Commit Graph

171 Commits

Author SHA1 Message Date
John Baldwin
a2677ff239 Don't bother clearing maps for static DMA allocations to NULL. Instead,
leave them as purely opaque values that are only set by bus_dmamem_alloc().
2014-06-17 18:10:06 +00:00
John Baldwin
f07894dbde Don't destroy bus_dma maps created by bus_dmamem_alloc(). In some cases,
don't create a map before calling bus_dmamem_alloc() (such maps were
leaked).  It is believed that the extra destroy of the map was generally
harmless since bus_dmamem_alloc() often uses special maps for which
bus_dmamap_destroy() is a no-op (e.g. on x86).

Reviewed by:	scottl
2014-06-10 20:25:45 +00:00
Alexander Motin
45f6d66569 Remove all legacy ATA code parts, not used since options ATA_CAM enabled in
most kernels before FreeBSD 9.0.  Remove such modules and respective kernel
options: atadisk, ataraid, atapicd, atapifd, atapist, atapicam.  Remove the
atacontrol utility and some man pages.  Remove useless now options ATA_CAM.

No objections:	current@, stable@
MFC after:	never
2013-04-04 07:12:24 +00:00
Alexander Motin
512a3aa005 Fix command timeout caused by data underrun during fetching ATAPI sense
data, introduced by r246713.  There are two places where ata_request is
filled in ATA_CAM: ata_cam_begin_transaction() and ata_cam_request_sense().
In the first case DMA should be done for addresses from the CCB. In second
case, DMA should be done to the different address, the address of the sense
buffer inside the CCB structure itself.
2013-02-22 21:43:21 +00:00
Konstantin Belousov
dd0b4fb6d5 Reform the busdma API so that new types may be added without modifying
every architecture's busdma_machdep.c.  It is done by unifying the
bus_dmamap_load_buffer() routines so that they may be called from MI
code.  The MD busdma is then given a chance to do any final processing
in the complete() callback.

The cam changes unify the bus_dmamap_load* handling in cam drivers.

The arm and mips implementations are updated to track virtual
addresses for sync().  Previously this was done in a type specific
way.  Now it is done in a generic way by recording the list of
virtuals in the map.

Submitted by:	jeff (sponsored by EMC/Isilon)
Reviewed by:	kan (previous version), scottl,
	mjacob (isp(4), no objections for target mode changes)
Discussed with:	     ian (arm changes)
Tested by:	marius (sparc64), mips (jmallet), isci(4) on x86 (jharris),
	amd64 (Fabian Keil <freebsd-listen@fabiankeil.de>)
2013-02-12 16:57:20 +00:00
Ulrich Spörlein
9a14aa017b Convert files to UTF-8 2012-01-15 13:23:18 +00:00
Marius Strobl
1ae5318fe8 - Allocate the DMA memory used for the work area as coherent as at least
the ataahci(4) and atamarvell(4) drivers share it between the host and
  the controller.
- Spell some zeros as BUS_DMA_WAITOK when used as bus_dmamem_alloc() flags.

MFC after:	2 weeks
2011-03-06 12:54:00 +00:00
Marius Strobl
1510a2b019 Several chipset drivers alter parameters relevant for the DMA tag creation,
i.e. alignment, max_address, max_iosize and segsize (only max_address is
thought to have an negative impact regarding this issue though), after
calling ata_dmainit() either directly or indirectly so these values have
no effect or at least no effect on the DMA tags and the defaults are used
for the latter instead. So change the drivers to set these parameters
up-front and ata_dmainit() to honor them.

Reviewd by:	mav
MFC after:	1 month
2010-11-28 18:53:29 +00:00
Alexander Motin
066f913a94 MFp4:
Introduce ATA_CAM kernel option, turning ata(4) controller drivers into
cam(4) interface modules. When enabled, this options deprecates all ata(4)
peripheral drivers (ad, acd, ...) and interfaces and allows cam(4) drivers
(ada, cd, ...) and interfaces to be natively used instead.

As side effect of this, ata(4) mode setting code was completely rewritten
to make controller API more strict and permit above change. While doing
this, SATA revision was separated from PATA mode. It allows DMA-incapable
SATA devices to operate and makes hw.ata.atapi_dma tunable work again.

