kernel/linux.git/drivers/ata/libata.h, branch v6.12.80

ata: libata-scsi: avoid Non-NCQ command starvation

2026-03-13T16:20:23+00:00

[ Upstream commit 0ea84089dbf62a92dc7889c79e6b18fc89260808 ] When a non-NCQ command is issued while NCQ commands are being executed, ata_scsi_qc_issue() indicates to the SCSI layer that the command issuing should be deferred by returning SCSI_MLQUEUE_XXX_BUSY. This command deferring is correct and as mandated by the ACS specifications since NCQ and non-NCQ commands cannot be mixed. However, in the case of a host adapter using multiple submission queues, when the target device is under a constant load of NCQ commands, there are no guarantees that requeueing the non-NCQ command will be executed later and it may be deferred again repeatedly as other submission queues can constantly issue NCQ commands from different CPUs ahead of the non-NCQ command. This can lead to very long delays for the execution of non-NCQ commands, and even complete starvation for these commands in the worst case scenario. Since the block layer and the SCSI layer do not distinguish between queueable (NCQ) and non queueable (non-NCQ) commands, libata-scsi SAT implementation must ensure forward progress for non-NCQ commands in the presence of NCQ command traffic. This is similar to what SAS HBAs with a hardware/firmware based SAT implementation do. Implement such forward progress guarantee by limiting requeueing of non-NCQ commands from ata_scsi_qc_issue(): when a non-NCQ command is received and NCQ commands are in-flight, do not force a requeue of the non-NCQ command by returning SCSI_MLQUEUE_XXX_BUSY and instead return 0 to indicate that the command was accepted but hold on to the qc using the new deferred_qc field of struct ata_port. This deferred qc will be issued using the work item deferred_qc_work running the function ata_scsi_deferred_qc_work() once all in-flight commands complete, which is checked with the port qc_defer() callback return value indicating that no further delay is necessary. This check is done using the helper function ata_scsi_schedule_deferred_qc() which is called from ata_scsi_qc_complete(). This thus excludes this mechanism from all internal non-NCQ commands issued by ATA EH. When a port deferred_qc is non NULL, that is, the port has a command waiting for the device queue to drain, the issuing of all incoming commands (both NCQ and non-NCQ) is deferred using the regular busy mechanism. This simplifies the code and also avoids potential denial of service problems if a user issues too many non-NCQ commands. Finally, whenever ata EH is scheduled, regardless of the reason, a deferred qc is always requeued so that it can be retried once EH completes. This is done by calling the function ata_scsi_requeue_deferred_qc() from ata_eh_set_pending(). This avoids the need for any special processing for the deferred qc in case of NCQ error, link or device reset, or device timeout. Reported-by: Xingui Yang Reported-by: Igor Pylypiv Fixes: bdb01301f3ea ("scsi: Add host and host template flag 'host_tagset'") Cc: stable@vger.kernel.org Signed-off-by: Damien Le Moal Reviewed-by: Niklas Cassel Reviewed-by: Martin K. Petersen Reviewed-by: John Garry Tested-by: Igor Pylypiv Tested-by: Xingui Yang Signed-off-by: Sasha Levin

ata: libata: Introduce ata_port_eh_scheduled()

2026-03-13T16:20:23+00:00

[ Upstream commit 7aae547bbe442affc4afe176b157fab820a12437 ] Introduce the inline helper function ata_port_eh_scheduled() to test if EH is pending (ATA_PFLAG_EH_PENDING port flag is set) or running (ATA_PFLAG_EH_IN_PROGRESS port flag is set) for a port. Use this helper in ata_port_wait_eh() and __ata_scsi_queuecmd() to replace the hardcoded port flag tests. No functional changes. Signed-off-by: Damien Le Moal Reviewed-by: Niklas Cassel Link: https://lore.kernel.org/r/20250704104601.310643-1-dlemoal@kernel.org Signed-off-by: Niklas Cassel Stable-dep-of: 0ea84089dbf6 ("ata: libata-scsi: avoid Non-NCQ command starvation") Signed-off-by: Sasha Levin

ata: libata: Remove ATA_DFLAG_ZAC device flag

2026-03-13T16:20:23+00:00

[ Upstream commit a0f26fcc383965e0522b81269062a9278bc802fe ] The ATA device flag ATA_DFLAG_ZAC is used to indicate if a devie is a host managed or host aware zoned device. However, this flag is not used in the hot path and only used during device scanning/revalidation and for inquiry and sense SCSI command translation. Save one bit from struct ata_device flags field by replacing this flag with the internal helper function ata_dev_is_zac(). This function returns true if the device class is ATA_DEV_ZAC (host managed ZAC device case) or if its identify data reports it supports the zoned command set (host aware ZAC device case). Signed-off-by: Damien Le Moal Reviewed-by: Hannes Reinecke Reviewed-by: Niklas Cassel Stable-dep-of: 0ea84089dbf6 ("ata: libata-scsi: avoid Non-NCQ command starvation") Signed-off-by: Sasha Levin

ata: libata: avoid long timeouts on hot-unplugged SATA DAS

2026-03-04T12:21:20+00:00

[ Upstream commit 151cabd140322205e27dae5c4bbf261ede0056e3 ] When a SATA DAS enclosure is connected behind a Thunderbolt PCIe switch, hot-unplugging the whole enclosure causes pciehp to tear down the PCI hierarchy before the SCSI layer issues SYNCHRONIZE CACHE and START STOP UNIT for the disks. libata still queues these commands and the AHCI driver tries to access the HBA registers even though the PCI channel is already offline. This results in a series of timeouts and error recovery attempts, e.g.: [ 824.778346] pcieport 0000:00:07.0: pciehp: Slot(14): Link Down [ 891.612720] ata8.00: qc timeout after 5000 msecs (cmd 0xec) [ 902.876501] ata8.00: qc timeout after 10000 msecs (cmd 0xec) [ 934.107998] ata8.00: qc timeout after 30000 msecs (cmd 0xec) [ 936.206431] sd 7:0:0:0: [sda] Synchronize Cache(10) failed: Result: hostbyte=DID_BAD_TARGET driverbyte=DRIVER_OK ... [ 1006.298356] ata1.00: qc timeout after 5000 msecs (cmd 0xec) [ 1017.561926] ata1.00: qc timeout after 10000 msecs (cmd 0xec) [ 1048.791790] ata1.00: qc timeout after 30000 msecs (cmd 0xec) [ 1050.890035] sd 0:0:0:0: [sdb] Synchronize Cache(10) failed: Result: hostbyte=DID_BAD_TARGET driverbyte=DRIVER_OK With this patch applied, the same hot-unplug looks like: [ 59.965496] pcieport 0000:00:07.0: pciehp: Slot(14): Link Down [ 60.002502] sd 7:0:0:0: [sda] Synchronize Cache(10) failed: Result: hostbyte=DID_BAD_TARGET driverbyte=DRIVER_OK ... [ 60.103050] sd 0:0:0:0: [sdb] Synchronize Cache(10) failed: Result: hostbyte=DID_BAD_TARGET driverbyte=DRIVER_OK In this test setup with two disks, the hot-unplug sequence shrinks from about 226 seconds (~3.8 minutes) between the Link Down event and the last SYNCHRONIZE CACHE failure to under a second. Without this patch the total delay grows roughly with the number of disks, because each disk gets its own SYNCHRONIZE CACHE and qc timeout series. If the underlying PCI device is already gone, these commands cannot succeed anyway. Avoid issuing them by introducing ata_adapter_is_online(), which checks pci_channel_offline() for PCI-based hosts. It is used from ata_scsi_find_dev() to return NULL, causing the SCSI layer to fail new commands with DID_BAD_TARGET immediately, and from ata_qc_issue() to bail out before touching the HBA registers. Since such failures would otherwise trigger libata error handling, ata_adapter_is_online() is also consulted from ata_scsi_port_error_handler(). When the adapter is offline, libata skips ap->ops->error_handler(ap) and completes error handling using the existing path, rather than running a full EH sequence against a dead adapter. With this change, SYNCHRONIZE CACHE and START STOP UNIT commands issued during hot-unplug fail quickly once the PCI channel is offline, without qc timeout spam or long libata EH delays. Suggested-by: Damien Le Moal Signed-off-by: Henry Tseng Signed-off-by: Damien Le Moal Signed-off-by: Sasha Levin

