Skip to contentSkip to conversation

One More NTFS Fuzzing Fix

PATCH In review Andrew Hamilton 2 participants 4 messages Jun 1, 2025 – Oct 24, 2025

lore
Share & tools
lore
#0 lore raw

cover One More NTFS Fuzzing Fix

I took one last pass at my attempts at ad-hoc fuzzing of NTFS with the goal of improving coverage and letting the fuzzer run for a while. After rebuilding afl++ to allow larger file inputs that are more representative of real NTFS file systems, it was uncovered that my last fix to address NTFS test regressions left a possible access violation in find_attr.

This fixes the last remaining fuzzing issue uncovered.

Confirmed that NTFS test cases still pass.

Andrew Hamilton (1):
fs/ntfs.c: Correct possible access violation on next_attribute

 grub-core/fs/ntfs.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)
Signature · 7 lines
2.39.5


_______________________________________________
Grub-devel mailing list
[email protected]
https://lists.gnu.org/mailman/listinfo/grub-devel
#1 lore raw

PATCH 1/1 fs/ntfs.c: Correct next_attribute validation

Improved ad-hoc fuzzing coverage releaved a possible access violation around line 342 of ntfs.c when accessing the attr_cur pointer due to possiblity of moving pointer 'next' beyond of the end of the valid buffer inside next_attribute. Prevent this for cases where full attribute validation is not performed (such as on attribute lists) by performing a sanity check on the newly calculated next pointer.

Fixes 06914b614 (fs/ntfs: Correct attribute vs attribute list validation)
---
 grub-core/fs/ntfs.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)
grub-core/fs/ntfs.c +6 -1
index 5b0a18f3d..9aff239c4 100644
@@ -233,7 +233,12 @@ next_attribute (grub_uint8_t *curr_attribute, void *end, bool validate)
     return NULL;
 
   next += u16at (curr_attribute, 4);
-  if (validate && validate_attribute (next, end) == false)
+  if (validate)
+  {
+    if (validate_attribute (next, end) == false)
+      return NULL;
+  }
+  else if (next >= (grub_uint8_t *)end)
     return NULL;
 
   return next;
Signature · 7 lines
2.39.5


_______________________________________________
Grub-devel mailing list
[email protected]
https://lists.gnu.org/mailman/listinfo/grub-devel
Reviewed-by on v1 1/1
#2 lore raw

Re: [PATCH 1/1] fs/ntfs.c: Correct next_attribute validation

Quoted 10 lines · Andrew Hamilton

Improved ad-hoc fuzzing coverage releaved a possible access violation around line 342 of ntfs.c when accessing the attr_cur pointer due to possiblity of moving pointer 'next' beyond of the end of the valid buffer inside next_attribute. Prevent this for cases where full attribute validation is not performed (such as on attribute lists) by performing a sanity check on the newly calculated next pointer.

Fixes 06914b614 (fs/ntfs: Correct attribute vs attribute list validation)

Daniel

_______________________________________________
Grub-devel mailing list
[email protected]
https://lists.gnu.org/mailman/listinfo/grub-devel

#3 lore raw
In reply to Andrew Hamilton

Hello,

Just re-raising this for consideration.

Thank you,
Andrew

Quoted 11 lines · Andrew Hamilton

I took one last pass at my attempts at ad-hoc fuzzing of NTFS with the goal of improving coverage and letting the fuzzer run for a while. After rebuilding afl++ to allow larger file inputs that are more representative of real NTFS file systems, it was uncovered that my last fix to address NTFS test regressions left a possible access violation in find_attr.

This fixes the last remaining fuzzing issue uncovered.

Confirmed that NTFS test cases still pass.

Quoted 10 lines

Andrew Hamilton (1):
fs/ntfs.c: Correct possible access violation on next_attribute

 grub-core/fs/ntfs.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

--
2.39.5

_______________________________________________
Grub-devel mailing list
[email protected]
https://lists.gnu.org/mailman/listinfo/grub-devel