Skip to contentSkip to conversation

fs/ntfs3: fix info-leak on partial LZNT decompress in ni_read_frame()

PATCH In review Samuel Page 2 participants 2 messages Jun 23 – Jul 28 NTFS3 filesystem

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 fs/ntfs3: fix info-leak on partial LZNT decompress in ni_read_frame()

ni_read_frame() decompresses an LZNT $DATA frame into the vmapped target pages and then trusts decompress_lznt()'s return value:

  unc_size = decompress_lznt(frame_ondisk, ondisk_size, frame_mem,
                             frame_size);
  if ((ssize_t)unc_size < 0)        err = unc_size;
  else if (!unc_size || unc_size > frame_size)  err = -EINVAL;

decompress_lznt() stops as soon as the compressed stream is exhausted (e.g. a zero chunk header) and returns the number of bytes it actually wrote, which may be far less than frame_size. The bytes between unc_size and frame_size are never written. The only memset() that follows zeroes the region beyond i_valid; when the frame lies entirely within the file's valid size that memset() does not run, so the gap retains whatever was in the just-vmapped pages. All pages are then marked uptodate and returned to userspace, disclosing uninitialized (recently-freed) kernel page memory. A crafted compressed file whose stream decompresses to only a few bytes leaks the remainder of every frame on a plain read(2), which is enough to recover kernel pointers and defeat KASLR.

Zero the [unc_size, frame_size) tail immediately after a successful LZNT decompress so the remainder reads back as zero.

Fixes 4342306f0f0d ("fs/ntfs3: Add file operations and implementation")
Assisted-by Bynario AI

---
Reproduced on a non-KASAN, KASLR-enabled arm64 guest: a crafted file whose 64 KiB LZNT frames each decompress to 8 bytes leaks the [8, 65536) tail of every frame on a plain read(2) - up to ~3.3 MiB of uninitialized page memory, including kernel-text return addresses from freed VMAP_STACK pages, from which the KASLR base was recovered and verified vs /proc/kallsyms. With the patch the same read returns zeroes; a zeroed control image (valid_size truncated so the existing memset runs) leaks nothing, isolating the cause.

 fs/ntfs3/frecord.c | 9 +++++++++
 1 file changed, 9 insertions(+)
fs/ntfs3/frecord.c +9 -0
index 7b035da63c12..cd11df75bdbc 100644
@@ -2443,6 +2443,15 @@ int ni_read_frame(struct ntfs_inode *ni, u64 frame_vbo, struct page **pages,
 			err = unc_size;
 		else if (!unc_size || unc_size > frame_size)
 			err = -EINVAL;
+		else if (unc_size < frame_size) {
+			/*
+			 * Partial decompress: zero the [unc_size, frame_size)
+			 * tail.  decompress_lznt() leaves it untouched, so
+			 * without this the freshly vmapped pages would expose
+			 * uninitialized kernel memory to userspace.
+			 */
+			memset(frame_mem + unc_size, 0, frame_size - unc_size);
+		}
 	}
 	if (!err && valid_size < frame_vbo + frame_size) {
 		size_t ok = valid_size - frame_vbo;
Signature · 1 line
2.54.0
#1 lore raw
Quoted 60 lines · Samuel Page

ni_read_frame() decompresses an LZNT $DATA frame into the vmapped target pages and then trusts decompress_lznt()'s return value:

   unc_size = decompress_lznt(frame_ondisk, ondisk_size, frame_mem,
                              frame_size);
   if ((ssize_t)unc_size < 0)        err = unc_size;
   else if (!unc_size || unc_size > frame_size)  err = -EINVAL;

decompress_lznt() stops as soon as the compressed stream is exhausted (e.g. a zero chunk header) and returns the number of bytes it actually wrote, which may be far less than frame_size. The bytes between unc_size and frame_size are never written. The only memset() that follows zeroes the region beyond i_valid; when the frame lies entirely within the file's valid size that memset() does not run, so the gap retains whatever was in the just-vmapped pages. All pages are then marked uptodate and returned to userspace, disclosing uninitialized (recently-freed) kernel page memory. A crafted compressed file whose stream decompresses to only a few bytes leaks the remainder of every frame on a plain read(2), which is enough to recover kernel pointers and defeat KASLR.

Zero the [unc_size, frame_size) tail immediately after a successful LZNT decompress so the remainder reads back as zero.

Fixes 4342306f0f0d ("fs/ntfs3: Add file operations and implementation")
Assisted-by Bynario AI

---
Reproduced on a non-KASAN, KASLR-enabled arm64 guest: a crafted file whose 64 KiB LZNT frames each decompress to 8 bytes leaks the [8, 65536) tail of every frame on a plain read(2) - up to ~3.3 MiB of uninitialized page memory, including kernel-text return addresses from freed VMAP_STACK pages, from which the KASLR base was recovered and verified vs /proc/kallsyms. With the patch the same read returns zeroes; a zeroed control image (valid_size truncated so the existing memset runs) leaks nothing, isolating the cause.

  fs/ntfs3/frecord.c | 9 +++++++++
  1 file changed, 9 insertions(+)
fs/ntfs3/frecord.c +9 -1
index 7b035da63c12..cd11df75bdbc 100644
@@ -2443,6 +2443,15 @@ int ni_read_frame(struct ntfs_inode *ni, u64 frame_vbo, struct page **pages,
                         err = unc_size;
                 else if (!unc_size || unc_size > frame_size)
                         err = -EINVAL;
+               else if (unc_size < frame_size) {
+                       /*
+                        * Partial decompress: zero the [unc_size, frame_size)
+                        * tail.  decompress_lznt() leaves it untouched, so
+                        * without this the freshly vmapped pages would expose
+                        * uninitialized kernel memory to userspace.
+                        */
+                       memset(frame_mem + unc_size, 0, frame_size - unc_size);
+               }
         }
         if (!err && valid_size < frame_vbo + frame_size) {
                 size_t ok = valid_size - frame_vbo;
--
Quoted 2 lines

2.54.0

Hello,

Sorry for the delay.
Your patch was applied, thank you.

Regards,
Konstantin