View Issue Details

Jump to NotesJump to History
IDProjectCategoryView StatusDate SubmittedLast Update
0001589channel: kernel/el10kernel-mlpublic2026-03-27 05:412026-04-06 14:16
Reporterfl0wfr Assigned Totoracat  
PrioritynormalSeveritymajorReproducibilityalways
Status closedResolutionno change required 
Platformx86_64OSRocky LinuxOS Version10.1
Summary0001589: mpt2sas/mpt3sas broken with kernel-ml-6.19.10-1.el10.elrepo.x86_64
DescriptionHello,

I updated from kernel-ml 6.19.9 to 6.19.10 and mpt2sas/mpt3sas stopped working at boot.
I get the following messages in dmesg and disks are not detected.

[ 1.232603] mpt3sas version 54.100.00.00 loaded
[ 1.232862] mpt3sas 0000:01:00.0: enabling device (0000 -> 0002)
[ 1.232945] mpt2sas_cm0: 64 BIT PCI BUS DMA ADDRESSING SUPPORTED, total mem (131493676 kB)
[ 1.232954] mpt2sas_cm0 fault info from func: _base_wait_for_iocstate
[ 1.233151] mpt2sas_cm0: fault_state(0x2655)!
[ 1.233156] mpt2sas_cm0: sending diag reset !!
[ 1.724192] mpt2sas_cm0: _base_diag_reset: failed going to ready state (ioc_state=0x40000000)
[ 1.724214] mpt2sas_cm0: System Register set:
[ 1.724309] mpt2sas_cm0: diag reset: FAILED
[ 1.724310] mpt2sas_cm0 fault info from func: mpt3sas_base_check_for_fault_and_issue_reset
[ 1.724312] mpt2sas_cm0: fault_state(0x1400)!
[ 1.724313] mpt2sas_cm0: sending diag reset !!
[ 1.968175] mpt2sas_cm0: _base_diag_reset: failed going to ready state (ioc_state=0x40000000)
[ 1.968179] mpt2sas_cm0: System Register set:
[ 1.968237] mpt2sas_cm0: diag reset: FAILED
[ 1.968299] mpt2sas_cm0: failure at drivers/scsi/mpt3sas/mpt3sas_scsih.c:12337/_scsih_probe()!

