ChangeSet@1.1332, 2004-03-30 15:08:53-05:00, jgarzik@redhat.com [libata] hand-merge upstream IDE driver changes ChangeSet@1.1302.16.56, 2004-03-27 00:22:26-05:00, jgarzik@redhat.com [libata] remove synchronize-cache dead code Just a placeholder, and now that a better way to do this is coming, we don't even need the [unused] placeholder anymore. ChangeSet@1.1302.16.55, 2004-03-26 22:14:43-05:00, jgarzik@redhat.com [libata] fix bug with READ(6) and WRITE(6) SCSI commands Must set the LBA bit on all transfers, not just {read,write}{10,16} scsi commands. ChangeSet@1.1302.16.54, 2004-03-26 20:41:47-05:00, jgarzik@redhat.com [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@1.1302.16.53, 2004-03-26 18:03:10-05:00, jgarzik@redhat.com [libata] more documentation libata-scsi.c should now be fully documented. ChangeSet@1.1302.16.52, 2004-03-26 04:04:43-05:00, jgarzik@redhat.com [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@1.1302.16.51, 2004-03-26 03:35:13-05:00, jgarzik@redhat.com [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@1.1302.16.50, 2004-03-26 03:21:48-05:00, jgarzik@redhat.com [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@1.1302.16.49, 2004-03-26 01:14:07-05:00, jgarzik@redhat.com [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@1.1302.16.48, 2004-03-25 17:24:15-05:00, jgarzik@redhat.com [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@1.1302.16.47, 2004-03-25 14:45:38-05:00, jgarzik@redhat.com [libata sata_via] sync with 2.6 ChangeSet@1.1302.16.46, 2004-03-25 14:35:06-05:00, jgarzik@redhat.com [libata] reduce diff with 2.6.x upstream ChangeSet@1.1302.16.45, 2004-03-25 14:19:27-05:00, jgarzik@redhat.com [libata] remove 'broken' tag from silicon image 311x driver ChangeSet@1.1302.16.44, 2004-03-25 01:57:59-05:00, jgarzik@redhat.com [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@1.1302.16.43, 2004-03-25 01:45:52-05:00, jgarzik@redhat.com [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@1.1302.16.42, 2004-03-25 00:53:40-05:00, jgarzik@redhat.com [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@1.1302.16.41, 2004-03-24 23:51:23-05:00, jgarzik@redhat.com [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@1.1302.16.40, 2004-03-24 13:07:52-05:00, jgarzik@redhat.com [libata] Fix probing bug in VIA, SiS drivers Some 2.6 code leaked into the 2.4 backport. Spotted by Sergey Vlasov. ChangeSet@1.1302.16.39, 2004-03-23 12:36:39-05:00, jgarzik@redhat.com [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@1.1302.16.38, 2004-03-23 11:53:43-05:00, jgarzik@redhat.com [libata ata_piix] clean up combined mode handling ChangeSet@1.1302.16.37, 2004-03-23 11:17:00-05:00, jgarzik@redhat.com [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@1.1302.16.36, 2004-03-23 09:21:14-05:00, jgarzik@redhat.com [libata] use kmap_atomic() rather than kmap() ChangeSet@1.1302.16.35, 2004-03-21 12:26:33-05:00, jgarzik@redhat.com [libata] Add SiS 964/180 SATA driver Written by Uwe Koziolek. Minor cleanups and 2.4 backport from me. ChangeSet@1.1302.16.34, 2004-03-19 10:11:50-05:00, jgarzik@redhat.com [libata] increase max from UDMA/100 to UDMA/133 for some drivers ServerWorks and Vitesse driver transfer mode maximums were incorrectly too low. ChangeSet@1.1302.16.33, 2004-03-18 13:23:59-05:00, jgarzik@redhat.com [libata] remove unneeded linux/config.h includes ChangeSet@1.1302.16.32, 2004-03-18 06:29:00-05:00, jgarzik@redhat.com [libata sata_sil] add post-set-mode hook to libata, use it Silicon Image has a register that indicates the data transfer mode (pio-old, pio-new, mdma or udma) that must be initialized after the SET FEATURES - XFER command has been issued. This requires a hook in libata core to set the register at the right time. Also, limit to UDMA5 due to scary comments in FreeBSD about chip errata. ChangeSet@1.1302.16.31, 2004-03-18 05:37:48-05:00, jgarzik@redhat.com [libata] API cleanup Remove ->phy_config hook, everybody used pata_phy_config. Rename function to ata_set_mode, and directly call internally. ChangeSet@1.1302.16.30, 2004-03-18 05:37:18-05:00, jgarzik@redhat.com [libata sata_sil] port init cleanup, mask SATA phy interrupts * don't define a constant for each of four sets of register block offsets, use a table instead. * mask all SATA phy interrupt events, for each port ChangeSet@1.1302.16.29, 2004-03-18 05:36:45-05:00, jgarzik@redhat.com [libata] update dma start/stop path to only set/clear the bits we care about Legacy IDE software has traditionally unconditionally cleared the bits in the IDE BMDMA control register, but at least one controller has important bits in the 'reserved' area that we don't want to touch. So now, we only set/clear the data-direction and dma-start bits, and leave the rest alone. ChangeSet@1.1302.16.28, 2004-03-17 05:11:23-05:00, jgarzik@redhat.com [libata] sata_via] follow ata_piix style of scsi_register_module calling ChangeSet@1.1302.16.27, 2004-03-17 05:08:21-05:00, jgarzik@redhat.com [libata] Several updates to via driver * remove unneeded linux/config.h include * reset and test for device presence using SATA phy registers (SCRs) * add code to make sure that channels and interrupts we are interested in, are enabled * rewrite PCI probe routine (svia_init_one) ChangeSet@1.1302.16.26, 2004-03-16 21:53:52-05:00, jgarzik@redhat.com [libata] sync with 2.6... minor cleanups ChangeSet@1.1302.16.25, 2004-03-16 14:47:51-05:00, jgarzik@redhat.com [libata] bump versions in the core and driver sata_sil ChangeSet@1.1302.16.24, 2004-03-15 11:44:24-05:00, jgarzik@redhat.com [libata] kill the warning everybody grumbles about ChangeSet@1.1302.16.23, 2004-03-15 11:06:42-05:00, jgarzik@redhat.com [libata] disable clustering by default, compiler fix (ata_host_intr defn.) Disable SCSI's clustering of S/G entries, typically reducing overall S/G list size. Although ATA S/G tables are getting smarter (64-bit address, 32-bit length, no DMA boundary besides 4G itself), working harder to generate large byte sizes for each S/G size is not necessarily in our best interest: * SATA consists of frames, much like ethernet, called "FIS". A Data FIS consists of up to 2048 dwords (8K) of data. We don't want generate large byte sizes in S/G entries, just to have the host controller break them up again into smaller chunks. * Reduces the possibility that ata_fill_sg (libata-core.c) must split an S/G into two pieces, because it straddles a 64K boundary. * Reduces the possibility that FIS-related errata workarounds must be enabled. One such exists in the Silicon Image driver (sata_sil), which is currently worked around by limiting the max-sectors-per-request to 15. Note: the above text was from 2.6.x kernel. In 2.4.x version of libata, clustering has always been disabled. We merely document that here. ChangeSet@1.1302.16.22, 2004-03-14 23:33:54-05:00, jgarzik@redhat.com [libata sata_sil] remove incorrect limit on drive quirk The original submittor of the quirk code noted that the errata of a lockup creating when the condition "sector % 15 == 1" is true only on chip revisions <= 1. This is limitation on chip revision is apparently not correct, therefore we apply to all affected drives in the quirk list. This is a sledgehammer fix, limiting max-sectors to 15. A much better fix would split the hardware request into two taskfiles, issuing an additional command, thereby avoiding "sectors % 15 == 1" Given the errata is limited to "sil 311x + ", it is hoped that some janitor cares enough to do a real fix. ChangeSet@1.1302.16.21, 2004-03-14 23:23:17-05:00, jgarzik@redhat.com [libata] raise ATA_MAX_PRD from 128 to 256, now that LIBATA_MAX_PRD exists ChangeSet@1.1302.16.20, 2004-03-12 21:22:37-05:00, jeremy@sgi.com [PATCH] update for sata_vsc Hch reviewed my sata_vsc driver and had a few comments. If you could please apply these changes, I'd appreciate it. The copyright is probably the most important for me (apparently I goofed with the OSL. thanks jeremy ChangeSet@1.1302.16.19, 2004-03-10 21:37:42-05:00, jgarzik@redhat.com [libata] make set_{pio,udma}mode hooks optional Only one driver actually uses them. ChangeSet@1.1302.16.18, 2004-03-09 04:50:00-05:00, jeremy@sgi.com [libata] Add new driver for Vitesse VSC-7174. ChangeSet@1.1302.16.17, 2004-03-09 04:35:48-05:00, jeremy@sgi.com [libata] misc fixes, and an export * export ata_host_intr * fix debugging printk * the write of Device Control register, which occurs between SATA phy reset and the read of device signature, apparently clobbers some of the register values (namely Error register) that are needed to determine whether or not the device passed diagnostics. Make this write follow the device diag checking and classification, instead. ChangeSet@1.1302.16.16, 2004-03-09 04:27:36-05:00, jeremy@sgi.com [libata] Split up shared IO register locations into individual components Most ATA host controllers follow a standard layout for the ATA shadow registers, where command/status, error/feature, and devctl/altstatus share a single bus I/O address, because one register of each pair is read-only, and the other is write-only. On the Vitesse/Intel chip, all registers are given distinction bus I/O addresses, which necessitates changing the libata data structures to cope with this. This simply involves storing a few more bus addresses. ChangeSet@1.1302.16.15, 2004-03-08 22:34:10-05:00, jgarzik@redhat.com [libata sata_promise] provide proper SCSI completion function Promise driver uses a custom error handling function, so we need the fix that was applied to the libata core: the SCSI error handling thread requires that we complete commands using a special completion function, since the normal one doesn't work inside the error handling thread. ChangeSet@1.1302.16.14, 2004-03-08 22:30:31-05:00, jgarzik@redhat.com [libata ata_piix] Make sure annoying BIOSen don't disable our interrupts ChangeSet@1.1302.16.13, 2004-03-03 02:35:26-05:00, jgarzik@redhat.com [libata ata_piix] Add yet another Intel ICH6 PCI id ChangeSet@1.1302.16.12, 2004-03-03 00:18:04-05:00, go@turbolinux.co.jp [libata ata_piix] Fix transposed Intel ICH6 PCI ids ChangeSet@1.1302.16.11, 2004-02-28 17:09:53-05:00, jgarzik@redhat.com [libata] Use scsi_finish_command as completion function, in our error handling thread callback. This also exports scsi_finish_command in the SCSI layer. Thanks much to James Bottomley and his patience, as this solution was figured out. ChangeSet@1.1302.16.10, 2004-02-26 22:36:57-05:00, jgarzik@redhat.com [libata ata_piix] Update PCI quirk with new Intel SATA devices And also make a note in ata_piix.c so this is not forgotten in the future. ChangeSet@1.1302.16.9, 2004-02-26 21:46:19-05:00, jgarzik@redhat.com [libata ata_piix] add support for ICH6 ChangeSet@1.1302.16.8, 2004-02-26 21:32:02-05:00, jgarzik@redhat.com [libata] limit S/G table size to 128 entries Agreeing with Ben H's recommendation, I reduce the max-hw-segments number to 128, to match max-phys-segments number. This shouldn't affect performance because 128*64K is far above the max transfer size for most current IDE devices. Even 128*4K is OK. ChangeSet@1.1302.16.7, 2004-02-25 22:43:33-05:00, jgarzik@redhat.com [libata] Much better s/g table fill routine My version wouldn't work for really large iommu merges. Contributed by Benjamin Herrenschmidt, with a tiny optimization from me. ChangeSet@1.1302.16.6, 2004-02-25 18:28:40-05:00, jgarzik@redhat.com [libata] bump versions ChangeSet@1.1302.16.5, 2004-02-25 18:26:47-05:00, dmilburn@redhat.com [libata sata_promise] Fix DIMM initialization on PCI-X bus ChangeSet@1.1302.16.4, 2004-02-25 18:19:26-05:00, jgarzik@redhat.com [libata] Limit s/g entries to 128, to ensure splitting is ok. ChangeSet@1.1302.16.3, 2004-02-25 16:55:12-05:00, jgarzik@redhat.com [libata sata_sil] fix 4-port support on SII 3114 ChangeSet@1.1302.16.2, 2004-02-25 16:53:55-05:00, jgarzik@redhat.com [libata] Guarantee that S/G entries do not straddle IDE DMA 64k boundary ChangeSet@1.1302.16.1, 2004-02-18 14:45:02-05:00, jgarzik@redhat.com Merge redhat.com:/spare/repo/linux-2.4.25 into redhat.com:/spare/repo/libata-2.4 ChangeSet@1.1302.7.3, 2004-02-14 13:52:32-05:00, jgarzik@redhat.com Cset exclude: jgarzik@redhat.com|ChangeSet|20040213173558|53502 ChangeSet@1.1302.7.2, 2004-02-13 13:06:41-05:00, jgarzik@redhat.com Bump libata, ata_piix to version 1.0. Also update copyrights for 2004. ChangeSet@1.1302.7.1, 2004-02-13 12:45:53-05:00, jgarzik@redhat.com Merge redhat.com:/spare/repo/linux-2.4 into redhat.com:/spare/repo/libata-2.4 ChangeSet@1.1267.40.8, 2004-02-13 12:35:58-05:00, jgarzik@redhat.com [libata] catch, and ack, spurious DMA interrupts Hardware issue on Intel ICH5 requires an additional ack sequence over and above the normal IDE DMA interrupt ack requirements. Issue described in post to freebsd list: http://www.mail-archive.com/freebsd-stable@freebsd.org/msg58421.html Since the bug workaround only requires a single additional PIO or MMIO read in the interrupt handler, it is applied to all chipsets using the standard libata interrupt handler. Credit for research the issue, creating the patch, and testing the patch all go to Jon Burgess. ChangeSet@1.1267.40.7, 2004-01-14 19:00:34-05:00, jgarzik@redhat.com Merge redhat.com:/spare/repo/linux-2.4 into redhat.com:/spare/repo/libata-2.4 ChangeSet@1.1267.33.6, 2004-01-14 18:49:11-05:00, benh@kernel.crashing.org [libata sata_svw] cleanup, better probing * use fewer magic numbers * probe all 4 ports, using standard SATA SCRs * limit udma mask to 0x3f * clean up PPC-specific procfs stuff ChangeSet@1.1267.33.5, 2004-01-14 18:35:56-05:00, arubin@atl.lmco.com [libata sata_sil] add pci id for Silicon Image 3512 ChangeSet@1.1267.33.4, 2004-01-14 18:21:32-05:00, normalperson@yhbt.net [libata sata_sil] cleaner, better version of errata workarounds No longer unfairly punishes non-errata Seagate and Maxtor drives. ChangeSet@1.1267.33.3, 2004-01-06 04:27:04-05:00, marchand@kde.org [libata sata_sil] add support for adaptec 1210sa, 4-port sii 3114 ChangeSet@1.1267.33.2, 2004-01-06 04:25:00-05:00, jgarzik@redhat.com [libata sata_svr] fix DRV_NAME to reflect actual driver filename ChangeSet@1.1267.33.1, 2004-01-06 03:43:20-05:00, jgarzik@redhat.com Merge redhat.com:/spare/repo/linux-2.4 into redhat.com:/spare/repo/libata-2.4 ChangeSet@1.1136.1.62, 2003-12-30 19:48:59-05:00, jgarzik@redhat.com [libata sata_sil] unmask interrupts during initialization Prudent in general, and needed for Adaptec BIOSes. ChangeSet@1.1136.1.61, 2003-12-19 20:50:14-05:00, jgarzik@redhat.com [libata] s/Scsi_Cmnd/struct scsi_cmnd/ ChangeSet@1.1136.1.60, 2003-12-19 20:48:03-05:00, jgarzik@redhat.com [libata] some cleanups suggested by Christoph * s/Scsi_Cmnd/struct scsi_cmnd/ * remove incorrect FIXME comments related to checking return values of certain SCSI mid layer functions. ChangeSet@1.1136.1.59, 2003-12-17 17:34:25-05:00, jgarzik@redhat.com [libata promise] fix another ugly bug For the SX4, only one Host DMA (local DIMM) engine is on the hardware, while there is an ATA engine for each SATA port. This means that Host DMA transactions must be queued. When previously fixing this problem (the driver had previously assumed an HDMA engine per port), I stored the HDMA packet queue in a per-port data structure. This was incorrect: this patch changes it to correctly use a per-host data structure, not a per-port structure. ChangeSet@1.1136.1.58, 2003-12-16 19:13:45-05:00, jgarzik@redhat.com [libata] move geometry code to libata-scsi ChangeSet@1.1136.1.57, 2003-12-16 19:07:04-05:00, Michael_E_Brown@Dell.com [libata] fake geometry for partition tables / setups that need such ChangeSet@1.1136.1.56, 2003-12-05 13:39:00-05:00, jgarzik@redhat.com [libata promise] Properly