From: tytso on
On Thu, Jun 10, 2010 at 01:10:51PM +0200, Andi Kleen wrote:
>
> No real bugs found, just various dead code removed I believe.
> Some review would be good.
>
> Found by gcc 4.6's new warnings
>
> Cc: tytso(a)mit.edu
> Cc: linux-ext4(a)vger.kernel.org
>
> Signed-off-by: Andi Kleen <ak(a)linux.intel.com>

I've reviewed this and pulled a fixed version into the ext4 patch
queue. You deleted the initializers for size and start in mballoc.c,
which introduced a bug (ironically you commented about start not
getting inititialized in a FIXME, when the patch deleted said
initialization right above the introduced FIXME comment):

> @@ -2922,8 +2916,9 @@ ext4_mb_normalize_request(struct ext4_al
> start_off = (loff_t)ac->ac_o_ex.fe_logical << bsbits;
> size = ac->ac_o_ex.fe_len << bsbits;
> }
> - orig_size = size = size >> bsbits;
> - orig_start = start = start_off >> bsbits;
> +
> + /* FIXME: start_off is not used for anything. bug? */
> + /* FIXME: start may be uninitialized? */
>
> /* don't cover already allocated blocks in selected range */
> if (ar->pleft && start <= ar->lleft) {

Also you only fixed one of two uses of the variable 'fidx' which you
removed here:

> @@ -1144,10 +1143,10 @@ static int ext4_ext_grow_indepth(handle_
> ext4_idx_store_pblock(curp->p_idx, newblock);
>
> neh = ext_inode_hdr(inode);
> - fidx = EXT_FIRST_INDEX(neh);
> ext_debug("new root: num %d(%d), lblock %d, ptr %llu\n",
> le16_to_cpu(neh->eh_entries), le16_to_cpu(neh->eh_max),
> - le32_to_cpu(fidx->ei_block), idx_pblock(fidx));
> + le32_to_cpu(EXT_FIRST_INDEX(neh)->ei_block),
> + idx_pblock(fidx));
>
> neh->eh_depth = cpu_to_le16(path->p_depth + 1);
> err = ext4_ext_dirty(handle, inode, curp);

Other than that the patch was fine. I've fixed these up and I'll
carry the patch in the ext4 tree.

Thanks,

- Ted
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/