Hi Henrik,
On Sat, May 18, 2013 at 7:24 AM, Henrik Nordström henrik@henriknordstrom.net wrote:
tor 2013-05-16 klockan 21:53 -0700 skrev Simon Glass:
Sorry I meant:
=>ext4ls host 0:3 Segmentation fault (core dumped)
It may not be your code, but I think the segfault is there.
It's crashing in ext4 for me.
#0 ext4fs_set_blk_dev (rbdd=0x6434a0 <host_devices>, info=info@entry=0x643980 <fs_partition>) at dev.c:56
#0 ext4fs_set_blk_dev (rbdd=0x6434a0 <host_devices>, info=info@entry=0x643980 <fs_partition>) at dev.c:56 56 get_fs()->total_sect = (info->size * info->blksz) >> 57 get_fs()->dev_desc->log2blksz; 58 get_fs()->dev_desc = rbdd;
p(gdb) p get_fs() $2 = (struct ext_filesystem *) 0x644410 <ext_fs> (gdb) p $2->dev_desc $5 = (block_dev_desc_t *) 0x0
Looks like a generic ext4 bug in the new block size handling code.
I think it just needs to be shuffled around a bit so the dev_desc is assigned before, not after.
Well that's a great bug to find.
It isn't related to your patch though - if you have time it would be nice to get a separate patch to fix that bug.
Regards Henrik
Regards, Simon