https://bugzilla.kernel.org/show_bug.cgi?id=187231 Bug ID: 187231 Summary: kernel panic during hpsa MSI plus tg3 MSI Product: IO/Storage Version: 2.5 Kernel Version: 4.8.6 Hardware: All OS: Linux Tree: Mainline Status: NEW Severity: normal Priority: P1 Component: SCSI Assignee: linux-scsi@xxxxxxxxxxxxxxx Reporter: kernelorg@xxxxxx Regression: No Created attachment 243801 --> https://bugzilla.kernel.org/attachment.cgi?id=243801&action=edit kernel 4.8.6 .config I'm not sure whether this is a SCSI / HPSA bug or a networking / tg3 driver bug. Both are seen in the stack dump. As the trigger seems to be HPSA I'm reporting as a SCSI issue here... I've been recently attempting to run mainline 4.8.x kernels, most recently 4.8.6, on our production HP DL 380 Intel servers. On several of them there is some related issue reported in https://bugzilla.kernel.org/show_bug.cgi?id=187221 where the HPSA driver on some of the hosts sometimes resets the logical device. I had seen that already with 4.4.x kernels, and again with 4.8.6. Now, specifically with 4.8.6, on the box which has the worst of these symptoms, I _additionally_ experienced multiple full kernel panics. The same box (with the same hpsa reset symtoms) had been running 4.4.x kernels before without such kernel panics. The panics then happened multiple times with about a day in between. On the last round I had the ILO SSH console running under screen with logging enabled, and was able to retrieve the following panic backtrace: [187283.903173] hpsa 0000:03:00.0: scsi 0:1:0:0: resetting logical Direct-Access HP LOGICAL VOLUME RAID-5 SSDSmartPathCap- En- Exp=1 [187314.331375] sd 0:1:0:0: rejecting I/O to offline device [187314.413441] sd 0:1:0:0: rejecting I/O to offline device [187314.854183] sd 0:1:0:0: rejecting I/O to offline device ... lots of these ... [187328.991285] sd 0:1:0:0: rejecting I/O to offline device [187328.991389] sd 0:1:0:0: rejecting I/O to offline device [187329.190166] sd 0:1:0:0: rejecting I/O to offline device [187329.271304] ffff88bd1a7e8000 ffff88bd1a7be500 ffff88bd7f483eb8 ffffffff8143 493f [187329.271304] Call Trace: [187329.271310] <IRQ> [187329.271310] [<ffffffffa002e332>] ? tg3_poll_msix+0xc2/0x160 [tg3] [187329.271311] [<ffffffff8143493f>] do_hpsa_intr_msi+0x8f/0x1c0 [187329.271314] [<ffffffff81148c46>] __handle_irq_event_percpu+0x66/0xe0 [187329.271315] [<ffffffff81148cde>] handle_irq_event_percpu+0x1e/0x50 [187329.271316] [<ffffffff81148d37>] handle_irq_event+0x27/0x50 [187329.271318] [<ffffffff8114bda5>] handle_edge_irq+0x65/0x140 [187329.271320] [<ffffffff81057255>] handle_irq+0x15/0x20 [187329.271321] [<ffffffff81057086>] do_IRQ+0x46/0xd0 [187329.271324] [<ffffffff816dc4fc>] common_interrupt+0x7c/0x7c [187329.271325] <EOI> [187329.271338] Code: 53 48 89 fb 48 83 ec 28 4c 8b a7 5c 02 00 00 4c 8b bf 40 0 2 00 00 4c 8b b7 38 02 00 00 4c 8b af 4c 02 00 00 49 8b 04 24 4c 89 e7 <48> 8b 8 0 98 00 00 00 48 89 45 c0 49 8b 87 d0 01 00 00 48 89 45 [187329.271339] RIP [<ffffffff81431417>] complete_scsi_command+0x37/0x8c0 [187329.271339] RSP <ffff88bd7f483e38> [187329.271339] CR2: 0000000000000098 [187329.271341] ---[ end trace 52898916f0da5c53 ]--- [187329.273413] Kernel panic - not syncing: Fatal exception in interrupt [187330.308465] Shutting down cpus with NMI [187330.308471] Kernel Offset: disabled [187330.919173] Rebooting in 300 seconds.. I'll attach my kernel .config. As this is a production system and so far the panics only hit with our usual (webserver and DB kvm machine) production load active, there's not much testing or bisecting I can do, but I didn't want to drop the issue unreported, either. Hope this helps somebody. If there is any more info I can provide, just ask what would be useful. (I'm back to running 4.4.x) -- You are receiving this mail because: You are the assignee for the bug. -- 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