As a comparison, I get the following messages on kernel-ml 6.19.9:
[ 1.177419] mpt3sas version 54.100.00.00 loaded
[ 1.177552] mpt3sas 0000:01:00.0: enabling device (0000 -> 0002)
[ 1.178189] mpt2sas_cm0: 64 BIT PCI BUS DMA ADDRESSING SUPPORTED, total mem (131493976 kB)
[ 1.178200] mpt2sas_cm0 fault info from func: _base_wait_for_iocstate
[ 1.178429] mpt2sas_cm0: fault_state(0x265d)!
[ 1.178434] mpt2sas_cm0: sending diag reset !!
[ 5.081679] mpt2sas_cm0: diag reset: SUCCESS
[ 5.140680] mpt2sas_cm0: CurrentHostPageSize is 0: Setting default host page size to 4k
[ 5.140687] mpt2sas_cm0: MSI-X vectors supported: 16
[ 5.140689] mpt2sas_cm0: 0 8 8
[ 5.140884] mpt2sas_cm0: High IOPs queues : disabled
[ 5.140886] mpt2sas0-msix0: PCI-MSI-X enabled: IRQ 339
[ 5.140886] mpt2sas0-msix1: PCI-MSI-X enabled: IRQ 340
[ 5.140887] mpt2sas0-msix2: PCI-MSI-X enabled: IRQ 341
[ 5.140887] mpt2sas0-msix3: PCI-MSI-X enabled: IRQ 342
[ 5.140888] mpt2sas0-msix4: PCI-MSI-X enabled: IRQ 343
[ 5.140888] mpt2sas0-msix5: PCI-MSI-X enabled: IRQ 344
[ 5.140889] mpt2sas0-msix6: PCI-MSI-X enabled: IRQ 345
[ 5.140889] mpt2sas0-msix7: PCI-MSI-X enabled: IRQ 346
[ 5.140889] mpt2sas_cm0: iomem(0x0000000084e40000), mapped(0x000000003d5e8f2d), size(65536)
[ 5.140892] mpt2sas_cm0: ioport(0x0000000000003000), size(256)
[ 5.195177] mpt2sas_cm0: CurrentHostPageSize is 0: Setting default host page size to 4k
[ 5.222680] mpt2sas_cm0: scatter gather: sge_in_main_msg(1), sge_per_chain(9), sge_per_io(128), chains_per_io(15)
[ 5.222842] mpt2sas_cm0: request pool(0x000000008222e548) - dma(0x20b000000): depth(10368), frame_size(128), pool_size(1296 kB)
[ 5.227984] mpt2sas_cm0: sense pool(0x000000009d5d2df7) - dma(0x1caa00000): depth(10107), element_size(96), pool_size (947 kB)
[ 5.228058] mpt2sas_cm0: reply pool(0x00000000e591b5f4) - dma(0x114400000): depth(10432), frame_size(128), pool_size(1304 kB)
[ 5.228063] mpt2sas_cm0: config page(0x0000000037a87ea1) - dma(0x207343000): size(512)
[ 5.228063] mpt2sas_cm0: Allocated physical memory: size(23840 kB)
[ 5.228064] mpt2sas_cm0: Current Controller Queue Depth(10104),Max Controller Queue Depth(10240)
[ 5.228064] mpt2sas_cm0: Scatter Gather Elements per IO(128)
[ 5.271758] mpt2sas_cm0: overriding NVDATA EEDPTagMode setting from 0 to 1
[ 5.272045] mpt2sas_cm0: LSISAS2308: FWVersion(17.00.01.00), ChipRevision(0x05)
[ 5.272047] mpt2sas_cm0: Protocol=(Initiator,Target), Capabilities=(TLR,EEDP,Snapshot Buffer,Diag Trace Buffer,Task Set Full,NCQ)
[ 5.273012] mpt2sas_cm0: sending port enable !!
[ 6.798029] mpt2sas_cm0: hba_port entry: 00000000c193358c, port: 255 is added to hba_port list
[ 6.799489] mpt2sas_cm0: host_add: handle(0x0001), sas_addr(0x500605b008dc1df0), phys(8)
[ 6.815412] mpt2sas_cm0: expander_add: handle(0x0009), parent(0x0001), sas_addr(0x50050cc10efa757f), phys(37)
[ 6.822935] mpt2sas_cm0: handle(0xa) sas_address(0x50050cc10efa754f) port_type(0x1)
[ 8.558741] mpt2sas_cm0: handle(0xb) sas_address(0x500605ba0146db3d) port_type(0x1)
[ 8.567368] mpt2sas_cm0: handle(0xc) sas_address(0x50000397285b7d46) port_type(0x1)
[ 8.571776] mpt2sas_cm0: handle(0xd) sas_address(0x50050cc10efa7554) port_type(0x1)
[ 8.815413] mpt2sas_cm0: handle(0xe) sas_address(0x50050cc10efa7555) port_type(0x1)
[ 9.058935] mpt2sas_cm0: handle(0xf) sas_address(0x500003973800546a) port_type(0x1)
