On Fri, May 20, 2011 at 2:45 PM, Mike Belopuhov <[email protected]> wrote: > On Fri, May 20, 2011 at 06:24 -0600, David Coppa wrote: >> Hi all, >> >> OpenBSD-current snapshot dated 16-May-2011: >> I get an always-reproducible panic with softraid crypto and >> kern.bufcachepercent >= 75, when untarring a tarball of the >> complete source tree. >> >> The disk layout is the following, with softraid crypto for >> all but /: >> >> /dev/sd0a on / type ffs (local) >> /dev/sd2f on /home type ffs (local, nodev, nosuid) >> /dev/sd2e on /usr type ffs (local, nodev) >> /dev/sd2d on /var type ffs (local, nodev, nosuid) >> /dev/sd3i on /mnt type msdos (local) >> >> Here's the trace: >> >> >> panic: sr_crypto_rw: no crypto op > > hi, > > although i'm not sure that this is a best solution, i can't > see why we should panic here. sr_scsi_cmd seems to cope with > stuffups just fine. > > Index: dev/softraid_crypto.c > =================================================================== > RCS file: /home/cvs/src/sys/dev/softraid_crypto.c,v > retrieving revision 1.65 > diff -u -p -r1.65 softraid_crypto.c > --- dev/softraid_crypto.c 6 Apr 2011 03:14:51 -0000 1.65 > +++ dev/softraid_crypto.c 20 May 2011 12:42:12 -0000 > @@ -1115,7 +1115,7 @@ sr_crypto_rw(struct sr_workunit *wu) > if (wu->swu_xs->flags & SCSI_DATA_OUT) { > crp = sr_crypto_getcryptop(wu, 1); > if (crp == NULL) > - panic("sr_crypto_rw: no crypto op"); > + return (1); > crp->crp_callback = sr_crypto_write; > crp->crp_opaque = wu; > s = splvm(); >
It now survives to several untarrings of src.tar with kern.bufcachepercent=90. cheers, David
