| Message ID | 5682584A.5030708@sandisk.com (mailing list archive) |
|---|---|
| State | Superseded |
| Headers | show |
On Tue, Dec 29, 2015 at 10:54:18AM +0100, Bart Van Assche wrote: > After having applied these changes the SRP initiator didn't receive any > RDMA completions anymore. I could remedy that by changing > "!test_and_set_bit()" into "test_and_set_bit()": Yes. I actually had this bug earlier, fixed it and managed to get it back during a rebase, d'oh. Reviewed-by: Christoph Hellwig <hch@lst.de> Can you resend it with a proper signoff? -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On 30/12/15 20:42, Christoph Hellwig wrote: > On Tue, Dec 29, 2015 at 10:54:18AM +0100, Bart Van Assche wrote: >> After having applied these changes the SRP initiator didn't receive any >> RDMA completions anymore. I could remedy that by changing >> "!test_and_set_bit()" into "test_and_set_bit()": > > Yes. I actually had this bug earlier, fixed it and managed to get > it back during a rebase, d'oh. I'm hitting an issue on a ppc64le box running linux-next, which according to git bisect is caused by this patch. It looks like I might be hitting a dodgy error path as well, as we seem to be trying to execute data. Any ideas? Andrew --- Sent SIGTERM to all processes Sent SIGKILL to all processes -> smp_release_cpus() spinning_secondaries = 47 <- smp_release_cpus() <- setup_system() sr 0:0:1:0: tag#0 Resetting device ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x6 t0 ata1.00: cmd a0/00:00:00:00:00/00:00:00:00:00/a0 tag 3 Test Unit Ready 00 00 00 00 00 00res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout) ata1.00: status: { DRDY } ata1: translated ATA stat/err 0xd1/00 to SCSI SK/ASC/ASCQ 0xb/47/00 sd 1:2:0:0: tag#0 Resetting device ata1.00: failed to set xfermode (err_mask=0x4) ipr 0003:04:00.0: Timed out waiting for aborted commands ipr 0003:04:00.0: Adapter being reset as a result of error recovery. ata1.00: failed to set xfermode (err_mask=0x4) ata1.00: failed to set xfermode (err_mask=0x4) ipr 0001:04:00.0: Adapter being reset as a result of error recovery. cpu 0x0: Vector: e40 (Emulation Assist) at [c000000000daf2e0] pc: c000000000e51ae8: dump_list_lock+0x0/0x4 lr: c0000000000f46e4: __wake_up_common+0x84/0xf0 sp: c000000000daf560 msr: 9000000102089033 current = 0xc000000000d6f500 paca = 0xc00000000fe00000 softe: 0 irq_happened: 0x01 pid = 0, comm = swapper/0 Linux version 4.4.0-next-20160118 (ajd@ka1) (gcc version 5.2.1 20150930 (GCC) ) #13 SMP Tue Jan 19 12:04:19 AEDT 2016 enter ? for help [link register ] c0000000000f46e4 __wake_up_common+0x84/0xf0 [c000000000daf560] c000000000da1100 pps_cdev_fops+0xc8/0x100 (unreliable) [c000000000daf5c0] c0000000000f5264 complete+0x54/0x90 [c000000000daf600] c00000000061f44c ata_qc_complete_internal+0x1c/0x30 [c000000000daf620] c000000000622828 __ata_qc_complete+0xb8/0x190 [c000000000daf660] c0000000005ef6e4 ipr_sata_eh_done+0x64/0x80 [c000000000daf680] c0000000005ef530 ipr_fail_all_ops+0x100/0x250 [c000000000daf740] c0000000005ffbf8 ipr_reset_restore_cfg_space+0x98/0x230 [c000000000daf7b0] c0000000005ed500 ipr_reset_ioa_job+0x80/0xf0 [c000000000daf7e0] c0000000005ebfac ipr_reset_timer_done+0xac/0xe0 [c000000000daf820] c00000000011eae4 call_timer_fn+0x54/0x180 [c000000000daf8b0] c00000000011ef2c run_timer_softirq+0x2ec/0x3a0 [c000000000daf980] c0000000000a4ee8 __do_softirq+0x188/0x3b0 [c000000000dafa70] c0000000000a5358 irq_exit+0xc8/0x100 [c000000000dafa90] c00000000001d894 timer_interrupt+0xa4/0xe0 [c000000000dafac0] c000000000002750 decrementer_common+0x150/0x180 --- Exception: 901 (Decrementer) at c000000000010364 arch_local_irq_restore+0x74/0x90 [c000000000dafdb0] c000000000dac000 init_thread_union+0x0/0x4000 (unreliable) [c000000000dafdd0] c000000000016be8 arch_cpu_idle+0x108/0x160 [c000000000dafe00] c0000000000f5594 default_idle_call+0x44/0x80 [c000000000dafe20] c0000000000f5a48 cpu_startup_entry+0x3d8/0x450 [c000000000dafee0] c00000000000bbe4 rest_init+0xa4/0xc0 [c000000000daff00] c000000000c14014 start_kernel+0x524/0x540 [c000000000daff90] c000000000008c60 start_here_common+0x20/0xa0 0:mon>
On 20/01/16 18:02, Andrew Donnellan wrote: > I'm hitting an issue on a ppc64le box running linux-next, which > according to git bisect is caused by this patch. Whoops, that should be linuxppc*-dev*@lists.ozlabs.org in the Cc. Andrew
diff --git a/lib/irq_poll.c b/lib/irq_poll.c index 43a3370..3a67019 100644 --- a/lib/irq_poll.c +++ b/lib/irq_poll.c @@ -29,7 +29,7 @@ void irq_poll_sched(struct irq_poll *iop) if (test_bit(IRQ_POLL_F_DISABLE, &iop->state)) return; - if (!test_and_set_bit(IRQ_POLL_F_SCHED, &iop->state)) + if (test_and_set_bit(IRQ_POLL_F_SCHED, &iop->state)) return; local_irq_save(flags);