initialize DIMM, on SX4 On-board DIMM should be sized and initialized by the driver. Previously, a single DIMM size was simply (and incorrectly) assumed, and initialization was presumed to have been done by the card's BIOS. Contributed by Promise, updated by David Milburn @ Red Hat. ChangeSet@1.1136.1.55, 2003-12-05 13:10:14-05:00, jgarzik@redhat.com [libata] fix use-after-free Fixes oops on unload that some were seeing. Caught by Jon Burgess. ChangeSet@1.1136.1.54, 2003-11-30 21:43:11-05:00, jgarzik@redhat.com Merge redhat.com:/spare/repo/linux-2.4 into redhat.com:/spare/repo/libata-2.4 ChangeSet@1.1136.1.53, 2003-11-30 11:55:36-05:00, grundler@parisc-linux.org [libata] use sg_dma_xxx macros Fixes build on some platforms, fixes issues on others. ChangeSet@1.1136.1.52, 2003-11-26 16:13:20-05:00, jgarzik@redhat.com [libata] Fix PDC20621: we only have one Host DMA engine, not one per port Whoops. So, we need to queue HDMA transactions internally. ChangeSet@1.1136.1.51, 2003-11-24 11:27:22-05:00, jgarzik@redhat.com [libata] bump versions for core and serverworks driver ChangeSet@1.1136.1.50, 2003-11-24 11:20:24-05:00, benh@kernel.crashing.org [libata] Fix flush of Device Control register to device Also add some ServerWorks-specific tweaks. ChangeSet@1.1136.1.49, 2003-11-24 10:32:33-05:00, jgarzik@redhat.com Merge redhat.com:/spare/repo/linux-2.4 into redhat.com:/spare/repo/libata-2.4 ChangeSet@1.1136.1.48, 2003-11-22 14:26:38-05:00, jgarzik@redhat.com Merge redhat.com:/spare/repo/linux-2.4 into redhat.com:/spare/repo/libata-2.4 ChangeSet@1.1136.48.32, 2003-11-14 19:52:14-05:00, pp@ee.oulu.fi [libata] add Promise PCI id ChangeSet@1.1136.48.31, 2003-11-10 10:42:24-05:00, jgarzik@redhat.com [libata promise] fixes suggested by Promise * flush host FIFO after sending data to DIMM window * don't set SCR addresses, as the hardware doesn't have SCRs (cosmetic) ChangeSet@1.1136.48.30, 2003-11-07 16:25:05-05:00, jgarzik@redhat.com [libata] fix Promise PCI posting bugs ChangeSet@1.1136.48.29, 2003-11-06 13:38:27-05:00, jgarzik@redhat.com Merge redhat.com:/spare/repo/linux-2.4 into redhat.com:/spare/repo/libata-2.4 ChangeSet@1.1136.7.67, 2003-11-06 13:35:56-05:00, jgarzik@redhat.com [libata] bump libata version ChangeSet@1.1136.7.66, 2003-11-06 13:31:37-05:00, jgarzik@redhat.com [libata] fix ugly Promise interrupt masking bug ChangeSet@1.1136.7.65, 2003-11-06 13:29:24-05:00, jgarzik@redhat.com [libata] cosmetic change to sync with 2.5 libata ChangeSet@1.1136.7.64, 2003-10-31 09:56:23-05:00, jgarzik@redhat.com [libata] add Promise SATA pci id ChangeSet@1.1136.7.63, 2003-10-31 09:55:23-05:00, jgarzik@redhat.com [libata] fix bugs in SATA reset code path ChangeSet@1.1136.7.62, 2003-10-28 11:04:47-05:00, jgarzik@redhat.com [libata] Add paranoia checks/settings suggested by Promise ChangeSet@1.1136.7.61, 2003-10-23 19:27:23-04:00, jgarzik@redhat.com [libata] PDC20621 hdma fixes ChangeSet@1.1136.7.60, 2003-10-22 22:55:17-04:00, jgarzik@redhat.com Merge redhat.com:/spare/repo/linux-2.4 into redhat.com:/spare/repo/libata-2.4 ChangeSet@1.1136.7.59, 2003-10-22 22:27:09-04:00, jgarzik@redhat.com [libata] fix Promise build on older compilers ChangeSet@1.1136.7.58, 2003-10-22 22:26:06-04:00, jgarzik@redhat.com [libata] kill timer when thread dies Noticed and fixed by David Woodhouse. ChangeSet@1.1136.7.57, 2003-10-22 18:02:31-04:00, jgarzik@redhat.com [libata] more pdc20621 work ChangeSet@1.1136.7.56, 2003-10-22 12:34:12-04:00, jgarzik@redhat.com [libata] fill in a lot more Promise PDC20621 support ChangeSet@1.1136.7.55, 2003-10-22 12:32:38-04:00, jgarzik@redhat.com [libata] add ->host_stop hook, and copy ->private_data from probe_ent ChangeSet@1.1136.7.54, 2003-10-22 01:02:37-04:00, jgarzik@redhat.com [libata] convert Promise to packetized DMA ChangeSet@1.1136.7.53, 2003-10-21 23:15:17-04:00, jgarzik@redhat.com [libata] add per-driver port init/shutdown hooks, with helper defaults ChangeSet@1.1136.7.52, 2003-10-21 22:39:06-04:00, jgarzik@redhat.com [libata] only reset if ATA_FLAG_SATA_RESET is present ChangeSet@1.1136.7.51, 2003-10-21 22:19:44-04:00, jgarzik@redhat.com [libata] move linux/ata.h to linux/libata.h ChangeSet@1.1136.7.50, 2003-10-20 20:56:33-04:00, jgarzik@redhat.com Merge redhat.com:/garz/repo/linux-2.4 into redhat.com:/garz/repo/libata-2.4 ChangeSet@1.1136.7.49, 2003-10-17 16:26:49-04:00, jgarzik@redhat.com [libata] bump promise version ChangeSet@1.1136.7.48, 2003-10-17 15:51:41-04:00, jgarzik@redhat.com [libata] PDC20621-specific register addresses ChangeSet@1.1136.7.47, 2003-10-17 14:06:04-04:00, jgarzik@redhat.com [libata] add PDC pci id, additional DMA timeout handling ChangeSet@1.1136.7.46, 2003-10-17 12:52:40-04:00, jgarzik@redhat.com [libata] fix PDC20621 chip offset ChangeSet@1.1136.7.45, 2003-10-17 12:19:17-04:00, jgarzik@redhat.com [libata] abstract out PDC20621, it's quite different ChangeSet@1.1136.7.44, 2003-10-17 12:01:15-04:00, jgarzik@redhat.com [libata] fix SATA phy reset ChangeSet@1.1136.7.43, 2003-10-17 11:59:22-04:00, jgarzik@redhat.com [libata] add ATA_FLAG_SATA_RESET, to support SATA bus reset ChangeSet@1.1136.7.42, 2003-10-17 08:53:24-04:00, jgarzik@redhat.com Merge http://gkernel.bkbits.net/libata-2.4 into