
+	static_assert((_offset) >= 0 && (_offset) < _DPRINTK_CLASS_DFLT, \
+		      "classmap use offset must be in 0..62");		\
 	static struct _ddebug_class_user __aligned(8) __used		\
 	__section("__dyndbg_class_users") _uname = {			\
 		.mod_name = KBUILD_MODNAME,				\
diff --git a/lib/dynamic_debug.c b/lib/dynamic_debug.c
index 2083a8546460..635df6edb4cf 100644
--- a/lib/dynamic_debug.c
+++ b/lib/dynamic_debug.c
@@ -320,7 +320,7 @@ static bool ddebug_match_desc(const struct ddebug_query *query,
 	/* site is class'd */
 	site_map = ddebug_find_map_by_class_id(di, dp->class_id);
 	if (!site_map) {
-		WARN_ONCE(1, "unknown class_id %d, check %s's CLASSMAP definitions",
+		pr_warn_ratelimited("unknown class_id %d, check %s's CLASSMAP definitions\n",
 			  dp->class_id, di->mod_name);
 		return false;
 	}
@@ -1404,9 +1404,8 @@ static void ddebug_apply_class_users(const struct _ddebug_info *di)
 	(__dst)->info._vec.len = __nc;					\
 })
 
-static int __maybe_unused
-ddebug_class_range_overlap(struct _ddebug_class_map *cm,
-			   u64 *reserved_ids)
+static int ddebug_class_range_overlap(struct _ddebug_class_map *cm,
+				      u64 *reserved_ids)
 {
 	u64 range = (((1ULL << cm->length) - 1) << cm->base);
 
@@ -1420,6 +1419,23 @@ ddebug_class_range_overlap(struct _ddebug_class_map *cm,
 	return 0;
 }
 
+static int ddebug_class_user_overlap(struct _ddebug_class_user *cli,
+				     u64 *reserved_ids)
+{
+	struct _ddebug_class_map *cm = cli->map;
+	int base = cm->base + cli->offset;
+	u64 range = (((1ULL << cm->length) - 1) << base);
+
+	if (range & *reserved_ids) {
+		pr_err("[%d..%d] (from %s) conflicts with %llx\n", base,
+		       base + cm->length - 1, cm->class_names[0],
+		       *reserved_ids);
+		return -EINVAL;
+	}
+	*reserved_ids |= range;
+	return 0;
+}
+
 /*
  * Allocate a new ddebug_table for the given module
  * and add it to the global list.
@@ -1430,6 +1446,7 @@ static int ddebug_add_module(struct _ddebug_info *di)
 	struct _ddebug_class_map *cm;
 	struct _ddebug_class_user *cli;
 	u64 reserved_ids = 0;
+	u64 bad_ids = 0;
 	int i;
 
 	if (!di->descs.len)
@@ -1454,10 +1471,33 @@ static int ddebug_add_module(struct _ddebug_info *di)
 	dd_mark_vector_subrange(i, cli, &dt->info, users, dt);
 	/* now di is stale */
 
-	/* insure 2+ classmaps share the per-module 0..62 class_id space */
-	for_subvec(i, cm, &dt->info, maps)
+	/* validate class map types and the per-module 0..62 class_id space */
+	for_subvec(i, cm, &dt->info, maps) {
+		if (cm->map_type > DD_CLASS_TYPE_LEVEL_NUM) {
+			pr_err("module %s has unknown classmap type %d\n", dt->info.mod_name, cm->map_type);
+			goto cleanup;
+		}
 		if (ddebug_class_range_overlap(cm, &reserved_ids))
 			goto cleanup;
+	}
+
+	for_subvec(i, cli, &dt->info, users)
+		if (ddebug_class_user_overlap(cli, &reserved_ids))
+			goto cleanup;
+
+	/* validate all class_ids against module's classmaps/users */
+	for (i = 0; i < dt->info.descs.len; i++) {
+		struct _ddebug *dp = &dt->info.descs.start[i];
+
+		if (dp->class_id == _DPRINTK_CLASS_DFLT)
+			continue;
+		if (bad_ids & (1ULL << dp->class_id))
+			continue;
+		if (!ddebug_find_map_by_class_id(&dt->info, dp->class_id)) {
+			pr_warn("module %s uses unknown class_id %d\n", dt->info.mod_name, dp->class_id);
+			bad_ids |= (1ULL << dp->class_id);
+		}
+	}
 
 	mutex_lock(&ddebug_lock);
 	list_add_tail(&dt->link, &ddebug_tables);
-- 
2.53.0


  parent reply	other threads:[~2026-03-13 13:24 UTC|newest]

