kernel/linux.git/drivers/scsi/libsas, branch v6.18.44

scsi: libsas: Fix HA resume deadlock and hisi_sas disk-wake race

2026-08-09T18:25:02+00:00

[ Upstream commit 3dbbbf656b850c9c8de05df6ad4a1dfc6ff02845 ] Commit fbefe22811c3 ("scsi: libsas: Don't always drain event workqueue for HA resume") introduced sas_resume_ha_no_sync() to avoid a deadlock: the PHYE_RESUME_TIMEOUT handler, running on the HA event workqueue, calls sas_deform_port() -> sas_destruct_devices(), which removes SCSI devices and waits for the host to become runtime-active. But the host cannot resume until sas_resume_ha() -> sas_drain_work() returns, and the drain is blocked on that very handler. However skipping the drain reintroduces a race: hisi_sas returns from resume before all PHY UP work and libsas discovery work finish. The controller may then autosuspend while disks are still waking up. The disks issue IO to a suspended controller, the IO fails, and the disks get disabled. Fix the deadlock at its source by moving the PHYE_RESUME_TIMEOUT notification to after sas_drain_work(). By then the host resume is about to complete, so device removal through device_link no longer blocks on the resume and the cycle is broken. With the deadlock gone, restore sas_resume_ha() (the draining variant) in hisi_sas and remove sas_resume_ha_no_sync(). The reorder is safe for the other libsas consumers (isci, pm8001, aic94xx, mvsas). During suspend, sas_suspend_devices() calls sas_notify_lldd_dev_gone() for each device, which sets dev->lldd_dev to NULL. When scsi_unblock_requests re-enables I/O in resume, any I/O to a timed-out phy's disk is immediately rejected by the LLDD before reaching hardware: isci returns SAS_DEVICE_UNKNOWN (mapped to DID_BAD_TARGET), and pm8001 returns SAS_PHY_DOWN (mapped to DID_NO_CONNECT). Both complete directly via scsi_done() without entering SCSI EH. This is identical in both the old and new ordering since lldd_dev_gone runs during suspend, before resume. The reorder only affects when the PHYE_RESUME_TIMEOUT handler runs (synchronized by sas_drain_work() vs. asynchronous after resume returns), not whether I/O can reach the device. aic94xx and mvsas do not register any PM ops and never reach this code path. Fixes: fbefe22811c3 ("scsi: libsas: Don't always drain event workqueue for HA resume") Signed-off-by: Xingui Yang Reviewed-by: John Garry Link: https://patch.msgid.link/20260716081145.3950172-1-yangxingui@huawei.com Signed-off-by: Martin K. Petersen Signed-off-by: Sasha Levin

scsi: Revert "scsi: libsas: Fix exp-attached device scan after probe failure scanned in again after probe failed"

2026-01-17T15:35:18+00:00

[ Upstream commit 278712d20bc8ec29d1ad6ef9bdae9000ef2c220c ] This reverts commit ab2068a6fb84751836a84c26ca72b3beb349619d. When probing the exp-attached sata device, libsas/libata will issue a hard reset in sas_probe_sata() -> ata_sas_async_probe(), then a broadcast event will be received after the disk probe fails, and this commit causes the probe will be re-executed on the disk, and a faulty disk may get into an indefinite loop of probe. Therefore, revert this commit, although it can fix some temporary issues with disk probe failure. Signed-off-by: Xingui Yang Reviewed-by: Jason Yan Reviewed-by: John Garry Link: https://patch.msgid.link/20251202065627.140361-1-yangxingui@huawei.com Signed-off-by: Martin K. Petersen Signed-off-by: Sasha Levin

Merge tag 'scsi-misc' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi

2025-10-04T02:17:48+00:00

Pull SCSI updates from James Bottomley: "Usual driver updates (ufs, mpi3mr, lpfc, pm80xx, mpt3sas) plus assorted cleanups and fixes. The only core update is to sd.c and is mostly cosmetic" * tag 'scsi-misc' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi: (105 commits) scsi: MAINTAINERS: Update FC element owners scsi: mpt3sas: Update driver version to 54.100.00.00 scsi: mpt3sas: Add support for 22.5 Gbps SAS link rate scsi: mpt3sas: Suppress unnecessary IOCLogInfo on CONFIG_INVALID_PAGE scsi: mpt3sas: Fix crash in transport port remove by using ioc_info() scsi: ufs: ufs-qcom: Add support for limiting HS gear and rate scsi: ufs: pltfrm: Add DT support to limit HS gear and gear rate scsi: ufs: ufs-qcom: Remove redundant re-assignment to hs_rate scsi: ufs: dt-bindings: Document gear and rate limit properties scsi: ufs: core: Fix data race in CPU latency PM QoS request handling scsi: libfc: Fix potential buffer overflow in fc_ct_ms_fill() scsi: storvsc: Remove redundant ternary operators scsi: ufs: core: Change MCQ interrupt enable flow scsi: smartpqi: Replace kmalloc() + copy_from_user() with memdup_user() scsi: hpsa: Replace kmalloc() + copy_from_user() with memdup_user() scsi: hpsa: Fix potential memory leak in hpsa_big_passthru_ioctl() scsi: lpfc: Copyright updates for 14.4.0.11 patches scsi: lpfc: Update lpfc version to 14.4.0.11 scsi: lpfc: Convert debugfs directory counts from atomic to unsigned int scsi: lpfc: Clean up extraneous phba dentries ...

