commit a4d72d2dd0cbc3ff20f66a9168dd68b191c57409 Author: Mike Christie Date: Sun Feb 22 17:27:04 2026 -0600 scsi: vhost-scsi: Report direction completion support This has vhost-scsi report that it supports direct completions. When using a worker task per queue or group of queues with fast backends then enabling direct completion and submissions increases performance 20-30% with workloads like: fio --filename=/dev/sdb --direct=1 --rw=randrw --bs=8K \ --ioengine=libaio --iodepth=128 --numjobs=$jobs As jobs matches and passes the number of vCPUs in the VM then the benefit increases. However, when using a single worker then queueing completions and submissions is best as the worker is busy handling mapping data and setting/tearing down commands. Signed-off-by: Mike Christie Link: https://patch.msgid.link/20260222232946.7637-5-michael.christie@oracle.com Signed-off-by: Martin K. Petersen commit e1502d990c8e26fa679b3253ff7db51483e6eb82 Author: Mike Christie Date: Sun Feb 22 17:27:03 2026 -0600 scsi: target: Allow userspace to set the completion type This allows userspace to request if we complete in the backend context or the frontend driver. It works the same as submission where you can write 0 to 2 to complete_type: 0 - Use the fabric driver's preference. 1 - Complete from the backend calling context if the fabric supports it. 2 - Queue the completion to LIO's completion workqueue. Signed-off-by: Mike Christie Link: https://patch.msgid.link/20260222232946.7637-4-michael.christie@oracle.com Signed-off-by: Martin K. Petersen commit 89663fb2e53822863de9cf4bca9636989da96615 Author: Mike Christie Date: Sun Feb 22 17:27:02 2026 -0600 scsi: target: Use driver completion preference by default This has us use the driver's completion preference by default. There is no behavior changes with this patch and we queue completion to LIO's completion workqueue by default. Signed-off-by: Mike Christie Link: https://patch.msgid.link/20260222232946.7637-3-michael.christie@oracle.com Signed-off-by: Martin K. Petersen commit 06933066d88a3093953b062922c016a67d2cdbf8 Author: Mike Christie Date: Sun Feb 22 17:27:01 2026 -0600 scsi: target: Add support for completing commands from backend context To complete a command several drivers just drop their reference and add it to list to be processed by a driver specific thread. So there's no need to go from backend context to the LIO thread then to the driver's thread. When avoiding the LIO thread, IOPS can increase from 20-30% for workloads like: fio --filename=/dev/sdb --direct=1 --rw=randrw --bs=8K \ --ioengine=libaio --iodepth=128 --numjobs=$jobs where increasing jobs increases the performance improvement (this is using NVMe drives with LIO's submit_type=1 to directly submit). Add the infrastructure so drivers and userspace can control how to complete a command like is done for the submission path. In this commit there is no behavior change and we continue to defer to the LIO workqueue thread. In the subsequent commits we will allow drivers to report what they support and allow userspace to control the behavior. Signed-off-by: Mike Christie Link: https://patch.msgid.link/20260222232946.7637-2-michael.christie@oracle.com Signed-off-by: Martin K. Petersen commit 7179e626b76eb42f2529c6f6dd6ba88ea2445372 Author: Swarna Prabhu Date: Wed Feb 18 20:37:42 2026 -0800 scsi: sd: Enable sector size > PAGE_SIZE in SCSI sd driver