d/freebsd-dev
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
108,286 Commits 72 Branches 135 Tags 3.2 GiB
b913aa0b7f
Commit Graph

749 Commits

Author SHA1 Message Date
Søren Schmidt
b913aa0b7f Dont sleep with lock held. 2004-10-10 13:24:39 +00:00
Søren Schmidt
8353d82bfd Add support for the ICH6 in legacy mode. 
The AHCI part is not supported yet, but is in the works.

5.3 RC1 candidate
 2004-10-09 16:27:13 +00:00
Søren Schmidt
6c35773729 Move the PC98 specific geometry "gunk" to geom_pc98.c where it belongs. 
This also adds support for bigger disks on the controller I have access to,
and maybe others if I understood the adhoc methods used on those.

Those with more PC98 bigdrive controllers it is hereby invited to add/fix
support for those in geom_pc98.c and not using #ifdef PC98 all over the place.
 2004-10-07 17:37:09 +00:00
Søren Schmidt
e59142714c Add SHARP to the pool of drives that doesn not need byteswapping of 
the model etc fields from identify.
 2004-10-07 11:43:25 +00:00
Søren Schmidt
6192895db8 Fix the PC98 lockups on boot. 
The interchannel locking for PC98 needed to be updated to match the
rest of the locking in ATA.
 2004-10-06 19:46:08 +00:00
Søren Schmidt
844acbc7a3 Always set half'n'half mode on ICH* chips. 2004-10-01 09:06:22 +00:00
Søren Schmidt
49686bbcaf Fix the serverworks modesetting code, of mask offset was wrong. 2004-10-01 09:04:53 +00:00
Søren Schmidt
8f0fb9584e Resurrect dump that broke with the last update. 2004-09-30 20:54:59 +00:00
Søren Schmidt
b5dee91f5f Remove the old ATA_*LOCK_CH macros that used atomic ops and use 
mutexes instead.
This closes the last (known) race issues in ATA which should fix
the various hangs etc seen on heavy loaded systems.

Change from using timeout functions to using callout functions in
the timeout code. This together with above closes the race that could
happen if timeout and device interrupt occured simultaniously.

Also fix the possible recursion in ata_reinit() on very dodgy
devices that could take us down in the probe.
 2004-09-26 11:48:43 +00:00
Søren Schmidt
c5b2c44ce8 Rearrange the order of I/O's in dma-start/stop, some chipsets are very 
picky on the order of this, especially in error situations.
 2004-09-26 11:42:42 +00:00
Søren Schmidt
a49513587b Ignore leading '_' in model name returned by devices. 
This make "_NEC" devices appear as "NEC" which is more corrent.
The reason is tha NEC originally screwed up on the byteorder in the
model string, so now that they have realized that they prefixed the '_'
so that not every ATA driver on the planet would call them "EN C" :)
 2004-09-16 09:35:01 +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
6c893545ad Release the hold on ata_delayed_attach earlier so we can use tsleep 
in the boot probe as well.