Thread overview: 113+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-03-13 13:19 [PATCH v11 00/65] Fix DRM_USE_DYNAMIC_DEBUG=y Jim Cromie
2026-03-13 13:19 ` [PATCH v11 01/65] dyndbg: fix NULL ptr on i386 due to section alignment Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:19 ` [PATCH v11 02/65] dyndbg: factor ddebug_match_desc out from ddebug_change Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:19 ` [PATCH v11 03/65] dyndbg: add stub macro for DECLARE_DYNDBG_CLASSMAP Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:19 ` [PATCH v11 04/65] docs/dyndbg: update examples \012 to \n Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:19 ` [PATCH v11 05/65] docs/dyndbg: explain flags parse 1st Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:19 ` [PATCH v11 06/65] test-dyndbg: fixup CLASSMAP usage error Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:19 ` [PATCH v11 07/65] dyndbg: reword "class unknown, " to "class:_UNKNOWN_" Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:19 ` [PATCH v11 08/65] dyndbg: make ddebug_class_param union members same size Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:19 ` [PATCH v11 09/65] dyndbg: drop NUM_TYPE_ARRAY Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:19 ` [PATCH v11 10/65] dyndbg: tweak pr_fmt to avoid expansion conflicts Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:19 ` [PATCH v11 11/65] dyndbg: reduce verbose/debug clutter Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:19 ` [PATCH v11 12/65] dyndbg: refactor param_set_dyndbg_classes and below Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:19 ` [PATCH v11 13/65] dyndbg: tighten fn-sig of ddebug_apply_class_bitmap Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:19 ` [PATCH v11 14/65] dyndbg: replace classmap list with a vector Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:19 ` [PATCH v11 15/65] dyndbg: macrofy a 2-index for-loop pattern Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:19 ` [PATCH v11 16/65] dyndbg, module: make proper substructs in _ddebug_info Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:19 ` [PATCH v11 17/65] dyndbg: hoist classmap-filter-by-modname up to ddebug_add_module Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:19 ` [PATCH v11 18/65] dyndbg: move mod_name down from struct ddebug_table to _ddebug_info Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:19 ` [PATCH v11 19/65] dyndbg-API: remove DD_CLASS_TYPE_(DISJOINT|LEVEL)_NAMES and code Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:19 ` [PATCH v11 20/65] selftests-dyndbg: add a dynamic_debug run_tests target Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:19 ` [PATCH v11 21/65] dyndbg: change __dynamic_func_call_cls* macros into expressions Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:19 ` [PATCH v11 22/65] dyndbg-API: replace DECLARE_DYNDBG_CLASSMAP Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:19 ` [PATCH v11 23/65] dyndbg: detect class_id reservation conflicts Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:19 ` [PATCH v11 24/65] dyndbg: check DYNAMIC_DEBUG_CLASSMAP_DEFINE args at compile-time Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:19 ` [PATCH v11 25/65] dyndbg-test: change do_prints testpoint to accept a loopct Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:19 ` [PATCH v11 26/65] dyndbg-API: promote DYNAMIC_DEBUG_CLASSMAP_PARAM to API Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:19 ` [PATCH v11 27/65] dyndbg: treat comma as a token separator Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:19 ` [PATCH v11 28/65] dyndbg: split multi-query strings with % Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:19 ` [PATCH v11 29/65] selftests-dyndbg: add test_mod_submod Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:19 ` [PATCH v11 30/65] dyndbg: resolve "protection" of class'd pr_debug Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:19 ` [PATCH v11 31/65] dyndbg: add DYNAMIC_DEBUG_CLASSMAP_USE_(dd_class_name, offset) Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:19 ` Jim Cromie [this message]
2026-03-13 21:00   ` Claude review: dyndbg: Harden classmap and callsite validation Claude Code Review Bot
