| Message ID | c40509001d8fedcd1a1ad7ad694766cd8ac92111.1589662192.git.mschiffer@universe-factory.net |
|---|---|
| State | Superseded |
| Delegated to: | Matthias Schiffer |
| Headers | show |
| Series | ucert fixes and cleanup | expand |
Matthias Schiffer <mschiffer@universe-factory.net> [2020-05-16 23:13:50]: Hi, > Some tools like ucert use concatenations of multiple blobs. Account for > this case by allowing the underlying buffer length to be greater than > the blog length. in order to prevent future regression, could you please add a test case for this as well? Probably into tests/cram/test_blob_parse.t, thanks! -- ynezz
diff --git a/blob.c b/blob.c index 528e717615d6..433becb904f5 100644 --- a/blob.c +++ b/blob.c @@ -277,7 +277,7 @@ blob_parse_untrusted(struct blob_attr *attr, size_t attr_len, struct blob_attr * return 0; len = blob_raw_len(attr); - if (len != attr_len) + if (attr_len < len) return 0; memset(data, 0, sizeof(struct blob_attr *) * max);
Some tools like ucert use concatenations of multiple blobs. Account for this case by allowing the underlying buffer length to be greater than the blog length. Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net> --- blob.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)