scsi: libsas: Add dev_parent_is_expander() helper

2025-08-19T02:08:30+00:00

Many libsas drivers check if the parent of the device is an expander. Create a helper that the libsas drivers will use in follow up commits. Suggested-by: Damien Le Moal Signed-off-by: Niklas Cassel Link: https://lore.kernel.org/r/20250814173215.1765055-15-cassel@kernel.org Reviewed-by: Damien Le Moal Reviewed-by: John Garry Reviewed-by: Jason Yan Signed-off-by: Martin K. Petersen

scsi: switch ->bios_param() to passing gendisk

2025-08-13T06:59:28+00:00

Instances are passed struct block_device *bdev argument; the only thing it is used for (if it's used in the first place) is bdev->bd_disk. Might as well pass that in the first place... Reviewed-by: Martin K. Petersen Reviewed-by: Christoph Hellwig Acked-by: Jens Axboe Signed-off-by: Al Viro

Merge tag 'scsi-misc' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi

2025-08-06T12:44:25+00:00

Pull more SCSI updates from James Bottomley: "This is mostly fixes and cleanups and code reworks that trickled in across the merge window and the weeks leading up. The only substantive update is the Mediatek ufs driver which accounts for the bulk of the additions" * tag 'scsi-misc' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi: (37 commits) scsi: libsas: Use a bool for sas_deform_port() second argument scsi: libsas: Move declarations of internal functions to sas_internal.h scsi: libsas: Make sas_get_ata_info() static scsi: libsas: Simplify sas_ata_wait_eh() scsi: libsas: Refactor dev_is_sata() scsi: sd: Make sd shutdown issue START STOP UNIT appropriately scsi: arm64: dts: mediatek: mt8195: Add UFSHCI node scsi: dt-bindings: mediatek,ufs: add MT8195 compatible and update clock nodes scsi: dt-bindings: mediatek,ufs: Add ufs-disable-mcq flag for UFS host scsi: ufs: ufs-mediatek: Add UFS host support for MT8195 SoC scsi: ufs: ufs-pci: Remove control of UIC Completion interrupt for Intel MTL scsi: ufs: core: Do not write interrupt enable register unnecessarily scsi: ufs: core: Set and clear UIC Completion interrupt as needed scsi: ufs: core: Remove duplicated code in ufshcd_send_bsg_uic_cmd() scsi: ufs: core: Move ufshcd_enable_intr() and ufshcd_disable_intr() scsi: ufs: ufs-pci: Remove UFS PCI driver's ->late_init() call back scsi: ufs: ufs-pci: Fix default runtime and system PM levels scsi: ufs: ufs-pci: Fix hibernate state transition for Intel MTL-like host controllers scsi: ufs: host: mediatek: Support FDE (AES) clock scaling scsi: ufs: host: mediatek: Support clock scaling with Vcore binding ...

scsi: libsas: Use a bool for sas_deform_port() second argument

2025-07-25T13:03:57+00:00

Change the type of the "gone" argument of sas_deform_port() from int to bool. Simliarly, to be consistent, do the same change to the function sas_unregister_domain_devices(). No functional changes. Signed-off-by: Damien Le Moal Link: https://lore.kernel.org/r/20250725015818.171252-6-dlemoal@kernel.org Reviewed-by: Johannes Thumshirn Reviewed-by: John Garry Reviewed-by: Jason Yan Signed-off-by: Martin K. Petersen

scsi: libsas: Move declarations of internal functions to sas_internal.h

2025-07-25T13:03:57+00:00

Move the declaration of all functions used only within libsas from include/scsi/sas_ata.h to drivers/scsi/libsas/sas_internal.h. No functional changes. Signed-off-by: Damien Le Moal Link: https://lore.kernel.org/r/20250725015818.171252-5-dlemoal@kernel.org Reviewed-by: Johannes Thumshirn Reviewed-by: John Garry Reviewed-by: Jason Yan Signed-off-by: Martin K. Petersen

scsi: libsas: Make sas_get_ata_info() static

2025-07-25T13:03:56+00:00

The function sas_get_ata_info() is used only in drivers/scsi/libsas/sas_ata.c. Remove its definition from include/scsi/sas_ata.h and make this function static. No functional changes. Signed-off-by: Damien Le Moal Link: https://lore.kernel.org/r/20250725015818.171252-4-dlemoal@kernel.org Reviewed-by: Johannes Thumshirn Reviewed-by: John Garry Reviewed-by: Jason Yan Signed-off-by: Martin K. Petersen

scsi: libsas: Simplify sas_ata_wait_eh()

2025-07-25T13:03:56+00:00

Simplify the code of sas_ata_wait_eh(), removing the local variable ap for the pointer to the device ata_port structure. The test using dev_is_sata() is also removed as all call sites of this function check if the device is a SATA one before calling this function. Signed-off-by: Damien Le Moal Link: https://lore.kernel.org/r/20250725015818.171252-3-dlemoal@kernel.org Reviewed-by: John Garry Reviewed-by: Jason Yan Signed-off-by: Martin K. Petersen