summaryrefslogtreecommitdiffstats
path: root/Documentation/ABI/testing/sysfs-ata
diff options
context:
space:
mode:
authorTejun Heo <tj@kernel.org>2026-09-14 22:12:34 -1000
committerTejun Heo <tj@kernel.org>2026-09-15 06:57:20 -1000
commita9e3760b0838299649c0d57cca44daaf40ba3c33 (patch)
tree95807ff0e511c05602ef34808cfc660303309057 /Documentation/ABI/testing/sysfs-ata
downloadlinux-stable-a9e3760b0838299649c0d57cca44daaf40ba3c33.tar.gz
linux-stable-a9e3760b0838299649c0d57cca44daaf40ba3c33.zip
sched_ext: Maintain an online cid mask in the scheduler arenagrafted
Schedulers on the default cid mapping treat [0, nr_online_cids) as the online set and restart on hotplug. Schedulers that install their own mapping with scx_bpf_cid_override() have no way to learn which cids are online: the count no longer identifies members and the CPU-form cpumask is unusable from cid programs. This is an obvious hole in the cid API. Add scx_bpf_online_cmask(), a kernel-maintained cmask in the scheduler's arena, allocated alongside the per-CPU scratch masks and populated after the cid mapping is finalized and before ops.init(), for child schedulers too. The pointer stays valid through ops.exit() with no reference to take. It is the arena offset as a void pointer, the same form struct_ops arena arguments arrive in. The verifier types the void return as a scalar for the program's arena cast. The mask follows the SCX hotplug notifications: seeded from cpu_active_mask and updated before ops.cid_online/offline() runs, so it lags cpu_online_mask only inside a hotplug transition. Updates walk the scheduler list under the lock that also serializes unlinking. Reads are live, not atomic snapshots. Root initialization excludes hotplug. v2: Reworded the getter kerneldoc (Andrea Righi). Signed-off-by: Tejun Heo <tj@kernel.org> Reviewed-by: Andrea Righi <arighi@nvidia.com>
Diffstat (limited to 'Documentation/ABI/testing/sysfs-ata')
-rw-r--r--Documentation/ABI/testing/sysfs-ata140
1 files changed, 140 insertions, 0 deletions
diff --git a/Documentation/ABI/testing/sysfs-ata b/Documentation/ABI/testing/sysfs-ata
new file mode 100644
index 000000000..3daecac48
--- /dev/null
+++ b/Documentation/ABI/testing/sysfs-ata
@@ -0,0 +1,140 @@
+What: /sys/class/ata_*
+Description:
+ Provide a place in sysfs for storing the ATA topology of the
+ system. This allows retrieving various information about ATA
+ objects.
+
+Files under /sys/class/ata_port
+-------------------------------
+
+For each port, a directory ataX is created where X is the ata_port_id of the
+port. The device parent is the ata host device.
+
+
+What: /sys/class/ata_port/ataX/nr_pmp_links
+What: /sys/class/ata_port/ataX/idle_irq
+Date: May, 2010
+KernelVersion: v2.6.37
+Contact: Gwendal Grignou <gwendal@chromium.org>
+Description:
+ nr_pmp_links: (RO) If a SATA Port Multiplier (PM) is
+ connected, the number of links behind it.
+
+ idle_irq: (RO) Number of IRQ received by the port while
+ idle [some ata HBA only].
+
+
+What: /sys/class/ata_port/ataX/port_no
+Date: May, 2013
+KernelVersion: v3.11
+Contact: Gwendal Grignou <gwendal@chromium.org>
+Description:
+ (RO) Host local port number. While registering host controller,
+ port numbers are tracked based upon number of ports available on
+ the controller. This attribute is needed by udev for composing
+ persistent links in /dev/disk/by-path.
+
+Files under /sys/class/ata_link
+-------------------------------
+
+Behind each port, there is a ata_link. If there is a SATA PM in the topology, 15
+ata_link objects are created.
+
+If a link is behind a port, the directory name is linkX, where X is ata_port_id
+of the port. If a link is behind a PM, its name is linkX.Y where X is
+ata_port_id of the parent port and Y the PM port.
+
+
+What: /sys/class/ata_link/linkX[.Y]/hw_sata_spd_limit
+What: /sys/class/ata_link/linkX[.Y]/sata_spd_limit
+What: /sys/class/ata_link/linkX[.Y]/sata_spd
+Date: May, 2010
+KernelVersion: v2.6.37
+Contact: Gwendal Grignou <gwendal@chromium.org>
+Description:
+ hw_sata_spd_limit: (RO) Maximum speed supported by the
+ connected SATA device.
+
+ sata_spd_limit: (RO) Maximum speed imposed by libata.
+
+ sata_spd: (RO) Current speed of the link
+ eg. 1.5, 3 Gbps etc.
+
+
+Files under /sys/class/ata_device
+---------------------------------
+
+Behind each link, up to two ata devices are created.
+The name of the directory is devX[.Y].Z where:
+- X is ata_port_id of the port where the device is connected,
+- Y the port of the PM if any, and
+- Z the device id: for PATA, there is usually 2 devices [0,1], only 1 for SATA.
+
+
+What: /sys/class/ata_device/devX[.Y].Z/spdn_cnt
+What: /sys/class/ata_device/devX[.Y].Z/gscr
+What: /sys/class/ata_device/devX[.Y].Z/ering
+What: /sys/class/ata_device/devX[.Y].Z/id
+What: /sys/class/ata_device/devX[.Y].Z/pio_mode
+What: /sys/class/ata_device/devX[.Y].Z/xfer_mode
+What: /sys/class/ata_device/devX[.Y].Z/dma_mode
+What: /sys/class/ata_device/devX[.Y].Z/class
+Date: May, 2010
+KernelVersion: v2.6.37
+Contact: Gwendal Grignou <gwendal@chromium.org>
+Description:
+ spdn_cnt: (RO) Number of times libata decided to lower the
+ speed of link due to errors.
+
+ gscr: (RO) Cached result of the dump of PM GSCR
+ register. Valid registers are:
+
+ 0: SATA_PMP_GSCR_PROD_ID,
+ 1: SATA_PMP_GSCR_REV,
+ 2: SATA_PMP_GSCR_PORT_INFO,
+ 32: SATA_PMP_GSCR_ERROR,
+ 33: SATA_PMP_GSCR_ERROR_EN,
+ 64: SATA_PMP_GSCR_FEAT,
+ 96: SATA_PMP_GSCR_FEAT_EN,
+ 130: SATA_PMP_GSCR_SII_GPIO
+
+ Only valid if the device is a PM.
+
+ ering: (RO) Formatted output of the error ring of the
+ device.
+
+ id: (RO) Cached result of IDENTIFY command, as
+ described in ATA8 7.16 and 7.17. Only valid if
+ the device is not a PM.
+
+ pio_mode: (RO) PIO transfer mode used by the device.
+ Mostly used by PATA devices.
+
+ xfer_mode: (RO) Current transfer mode. Mostly used by
+ PATA devices.
+
+ dma_mode: (RO) DMA transfer mode used by the device.
+ Mostly used by PATA devices.
+
+ class: (RO) Device class. Can be "ata" for disk,
+ "atapi" for packet device, "pmp" for PM, or
+ "none" if no device was found behind the link.
+
+
+What: /sys/class/ata_device/devX[.Y].Z/trim
+Date: May, 2015
+KernelVersion: v4.10
+Contact: Gwendal Grignou <gwendal@chromium.org>
+Description:
+ (RO) Shows the DSM TRIM mode currently used by the device. Valid
+ values are:
+
+ unsupported: Drive does not support DSM TRIM
+
+ unqueued: Drive supports unqueued DSM TRIM only
+
+ queued: Drive supports queued DSM TRIM
+
+ forced_unqueued: Drive's queued DSM support is known to
+ be buggy and only unqueued TRIM commands
+ are sent