The WRITE SAME(16) and WRITE SAME(10) SCSI commands use a page from a dedicated mempool (sd_page_pool) for their payload. This pool was initialized to allocate single pages, which was sufficient as long as the device sector size did not exceed the PAGE_SIZE. Given that block layer now supports block size upto 64KB, i.e. beyond PAGE_SIZE, initialize a large page pool in sd_probe() if a higher sector device is attached, ensuring atomicity. Adapt sd_set_special_bvec() to use large page pool when a higher sector size device is attached. Hence enable sector sizes > PAGE_SIZE in SCSI sd driver. Reviewed-by: Damien Le Moal Signed-off-by: Swarna Prabhu Co-developed-by: Pankaj Raghav Signed-off-by: Pankaj Raghav Link: https://patch.msgid.link/20260219043741.276729-2-sw.prabhu6@gmail.com Signed-off-by: Martin K. Petersen commit 50209dec14f8c594a9ef26237b7e7ddd39e12a40 Author: Yang Erkun Date: Tue Jan 27 14:20:44 2026 +0800 scsi: sg: Remove deprecated sg-big-buff These deprecated sysctl has been broken since commit 26d1c80fd61e ("scsi/sg: move sg-big-buff sysctl to scsi/sg.c") and nobody has found this. I believe it's time to remove it, which will allow us to clean up a significant amount of code. Signed-off-by: Yang Erkun Reviewed-by: Bart Van Assche Link: https://patch.msgid.link/20260127062044.3034148-4-yangerkun@huawei.com Signed-off-by: Martin K. Petersen commit d06a310b45e153872033dd0cf19d5a2279121099 Author: Yang Erkun Date: Tue Jan 27 14:20:43 2026 +0800 scsi: sg: Resolve soft lockup issue when opening /dev/sgX The parameter def_reserved_size defines the default buffer size reserved for each Sg_fd and should be restricted to a range between 0 and 1,048,576 (see https://tldp.org/HOWTO/SCSI-Generic-HOWTO/proc.html). Although the function sg_proc_write_dressz enforces this limit, it is possible to bypass it by directly modifying the module parameter as shown below, which then causes a soft lockup: echo -1 > /sys/module/sg/parameters/def_reserved_size exec 4<> /dev/sg0 watchdog: BUG: soft lockup - CPU#5 stuck for 26 seconds! [bash:537] Modules loaded: CPU: 5 UID: 0 PID: 537 Command: bash, kernel version 6.19.0-rc3+ #134, PREEMPT disabled Hardware: QEMU Standard PC (i440FX + PIIX, 1996), BIOS version 1.16.1-2.fc37 dated 04/01/2014 ... Call Trace: sg_build_reserve+0x5c/0xa0 sg_add_sfp+0x168/0x270 sg_open+0x16e/0x340 chrdev_open+0xbe/0x230 do_dentry_open+0x175/0x480 vfs_open+0x34/0xf0 do_open+0x265/0x3d0 path_openat+0x110/0x290 do_filp_open+0xc3/0x170 do_sys_openat2+0x71/0xe0 __x64_sys_openat+0x6d/0xa0 do_syscall_64+0x62/0x310 entry_SYSCALL_64_after_hwframe+0x76/0x7e The fix is to use module_param_cb to validate and reject invalid values assigned to def_reserved_size. Fixes: 6460e75a104d ("[SCSI] sg: fixes for large page_size") Signed-off-by: Yang Erkun Reviewed-by: Bart Van Assche Link: https://patch.msgid.link/20260127062044.3034148-3-yangerkun@huawei.com Signed-off-by: Martin K. Petersen commit 3033c471aaf675254efaa0da431e95d91a104b41 Author: Yang Erkun Date: Tue Jan 27 14:20:42 2026 +0800 scsi: sg: Fix sysctl sg-big-buff register during sg_init() Commit 26d1c80fd61e ("scsi/sg: move sg-big-buff sysctl to scsi/sg.c") made a mistake. sysctl sg-big-buff was not created because the call to register_sg_sysctls() was placed on the wrong code path. Fixes: 26d1c80fd61e ("scsi/sg: move sg-big-buff sysctl to scsi/sg.c") Signed-off-by: Yang Erkun Reviewed-by: Bart Van Assche Link: https://patch.msgid.link/20260127062044.3034148-2-yangerkun@huawei.com Signed-off-by: Martin K. Petersen commit 94c125bafa00042daf6d63b4fdd78384abc121fc Author: Igor Pylypiv Date: Mon Feb 9 13:21:51 2026 -0800 scsi: core: Add 'serial' sysfs attribute for SCSI/SATA Add a 'serial' sysfs attribute for SCSI and SATA devices. This attribute exposes the Unit Serial Number, which is derived from the Device Identification Vital Product Data (VPD) page 0x80. Whitespace is stripped from the retrieved serial number to handle the different alignment (right-aligned for SCSI, potentially left-aligned for SATA). As noted in SAT-5 10.5.3, "Although SPC-5 defines the PRODUCT SERIAL NUMBER field as right-aligned, ACS-5 does not require its SERIAL NUMBER field to be right-aligned. Therefore, right-alignment of the PRODUCT SERIAL NUMBER field for the translation is not assured." This attribute is used by tools such as lsblk to display the serial number of block devices. [mkp: length adjustment] Signed-off-by: Igor Pylypiv Reviewed-by: Bart Van Assche Reviewed-by: Hannes Reinecke Link: https://patch.msgid.link/20260209212151.342151-1-ipylypiv@google.com Signed-off-by: Martin K. Petersen commit 690d41fae92f0f255b1059d586bf064c63b5bfc3 Author: Pradeep P V K Date: Wed Feb 11 18:59:24 2026 +0530 scsi: ufs: qcom,sc7180-ufshc: dt-bindings: Add UFSHC compatible for x1e80100 Add UFS Host Controller (UFSHC) compatible for x1e80100 SoC. Use SM8550 as a fallback since x1e80100 is fully compatible with it. Qualcomm UFSHC is no longer compatible with JEDEC UFS-2.0 binding. Avoid using the "jedec,ufs-2.0" string in the compatible property. [mkp: conflict resolution] Acked-by: Manivannan Sadhasivam Reviewed-by: Krzysztof Kozlowski Signed-off-by: Pradeep P V K Link: https://patch.msgid.link/20260211132926.3716716-2-pradeep.pragallapati@oss.qualcomm.com Signed-off-by: Martin K. Petersen commit cf44b6369b8350e46e66bb69ef975c5aa22cec5e Author: Luca Weiss Date: Mon Jan 12 14:53:15 2026 +0100 scsi: ufs: qcom,sc7180-ufshc: dt-bindings: Document the Milos UFS Controller Document the UFS Controller on the Milos SoC. Reviewed-by: Krzysztof Kozlowski Signed-off-by: Luca Weiss Link: https://patch.msgid.link/20260112-milos-ufs-v2-2-d3ce4f61f030@fairphone.com Signed-off-by: Martin K. Petersen commit 5e0d4fdb98f3ab4c25aabda00b31dd5d4a806638 Author: Randy Dunlap Date: Tue Feb 24 15:49:54 2026 -0800 scsi: lpfc: ELIMINATE kernel-doc warnings in lpfc.h Avoid all kernel-doc warnings in lpfc.h: - Use the correct function parameter name - Add a '*' to a kernel-doc line - Repair the function Returns: comments Fixes these warnings: Warning: drivers/scsi/lpfc/lpfc.h:1674 No description found for return value of 'lpfc_next_online_cpu' Warning: drivers/scsi/lpfc/lpfc.h:1686 No description found for return value of 'lpfc_next_present_cpu' Warning: drivers/scsi/lpfc/lpfc.h:1700 function parameter 'eq' not described in 'lpfc_sli4_mod_hba_eq_delay' Warning: drivers/scsi/lpfc/lpfc.h:1755 bad line: -------------------------- Warning: drivers/scsi/lpfc/lpfc.h:1759 No description found for return value of 'lpfc_is_vmid_enabled' Signed-off-by: Randy Dunlap Reviewed-by: Justin Tee Link: https://patch.msgid.link/20260224234954.3606638-1-rdunlap@infradead.org Signed-off-by: Martin K. Petersen commit e521b77688365e0ed495baa6dae4905428a9f517 Author: Bart Van Assche Date: Mon Feb 23 