[ 9.064125] mpt2sas_cm0: handle(0x10) sas_address(0x50000397380038ce) port_type(0x1)
[ 9.068608] mpt2sas_cm0: handle(0x11) sas_address(0x5000039738003942) port_type(0x1)
[ 9.072982] mpt2sas_cm0: handle(0x12) sas_address(0x5000039cf83a8afe) port_type(0x1)
[ 9.079189] mpt2sas_cm0: handle(0x13) sas_address(0x50050cc10efa755c) port_type(0x1)
[ 9.306790] mpt2sas_cm0: handle(0x14) sas_address(0x50050cc10efa755d) port_type(0x1)
[ 9.562041] mpt2sas_cm0: handle(0x15) sas_address(0x50000397285adf02) port_type(0x1)
[ 9.566505] mpt2sas_cm0: handle(0x16) sas_address(0x5000039cf83b1e9a) port_type(0x1)
[ 9.572710] mpt2sas_cm0: handle(0x17) sas_address(0x50050cc10efa7560) port_type(0x1)
[ 9.806129] mpt2sas_cm0: handle(0x18) sas_address(0x50050cc10efa7561) port_type(0x1)
[ 10.068218] mpt2sas_cm0: handle(0x19) sas_address(0x50050cc10efa757e) port_type(0x1)
[ 13.056522] mpt2sas_cm0: port enable: SUCCESS
[ 22.610251] intel_cstate(E) snd(E) platform_profile(E) mei_me(E) i2c_designware_core(E) usb_storage(E) rfkill(E) intel_uncore(E) pmt_crashlog(E) rc_core(E) ch341(E) soundcore(E) wmi_bmof(E) ccp(E) i2c_i801(E) mei(E) pcspkr(E) ses(E) i2c_smbus(E) idma64(E) enclosure(E) ie31200_edac(E) intel_pmc_core(E) joydev(E) pmt_telemetry(E) pmt_discovery(E) pmt_class(E) intel_pmc_ssram_telemetry(E) acpi_pad(E) acpi_tad(E) sg(E) loop(E) xfs(E) sd_mod(E) cxd2878(OE) xe(E) raid1(E) drm_ttm_helper(E) ttm(E) drm_client_lib(E) i2c_algo_bit(E) drm_suballoc_helper(E) drm_buddy(E) gpu_sched(E) drm_gpuvm(E) drm_exec(E) drm_gpusvm_helper(E) drm_display_helper(E) drm_kms_helper(E) tbsecp3(OE) tas2101(OE) drm(E) gx1133(OE) i2c_mux(E) nvme(E) mpt3sas(E) apex(OE) dvb_core(OE) igc(E) nvme_core(E) gasket(OE) ghash_clmulni_intel(E) mc(E) cec(E) raid_class(E) scsi_transport_sas(E) intel_vsec(E) video(E) wmi(E) pinctrl_alderlake(E) sunrpc(E) dm_mirror(E) dm_region_hash(E) dm_log(E) iscsi_tcp(E) libiscsi_tcp(E) libiscsi(E) scsi_transport_iscsi(E)
[ 22.674305] intel_cstate(E) snd(E) platform_profile(E) mei_me(E) i2c_designware_core(E) usb_storage(E) rfkill(E) intel_uncore(E) pmt_crashlog(E) rc_core(E) ch341(E) soundcore(E) wmi_bmof(E) ccp(E) i2c_i801(E) mei(E) pcspkr(E) ses(E) i2c_smbus(E) idma64(E) enclosure(E) ie31200_edac(E) intel_pmc_core(E) joydev(E) pmt_telemetry(E) pmt_discovery(E) pmt_class(E) intel_pmc_ssram_telemetry(E) acpi_pad(E) acpi_tad(E) sg(E) loop(E) xfs(E) sd_mod(E) cxd2878(OE) xe(E) raid1(E) drm_ttm_helper(E) ttm(E) drm_client_lib(E) i2c_algo_bit(E) drm_suballoc_helper(E) drm_buddy(E) gpu_sched(E) drm_gpuvm(E) drm_exec(E) drm_gpusvm_helper(E) drm_display_helper(E) drm_kms_helper(E) tbsecp3(OE) tas2101(OE) drm(E) gx1133(OE) i2c_mux(E) nvme(E) mpt3sas(E) apex(OE) dvb_core(OE) igc(E) nvme_core(E) gasket(OE) ghash_clmulni_intel(E) mc(E) cec(E) raid_class(E) scsi_transport_sas(E) intel_vsec(E) video(E) wmi(E) pinctrl_alderlake(E) sunrpc(E) dm_mirror(E) dm_region_hash(E) dm_log(E) iscsi_tcp(E) libiscsi_tcp(E) libiscsi(E) scsi_transport_iscsi(E)
[ 22.807775] mpt3sas 0000:01:00.0: invalid VPD tag 0x00 (size 0) at offset 0; assume missing optional EEPROM
Steps To ReproduceJust upgrade from kernel-ml 6.19.9 to 6.19.10 and reboot.
Additional InformationHBA model is a Broadcom / LSI SAS2308 PCI-Express Fusion-MPT SAS-2 (rev 05)
TagsNo tags attached.

