* XCOFF archive sanity check
@ 2023-03-19 11:53 Alan Modra
0 siblings, 0 replies; only message in thread
From: Alan Modra @ 2023-03-19 11:53 UTC (permalink / raw)
To: binutils
XCOFF archive elements are in a linked list. Add a little more sanity
checking. This of course doesn't stop the fuzzers finding a way to
make a loop, but this check is cheap.
* coff-rs6000.c (_bfd_xcoff_openr_next_archived_file): Sanity
check that next element isn't pointing back to the header.
diff --git a/bfd/coff-rs6000.c b/bfd/coff-rs6000.c
index 4b7b5d315df..735d434951e 100644
--- a/bfd/coff-rs6000.c
+++ b/bfd/coff-rs6000.c
@@ -1714,8 +1714,11 @@ _bfd_xcoff_openr_next_archived_file (bfd *archive, bfd *last_file)
laststart -= SIZEOF_AR_HDR + arel->extra_size;
}
- /* Sanity check that we aren't pointing into the previous element. */
- if (filestart != 0 && filestart >= laststart && filestart < lastend)
+ /* Sanity check that we aren't pointing into the previous element,
+ or into the header. */
+ if (filestart != 0
+ && (filestart < SIZEOF_AR_FILE_HDR
+ || (filestart >= laststart && filestart < lastend)))
{
bfd_set_error (bfd_error_malformed_archive);
return NULL;
@@ -1747,8 +1750,11 @@ _bfd_xcoff_openr_next_archived_file (bfd *archive, bfd *last_file)
laststart -= SIZEOF_AR_HDR_BIG + arel->extra_size;
}
- /* Sanity check that we aren't pointing into the previous element. */
- if (filestart != 0 && filestart >= laststart && filestart < lastend)
+ /* Sanity check that we aren't pointing into the previous element
+ or into the header. */
+ if (filestart != 0
+ && (filestart < SIZEOF_AR_FILE_HDR_BIG
+ || (filestart >= laststart && filestart < lastend)))
{
bfd_set_error (bfd_error_malformed_archive);
return NULL;
--
Alan Modra
Australia Development Lab, IBM
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2023-03-19 11:53 UTC | newest]
Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-03-19 11:53 XCOFF archive sanity check Alan Modra
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).