redhat.com:/garz/repo/libata-2.4 ChangeSet@1.1136.32.1, 2003-10-14 11:18:20-04:00, jgarzik@redhat.com Merge redhat.com:/garz/repo/linux-2.4 into redhat.com:/garz/repo/libata-2.4 ChangeSet@1.1136.7.41, 2003-10-13 16:28:19-04:00, jgarzik@redhat.com [libata] bump promise version ChangeSet@1.1136.7.40, 2003-10-13 16:26:24-04:00, jgarzik@redhat.com [libata] initial 20621 support ChangeSet@1.1136.7.39, 2003-10-10 21:28:20-04:00, jgarzik@redhat.com [libata] fix another typo in Promise SATA; bump version ChangeSet@1.1136.7.38, 2003-10-10 19:20:31-04:00, jgarzik@redhat.com [libata] add another Promise PCI id ChangeSet@1.1136.7.37, 2003-10-10 18:52:44-04:00, jgarzik@redhat.com Merge kernel.bkbits.net:libata-2.4 into redhat.com:/garz/repo/libata-2.4 ChangeSet@1.1136.10.7, 2003-10-10 18:48:46-04:00, jgarzik@redhat.com [libata] use PCI_VENDOR_ID_PROMISE constant from pci_ids.h ChangeSet@1.1136.10.6, 2003-10-10 17:21:32-04:00, jgarzik@redhat.com [libata] fix typos that prevents 4-port Promise support from working ChangeSet@1.1136.10.5, 2003-10-10 17:08:58-04:00, jgarzik@redhat.com [libata] add Promise SATA support ChangeSet@1.1136.7.36, 2003-10-10 14:54:15-04:00, jgarzik@redhat.com Merge redhat.com:/garz/repo/linux-2.4 into redhat.com:/garz/repo/libata-2.4 ChangeSet@1.1136.23.1, 2003-10-10 12:20:13-04:00, jgarzik@redhat.com Merge ChangeSet@1.1136.10.4, 2003-10-10 05:17:54-04:00, jgarzik@redhat.com [libata] ->eng_timeout hook, ata_qc_complete export, SATA phy reset * Add ->eng_timeout hook, to allow host controllers to replace the standard one with their own. * export ata_qc_complete for use in low-level drivers. * In standard helper sata_phy_reset, actually reset the phy and disable partial/slumber PM modes. Then wait for a device to appear. ChangeSet@1.1136.10.3, 2003-10-05 18:52:33-04:00, jgarzik@redhat.com [libata] bump lib version ChangeSet@1.1136.10.2, 2003-10-05 18:40:42-04:00, jgarzik@redhat.com [libata] use request_region for legacy mode; better integration w/ IDE driver ChangeSet@1.1136.10.1, 2003-10-05 17:19:38-04:00, jgarzik@redhat.com Merge kernel.bkbits.net:libata-2.4 into redhat.com:/garz/repo/libata-2.4 ChangeSet@1.1120.24.9, 2003-09-26 21:44:03-04:00, jgarzik@redhat.com [libata] docbook docs fixes for newly-renamed drivers ChangeSet@1.1120.24.8, 2003-09-26 21:38:17-04:00, jgarzik@redhat.com [libata] bump versions ChangeSet@1.1120.24.7, 2003-09-26 21:36:18-04:00, jgarzik@redhat.com [libata] fix dependencies in Config.in ChangeSet@1.1120.24.6, 2003-09-26 19:59:28-04:00, jgarzik@redhat.com [libata] rename config option s/ATA/SATA/. remove pata, atapi config options. Also, rename drivers: sata_sil, sata_svw, sata_via ChangeSet@1.1120.24.5, 2003-09-26 19:27:23-04:00, jgarzik@redhat.com [libata] break post-IDENTIFY step out into ->phy_config hook. ChangeSet@1.1120.24.4, 2003-09-26 19:08:18-04:00, jgarzik@redhat.com [libata] switch K2, siimage to using ATA software reset ChangeSet@1.1120.24.3, 2003-09-26 18:52:24-04:00, jgarzik@redhat.com [libata] abstract out pre-IDENTIFY step into ->phy_reset hook ChangeSet@1.1120.24.2, 2003-09-26 18:33:17-04:00, jgarzik@redhat.com [libata] move force-pio, config-dma thread states to ata_port_reset() ChangeSet@1.1120.24.1, 2003-09-26 17:44:24-04:00, jgarzik@redhat.com Merge redhat.com:/garz/repo/linux-2.4 into redhat.com:/garz/repo/libata-2.4 ChangeSet@1.1120.18.6, 2003-09-26 17:33:54-04:00, jgarzik@redhat.com [libata] Simplify probe phase. Eliminate needless successive switching through thread states, and consolidate the port reset/probe code into a new function, ata_port_reset. This also begins to lay the groundwork for the upcoming pre- and post-IDENTIFY hooks ->phy_reset and ->phy_config. ChangeSet@1.1120.18.5, 2003-09-26 00:34:16-04:00, jgarzik@redhat.com [libata] create ->scr_{read,write} hooks, use them to implement generic phy device presence detection. ChangeSet@1.1120.18.4, 2003-09-25 23:39:46-04:00, jgarzik@redhat.com [libata] store mmio_base in host_set; iounmap() in libata core. Now that the MMIO base, that is mapped by the low-level driver at startup, is stored in struct ata_host_set, we can call iounmap() on the proper value in ata_pci_remove_one(). This allows the removal of driver-specific k2_sata_remove_one and sil_remove_one, and kills much duplicate code. ChangeSet@1.1120.18.3, 2003-09-24 22:22:27-04:00, jgarzik@redhat.com [libata] new ->fill_sg hook ChangeSet@1.1120.18.2, 2003-09-24 20:13:12-04:00, jgarzik@redhat.com [libata] make it easy for low-level drivers to provide their own irq handlers ChangeSet@1.1120.18.1, 2003-09-24 18:58:19-04:00, jgarzik@redhat.com Merge redhat.com:/garz/repo/linux-2.4 into redhat.com:/garz/repo/libata-2.4 ChangeSet@1.1120.8.13, 2003-09-24 18:40:43-04:00, jgarzik@redhat.com [libata] add exec_command, check_status hooks Contributed by Benjamin Herrenschmidt. ChangeSet@1.1120.8.12, 2003-09-23 23:51:25-04:00, jgarzik@redhat.com [libata] add a 'thanks' chapter ChangeSet@1.1120.8.11, 2003-09-21 19:27:56-04:00, jgarzik@redhat.com [libata] bump version to libata 0.72 ChangeSet@1.1120.8.10, 2003-09-21 19:14:30-04:00, jgarzik@redhat.com [libata] support post-IDENTIFY errata fixups ChangeSet@1.1120.8.9, 2003-09-21 02:00:55-04:00, jgarzik@redhat.com [libata] mark various things