Activities

toracat

2026-03-27 20:33

administrator   ~0010836

Acknowledged.

toracat

2026-03-29 17:02

administrator   ~0010839

I compared the kernel source code between v6.19.9 and v6.19.10. Looking at the /drivers/scsi/mpt3sas/ directory, I see no change. Even in the whole /drivers/scsi/ directory, the code remained the same. Also, our kernel config files were unchanged.

It looks like this is not something we (elrepo) can help with. One thing you may want to try will be to ask upstream at kernel.org. However given the fact the hardware LSI SAS2308 is no longer supported, you probably cannot expect any help from them. Another option will be to contact Broadcom. In fact a Broadcom person made such a suggestion ( https://bugzilla.kernel.org/show_bug.cgi?id=209177#c11 ) regarding an unsupported card.

fl0wfr

2026-03-30 03:06

reporter   ~0010840

Thanks @toracat for your feedback. Could it be related to changes in kernel regarding IOMMU? This is the only custom option in GRUB, so I'm a bit clueless regarding this issue.

toracat

2026-03-30 13:49

administrator   ~0010841

@fl0wfr

Can you share more details about "the only custom option in GRUB"?

fl0wfr

2026-03-30 13:54

reporter   ~0010842

Sure:
GRUB_CMDLINE_LINUX="rd.md.uuid=<redacted> rd.lvm.lv=root/root rd.md.uuid=<redacted> rhgb quiet panic=10 crashkernel=no selinux=0 intel_iommu=on iommu=pt pcie_acs_override=downstream,multifunction xe.max_vfs=2 xe.force_probe=e212

toracat

2026-03-30 19:59

administrator   ~0010843

Do you need those iommu-related flags to boot your system? If not, can you test-boot v6.19.9 and v6.19.10 without them?

fl0wfr

2026-03-31 04:06

reporter   ~0010844

After doing more tests, I realized that the host needs to be properly shut down instead of being rebooted. It's like some states are still stored in the HBA memory which are incompatible with the new kernel.
After performing a shutdown, I can boot to kernel 6.19.10 and perform several reboots without any issue.
Sorry for the wrongly created issue and thank you for your help.

toracat

2026-03-31 13:13

administrator   ~0010846

Glad to hear you have resolved the issue.

I've seen a situation where a "real cold" reboot was required, but never imagined what seemed to be a minor update to the kernel caused it.

Issue History

Date Modified Username Field Change
2026-03-27 05:41 fl0wfr New Issue
2026-03-27 05:41 fl0wfr Status new => assigned
2026-03-27 05:41 fl0wfr Assigned To => toracat
2026-03-27 20:33 toracat Status assigned => acknowledged
2026-03-27 20:33 toracat Note Added: 0010836
2026-03-29 17:02 toracat Note Added: 0010839
2026-03-30 03:06 fl0wfr Note Added: 0010840
2026-03-30 13:49 toracat Note Added: 0010841
2026-03-30 13:54 fl0wfr Note Added: 0010842
2026-03-30 19:59 toracat Note Added: 0010843
2026-03-31 04:06 fl0wfr Note Added: 0010844
2026-03-31 13:13 toracat Note Added: 0010846
2026-03-31 13:13 toracat Status acknowledged => resolved
2026-03-31 13:13 toracat Resolution open => no change required
2026-04-06 14:16 toracat Status resolved => closed