also sprach martin f krafft <madduck@xxxxxxxxxx> [2007.09.30.1234 +0100]:
> Oh well, I think this is an amd64-specific problem. Daniel, are you
> around today to debug this? Or anyone else with amd64? I don't have
> an amd64 machine around to test this for another three weeks, so I'd
> really appreciate if someone else stepped in.
Okay, I did find one and I can reproduce. First thing to note:
#535 0x000000000041f07c in load_super1 (st=0x634030, fd=8, sbp=0x7fff9f4fefd0,
devname=0x0) at super1.c:1005
#536 0x000000000041f07c in load_super1 (st=0x634030, fd=8, sbp=0x7fff9f4fefd0,
devname=0x0) at super1.c:1005
load_super1 apparently recurses infinitely. Looking at the code:
static int load_super1(struct supertype *st, int fd, void **sbp, char *devname)
{
unsigned long long dsize;
unsigned long long sb_offset;
struct mdp_superblock_1 *super;
int uuid[4];
struct bitmap_super_s *bsb;
struct misc_dev_info *misc;
if (st->ss == NULL || st->minor_version == -1) {
int bestvers = -1;
struct supertype tst;
__u64 bestctime = 0;
/* guess... choose latest ctime */
tst.ss = &super1;
for (tst.minor_version = 0; tst.minor_version <= 2 ; tst.minor_version++) {
switch(load_super1(st, fd, sbp, devname)) {
I can't help but note that there is no way to break out of this
loop if (st->ss == NULL || st->minor_version == -1) is true when
it's called the first time.
So it turns out that I think Neil simply forgot to replace the first
argument by &tst in commit a40b4fe, as the forthcoming patch does.
--
.''`. martin f. krafft <madduck@xxxxxxxxxx>
: :' : proud Debian developer, author, administrator, and user
`. `'` http://people.debian.org/~madduck - http://debiansystem.info
`- Debian - when you have better things to do than fixing systems
because light travels faster than sound,
some people appear to be intelligent,
until you hear them speak.
-
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