CONFIG_EXPERIMENTAL or (BROKEN) ChangeSet@1.1120.8.8, 2003-09-21 01:23:46-04:00, jgarzik@redhat.com [libata] add support for Silicon Image 3112 SATA ChangeSet@1.1120.8.7, 2003-09-20 21:02:43-04:00, jgarzik@redhat.com [libata 2.4] update docbook makefile to reference new libata-*.c filenames ChangeSet@1.1120.8.6, 2003-09-20 21:02:00-04:00, jgarzik@redhat.com [libata] remove mentions of "ata-scsi" from docbook docs ChangeSet@1.1120.8.5, 2003-09-20 20:51:47-04:00, jgarzik@redhat.com [libata] correct filenames embedded within the DocBook docs ChangeSet@1.1120.8.4, 2003-09-20 20:01:48-04:00, jgarzik@redhat.com [libata] minor changes; sync with 2.5 libata. ChangeSet@1.1120.8.3, 2003-09-20 19:25:51-04:00, jgarzik@redhat.com [libata] Add driver for Broadcom (Apple K2) SATA controller Contributed by Benjamin Herrenschmidt ChangeSet@1.1120.8.2, 2003-09-20 19:03:02-04:00, jgarzik@redhat.com [libata] fix major bug in master/slave operation Pass proper device (0/1) to taskfile init at all times. Also, remove ap->devsel, as it's no longer needed. This allows additional simplification of __ata_dev_select. Issue isolated by Dell. Patch by Matt Domsch, with additional simplifications by me. ChangeSet@1.1120.8.1, 2003-09-20 18:31:20-04:00, jgarzik@redhat.com Merge redhat.com:/garz/repo/linux-2.4 into redhat.com:/garz/repo/libata-2.4 ChangeSet@1.1109.2.24, 2003-09-20 18:27:30-04:00, jgarzik@redhat.com [libata] support hardware with non-standard ATA command register spacing. Don't assume that one ATA command register follows another, in the ATA command block. Rough cut from the PPC guys, polished up by me. ChangeSet@1.1109.2.23, 2003-09-20 17:21:51-04:00, jgarzik@redhat.com [libata] minor fixes from the PPC guys ChangeSet@1.1109.2.22, 2003-09-20 16:54:38-04:00, jgarzik@redhat.com [libata] mark exported symbols EXPORT_SYMBOL_GPL As the comment indicates, this is mainly to emphasize that libata is tightly tied to the low-level drivers it supports. The in-kernel libata API/ABI is likely to continue changing for the short-term. ChangeSet@1.1109.2.21, 2003-09-14 22:29:50-04:00, jgarzik@redhat.com Merge kernel.bkbits.net:libata-2.4 into redhat.com:/garz/repo/libata-2.4 ChangeSet@1.1109.2.20, 2003-09-12 22:05:33-04:00, jgarzik@redhat.com Merge redhat.com:/garz/repo/linux-2.4 into redhat.com:/garz/repo/libata-2.4 ChangeSet@1.1063.2.41, 2003-09-02 06:10:16-04:00, jgarzik@redhat.com [libata] Now that ATA software reset (SRST) is reliable, make it default ChangeSet@1.1063.2.40, 2003-09-02 05:22:15-04:00, jgarzik@redhat.com Merge redhat.com:/garz/repo/linux-2.4 into redhat.com:/garz/repo/libata-2.4 ChangeSet@1.1063.2.39, 2003-08-31 19:30:52-04:00, jgarzik@redhat.com Merge redhat.com:/garz/repo/marcelo-2.4 into redhat.com:/garz/repo/atascsi-2.4 ChangeSet@1.1063.2.38, 2003-08-30 20:17:14-04:00, jgarzik@redhat.com [libata] mark pci_bits initializers with U and UL suffixes ChangeSet@1.1063.2.37, 2003-08-30 20:16:41-04:00, jgarzik@redhat.com [libata 2.4] correct ordering of struct pci_bits members ChangeSet@1.1063.2.36, 2003-08-30 17:41:20-04:00, jgarzik@redhat.com Merge redhat.com:/garz/repo/marcelo-2.4 into redhat.com:/garz/repo/atascsi-2.4 ChangeSet@1.1063.35.2, 2003-08-30 14:04:18-04:00, jgarzik@redhat.com [libata] remove last bits of 2.5 compat code from linux/ata.h Marcelo merged these SCSI opcode definitions into scsi/scsi.h. ChangeSet@1.1063.35.1, 2003-08-30 13:56:50-04:00, jgarzik@redhat.com Merge redhat.com:/garz/repo/marcelo-2.4 into redhat.com:/garz/repo/atascsi-2.4 ChangeSet@1.1063.31.18, 2003-08-30 04:43:59-04:00, jgarzik@redhat.com Ignore drivers/char/defkeymap.c. ChangeSet@1.1063.31.17, 2003-08-30 04:40:21-04:00, jgarzik@redhat.com Ignore auto-generated drivers/char/defkeymap.c. ChangeSet@1.1063.31.16, 2003-08-29 22:57:25-04:00, jgarzik@redhat.com [libata] store s/g ptr in ata_queued_cmd; abstract out more scsi bits ChangeSet@1.1063.31.15, 2003-08-29 22:03:24-04:00, jgarzik@redhat.com ignore some files ChangeSet@1.1063.31.14, 2003-08-29 21:52:37-04:00, jgarzik@redhat.com [libata] rename atascsi-dev docbook to libata ChangeSet@1.1063.31.13, 2003-08-29 21:47:56-04:00, jgarzik@redhat.com [libata] pci_bits for via, docbook for via, pci_bits un-C99 for piix ChangeSet@1.1063.31.12, 2003-08-29 19:57:36-04:00, jgarzik@redhat.com [libata] s/host/port/ ChangeSet@1.1063.31.11, 2003-08-29 19:18:25-04:00, jgarzik@redhat.com [libata] s/sata_via/ata_via/ ChangeSet@1.1063.31.10, 2003-08-29 19:14:49-04:00, jgarzik@redhat.com [libata] s/board/port_info/ in several places ChangeSet@1.1063.31.9, 2003-08-29 18:48:28-04:00, jgarzik@redhat.com [libata] rename *host_info to *port_op[eration]s* ChangeSet@1.1063.31.8, 2003-08-29 14:27:36-04:00, jgarzik@redhat.com [libata] tone down printk severity of a few messages ChangeSet@1.1063.31.7, 2003-08-29 14:13:29-04:00, jgarzik@redhat.com [libata] better debugging output in ata_dev_identify ChangeSet@1.1063.31.6, 2003-08-29 13:54:13-04:00, jgarzik@redhat.com [libata] check BIOS "port enabled" bit, in piix ->port_probe hooks ChangeSet@1.1063.31.5, 2003-08-29 02:06:22-04:00, jgarzik@redhat.com [libata] add VIA SATA driver ChangeSet@1.1063.31.4, 2003-08-29 00:28:15-04:00, jgarzik@redhat.com [libata] ATA software reset, ata_busy_sleep