Suggested by:	gibbs
 2004-09-09 13:25:46 +00:00
Søren Schmidt
0f7cfb8473 Introduce ata_udelay() that uses tsleep instead of DELAY if possible. 
In places where we have long delays that doesn't depend on too accurate
timing, use ata_udelay() instead of DELAY() so we dont uselessly spin
the CPU if not nessesary;
 2004-09-03 12:10:44 +00:00
Søren Schmidt
7a6d2ca81b Sync promise_mio_command with WIP on the Promise line of SATA controllers. 2004-09-03 08:23:04 +00:00
Søren Schmidt
df6a77ddcb Cleanup the defines used for various chipsets. 
Add new nVidia controllers.
 2004-09-03 07:37:53 +00:00
Søren Schmidt
8ea3547a15 Cleanup the storing and printing of the device transfermode for SATA. 2004-09-01 12:15:44 +00:00
Søren Schmidt
d3ef327c8d remove unused prototype 2004-08-27 22:14:45 +00:00
Søren Schmidt
dc26948b0c Fix the handling of "inflight" requests when doing reinit's. 
Add missing untimeout that would get lost in handling of some
error situations, and caused what looked like random timeouts
afterwards when the timeout fired.
 2004-08-27 14:48:32 +00:00
Søren Schmidt
33203bee0c Set and reset the right bit for UDMA enable/disable on the ICH* chips. 2004-08-27 12:54:58 +00:00
Søren Schmidt
71d7101fac Rearrange the call to disk_destroy. 
Suggested by:	phk
 2004-08-27 12:03:18 +00:00
Søren Schmidt
1c01d05f9c Sii_reset needs to wait up to 1 second to get slow disks with it so 
use tsleep instead of DELAY.
 2004-08-24 20:11:26 +00:00
Søren Schmidt
741d64783e Always pick up giant before returning from an ioctl call. 2004-08-24 15:09:05 +00:00
Søren Schmidt
5a8f8e353e Properly check malloc returns. 2004-08-24 10:39:00 +00:00
Søren Schmidt
d607fde675 Workaround devices that responds with registers as *both* master & slave, 
but fail utterly when we try to talk to the "fake" device.
 2004-08-22 15:54:08 +00:00
Søren Schmidt
001eba0b3c Correct the args to busdma, mostly cosmetic. 2004-08-20 19:05:22 +00:00
Søren Schmidt
d04d789463 DONT PANIC 
Only call dmainit() if there is a valid busmaster resource.
 2004-08-20 06:19:25 +00:00
Søren Schmidt
5f7a48ef1e Improve (hopefully) on the workaround code for devices that doesn't 
interrupt when command is done, ie some ATAPI CD drives with no
media loaded.
 2004-08-16 09:32:35 +00:00
Søren Schmidt
fe76f7b6c8 Fix Promise sx4/sx4000 support that broke on the latest race fixes. 2004-08-16 09:29:46 +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
ed0c8a7dc7 Fix bad free() usage. 
Submitted by:	Pawel Jakub Dawidek
 2004-08-09 14:22:58 +00:00
Søren Schmidt
74f492f9bf Close a race in ata_reinit(). 2004-08-09 12:02:32 +00:00
John-Mark Gurney
3633dce05f make sure that ioctl's to ata use the proper cmd IOCATA... 
Reviewed by:	sos
 2004-08-09 06:47:53 +00:00
Poul-Henning Kamp
5721c9c76a Tag all geom classes in the tree with a version number. 2004-08-08 07:57:53 +00:00
Søren Schmidt
ba0b5c1c3b Correct the last commit so it works in error situations as well. 2004-08-07 12:49:28 +00:00
Nate Lawson
c68afc9d21 Fix a panic in ata_generic_transaction(). The DMA pointer of the channel 
was being unconditionally dereferenced but was NULL for PIO requests.
Check the request flags for a DMA transaction before dereferencing.

Reported by:	ceri
Tested by:	Radek Kozlowski <radek -at- raadradd.com>
 2004-08-06 22:23:53 +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
d5a9dcbf23 Add firmware revision to probe printf. 2004-08-05 21:11:33 +00:00
Søren Schmidt
601d3a344e Change the default to switch on DMA on ATAPI devices if they can 
do UDMA2 (ATA33) mode and beyond.
 2004-08-01 12:31:38 +00:00
Søren Schmidt
eaf8d2de9e Fix the panic (""memory modified after free") when ata_getparam() fails 
and retries.

Found by:	Nate Lawson
 2004-07-31 10:29:29 +00:00
Søren Schmidt
b4d89b2e0f Fix the reset sequence for sx4/sx4000 type controllers, it could hang the 
chip in an indefinite state.
 2004-07-30 13:33:09 +00:00
Søren Schmidt
a4c3e742b7 Refine the wait for ATAPI_RESET. 
Properly wait for not busy and introduce a timeout for devices not
setting busy (as they should).
Leave a printf in there that states how long the wait was, as I'd like
to get an idea of the variations here. The time needed seems also to be
affected by whether a medium is present or not.
 2004-07-24 19:03:28 +00:00
Søren Schmidt
426fc84503 Dont expect interrupt from ATAPI_RESET, it doesn't deliver one. 2004-07-23 17:01:47 +00:00
Søren Schmidt
4ad91b11e7 Dont set "address setup timing" on newer VIA chips. 
closes PR:	69180
 2004-07-17 14:48:46 +00:00
Søren Schmidt
cc97de873c Attempt to handle suspend/resume better. 2004-07-12 10:50:50 +00:00
Søren Schmidt
3e3807252b Remove leftover debug output. 2004-06-30 09:22:47 +00:00
Søren Schmidt
dfc7e0081f Add support for LSI type software RAID's. 
Made possible by: John Cagle @ HP
 2004-06-25 21:21:59 +00:00
Søren Schmidt
29dc49f725 Use the right ordering of args on mtx_init(). No functional changes 
since the args in question was all zero's.

Found by: Jimmy Olgeni <olgeni@FreeBSD.org>
 2004-06-22 11:18:25 +00:00
Brian Feldman
5fd2bb173d Call the detach function with g_waitfor_event() so that it can access 
the GEOM topology.

There are still issues with not detaching from cam correctly such that
upon a device detach there's an invalid pointer dereference from the
later call to cam_rescan().
 2004-06-21 20:42:03 +00:00