14:00:30 2026 -0800 scsi: megaraid_sas: Protect more code with instance->reset_mutex megasas_get_device_list() and megasas_get_snapdump_properties() may unlock instance->reset_mutex indirectly. Hence, hold reset_mutex while calling these functions. Cc: Kashyap Desai Cc: Sumit Saxena Cc: Shivasharan S Cc: Chandrakanth patil Cc: James E.J. Bottomley Cc: Martin K. Petersen Cc: megaraidlinux.pdl@broadcom.com Cc: linux-scsi@vger.kernel.org Signed-off-by: Bart Van Assche Link: https://patch.msgid.link/20260223220102.2158611-31-bart.vanassche@linux.dev Signed-off-by: Martin K. Petersen commit fc803a39c42ad3887796cb3afbf7bdd4221199bf Author: Bart Van Assche Date: Mon Feb 23 14:00:29 2026 -0800 scsi: fnic: Make fnic_queuecommand() easier to analyze Move a spin_unlock_irqrestore() call such that the io_lock_acquired variable can be eliminated. This patch prepares for enabling the Clang thread-safety analyzer. Cc: Satish Kharat Cc: Sesidhar Baddela Cc: Karan Tilak Kumar Cc: James E.J. Bottomley Cc: Martin K. Petersen Cc: linux-scsi@vger.kernel.org Signed-off-by: Bart Van Assche Reviewed-by: Karan Tilak Kumar Link: https://patch.msgid.link/20260223220102.2158611-30-bart.vanassche@linux.dev Signed-off-by: Martin K. Petersen commit 47e088c9d1a06e0f762ac7ea62ae48c9e16c4def Author: Karan Tilak Kumar Date: Tue Feb 17 14:39:43 2026 -0800 scsi: fnic: Bump up version number Bump up version number. Tested-by: Karan Tilak Kumar Reviewed-by: Sesidhar Baddela Reviewed-by: Arulprabhu Ponnusamy Reviewed-by: Gian Carlo Boffa Reviewed-by: Arun Easi Reviewed-by: Hannes Reinecke Signed-off-by: Karan Tilak Kumar Link: https://patch.msgid.link/20260217223943.7938-5-kartilak@cisco.com Signed-off-by: Martin K. Petersen commit 927b5282df6463fea8eeb1342e58cec0fa03b35e Author: Karan Tilak Kumar Date: Tue Feb 17 14:39:42 2026 -0800 scsi: fnic: Refactor in_remove flag and call to fnic_fcpio_reset() Modify logic to remove unnecessary acquire/release of spinlock to set in_remove flag. There's also no need to check for init status to call fnic_fcpio_reset. Tested-by: Karan Tilak Kumar Reviewed-by: Sesidhar Baddela Reviewed-by: Arulprabhu Ponnusamy Reviewed-by: Gian Carlo Boffa Reviewed-by: Arun Easi Reviewed-by: Hannes Reinecke Signed-off-by: Karan Tilak Kumar Co-developed-by: Hannes Reinecke Link: https://patch.msgid.link/20260217223943.7938-4-kartilak@cisco.com Signed-off-by: Martin K. Petersen commit 31eda39bfd468a0fbabc0179e913c0755377e3b5 Author: Karan Tilak Kumar Date: Tue Feb 17 14:39:41 2026 -0800 scsi: fnic: Rename fnic_scsi_fcpio_reset() The function has no dependency on SCSI/FCP, so rename it to fnic_fcpio_reset() and move it to fnic_fcs.c Tested-by: Karan Tilak Kumar Reviewed-by: Sesidhar Baddela Reviewed-by: Arulprabhu Ponnusamy Reviewed-by: Gian Carlo Boffa Reviewed-by: Arun Easi Reviewed-by: Hannes Reinecke Reviewed-by: Lee Duncan Signed-off-by: Karan Tilak Kumar Co-developed-by: Hannes Reinecke Link: https://patch.msgid.link/20260217223943.7938-3-kartilak@cisco.com Signed-off-by: Martin K. Petersen commit a59d1caf1ded07e38a0f6e98c6491a75faedd70f Author: Karan Tilak Kumar Date: Tue Feb 17 14:39:40 2026 -0800 scsi: fnic: Do not use GFP_ZERO for mempools One cannot use the GFP_ZERO flag for mempool allocation, so use memset() instead. Tested-by: Karan Tilak Kumar Reviewed-by: Sesidhar Baddela Reviewed-by: Arulprabhu Ponnusamy Reviewed-by: Gian Carlo Boffa Reviewed-by: Arun Easi Reviewed-by: Hannes Reinecke Reviewed-by: Lee Duncan Signed-off-by: Karan Tilak Kumar Co-developed-by: Hannes Reinecke Link: https://patch.msgid.link/20260217223943.7938-2-kartilak@cisco.com Signed-off-by: Martin K. Petersen commit 0e07baae55bc319e4e9559fee352b9252a467db6 Author: Karan Tilak Kumar Date: Tue Feb 17 14:39:39 2026 -0800 scsi: fnic: Use mempool for receive frames The receive frames are constantly replenished so we should rather use a mempool here. fip_frame_queue is an rxq. Deallocate it in fnic_free_rxq(). Tested-by: Karan Tilak Kumar Reviewed-by: Sesidhar Baddela Reviewed-by: Arulprabhu Ponnusamy Reviewed-by: Gian Carlo Boffa Reviewed-by: Arun Easi Reviewed-by: Hannes Reinecke Signed-off-by: Karan Tilak Kumar Co-developed-by: Hannes Reinecke Link: https://patch.msgid.link/20260217223943.7938-1-kartilak@cisco.com Signed-off-by: Martin K. Petersen commit f707860ebc84dd07148c3855e2f0fa9f41a3ed4a Author: Peter Wang Date: Tue Feb 10 15:17:30 2026 +0800 scsi: ufs: core: Support UFSHCI 4.1 CQ entry tag The UFSHCI 4.1 specification introduces a new completion queue (CQ) entry format, allowing the tag to be obtained directly. Signed-off-by: Peter Wang Reviewed-by: Bean Huo Reviewed-by: Bart Van Assche Link: https://patch.msgid.link/20260210071834.1837878-1-peter.wang@mediatek.com Signed-off-by: Martin K. Petersen commit 3abe4113e784b4a1135fe0e89828afecbfebf862 Author: Peter Wang Date: Tue Feb 10 14:41:44 2026 +0800 scsi: ufs: core: Add debug log for MCQ command timeout It is difficult to debug situations where an MCQ command timeout occurs, the corresponding CQ tag response is received, but the request is not completed. Add a one-line log to indicate when the CQ entry is abnormal. Signed-off-by: Peter Wang Reviewed-by: Bart Van Assche Link: https://patch.msgid.link/20260210070837.1820710-3-peter.wang@mediatek.com Signed-off-by: Martin K. Petersen commit 01517654bc258efb2610e53e99fa4ef641d134b9 Author: Peter Wang Date: Tue Feb 10 14:41:43 2026 +0800 scsi: ufs: core: Add debug log for UIC command timeout It is difficult to debug when a UIC command timeout occurs simultaneously with a UIC command complete interrupt. Currently, we only see the timeout log without any debug information, making it unclear whether the UFS device failed to respond or the host entered an incorrect state. Add a one-line log to cover this situation. Signed-off-by: Peter Wang Reviewed-by: Bart Van Assche Link: https://patch.msgid.link/20260210070837.1820710-2-peter.wang@mediatek.com Signed-off-by: Martin K. Petersen commit 931c105de11c2e0e9675fbc061b16a9c5f134dcf Author: Thorsten Blum Date: Tue Feb 24 15:48:27 2026 +0100 scsi: BusLogic: Replace deprecated strcpy() + strcat() in blogic_rdconfig() strcpy() is deprecated [1] and using strcat() is discouraged. Replace them with scnprintf(). No functional changes. Link: https://www.kernel.org/doc/html/latest/process/deprecated.html#strcpy [1] Signed-off-by: Thorsten Blum Link: https://patch.msgid.link/20260224144828.585577-1-thorsten.blum@linux.dev Signed-off-by: Martin K. Petersen commit 6446e8c2b6c76b4d253c46352540ca1f5fa3f854 Author: Justin Tee Date: Thu Feb 12 13:30:08 2026 -0800 scsi: lpfc: Update lpfc version to 14.4.0.14 Update lpfc version to 14.4.0.14. Signed-off-by: Justin Tee Link: https://patch.msgid.link/20260212213008.149873-14-justintee8345@gmail.com Signed-off-by: Martin K. Petersen