Thread (77 messages) 77 messages, 6 authors, 2025-11-18

Re: [PATCH v3 36/50] functionfs: switch to simple_remove_by_name()

From: bot+bpf-ci@kernel.org
Date: 2025-11-11 07:53:17
Also in: bpf, linux-efi, linux-fsdevel, linux-mm, linux-usb, ocfs2-devel, selinux

quoted hunk ↗ jump to hunk
diff --git a/drivers/usb/gadget/function/f_fs.c b/drivers/usb/gadget/function/f_fs.c
index 47cfbe41f..43dcd39b7 100644
--- a/drivers/usb/gadget/function/f_fs.c
+++ b/drivers/usb/gadget/function/f_fs.c
[ ... ]
quoted hunk ↗ jump to hunk
@@ -2323,6 +2316,7 @@ static int ffs_epfiles_create(struct ffs_data *ffs)
 {
 	struct ffs_epfile *epfile, *epfiles;
 	unsigned i, count;
+	int err;

 	count = ffs->eps_count;
 	epfiles = kcalloc(count, sizeof(*epfiles), GFP_KERNEL);
@@ -2339,12 +2333,11 @@ static int ffs_epfiles_create(struct ffs_data *ffs)
 			sprintf(epfile->name, "ep%02x", ffs->eps_addrmap[i]);
 		else
 			sprintf(epfile->name, "ep%u", i);
-		epfile->dentry = ffs_sb_create_file(ffs->sb, epfile->name,
-						 epfile,
-						 &ffs_epfile_operations);
-		if (!epfile->dentry) {
+		err = ffs_sb_create_file(ffs->sb, epfile->name,
+					 epfile, &ffs_epfile_operations);
+		if (err) {
 			ffs_epfiles_destroy(epfiles, i - 1);
-			return -ENOMEM;
+			return err;
 		}
 	}
@@ -2355,13 +2348,11 @@ static int ffs_epfiles_create(struct ffs_data *ffs)
 static void ffs_epfiles_destroy(struct ffs_epfile *epfiles, unsigned count)
 {
 	struct ffs_epfile *epfile = epfiles;
+	struct dentry *root = epfile->ffs->sb->s_root;
Can this dereference uninitialized memory in ffs_epfiles_destroy()?

When ffs_epfiles_create() calls ffs_epfiles_destroy(epfiles, i - 1) after
the first ffs_sb_create_file() call fails (when i=1), it passes count=0.
The initialization loop starts at i=1, so epfiles[0].ffs is never
initialized. The kcalloc() zeros the memory, making epfiles[0].ffs NULL.

The new code unconditionally dereferences epfile->ffs->sb->s_root before
the loop, causing a NULL pointer dereference when count=0. The old code
was safe because the dereference only happened inside the loop.
 	for (; count; --count, ++epfile) {
 		BUG_ON(mutex_is_locked(&epfile->mutex));
-		if (epfile->dentry) {
-			simple_recursive_removal(epfile->dentry, NULL);
-			epfile->dentry = NULL;
-		}
+		simple_remove_by_name(root, epfile->name, NULL);
 	}

 	kfree(epfiles);

---
AI reviewed your patch. Please fix the bug or email reply why it's not a bug.
See: https://github.com/kernel-patches/vmtest/blob/master/ci/claude/README.md

CI run summary: https://github.com/kernel-patches/bpf/actions/runs/19257803162
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help