Also allow ata(4) controller drivers (except some specific or broken ones)
to handle larger data transfers. Previous constraint of 64K was artificial
and is not really required by PCI ATA BM specification or hardware.

Submitted by:	nwitehorn (powerpc part)
2009-12-06 00:10:13 +00:00
Alexander Motin
ebbb35ba70 MFp4:
- Remove most of direct relations between ATA(4) peripherial and controller
levels. It makes logic more transparent and is a mandatory step to wrap
ATA(4) controller level into ATA-native CAM SIM.
- Tune AHCI and SATA2 SiI drivers memory allocation a bit to allow bigger
I/O transaction sizes without additional cost.
2009-10-31 13:24:14 +00:00
Alexander Motin
4b861b1569 Remove constraint, requiring request data to fulfill controller's alignment
requirements. It is busdma task, to manage proper alignment by loading
data to bounce buffers.

PR:		kern/127316
Reviewed by:	current@
Tested by:	Ryan Rogers
2009-09-06 14:23:26 +00:00
Rafal Jaworowski
93588d5c28 Move non-PCI prototypes from ata-pci.h -> ata-all.h.
This removes unnecessary PCI #includes dependency for systems with ATA
controllers living at non-PCI buses.

Submitted by:	Piotr Ziecik
Obtained from:	Semihalf
2009-06-24 15:38:17 +00:00
Alexander Motin
78d154163c Quite mechanical ch_detach implementations for all atapci subdrivers.
Some dmainit call fixes for previous commit.
2009-02-19 00:32:55 +00:00
John Baldwin
03befed82c Restore the default maximum segment size for the bus dma tag to 64k as it
is in 6.x and 7.x.  The typo caused 64k transactions to be unnecessarily
split up into two PRD entries.
2008-10-21 18:51:55 +00:00
Søren Schmidt
13014ca04a This is the roumored ATA modulerisation works, and it needs a little explanation.
If you just config KERNEL as usual there should be no apparent changes, you'll get all chipset support code compiled in.

However there is now a way to only compile in code for chipsets needed on a pr vendor basis. ATA now has the following "device" entries:

atacore:	ATA core functionality, always needed for any ATA setup

atacard:	CARDBUS support
atacbus:	PC98 cbus support
ataisa:		ISA bus support
atapci:		PCI bus support only generic chipset support.

ataahci:	AHCI support, also pulled in by some vendor modules.

ataacard, ataacerlabs, ataadaptec, ataamd, ataati, atacenatek, atacypress, atacyrix, atahighpoint, ataintel, ataite, atajmicron, atamarvell, atamicron, atanational, atanetcell, atanvidia, atapromise, ataserverworks, atasiliconimage, atasis, atavia;	Vendor support, ie atavia for VIA chipsets

atadisk:	ATA disk driver
ataraid:	ATA softraid driver

atapicd:	ATAPI cd/dvd driver
atapifd:	ATAPI floppy/flashdisk driver
atapist:	ATAPI tape driver

atausb:		ATA<>USB bridge
atapicam:	ATA<>CAM bridge

This makes it possible to config a kernel with just VIA chipset support by having the following ATA lines in the kernel config file:

device          atacore
device          atapci
device          atavia

And then you need the atadisk, atapicd etc lines in there just as usual.

If you use ATA as modules loaded at boot there is few changes except the rename of the "ata" module to "atacore", things looks just as usual.
However under atapci you now have a whole bunch of vendor specific drivers, that you can kldload individually depending on you needs. Drivers have the same names as used in the kernel config explained above.
2008-10-09 12:56:57 +00:00
Søren Schmidt
3d970c5c0e Fix the breakage that caused AHCI devices to vanish. Editor droppings :(
Put the ATAPI device numbering back to the old ways.
2008-04-18 15:15:04 +00:00
Søren Schmidt
1b164f3101 Alloc two DMA slots pr default, silly me forgot that slaves still exists. 2008-04-17 18:11:47 +00:00
Søren Schmidt
104c094e06 Go back to preallocating everything possible on init.
This avoids calling busdma in the request processing path which caused a traumatic performance degradation.
Allocation has be postponed to after we know how many devices we possible can have on portmulitpliers to save some space.
2008-04-17 12:29:35 +00:00
Søren Schmidt
d90a6aaeb2 Fix problem with slave devices.
Fix or rather bring ENOMEM problems back to the state it was before.
Temporarily disable PortMultipliers on AHCI devices.
2008-04-14 18:34:24 +00:00
Søren Schmidt
dca5e1abd5 Fix the brokenness in the former commit, sorry for the mess.
The problem is that the PM support is part of a much larger WIP here, but due to popular demand I decided to get some of it imported.

Also I forgot the mention:

HW sponsored by: Vitsch Electronics / VEHosting
2008-04-11 11:30:27 +00:00
Søren Schmidt
9f82379c24 Add experimental support for SATA Port Multipliers
Support is working on the Silicon Image SiI3124/3132.
Support is working on some AHCI chips but far from all.

Remember this is WIP, so test reports and (constructive) suggestions are welcome!
2008-04-10 13:05:05 +00:00
Søren Schmidt
4c088dcd6c Implement a workaround of the datacorruption problem on serverworks HT1000 chipsets.
The HT1000 DMA engine seems to not always like 64K transfers and sometimes barfs data all over memory leading to instant chrash and burn.
Also fix 48bit adressing issues, apparently newer chips needs 16bit writes and not the usual fifo thing.

HW donated by: Travis Mikalson at TerraNovaNet
2007-12-13 11:47:36 +00:00
Kevin Lo
2f3ad9ba29 Fix KASSERT messages. 2007-11-20 04:52:19 +00:00
Søren Schmidt
34cf71f7ef Try to workaound silicon bugs in Promise gen2 (ie TX4) chips
Initial patch by Alexander Sabourenkov who found it in Promise's own driver.

Further fixes and sanity checks by yours truely.
2007-11-19 20:47:31 +00:00
Søren Schmidt
ae4ce3ceef OK, this is not my day, fix the former fix :/ 2007-04-08 21:53:52 +00:00
Søren Schmidt
f27a14650f Hopefully unbreak the 64bit DMA support this time. 2007-04-08 19:18:51 +00:00
Søren Schmidt
cd945eed47 Dont zero out 64BIT flag on DMA ops. 2007-04-08 15:31:39 +00:00
Søren Schmidt
16194fc40b Add support for 64bit addressing to AHCI and Marvell controllers.
Munged into ATA shape and Marvell specifics my yours truely.

Submitted by: jhb
2007-04-06 16:18:59 +00:00
Søren Schmidt
129230b816 Update copyright headers. 2007-02-21 19:07:19 +00:00
John-Mark Gurney
378f231e7d add a newbus method for obtaining the bus's bus_dma_tag_t... This is
required by arches like sparc64 (not yet implemented) and sun4v where there
are seperate IOMMU's for each PCI bus...  For all other arches, it will
end up returning NULL, which makes it a no-op...

Convert a few drivers (the ones we've been working w/ on sun4v) to the
new convection...  Eventually all drivers will need to replace the parent
tag of NULL, w/ bus_get_dma_tag(dev), though dev is usually different for
each driver, and will require hand inspection...

Reviewed by:	scottl (earlier version)
2006-09-03 00:27:42 +00:00
Søren Schmidt
e2bf77c5c2 Get rid of the advertising clause in the copyright. 2006-01-05 21:27:19 +00:00
Søren Schmidt
b39424ba7a Dont use the BUS_DMA_ALLOCNOW flag. Instead use BUS_DMA_NOWAIT and return
ENOMEM to the upper layers if we run out of memory.

This solves part of the trouble with running on >4GB memory systems.
2005-12-05 22:31:55 +00:00
Robert Watson
5bb84bc84b Normalize a significant number of kernel malloc type names:
- Prefer '_' to ' ', as it results in more easily parsed results in
  memory monitoring tools such as vmstat.

- Remove punctuation that is incompatible with using memory type names
  as file names, such as '/' characters.

- Disambiguate some collisions by adding subsystem prefixes to some
  memory types.

- Generally prefer lower case to upper case.

- If the same type is defined in multiple architecture directories,
  attempt to use the same name in additional cases.

Not all instances were caught in this change, so more work is required to
finish this conversion.  Similar changes are required for UMA zone names.
2005-10-31 15:41:29 +00:00
Søren Schmidt
40fdf81237 Add support for setting the SG list segment size.
Use this for the SiI3112 workaround to get rid of the "oversized DMA" errors.