commit 107cb8ed4f44eeb17f9624e183603c0b885ef039 Author: Justin Tee Date: Thu Feb 12 13:30:07 2026 -0800 scsi: lpfc: Update copyright year string for 2026 Update copyright string to 2026 for this version patch set. Signed-off-by: Justin Tee Link: https://patch.msgid.link/20260212213008.149873-13-justintee8345@gmail.com Signed-off-by: Martin K. Petersen commit 5807d96c46d5ccfb4c247f16653e616e8c90ae06 Author: Justin Tee Date: Thu Feb 12 13:30:06 2026 -0800 scsi: lpfc: Restrict first burst to non-FCoE and SLI4 adapters only First burst is only supported on adapters running in SLI4 mode and that are non-FCoE based. Include sli_rev and FCoE mode checks before setting the write transfer ready disabled bit in PRLIs. Signed-off-by: Justin Tee Link: https://patch.msgid.link/20260212213008.149873-12-justintee8345@gmail.com Signed-off-by: Martin K. Petersen commit 9714c5463fd1d963fe30193ed75b9578e84278ab Author: Justin Tee Date: Thu Feb 12 13:30:05 2026 -0800 scsi: lpfc: Update class of service bit field to 3 bits for WQE submissions WQE submissions only require a 3 bit field when specifying the class of service to use. So, update WQE submission paths to use a 3 bit field instead of 0x0f as the bit mask. A NLP_FCP_CLASS_MASK bitmask is defined to ensure only a 3 bit mask is used. Signed-off-by: Justin Tee Link: https://patch.msgid.link/20260212213008.149873-11-justintee8345@gmail.com Signed-off-by: Martin K. Petersen commit 559a6c2ab097695f7b3cd4fdd5f6aca81ae3da09 Author: Justin Tee Date: Thu Feb 12 13:30:04 2026 -0800 scsi: lpfc: Add clean up of aborted NVMe commands during PCI fcn reset When handling a PCI function reset, notification to the NVMe transport layer is skipped for outstanding aborted NVMe I/O. Introduce a new routine called lpfc_nvme_flush_abts_list(), which notifies upper NVMe transport layer of outstanding aborted NVMe I/O that are not planned to be completed normally due to a PCI function reset request. Signed-off-by: Justin Tee Link: https://patch.msgid.link/20260212213008.149873-10-justintee8345@gmail.com Signed-off-by: Martin K. Petersen commit 2da10bcaa58a389ca60f8e788180e0dca00739bc Author: Justin Tee Date: Thu Feb 12 13:30:03 2026 -0800 scsi: lpfc: Fix incorrect txcmplq_cnt during cleanup in lpfc_sli_abort_ring() When a port is offline in lpfc_sli_abort_ring, the phba->txcmplq is cleared but the phba->txcmplq_cnt is not reset to zero. This can sometimes result in a phba->txcmplq_cnt that never reaches zero, which hangs the cleanup process. Update lpfc_sli_abort_ring so that txcmplq_cnt is reset to zero and also ensure that the LPFC_IO_ON_TXCMPLQ flag is properly cleared. Signed-off-by: Justin Tee Link: https://patch.msgid.link/20260212213008.149873-9-justintee8345@gmail.com Signed-off-by: Martin K. Petersen commit 6b0bcf4b6430688984fe1ee69fce7165a3e24b92 Author: Justin Tee Date: Thu Feb 12 13:30:02 2026 -0800 scsi: lpfc: Cleanup error exit paths in lpfc_fdmi_cmd() and associated messages Error labels in lpfc_fdmi_cmd() accidentally return success status and can potentially leak memory. Change error exit path status to return a non-zero value using a common exit path for failure cases. The error path also frees allocated memory and provides logging. Signed-off-by: Justin Tee Link: https://patch.msgid.link/20260212213008.149873-8-justintee8345@gmail.com Signed-off-by: Martin K. Petersen commit f6bfb8d149336661bb80e62980da9a45b920403c