Hi Liam, On Mon, 2017-01-23 at 10:38 -0500, Liam R. Howlett wrote: > < removed most of dmesg > > > > [ 64.557099] ehci_hcd: USB 2.0 'Enhanced' Host Controller (EHCI) > > Driver > > [ 64.633786] qla2xxx [0000:00:00.0]-0005: : QLogic Fibre Channel > > HBA Driver: 8.07.00.38-k. > > [ 64.633966] PCI: Enabling device: (0001:00:04.0), cmd 3 > > [ 64.634261] qla2xxx [0001:00:04.0]-001d: : Found an ISP2200 irq > > 20 iobase 0x000007fd00100000. > > [ 64.647517] sym0: No NVRAM, ID 7, Fast-20, SE, parity checking > > [ 64.652483] ohci_hcd: USB 1.1 'Open' Host Controller (OHCI) > > Driver > > [ 64.655670] qla2xxx [0001:00:04.0]-0050:1: No matching ROM > > signature. > > Is this normal? Comparing with a old kernel that boots well. 3.16.0-0.bpo.4-sparc64-smp #1 SMP Debian 3.16.7-ckt25-2~bpo70+1 (2016-04-12). I am getting ... so that looks the same. [ 58.792508] qla2xxx [0001:00:04.0]-0050:0: No matching ROM signature. > > [ 64.656401] ehci-pci: EHCI PCI platform driver > > [ 64.657269] ohci-pci: OHCI PCI platform driver > > [ 64.664424] sym0: SCSI BUS has been reset. > > [ 64.667307] scsi host0: sym-2.2.3 > > [ 64.679180] PCI: Enabling device: (0000:00:06.1), cmd 147 > > [ 64.680362] sym1: <875> rev 0x37 at pci 0000:00:06.1 irq 17 > > [ 64.713347] gem 0000:00:05.1 enp0s5f1: renamed from eth0 > > [ 64.758542] qla2xxx [0001:00:04.0]-0064:1: Inconsistent NVRAM > > detected: checksum=0x0 id= > > [ 64.764091] qla2xxx [0001:00:04.0]-0069:1: NVRAM configuration > > failed. > > Does this happen in the success case? Yes a success case, booted 3.16.0 does. [ 58.895901] qla2xxx [0001:00:04.0]-0069:0: NVRAM configuration failed. [ 64.786902] qla2xxx 0001:00:04.0: firmware: direct-loading firmware ql2200_fw.bin > > [ 64.833101] sym1: No NVRAM, ID 7, Fast-20, SE, parity checking > > [ 64.843136] sym1: SCSI BUS has been reset. > > [ 64.845936] scsi host2: sym-2.2.3 Witch does, nicely ... [ 58.886906] qla2xxx [0001:00:04.0]-0064:0: Inconsistent NVRAM detected: checksum=0x0 id= [ 58.889233] PCI: Enabling device: (0000:00:06.0), cmd 147 [ 58.889959] qla2xxx [0001:00:04.0]-0065:0: Falling back to functioning (yet invalid -- WWPN) defaults. [ 58.890409] sym0: <875> rev 0x37 at pci 0000:00:06.0 irq 16 [ 58.895901] qla2xxx [0001:00:04.0]-0069:0: NVRAM configuration failed. [ 58.911709] qla2xxx 0001:00:04.0: firmware: direct-loading firmware ql2200_fw.bin [ 58.985621] sym0: No NVRAM, ID 7, Fast-20, SE, parity checking [ 58.995808] sym0: SCSI BUS has been reset. <snap> and some later on ... [ 69.700087] qla2xxx [0001:00:04.0]-00fb:0: QLogic QLA22xx - . [ 69.703176] qla2xxx [0001:00:04.0]-00fc:0: ISP2200: PCI (66 MHz) @ 0001:00:04.0 hdma- host#=0 fw=2.02.08 TP. [ 70.244468] scsi 0:0:0:0: Direct- Access SEAGATE ST373307FSUN72G 0207 PQ: 0 ANSI: 3 [ 70.252898] scsi 0:0:1:0: Direct-Access FUJITSU MAP3735F SUN72G 1201 PQ: 0 ANSI: 4 [ 74.726434] sd 0:0:0:0: [sda] 143374738 512-byte logical blocks: (73.4 GB/68.3 GiB) [ 74.729661] sd 0:0:1:0: [sdb] 143374738 512-byte logical blocks: (73.4 GB/68.3 GiB) > > [ 78.162677] ERROR(0): Cheetah error trap taken > > afsr[0000080000000000] afar[000007fd00100040] TL1(0) > > [ 78.165632] ERROR(0): TPC[101ade8c] TNPC[101ade90] O7[101ade80] > > TSTATE[9911001603] > > [ 78.168591] ERROR(0): > > [ 78.168988] TPC<qla2x00_mailbox_command+0x8ac/0xec0 [qla2xxx]> > > [ 78.171864] ERROR(0): M_SYND(0), E_SYND(0) > > [ 78.174808] ERROR(0): Highest priority error (0000080000000000) > > "Bus error response from system bus" > > [ 