updates * First check-status loop maximum iterations increased from 3 to 300, to be optimistic / aggressive. * Remove ATA_NIEN bit during ATA software reset. * Write standard Device Control register, prior to writing the value or'd with ATA_SRST. This ensures a known value in the Device Control register, prior to software reset. * Be sure to pause the minimum amount of time, before checking status (per spec). ChangeSet@1.1063.31.3, 2003-08-28 22:07:28-04:00, jgarzik@redhat.com [libata] remove scsi_[un]register_host() compat, was only used in 2.4 ChangeSet@1.1063.31.2, 2003-08-28 22:00:04-04:00, jgarzik@redhat.com [libata] remove unneeded 2.6 compat irqreturn_t definitions ChangeSet@1.1063.31.1, 2003-08-28 21:55:21-04:00, jgarzik@redhat.com Merge redhat.com:/garz/repo/marcelo-2.4 into redhat.com:/garz/repo/atascsi-2.4 ChangeSet@1.1063.20.8, 2003-08-28 21:54:13-04:00, jgarzik@redhat.com [libata] bump libata and ata_piix versions ChangeSet@1.1063.20.7, 2003-08-28 16:11:24-04:00, jgarzik@redhat.com [libata] remove a couple scsi mentions, in libata-core.c ChangeSet@1.1063.20.6, 2003-08-27 21:39:22-04:00, jgarzik@redhat.com [libata] move ata_ and atapi_scsi_queuecmd to libata-scsi.c ChangeSet@1.1063.20.5, 2003-08-27 21:29:38-04:00, jgarzik@redhat.com [libata] add copyright and legal gunk to libata.h ChangeSet@1.1063.20.4, 2003-08-27 21:22:25-04:00, jgarzik@redhat.com [libata] move some more functions to libata-scsi.c ChangeSet@1.1063.20.3, 2003-08-27 20:58:24-04:00, jgarzik@redhat.com [libata] create libata-scsi.c, move *_scsiop_* there ChangeSet@1.1063.20.2, 2003-08-27 20:33:46-04:00, jgarzik@redhat.com [libata] make libata.o a multi-part object libata.c is renamed to libata-core.c. ChangeSet@1.1063.20.1, 2003-08-25 20:34:50-04:00, jgarzik@redhat.com Merge redhat.com:/garz/repo/marcelo-2.4.22 into redhat.com:/garz/repo/atascsi-2.4 ChangeSet@1.930.151.69, 2003-08-16 23:31:29-04:00, jgarzik@redhat.com [libata] turn some enum constants into #defines (sync with 2.6 libata) ChangeSet@1.930.151.68, 2003-08-16 22:13:49-04:00, jgarzik@redhat.com [libata] renaming and moving around internal funcs; no code changes. ChangeSet@1.930.151.67, 2003-08-16 21:36:10-04:00, jgarzik@redhat.com [libata] eliminate bus_state in favor of tf->protocol Also misc. minor cleanups. ChangeSet@1.930.151.66, 2003-08-16 14:36:30-04:00, jgarzik@redhat.com [libata] more docbook docs ChangeSet@1.930.151.65, 2003-08-16 13:48:30-04:00, jgarzik@redhat.com [libata] add more docbook documentation ChangeSet@1.930.151.64, 2003-08-16 11:19:58-04:00, jgarzik@redhat.com [libata] fix spinlock bug; finish mmio-iziation except for {in,out}sl ChangeSet@1.930.151.63, 2003-08-16 10:39:09-04:00, jgarzik@redhat.com [libata] mmio-ize dma_start; dma start/complete simplifications ChangeSet@1.930.151.62, 2003-08-16 09:47:39-04:00, jgarzik@redhat.com [libata] add ata_tag_valid helper; use it. ChangeSet@1.930.151.61, 2003-08-15 22:10:21-04:00, jgarzik@redhat.com [libata] QCFLAG cleanup, beginnings of ATA protocol recognition ChangeSet@1.930.151.60, 2003-08-15 01:24:31-04:00, jgarzik@redhat.com [libata] delete dead code, sort struct ata_taskfile by usage in hot path ChangeSet@1.930.151.59, 2003-08-15 01:08:24-04:00, jgarzik@redhat.com [libata] remove ap->dmactl member, mmio-ize ata_dma_complete ChangeSet@1.930.151.58, 2003-08-14 11:53:47-04:00, jgarzik@redhat.com [libata] more mmio work ChangeSet@1.930.151.57, 2003-08-14 02:21:34-04:00, jgarzik@redhat.com [libata] begin work on cable detect ChangeSet@1.930.151.56, 2003-08-14 01:06:36-04:00, jgarzik@redhat.com [atascsi] remove __devinit and __devinitdata ChangeSet@1.930.151.55, 2003-08-14 01:02:16-04:00, jgarzik@redhat.com [atascsi] support mmio as well as pio for many operations ChangeSet@1.930.151.54, 2003-08-13 23:47:59-04:00, jgarzik@redhat.com [atascsi] support no-legacy use-ata-srst host flags ChangeSet@1.930.151.53, 2003-08-13 17:53:48-04:00, jgarzik@redhat.com [atascsi] remove redundant args in internal functions ChangeSet@1.930.151.52, 2003-08-12 22:57:17-04:00, jgarzik@redhat.com [atascsi] bump versions ChangeSet@1.930.151.51, 2003-08-12 22:52:22-04:00, jgarzik@redhat.com [atascsi] fix lba48 support, and printk 'lba48' when we see it ChangeSet@1.930.151.50, 2003-08-12 21:16:42-04:00, jgarzik@redhat.com [atascsi] remove ATA_TFLAG_PIO (redundant), move more code into ata_qc_issue ChangeSet@1.930.151.49, 2003-08-12 19:45:16-04:00, jgarzik@redhat.com [atascsi] add ata_qc_issue, and cleanups associated with that ChangeSet@1.930.151.48, 2003-08-12 18:30:10-04:00, jgarzik@redhat.com [atascsi] add ata_eng_schedule, ata_qc_push, other smallish items ChangeSet@1.930.151.47, 2003-08-12 17:41:06-04:00, jgarzik@redhat.com [atascsi] clean up ata_qc_complete, add qc callbacks ChangeSet@1.930.151.46, 2003-08-12 02:04:01-04:00, jgarzik@redhat.com [atascsi] abstract out taskfile load ChangeSet@1.930.151.45, 2003-08-12 01:36:57-04:00, jgarzik@redhat.com [atascsi] support the hideous "combined mode" ChangeSet@1.930.151.44, 2003-08-08 22:38:55-04:00, jgarzik@sata.normnet.org [atascsi] import cosmetic changes from 2.6 Also remove a useless debugging printk. ChangeSet@1.930.151.43, 2003-08-08 20:53:20-04:00, jgarzik@sata.normnet.org [atascsi] obtain host_set lock