2026-03-13 13:19 ` [PATCH v11 33/65] docs/dyndbg: add classmap info to howto Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:19 ` [PATCH v11 34/65] drm: use correct ccflags-y spelling Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:20 ` [PATCH v11 35/65] drm-dyndbg: adapt drm core to use dyndbg classmaps-v2 Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:20 ` [PATCH v11 36/65] drm-dyndbg: adapt DRM to invoke DYNAMIC_DEBUG_CLASSMAP_PARAM Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:20 ` [PATCH v11 37/65] drm-print: modernize an archaic comment Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:20 ` [PATCH v11 38/65] drm-print: fix config-dependent unused variable Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:20 ` [PATCH v11 39/65] drm-dyndbg: DRM_CLASSMAP_USE in amdgpu driver Jim Cromie
2026-03-13 13:20 ` [PATCH v11 40/65] drm-dyndbg: DRM_CLASSMAP_USE in i915 driver Jim Cromie
2026-03-13 13:20 ` [PATCH v11 41/65] drm-dyndbg: DRM_CLASSMAP_USE in drm_crtc_helper Jim Cromie
2026-03-13 13:20 ` [PATCH v11 42/65] drm-dyndbg: DRM_CLASSMAP_USE in drm_dp_helper Jim Cromie
2026-03-13 13:20 ` [PATCH v11 43/65] drm-dyndbg: DRM_CLASSMAP_USE in nouveau Jim Cromie
2026-03-13 13:20 ` [PATCH v11 44/65] drm-dyndbg: add DRM_CLASSMAP_USE to Xe driver Jim Cromie
2026-03-13 13:20 ` [PATCH v11 45/65] drm-dyndbg: add DRM_CLASSMAP_USE to virtio_gpu Jim Cromie
2026-03-13 13:20 ` [PATCH v11 46/65] drm-dyndbg: add DRM_CLASSMAP_USE to simpledrm Jim Cromie
2026-03-13 13:20 ` [PATCH v11 47/65] drm-dyndbg: add DRM_CLASSMAP_USE to bochs Jim Cromie
2026-03-13 13:20 ` [PATCH v11 48/65] drm-dyndbg: add DRM_CLASSMAP_USE to etnaviv Jim Cromie
2026-03-13 13:20 ` [PATCH v11 49/65] drm-dyndbg: add DRM_CLASSMAP_USE to gma500 driver Jim Cromie
2026-03-13 13:20 ` [PATCH v11 50/65] drm-dyndbg: add DRM_CLASSMAP_USE to radeon Jim Cromie
2026-03-13 13:20 ` [PATCH v11 51/65] drm-dyndbg: add DRM_CLASSMAP_USE to vmwgfx driver Jim Cromie
2026-03-13 13:20 ` [PATCH v11 52/65] drm-dyndbg: add DRM_CLASSMAP_USE to vkms driver Jim Cromie
2026-03-13 13:20 ` [PATCH v11 53/65] drm-dyndbg: add DRM_CLASSMAP_USE to udl driver Jim Cromie
2026-03-13 13:20 ` [PATCH v11 54/65] drm-dyndbg: add DRM_CLASSMAP_USE to mgag200 driver Jim Cromie
2026-03-13 13:20 ` [PATCH v11 55/65] drm-dyndbg: add DRM_CLASSMAP_USE to the gud driver Jim Cromie
2026-03-13 13:20 ` [PATCH v11 56/65] drm-dyndbg: add DRM_CLASSMAP_USE to the qxl driver Jim Cromie
2026-03-13 13:20 ` [PATCH v11 57/65] drm-dyndbg: add DRM_CLASSMAP_USE to the drm_gem_shmem_helper driver Jim Cromie
2026-03-13 13:20 ` [PATCH v11 58/65] accel: add -DDYNAMIC_DEBUG_MODULE to subdir-ccflags Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:20 ` [PATCH v11 59/65] accel/ivpu: implement IVPU_DBG_* as a dyndbg classmap Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:20 ` [PATCH v11 60/65] drm: restore CONFIG_DRM_USE_DYNAMIC_DEBUG un-BROKEN Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:20 ` [PATCH v11 61/65] accel/ethosu: enable drm.debug control Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:20 ` [PATCH v11 62/65] accel/rocket: " Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:20 ` [PATCH v11 63/65] drm_buddy: fix 64-bit truncation in power-of-2 rounding Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:20 ` [PATCH v11 64/65] drm_print: fix drm_printer dynamic debug bypass Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 13:20 ` [PATCH v11 65/65] drm_vblank: use dyndbg's static-key to avoid flag-check Jim Cromie
2026-03-13 21:00   ` Claude review: " Claude Code Review Bot
2026-03-13 21:00 ` Claude review: Fix DRM_USE_DYNAMIC_DEBUG=y Claude Code Review Bot

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20260313132103.2529746-33-jim.cromie@gmail.com \
    --to=jim.cromie@gmail.com \
    --cc=airlied@gmail.com \
    --cc=akpm@linux-foundation.org \
    --cc=amd-gfx@lists.freedesktop.org \
    --cc=arunpravin.paneerselvam@amd.com \
    --cc=christian.koenig@amd.com \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=gregkh@linuxfoundation.org \
    --cc=intel-gfx@lists.freedesktop.org \
    --cc=jani.nikula@intel.com \
    --cc=jbaron@akamai.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=louis.chauvet@bootlin.com \
    --cc=maarten.lankhorst@linux.intel.com \
    --cc=matthew.auld@intel.com \
    --cc=mripard@kernel.org \
    --cc=pmladek@suse.com \
    --cc=simona@ffwll.ch \
    --cc=skhan@linuxfoundation.org \
    --cc=tzimmermann@suse.de \
    --cc=ukaszb@chromium.org \
    --cc=ville.syrjala@linux.intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox