[sata] libata update

From: Jeff Garzik
Date: Fri Mar 26 2004 - 21:29:44 EST


More work on libata... human-readable summary:
* scsi portion of command submission path now a lot more lightweight. Should reduce CPU usage a bit.
* improved documentation (see below)
* New SiS SATA driver.
* Promise: minor improvements in locking, error handling, and initialization. Promise users, please test.
* Intel ICH5: minor improvements in probing and combined mode handling. Intel users, please test.

This is going to Linus after 2.6.5 is released.

BK repositories:
http://gkernel.bkbits.net/libata-2.[46]

Patches:
http://www.kernel.org/pub/linux/kernel/people/jgarzik/libata/2.4.25-libata15.patch.bz2
http://www.kernel.org/pub/linux/kernel/people/jgarzik/libata/2.6.5-rc2-bk6-libata2.patch.bz2

And also, there are some PDF docs generated from the source code. Although this is always available via "make pdfdocs", I also post this document at
http://www.kernel.org/pub/linux/kernel/people/jgarzik/libata/libata.pdf


I have attached the short and long changelogs for the above patches, versus upstream 2.6.x.




<uwe.koziolek:gmx.net>:
o [libata] Add driver for SiS 964/180 SATA

Jeff Garzik:
o [libata sata_promise] better locking and error handling
o [libata] more documentation
o [libata] more cmd queue path cleanups
o [libata] more command queue path cleanup
o [libata] clean up command queue/submit path a bit
o [libata sata_promise] minor initialization updates
o [libata] documentation, and a couple tiny cleanups
o [libata] use scsi host lock
o [libata] reduce diff with 2.4 libata backport
o [libata] pci_dma_error() was renamed to pci_dma_mapping_error()
o [ata] move some generic stuff linux/libata.h -> linux/ata.h
o [libata] consolidate data transfer mode handling
o [libata] set up some of the per-command data beforehand
o [libata sata_promise] check for PATA port on PDC20375
o [libata ata_piix] fix combined mode device detection
o [libata ata_piix] clean up combined mode handling
o [libata ata_piix] do not disable SATA port on module unload
o [libata] use kmap_atomic() rather than kmap()
o [libata] use new pci_dma_error() to check for pci_map_single() failure
o [libata sata_sis] minor cleanups

ChangeSet@xxxxxx, 2004-03-26 20:40:58-05:00, jgarzik@xxxxxxxxxx
[libata sata_promise] better locking and error handling

* Prefer spin_lock() to spin_lock_irq() in interrupt handler
* Reset each port, before probing the SATA phy
* Reset port when every time an error occurs

ChangeSet@xxxxxxxxxxx, 2004-03-26 18:02:34-05:00, jgarzik@xxxxxxxxxx
[libata] more documentation

libata-scsi.c should now be fully documented.

ChangeSet@xxxxxxxxxxx, 2004-03-26 04:04:17-05:00, jgarzik@xxxxxxxxxx
[libata] more cmd queue path cleanups

Final part in tonight's series of changes to clean up the
command queueing path.

The simulate-ATA-over-SCSI code is moved to a new function,
ata_scsi_simulate(), and ata_scsi_rw_queue() is renamed to
ata_scsi_translate(). With the SCSI CDB debugging dump function
also moved into its own function, the queue-command path is now
nice, compact, and readable.

ChangeSet@xxxxxxxxxxx, 2004-03-26 03:32:10-05:00, jgarzik@xxxxxxxxxx
[libata] more command queue path cleanup

A new helper ata_scsi_xlat_possible(), and the command queue path
gets a bit more compact.

As side effects we kill the 'cmd_size' argument from two functions,
and mark ata_scsi_rw_queue() as static, as its only needed
in libata-scsi.c.

ChangeSet@xxxxxxxxxxx, 2004-03-26 03:20:47-05:00, jgarzik@xxxxxxxxxx
[libata] clean up command queue/submit path a bit

This change is part of a series that compartmentalizes and consolidates
ATA taskfile submission.

Here, the device-location-related checks are moved out of the ->queuecommand()
hook and into an inline helper function.

ChangeSet@xxxxxxxxxx, 2004-03-26 01:13:20-05:00, jgarzik@xxxxxxxxxx
[libata sata_promise] minor initialization updates

* remove incorrect PATA port check
* enable undocumented bit 13 in flash control register,
because the Promise driver does so.
* wait 10 ms after setting TBG mode, for the same reason.

ChangeSet@xxxxxxxxxx, 2004-03-25 17:23:41-05:00, jgarzik@xxxxxxxxxx
[libata] documentation, and a couple tiny cleanups

Add more per-function source code documentation. Some of this stuff
is esoteric ATA crapola, and definitely needed to be documented.

Also, two tiny cleanups spotted while documenting:
* kill unused arg from internal function ata_dev_try_classify()
* kill unused return value from ata_dev_id_string()

ChangeSet@xxxxxxxxxx, 2004-03-25 14:40:17-05:00, jgarzik@xxxxxxxxxx
[libata] use scsi host lock

In 2.4 we release io_request_lock and take our own per-host lock,
in the ->queuecommand() hook. In 2.6, the SCSI layer provides a
useful to simply use the lock we already have, via scsi_assign_lock().

ChangeSet@xxxxxxxxxx, 2004-03-25 14:36:59-05:00, jgarzik@xxxxxxxxxx
[libata] reduce diff with 2.4 libata backport

ChangeSet@xxxxxxxxxx, 2004-03-25 14:36:27-05:00, jgarzik@xxxxxxxxxx
[libata] pci_dma_error() was renamed to pci_dma_mapping_error()

ChangeSet@xxxxxxxxxxxx, 2004-03-25 01:57:34-05:00, jgarzik@xxxxxxxxxx
[ata] move some generic stuff linux/libata.h -> linux/ata.h

struct ata_taskfile is generic, and so far its flags (ATA_TFLAG_xxx)

Also, move ATA_PROT_xxx definitions into their own enum.

ChangeSet@xxxxxxxxxxxx, 2004-03-25 01:44:08-05:00, jgarzik@xxxxxxxxxx
[libata] consolidate data transfer mode handling

The various ways you can send data to/from your ATA device is
known as the ATA taskfile protocol: PIO single sector, PIO
multiple sector, DMA, DMA TCQ, DMA NCQ, ...

Prior to this change, the data direction (read/write) was encoded
implicitly into the ATA_PROT_xxx value itself. This increased
complexity in some areas, and inhibited flexibility in others.

This change separates data direction from taskfile protocol, and also
moves the data direction flag (ATA_QCFLAG_WRITE) down to a lower
level (ATA_TFLAG_WRITE).

ChangeSet@xxxxxxxxxxxx, 2004-03-25 00:53:07-05:00, jgarzik@xxxxxxxxxx
[libata] set up some of the per-command data beforehand

The data transfer mode and the set of read/write commands we generate
during normal operation remains constant until we change the data
transfer mode.

This removes a series of branches in the read/write fast path,
and in general cleans up that particular spot of code.

ChangeSet@xxxxxxxxxxxx, 2004-03-24 23:50:34-05:00, jgarzik@xxxxxxxxxx
[libata sata_promise] check for PATA port on PDC20375

We don't handle it yet, but this prints out a message in its presence,
permitting verification of the check and informing users why their
PATA device is not recognized.

ChangeSet@xxxxxxxxxxxx, 2004-03-23 12:35:54-05:00, jgarzik@xxxxxxxxxx
[libata ata_piix] fix combined mode device detection

SATA port detection should not have assumed that a single SATA port
mapped to a single struct ata_port. Combined mode breaks this
assumption.

Change code to simply detect if one or more devices are present
on the struct ata_port, which is what we really wanted to do.

ChangeSet@xxxxxxxxxxxx, 2004-03-23 11:52:00-05:00, jgarzik@xxxxxxxxxx
[libata ata_piix] clean up combined mode handling

ChangeSet@xxxxxxxxxxxx, 2004-03-23 11:14:06-05:00, jgarzik@xxxxxxxxxx
[libata ata_piix] do not disable SATA port on module unload

We were disabling the SATA port, but not enabling it on module load.
So, modprobe+rmmod+modprobe would fail.

ChangeSet@xxxxxxxxxxxx, 2004-03-23 09:20:14-05:00, jgarzik@xxxxxxxxxx
[libata] use kmap_atomic() rather than kmap()

ChangeSet@xxxxxxxxxxxx, 2004-03-22 23:40:01-05:00, jgarzik@xxxxxxxxxx
[libata] use new pci_dma_error() to check for pci_map_single() failure

ChangeSet@xxxxxxxxxxx, 2004-03-21 12:15:16-05:00, jgarzik@xxxxxxxxxx
[libata sata_sis] minor cleanups

ChangeSet@xxxxxxxxxxx, 2004-03-21 11:55:35-05:00, uwe.koziolek@xxxxxxx
[libata] Add driver for SiS 964/180 SATA.