MFC to 6.0 candidate.
2005-10-06 15:44:07 +00:00
Søren Schmidt
eeda55ce8e Reshape the dma code to be a bit more flexible so it can cope with
new HW that has new and different demands.
Fix a few nits in former commit in this cleanup crusade.

Sponsored by:	pair.com
2005-05-03 07:55:07 +00:00
Søren Schmidt
9ec5e87f63 Update on the last commit, the dma* funciton needs to be called with
a channel device, not an ata device, or we'll be out of luck in
reset/timeout where we dont have a device.
2005-05-01 12:24:45 +00:00
Søren Schmidt
0068f98f88 Take newbusification one step further, ie use the device_t more consequently
all way through the code down the layers, instead of the mix'n'match that
resulted from the conversion done earlier.

Sponsored by:	pair.com
2005-04-30 16:22:07 +00:00
Søren Schmidt
3c8b1aa784 Put the BUS_DMASYNC_PREWRITE in the rigth position. 2005-04-08 12:16:51 +00:00
Søren Schmidt
8ca4df3299 This is the much rumoured ATA mkIII update that I've been working on.
o       ATA is now fully newbus'd and split into modules.
        This means that on a modern system you just load "atapci and ata"
        to get the base support, and then one or more of the device
        subdrivers "atadisk atapicd atapifd atapist ataraid".
        All can be loaded/unloaded anytime, but for obvious reasons you
        dont want to unload atadisk when you have mounted filesystems.

o       The device identify part of the probe has been rewritten to fix
        the problems with odd devices the old had, and to try to remove
        so of the long delays some HW could provoke. Also probing is done
	without the need for interrupts, making earlier probing possible.

o       SATA devices can be hot inserted/removed and devices will be created/
        removed in /dev accordingly.
	NOTE: only supported on controllers that has this feature:
	Promise and Silicon Image for now.
	On other controllers the usual atacontrol detach/attach dance is
	still needed.

o	Support for "atomic" composite ATA requests used for RAID.

o       ATA RAID support has been rewritten and and now supports these
        metadata formats:
                 "Adaptec HostRAID"
                 "Highpoint V2 RocketRAID"
                 "Highpoint V3 RocketRAID"
                 "Intel MatrixRAID"
                 "Integrated Technology Express"
                 "LSILogic V2 MegaRAID"
                 "LSILogic V3 MegaRAID"
                 "Promise FastTrak"
                 "Silicon Image Medley"
		 "FreeBSD PseudoRAID"

o       Update the ioctl API to match new RAID levels etc.

o       Update atacontrol to know about the new RAID levels etc
        NOTE: you need to recompile atacontrol with the new sys/ata.h,
        make world will take care of that.
	NOTE2: that rebuild is done differently from the old system as
	the rebuild is now done piggybacked on read requests to the
	array, so atacontrol simply starts a background "dd" to rebuild
	the array.

o       The reinit code has been worked over to be much more robust.

o       The timeout code has been overhauled for races.

o	Support of new chipsets.

o       Lots of fixes for bugs found while doing the modulerization and
        reviewing the old code.

Missing or changed features from current ATA:

o       atapi-cd no longer has support for ATAPI changers. Todays its
        much cheaper and alot faster to copy those CD images to disk
        and serve them from there. Besides they dont seem to be made
        anymore, maybe for that exact reason.

o       ATA RAID can only read metadata from all the above metadata formats,
	not write all of them (Promise and Highpoint V2 so far). This means
	that arrays can be picked up from the BIOS, but they cannot be
	created from FreeBSD. There is more to it than just the missing
	write metadata support, those formats are not unique to a given
	controller like Promise and Highpoint formats, instead they exist
	for several types, and even worse, some controllers can have
	different formats and its impossible to tell which one.
	The outcome is that we cannot reliably create the metadata of those
	formats and be sure the controller BIOS will understand it.
	However write support is needed to update/fail/rebuild the arrays
	properly so it sits fairly high on the TODO list.

o       So far atapicam is not supported with these changes. When/if this
	will change is up to the maintainer of atapi-cam so go there for
	questions.

