ext4: Annotate struct fname with __counted_by()
PATCH v2 1/1 ext4: Annotate struct fname with __counted_by()
Add the __counted_by compiler attribute to the flexible array member name to improve access bounds-checking via CONFIG_UBSAN_BOUNDS and CONFIG_FORTIFY_SOURCE.
---
Changes in v2:
- Split the patch into two separate patches as suggested by Greg KH
- Link to v1: https://lore.kernel.org/r/[email protected]/
--- fs/ext4/dir.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
fs/ext4/dir.c +1 -1
@@ -418,7 +418,7 @@ struct fname {__u32 inode;__u8 name_len;__u8 file_type; -char name[]; + char name[] __counted_by(name_len); };/*
Signature · 1 line
2.47.0
Quoted 5 lines · Thorsten Blum
Add the
__counted_bycompiler attribute to the flexible array member name to improve access bounds-checking viaCONFIG_UBSAN_BOUNDSandCONFIG_FORTIFY_SOURCE.
Applied, thanks!
[1/1] ext4: Annotate struct fname with __counted_by()
commit: de183b2baf90f0acc1854a3998c14b8b228f9643
By the way, in general, you don't need to resend patches unless you need to rebase them to fix patch conflicts; I track requested patches using patchwork, and I tend to process and review patches in batches.
If you are concerned that I might have missed a patch, feel free to reply to the original patch with a ping. I'll see it in my inbox, and even if it had gotten lost in my inbox, I can find the original patch using lore.kernel.org or patchwork.
Thanks!
Best regards,
Signature · 1 line
Theodore Ts'o <[email protected]>
Quoted 8 lines · Theodore Ts'o
By the way, in general, you don't need to resend patches unless you need to rebase them to fix patch conflicts; I track requested patches using patchwork, and I tend to process and review patches in batches.
If you are concerned that I might have missed a patch, feel free to reply to the original patch with a ping. I'll see it in my inbox, and even if it had gotten lost in my inbox, I can find the original patch using lore.kernel.org or patchwork.
Thanks for letting me know.
Best regards,
Thorsten