ata: libata: Improve CDL resource management

2024-09-07T01:16:56+00:00

The ncq_sense_buf buffer field of struct ata_port is allocated and used only for devices that support the Command Duration Limits (CDL) feature. However, the cdl buffer of struct ata_device, which is used to cache the command duration limits log page for devices supporting CDL is always allocated as part of struct ata_device, which is wasteful of memory for devices that do not support this feature. Clean this up by defining both buffers as part of the new ata_cdl structure and allocating this structure only for devices that support the CDL feature. This new structure is attached to struct ata_device using the cdl pointer. The functions ata_dev_init_cdl_resources() and ata_dev_cleanup_cdl_resources() are defined to manage this new structure allocation, initialization and freeing when a port is removed or a device disabled. ata_dev_init_cdl_resources() is called from ata_dev_config_cdl() only for devices that support CDL. ata_dev_cleanup_cdl_resources() is called from ata_dev_free_resources() to free the ata_cdl structure when a device is being disabled by EH. Note that the name of the former cdl log buffer of struct ata_device is changed to desc_log_buf to make it clearer that it is a buffer for the limit descriptors log page. This change reduces the size of struct ata_device, thus reducing memory usage for ATA devices that do not support the CDL feature. Signed-off-by: Damien Le Moal Reviewed-by: Niklas Cassel

ata: libata: Introduce ata_dev_free_resources

2024-09-07T01:16:55+00:00

Introduce the function ata_dev_free_resources() to free the resources allocated to support a device features. For now, this function is reduced to calling zpodd_exit() for devices that have this feature enabled. ata_dev_free_resources() is called from ata_eh_dev_disable() as this function is always called for all devices attached to a port that is being detached and for devices that are being disabled due to being removed (detached) from the system or due to errors. With this change, the call to zpodd_exit() done in ata_port_detach() and ata_scsi_handle_link_detach() are removed as these functions remove all devices attached to the link or port using libata EH, thus resulting in ata_eh_dev_disable() being called and the zpodd_exit() function being executed. Signed-off-by: Damien Le Moal Reviewed-by: Niklas Cassel

ata: libata: Rename ata_eh_read_sense_success_ncq_log()

2024-09-07T01:16:55+00:00

The function ata_eh_read_sense_success_ncq_log() does more that just reading the sense data for successful NCQ commands log page as it also sets the sense data for all commands listed in the log page. Rename this function to ata_eh_get_ncq_success_sense() to better describe what the function does. Furthermore, since this function is only called from ata_eh_get_success_sense() in libata-eh.c, there is no need to export it and its declaration can be moved to drivers/ata/libata.h. To be consistent with this change, the function ata_eh_read_sense_success_non_ncq() is also renamed to ata_eh_get_non_ncq_success_sense(). Signed-off-by: Damien Le Moal Reviewed-by: Niklas Cassel Reviewed-by: Hannes Reinecke

ata: libata: Move sata_down_spd_limit() to libata-sata.c

2024-09-07T01:16:55+00:00

Move the definition of the function sata_down_spd_limit() to libata-sata.c where it belongs, together with sata_set_spd(). The helper function ata_sstatus_online() is also changed to be an inline function defined in drivers/ata/libata.h. Signed-off-by: Damien Le Moal Reviewed-by: Niklas Cassel Reviewed-by: Hannes Reinecke

ata: libata: Add helper ata_eh_decide_disposition()

2024-08-29T01:06:11+00:00

Every time I see libata code calling scsi_check_sense(), I get confused why the code path that is working fine for SCSI code, is not sufficient for libata code. The reason is that SCSI usually gets the sense data as part of the completion, and will thus automatically call scsi_check_sense(), which will set the SCSI ML byte (if any). However, for libata queued commands, we always need to fetch the sense data via SCSI EH, and thus do not get the luxury of having scsi_check_sense() called automatically. Add a new helper, ata_eh_decide_disposition(), that has a ata_eh_ prefix to more clearly highlight that this is only needed for code called by EH, while also having a similar name to scsi_decide_disposition(), such that it is easier to compare the libata code with the equivalent SCSI code. Also add a big kdoc comment explaining why this helper is called/needed in the first place. Signed-off-by: Niklas Cassel Signed-off-by: Damien Le Moal

ata: libata: Remove obsolete function declarations

2024-08-26T23:04:53+00:00

The function ata_schedule_scsi_eh() was removed with commit f8bbfc247efb ("[PATCH] SCSI: make scsi_implement_eh() generic API for SCSI transports"), and the function ata_sff_irq_clear() was removed with commit 37f65b8bc262("libata-sff: ata_sff_irq_clear() is BMDMA specific"). Remove the now useless declarations of these functions in drivers/ata/libata.h and include/linux/libata.h. Signed-off-by: Gaosheng Cui Signed-off-by: Damien Le Moal