HW donated by:  Webveveriet AS
HW donated by:  Frode Nordahl
HW donated by:  Yahoo!
HW donated by:  Sentex
Patience by:	Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
Scott Long
5f26d98581 Don't set the BUS_DMA_ALLOCNOW flag for the parent tag or the tags that are
used for static memory allocations.

Discussed with: sos
2004-11-17 11:27:30 +00:00
Søren Schmidt
1d535cd590 Reduce the amount of memory reported to busdma.
This made the requirements for bouncebuffers too big with PAE.
Cleanup the way size defines for transfers are implemented.
2004-09-10 10:31:37 +00:00
Søren Schmidt
001eba0b3c Correct the args to busdma, mostly cosmetic. 2004-08-20 19:05:22 +00:00
Søren Schmidt
8b68793ab3 Allow the use of a supplied function to set the PRD table. This is
needed for new chips that supports 64bit addressing.
2004-08-13 08:14:27 +00:00
Søren Schmidt
d53fc3e41c Change the order of ata_dmainit/ata_allocate in preparation of
supporting new chipsets where this is needed.
2004-08-12 08:20:36 +00:00
Søren Schmidt
7e6d75884f Try to narrow down the race window on HW that does not have ways to
poll for which channel actually pulled the irq line.
2004-08-05 21:13:41 +00:00
Søren Schmidt
f2972d7eb8 Add support for the Promise command sequencer present on all modern Promise
controllers (PDC203** PDC206**).

This also adds preliminary support for the Promise SX4/SX4000 but *only*
as a "normal" Promise ATA controller (ATA RAID's are supported though
but only RAID0, RAID1 and RAID0+1).

This cuts off yet another 5-8% of the command overhead on promise controllers,
making them the fastest we have ever had support for.

Work is now continuing to add support for this in ATA RAID, to accellerate
ATA RAID quite a bit on these controllers, and especially the SX4/SX4000
series as they have quite a few tricks in there..

This commit also adds a few fixes to the SATA code needed for proper support.
2004-04-13 09:44:20 +00:00
Søren Schmidt
5df3ca789c Use UMA instead of plain malloc for getting ATA request storage.
This gives +10% performance on simple tests, so definitly worth it.
A few percent more could be had by not using M_ZERO'd alloc's, but
we then need to clear fields all over the place to be safe, and
that was deemed not worth the trouble (and it makes life dangerous).
2004-01-14 21:26:35 +00:00
Søren Schmidt
a7a120f649 Overhaul of the timeout/reinit framework. This should clear up most
of the leftovers from the old version that really doesn't work anymore.

Add a reset function for host-end of the ATA channel. This is needed
for the SiI3112 in order to whack it back to reality if a device
locks up the SATA interface (thereby preventing that we can reset the
device). The result is that ATA now recovers from the timeouts that
happens with the SiI3112A and more or less all disks based on old
PATA electronics with a Marvell PATA->SATA converter. This includes
lots of the popular SATA dongles and the WDC Raptor disks..
2004-01-11 22:08:34 +00:00
Søren Schmidt
b437f21e88 Workaround for errata on early versions of the sii3112.
Approved by: re@
2003-11-28 19:01:28 +00:00
Søren Schmidt
80344be509 Fix the DMA problem that most severely hit on the DS3112a SATA chip
in connection with Marvell based SATA->PATA dongles.

The problem was caused by a combination of things working
together to make it hard to spot...

The ATA driver has always started the ATA command, then build
the SG list for DMA and then finally started the DMA engine.
While this is according to specs, it poses a potential
problem as some controllers apparently do not allow for unlimitted
time between starting the ATA command and starting the DMA engine.

At about the same time as ATAng was committed there were lots
of other changes applied, some of which was locking in parts
that causes the busdma load functions to take significantly
longer to load the SG list.

This pushed the time spent between starting the ATA command and
starting the DMA engine over the hill for some controllers
(especially the Silicon Image DS3112a) and caused what looked
like lost interrupts.

The solution is to get all the SG list work or rather all
busdma related stuff done before we even try to start anything.

This has the nice side effect of seperating busdma out the
way it should be, so the working of the ATA machinery is not
cluttered up with busdma droppings, making the code easier
to read and understand.
2003-10-21 19:20:37 +00:00