Messages in this thread | | | From | Roderich Schupp <> | Date | Wed, 12 Aug 1998 21:42:39 +0200 | Subject | Re: ENOSYS, CD-ROM, and SMP oopses |
| |
Meelis Roos <mroos@tartu.cyber.ee> said: > G> mount("/dev/hdc", "/mnt/cdrom", "iso9660", 0xc0ed0000, 0x80541f0) = -1 > G> ENOSYS (Function not implemented) > G> mount("/dev/hdc", "/mnt/cdrom", "iso9660", 0xc0ed0000, 0x80541f0) = -1 > G> ENOSYS (Function not implemented) > G> mount("/dev/hdc", "/mnt/cdrom", "iso9660", 0xc0ed0000, 0x80541f0) = -1 > G> ENOTBLK (Block device required)
> This looks _very similar_ to the amd mounting problems that happen > since 114. > I got ENOSYS errors on mount syscall too. UP kernel.
Folks, don't get distracted by the ENOSYS. I've been seeing this ENOSYS for quite a while now. I first discovered it when chasing some devfs bugs (since fixed), but I think it has nothing to do with devfs at all. In my case the syscall was a mknod and the strace looked something like this
mknod(...) = ENOSYS mknod(...) = ENOSYS mknod(...) = 0 (or the "correct" error)
I instrumented the sys_mknod system call and this showed that it was only entered _once_ (and from the return value this should have been the _last_ line above). This is on an UP kernel. I therefore think the first two lines reported by strace are bogus. However, it seems to have something to do with kmod, because I can trigger this with almost any syscall on a device that leads to a call to kmod. Maybe strace gets disturbed by the kmod process - it's forked inside the kernel so strace doesn't know about it, but is technically a child of the process causing the kmod.
Anyway, the ENOSYS has probably nothing to do with your real problem.
Cheers, Roderich
-- Vinny: Is it possible, the two utes... Judge: Eh, the two what? Uh, uh, what was that word? Vinny: Uh, what word? Judge: Two what? Vinny: What? Judge: Uh, did you say 'Utes'? Vinny: Yeah, two utes. Judge: What is a ute? Vinny: Oh, excuse me, your honor. Two YOUTHS.
Roderich Schupp mailto:rsch@ExperTeam.de ExperTeam GmbH http://www.experteam.de/ Munich, Germany linux:2.1.116
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html
| |