Author: Justin Tee Date: Thu Feb 12 13:30:01 2026 -0800 scsi: lpfc: Remove unnecessary ndlp kref get in lpfc_check_nlp_post_devloss When NLP_IN_RECOV_POST_DEV_LOSS is set, the initial node reference remains held while recovery is in progress. Taking a reference when NLP_IN_RECOV_POST_DEV_LOSS is cleared results in an additional reference being held. This causes an extra reference when cleaning up lpfc_vport instances. Thus, remove the extraneous ndlp kref get in lpfc_check_nlp_post_devloss. Signed-off-by: Justin Tee Link: https://patch.msgid.link/20260212213008.149873-7-justintee8345@gmail.com Signed-off-by: Martin K. Petersen commit 70b468d41b822e4b510761120be3924df966a62d Author: Justin Tee Date: Thu Feb 12 13:30:00 2026 -0800 scsi: lpfc: Reduce pointer chasing when accessing vmid_flag For all FLOGI completions, the vport->phba->pport pointer is actually a pointer to the original vport pointer because FLOGIs always complete on the physical lpfc_vport object. Thus, we can reduce the vport->phba->pport->vmid_flag dereference to simply vport->vmid_flag. Signed-off-by: Justin Tee Link: https://patch.msgid.link/20260212213008.149873-6-justintee8345@gmail.com Signed-off-by: Martin K. Petersen commit f8c599ad90f53dbe2246935f90ff49693c26b34f Author: Justin Tee Date: Thu Feb 12 13:29:59 2026 -0800 scsi: lpfc: Use min_t() instead of min() in lpfc_sli4_driver_resource_setup The member called cfg_sg_dma_buf_size is declared as a u32, while the min comparator's second argument called SLI4_PAGE_SIZE is a #define. Proper comparison should be using the same type, therefore change to use min_t. Signed-off-by: Justin Tee Link: https://patch.msgid.link/20260212213008.149873-5-justintee8345@gmail.com Signed-off-by: Martin K. Petersen commit 5f442e54e9ef662aaad736ca1af13f20d0448f08 Author: Justin Tee Date: Thu Feb 12 13:29:58 2026 -0800 scsi: lpfc: Add log messages to fabric login error labels Should fabric login or related initialization mailbox commands fail, there are no log messages to notify which step encountered an issue. Update error label paths to log when unexpected fabric login issues occur. Signed-off-by: Justin Tee Link: https://patch.msgid.link/20260212213008.149873-4-justintee8345@gmail.com Signed-off-by: Martin K. Petersen commit b4082ac8e62ca669beabddff47238cdb33ea47dc Author: Justin Tee Date: Thu Feb 12 13:29:57 2026 -0800 scsi: lpfc: Log discarded and insufficient RQE buffer events An RCQE with statuses indicating that an RQE is dropped or when there are insufficient buffers to receive new RQEs are currently occuring silently. Add a new log message to warn when such events occur. Signed-off-by: Justin Tee Link: https://patch.msgid.link/20260212213008.149873-3-justintee8345@gmail.com Signed-off-by: Martin K. Petersen commit 8ecb3ec244acd7db2a6c071d53eb4870619fb5e6 Author: Justin Tee Date: Thu Feb 12 13:29:56 2026 -0800 scsi: lpfc: Update log message when ndlp kref get is unsuccessful If kref_get_unless_zero on ndlp->kref is unsuccessful, then there's no point to log kref_read(&ndlp->kref) because it will of course be zero. In such cases, ndlp->vport would also be an invalid pointer. Thus, use pr_info() instead of lpfc_printf_vlog() to log when a kref get is attempted on an ndlp with a zero kref. Signed-off-by: Justin Tee Link: https://patch.msgid.link/20260212213008.149873-2-justintee8345@gmail.com Signed-off-by: Martin K. Petersen