On Sun, 15 Apr 2007 13:46:16 -0700 bugme-daemon@xxxxxxxxxxxxxxxxxxx wrote: > http://bugzilla.kernel.org/show_bug.cgi?id=8329 > > Summary: mtplan.c: kmalloc while spinlock is held > Kernel Version: 2.6.20 > Status: NEW > Severity: normal > Owner: jgarzik@xxxxxxxxx > Submitter: matthias.kaehlcke@xxxxxxxxx > > > Problem Description: > > under some conditions mpt_lan_type_trans() can do a kmalloc(GFP_KERNEL) while > the spinlock bad_naa_lock is held > > drivers/message/fusion/mptlan.c: > > ... > write_lock_irq(&bad_naa_lock); > for (prevnh = nh = mpt_bad_naa; nh != NULL; > prevnh=nh, nh=nh->next) { > if ((nh->ieee[0] == fch->saddr[0]) && > (nh->ieee[1] == fch->saddr[1]) && > (nh->ieee[2] == fch->saddr[2]) && > (nh->ieee[3] == fch->saddr[3]) && > (nh->ieee[4] == fch->saddr[4]) && > (nh->ieee[5] == fch->saddr[5])) { > found = 1; > dlprintk ((KERN_INFO "mptlan/type_trans: ARP Re" > "q/Rep w/ bad NAA from system already" > " in DB.\n")); > break; > } > } > > if ((!found) && (nh == NULL)) { > > nh = kmalloc(sizeof(struct NAA_Hosed), GFP_KERNEL); > ... > > AFAIK this should not be done as kmalloc could sleep while holding the spinlock > yup. And the fact that nobody has ever reported a might_sleep() warning coming out of there makes me think that it is dead code. - To unsubscribe from this list: send the line "unsubscribe linux-scsi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