78.177788] ERROR(0): D-cache idx[0] tag[0000000000000000] > > utag[0000000000000000] stag[0000000000000000] > > [ 78.180771] ERROR(0): D-cache data0[0000000000000000] > > data1[0000000000000000] data2[0000000000000000] > > data3[0000000000000000] > > [ 78.183808] ERROR(0): I-cache idx[0] tag[0000000000000000] > > utag[0000000000000000] stag[0000000000000000] u[0000000000000000] > > l[0000000000000000] > > [ 78.186839] ERROR(0): I-cache INSN0[0000000000000000] > > INSN1[0000000000000000] INSN2[0000000000000000] > > INSN3[0000000000000000] > > [ 78.189899] ERROR(0): I-cache INSN4[0000000000000000] > > INSN5[0000000000000000] INSN6[0000000000000000] > > INSN7[0000000000000000] > > [ 78.192971] ERROR(0): E-cache idx[100040] tag[00000000e48dc920] > > [ 78.196010] ERROR(0): E-cache data0[0000000000000000] > > data1[0000000000000000] data2[0000000000000000] > > data3[0000000000000000] > > [ 78.199157] Kernel panic - not syncing: Irrecoverable deferred > > error trap. > > [ 78.199157] > > [ 78.205490] CPU: 0 PID: 80 Comm: systemd-udevd Not tainted > > 4.9.0-1-sparc64-smp #1 Debian 4.9.2-2 > > [ 78.208768] Call Trace: > > [ 78.212074] [000000000056b7a8] panic+0xe8/0x298 > > [ 78.215400] [0000000000429b8c] > > cheetah_deferred_handler+0x1ec/0x460 > > [ 78.218727] [0000000000405e44] c_deferred+0x18/0x24 > > [ 78.222092] [00000000101ade8c] > > qla2x00_mailbox_command+0x8ac/0xec0 [qla2xxx] > > [ 78.225391] [00000000101b04e8] qla2x00_init_firmware+0xe8/0x1e0 > > [qla2xxx] > > [ 78.228692] [00000000101a53ec] qla2x00_init_rings+0x3ac/0x400 > > [qla2xxx] > > [ 78.231985] [00000000101ac410] > > qla2x00_initialize_adapter+0x470/0x6e0 [qla2xxx] > > [ 78.235306] [000000001019e870] qla2x00_probe_one+0xff0/0x29a0 > > [qla2xxx] > > [ 78.238540] [0000000000766d60] pci_device_probe+0x80/0x100 > > [ 78.241858] [00000000007e6480] driver_probe_device+0x180/0x420 > > [ 78.245132] [00000000007e6820] __driver_attach+0x100/0x120 > > [ 78.248395] [00000000007e3e9c] bus_for_each_dev+0x5c/0xa0 > > [ 78.251625] [00000000007e5b7c] driver_attach+0x1c/0x40 > > [ 78.254818] [00000000007e5564] bus_add_driver+0x164/0x2a0 > > [ 78.258016] [00000000007e7314] driver_register+0x74/0x120 > > [ 78.261209] [0000000000765638] __pci_register_driver+0x38/0x60 > > [ 78.264419] Press Stop-A (L1-A) to return to the boot prom > > [ 78.267612] ---[ end Kernel panic - not syncing: Irrecoverable > > deferred error trap. > > [ 78.267612] > > [ 291.373806] random: crng init done > > I am not familiar with cheetah or the qla2xxx card, but it looks like > qla2x00_mailbox_command is accessing the PCI bus which is not mapped. > Have a look at trap_64 in cheetah_deferred_handler. There is a > pci_poke_faulted variable that is used to flag these errors and to > skip the instruction. From a quick look at the driver, this shouldn't > be happening. The PCI space should be configured first. I would > enable ql_dbg output to see more of what is going on. Perhaps one of > the messages above indicate an issue and the return value isn't being > validated correctly? Or perhaps the error path assumes it is safe to > access the PCI bus when it's not safe? If someone could dig into trap_64 in cheetah_deferred_handler? I am able to install the dbgsym of the linux-image package and set debug on. That will be in the next e-mail if it does what we are looking for. -- To unsubscribe from this list: send the line "unsubscribe sparclinux" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