during scsi layer queuecommand ChangeSet@1.930.151.42, 2003-08-03 18:02:54-04:00, jgarzik@redhat.com [atascsi] do a no-intr poll for completion, when setting dev xfer mode ChangeSet@1.930.151.41, 2003-08-03 13:30:37-04:00, jgarzik@redhat.com [atascsi] clean up ata_queued_cmd handling most of TCQ infrastructure now done. ChangeSet@1.930.151.40, 2003-08-03 00:14:35-04:00, jgarzik@redhat.com [atascsi] better dprintk, add 'qactive' member needed for TCQ ChangeSet@1.930.151.39, 2003-07-17 18:40:28-04:00, jgarzik@redhat.com [atascsi] fix sata module build ChangeSet@1.930.151.38, 2003-07-17 16:37:38-04:00, jgarzik@redhat.com [atascsi] bump versions ChangeSet@1.930.151.37, 2003-07-17 16:34:54-04:00, jgarzik@redhat.com [atascsi] sync with 2.5 some more ChangeSet@1.930.151.36, 2003-07-17 15:59:23-04:00, jgarzik@redhat.com [atascsi] fix unregister, fix unload, sync with 2.5 ChangeSet@1.930.151.35, 2003-07-16 07:47:34-04:00, jgarzik@redhat.com [atascsi] fix module unload ChangeSet@1.930.151.34, 2003-07-04 23:39:08-04:00, jgarzik@redhat.com [atascsi] fix bus reset (for now) This is the magic sequence that seems to work so far. ChangeSet@1.930.151.33, 2003-07-04 21:03:38-04:00, jgarzik@redhat.com [atascsi] abstract out post-reset device quiesce ChangeSet@1.930.151.32, 2003-07-04 20:00:34-04:00, jgarzik@redhat.com [atascsi] cleanups, disable software reset, start cbl type support ChangeSet@1.930.151.31, 2003-07-02 19:12:21-04:00, jgarzik@redhat.com [atascsi] host_info cleanup, flag piix pata as being slave-capable ChangeSet@1.930.151.30, 2003-07-02 08:37:04-04:00, jgarzik@redhat.com [atascsi] get ATA soft-reset working ChangeSet@1.930.151.29, 2003-07-02 07:41:29-04:00, jgarzik@redhat.com [atascsi] separate out EDD code into a separate function ChangeSet@1.930.151.28, 2003-07-02 06:29:37-04:00, jgarzik@redhat.com [atascsi] clean up device-select code ChangeSet@1.930.151.27, 2003-07-02 06:11:22-04:00, jgarzik@redhat.com [atascsi] better diagnostics in dev-identify path ChangeSet@1.930.151.26, 2003-07-02 02:16:11-04:00, jgarzik@redhat.com [atascsi] polling id-device, add atapi delays, and better err handling ChangeSet@1.930.151.25, 2003-07-02 01:18:41-04:00, jgarzik@redhat.com [atascsi] update bus reset includes better device 1 probing courtesy Hale Landis's atadrvr ChangeSet@1.930.151.24, 2003-07-01 22:57:11-04:00, jgarzik@redhat.com [atascsi] new helpers, abstract out host_init, remove __init tags ChangeSet@1.930.151.23, 2003-07-01 22:30:33-04:00, jgarzik@redhat.com [atascsi] better select-device behavior ChangeSet@1.930.151.22, 2003-07-01 02:05:09-04:00, jgarzik@redhat.com [atascsi] properly kmap during pio data xfer ChangeSet@1.930.151.21, 2003-07-01 01:37:03-04:00, jgarzik@redhat.com [atascsi] remove read poisoning debugging feature ChangeSet@1.930.151.20, 2003-07-01 01:34:09-04:00, jgarzik@redhat.com [atascsi 2.4] initialize page, offset members of scatterlist ChangeSet@1.930.151.19, 2003-07-01 00:05:58-04:00, jgarzik@redhat.com [atascsi] pass zero for packet cmd byte count limit when DMAing ChangeSet@1.930.151.18, 2003-06-30 19:35:36-04:00, jgarzik@redhat.com [atascsi] bump versions ChangeSet@1.930.151.17, 2003-06-29 17:58:18-04:00, jgarzik@redhat.com [atascsi] fix Config.in bugs, add atapi config option ChangeSet@1.930.151.16, 2003-06-29 17:50:46-04:00, jgarzik@redhat.com atascsi: fix gcc 3.3 warning ChangeSet@1.930.151.15, 2003-06-29 17:34:20-04:00, jgarzik@redhat.com atascsi: support mode sense/select on ATA devices new ata_bad_{scsiop,cdb} helpers trim trailing whitespace prepare for synchronize-cache scsicmd support return most fixed data to request buffer via memcpy ChangeSet@1.930.151.14, 2003-06-29 05:08:50-04:00, jgarzik@redhat.com [atascsi] debugging output tweaking ChangeSet@1.930.151.13, 2003-06-29 01:48:44-04:00, jgarzik@redhat.com [atascsi] atapi starts to work (bug fixes and addnl. debugging) ChangeSet@1.930.151.12, 2003-06-29 00:29:13-04:00, jgarzik@redhat.com [atascsi] begin atapi support ChangeSet@1.930.151.11, 2003-06-28 18:49:41-04:00, jgarzik@redhat.com [atascsi] increase max ich5 sata speed to udma6 ChangeSet@1.930.151.10, 2003-06-28 16:23:30-04:00, jgarzik@redhat.com [atascsi] fill in more inline docs ChangeSet@1.930.151.9, 2003-06-28 15:11:55-04:00, jgarzik@redhat.com [atascsi] rewrite bus reset to be polling, and to probe device 1 ChangeSet@1.930.151.8, 2003-06-28 10:47:54-04:00, jgarzik@redhat.com [atascsi] add docbook docs ChangeSet@1.930.151.7, 2003-06-27 23:49:13-04:00, jgarzik@redhat.com [atascsi] s/ata_host/ata_port/ ChangeSet@1.930.151.6, 2003-06-25 07:36:22-04:00, jgarzik@redhat.com atascsi: include linux/highmem.h, because we use kmap ChangeSet@1.930.151.5, 2003-06-25 07:35:57-04:00, jgarzik@redhat.com atascsi: new PCI ids for Intel ICH5 ChangeSet@1.930.151.4, 2003-06-19 20:49:51-04:00, jgarzik@redhat.com Merge redhat.com:/garz/repo/marcelo-2.4.21 into redhat.com:/garz/repo/atascsi-2.4 ChangeSet@1.930.151.3, 2003-06-09 21:48:14-04:00, jgarzik@redhat.com Merge kernel.bkbits.net:atascsi-2.4 into redhat.com:/garz/repo/atascsi-2.4 ChangeSet@1.930.151.2, 2003-06-09 11:18:41-04:00, jgarzik@redhat.com fix for current scsi includes ChangeSet@1.930.151.1, 2003-06-09 11:14:45-04:00, jgarzik@redhat.com add ata-scsi driver