On Tue, Oct 25, 2022 at 03:57:06PM -0700, James Smart wrote:
An issue was identified with lpfc in the LTS 5.15 kernel. There is an FLOGI failure which prevents FC link bringup.
In the past several kernel releases, we have been reworking areas of the driver to fix issues in the broader design rather than continuing to create a patchwork on an issue-by-issue basis. This means there are a lot of inter-related patches.
In this case, it appears that a portion of the "path split" rework was pulled into 5.15, and the portion that wasn't picked up introduced the error.
This patch set reverts the patches for the partial pull in, then adds the full rework set and all known fixes. lpfc ends up in a state somewhat close to 5.18.y
-- james
This patch set was created via the following:
# Revert prior partial "path split" patches git revert 17bf429b913b 6e99860de6f4 9a570069cdbb 2b5ef6430c21 b4543dbea84c c56cc7fefc31 1c5e670d6a5a
# Pick up full patch set for "path split" git cherry-pick a680a9298e7b 1b64aa9eae28 561341425bcc 6831ce129f19 cad93a089031 3bea83b68d54 3f607dcb43f1 e0367dfe90d6 9d41f08aa2eb 351849800157 2d1928c57df6 61910d6a5243 3512ac094293 31a59f75702f 0e082d926f59
# Pick up atomic_inc VMID fix git cherry-pick 0948a9c53860
# Pick up known fixes on "path split" git cherry-pick 7294a9bcaa7e c26bd6602e1d c2024e3b33ee cc28fac16ab7 775266207105 84c6f99e3907 596fc8adb171 44ba9786b673 24e1f056677e e27f05147bff
James Smart (33): Revert "scsi: lpfc: Resolve some cleanup issues following SLI path refactoring" Revert "scsi: lpfc: Fix element offset in __lpfc_sli_release_iocbq_s4()" Revert "scsi: lpfc: Fix locking for lpfc_sli_iocbq_lookup()" Revert "scsi: lpfc: Remove extra atomic_inc on cmd_pending in queuecommand after VMID" Revert "scsi: lpfc: SLI path split: Refactor SCSI paths" Revert "scsi: lpfc: SLI path split: Refactor fast and slow paths to native SLI4" Revert "scsi: lpfc: SLI path split: Refactor lpfc_iocbq" scsi: lpfc: SLI path split: Refactor lpfc_iocbq scsi: lpfc: SLI path split: Refactor fast and slow paths to native SLI4 scsi: lpfc: SLI path split: Introduce lpfc_prep_wqe scsi: lpfc: SLI path split: Refactor base ELS paths and the FLOGI path scsi: lpfc: SLI path split: Refactor PLOGI/PRLI/ADISC/LOGO paths scsi: lpfc: SLI path split: Refactor the RSCN/SCR/RDF/EDC/FARPR paths scsi: lpfc: SLI path split: Refactor LS_ACC paths scsi: lpfc: SLI path split: Refactor LS_RJT paths scsi: lpfc: SLI path split: Refactor FDISC paths scsi: lpfc: SLI path split: Refactor VMID paths scsi: lpfc: SLI path split: Refactor misc ELS paths scsi: lpfc: SLI path split: Refactor CT paths scsi: lpfc: SLI path split: Refactor SCSI paths scsi: lpfc: SLI path split: Refactor Abort paths scsi: lpfc: SLI path split: Refactor BSG paths scsi: lpfc: Remove extra atomic_inc on cmd_pending in queuecommand after VMID scsi: lpfc: Fix broken SLI4 abort path scsi: lpfc: Fix locking for lpfc_sli_iocbq_lookup() scsi: lpfc: Remove redundant lpfc_sli_prep_wqe() call scsi: lpfc: Fix split code for FLOGI on FCoE scsi: lpfc: Correct BDE DMA address assignment for GEN_REQ_WQE scsi: lpfc: Fix element offset in __lpfc_sli_release_iocbq_s4() scsi: lpfc: Fix dmabuf ptr assignment in lpfc_ct_reject_event() scsi: lpfc: Correct BDE type for XMIT_SEQ64_WQE in lpfc_ct_reject_event() scsi: lpfc: Resolve some cleanup issues following abort path refactoring scsi: lpfc: Resolve some cleanup issues following SLI path refactoring
drivers/scsi/lpfc/lpfc.h | 56 +- drivers/scsi/lpfc/lpfc_bsg.c | 303 ++--- drivers/scsi/lpfc/lpfc_crtn.h | 21 +- drivers/scsi/lpfc/lpfc_ct.c | 338 +++-- drivers/scsi/lpfc/lpfc_els.c | 1378 +++++++++++++------- drivers/scsi/lpfc/lpfc_hbadisc.c | 44 +- drivers/scsi/lpfc/lpfc_hw.h | 14 +- drivers/scsi/lpfc/lpfc_hw4.h | 29 + drivers/scsi/lpfc/lpfc_nportdisc.c | 98 +- drivers/scsi/lpfc/lpfc_nvme.c | 11 +- drivers/scsi/lpfc/lpfc_scsi.c | 2 +- drivers/scsi/lpfc/lpfc_sli.c | 1953 +++++++++++++--------------- drivers/scsi/lpfc/lpfc_sli.h | 5 +- 13 files changed, 2308 insertions(+), 1944 deletions(-)
Ick, this is crazy and way too much work for stable kernels, don't you think?
How about I just take the reverts and stop there?
Also, when sending in patches for stable kernels, I need the upstream git id in the changelog text somewhere. I don't see that here, so I couldn't take these as-is anyway :(
thanks,
greg k-h