> Thanks for spotting, but I have to disagree with analysis.
> a) it's a procfs bug. We leave ->i_blocks uninitialized and inode
> constructor leaves it in undefined state.
> b) I'd rather fix it once in fs/inode.c::clean_inode() instead of
> hunting similar bugs down again and again.
>
> See if the patch below works for you. It makes sure that inodes passed to
> ->read_inode() or returned by new_inode() have zero in i_blocks and removes
> the redundant assignments in filesystems. Warning: it's completely untested.
Nikita sent you an email saying that he though you should fix it in the manner
you fixed it above, and then posted a patch which was localized in its impact to
just ReiserFS because he was sure it was correct.
Nikita should be more aggressive in patching VFS rather than asking you to do
it, but I don't think you can claim to be differing with his analysis.
Hans
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/