d/freebsd-skq
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
46,828 Commits 4 Branches 49 Tags
Commit Graph

294 Commits

Author SHA1 Message Date
n_hibma
914df027ab Please welcome the URio driver. Written by 
Iwasa Kazmi <kzmi\@ca2.so-net.ne.jp>
 2000-03-16 09:16:14 +00:00
n_hibma
8dc02eb79c Add the TD's to the hash when they are created not when they are used. 
This makes it possible to handle the done queue even when the end of tail
TD has ended up on the done queue as well (the xfer in this case is NULL
but the next field is valid, so we simply skip that element.

Make sure we mark all the handled transfers as handled.

Stop using powerhooks. They are a NetBSD-ism.
 2000-03-15 22:27:24 +00:00
n_hibma
b5dac19635 Add an ID for the SiS 5571. 
Remove the unnecessary use of parent.
 2000-03-15 22:26:17 +00:00
n_hibma
05fb95dedc powerhooks no longer used 2000-03-15 22:15:21 +00:00
n_hibma
c9949b5ab1 Make the makefile work properly. 2000-03-15 22:13:50 +00:00
n_hibma
2255c5b248 Do not use the powerhooks at all, instead of abusing and empty #define. 2000-03-15 22:12:47 +00:00
n_hibma
011df3599c Never stop the control pipe 2000-03-15 22:09:53 +00:00
n_hibma
2247521d20 Regen. 2000-03-15 22:04:01 +00:00
n_hibma
c90a664a70 Add Y-E Data floppy drive. 2000-03-15 21:53:09 +00:00
n_hibma
88080d726e Update the e-mail address to a valid one. 2000-03-15 21:50:41 +00:00
gehenna
6305007817 Add vendor/device ID for Corega FEther USB-TX. 
Submitted by:	iwasaki
Approved by:	jkh
 2000-03-09 16:35:21 +00:00
gehenna
080d706712 Regen. 2000-03-09 16:29:49 +00:00
gehenna
e03f01682d Add Corega FEther USB-TX. 
Submitted by:	iwasaki
Approved by:	jkh
 2000-03-09 16:28:58 +00:00
n_hibma
c68b859120 Fix flushing of the clists before deleting them. Stops the machine 
from panicing when disconnecting a USB Wacom Graphire tablet.

jhk approved the fix some time ago, but I forgot to actually commit the
thing.

Reported by:	Dirk van Gulik <dirkx@webweaving.org>
Approved by:	jhk
 2000-03-02 16:01:53 +00:00
n_hibma
d0fd416b57 Move the warning on wrong IRQ values to the right place, before any 
resource allocation is attempted.

It will present the user with a message that he has to switch on USB
support in his BIOS.
 2000-02-20 14:22:44 +00:00
n_hibma
e8b624173a Correctly handle the conversion from virtual to physical addresses. The 
problem was basically (for offset > 4096):

	vtophys(addr) + offset != vtophys(addr + offset)

Also, use TD's with a maximum size of 4k instead of 8kb for OHCI
controllers.

This problem occurs in drivers that use large transfer sizes:
umass, host2host and ethernet with jumbo frames.
 2000-02-10 18:50:19 +00:00
n_hibma
8fafd8e769 Add PCI Id's for i810 chipsets. 
PR:		16517
Submitted by:	SAKIYAMA Nobuo <sakichan@lares.dti.ne.jp>
Approved by:	jhk
 2000-02-07 12:50:33 +00:00
n_hibma
f07fcd07c8 Don't use tsleep when cold booting. It is called before bus->use_polling 
is initialised by usb_init.

This might solve problems with some controllers not being initiliased
properly, because a delay was effectively a tsleep that returned
immediately.

Approved by:	jhk
 2000-02-06 14:59:00 +00:00
n_hibma
adad9ac78e Disable the generation of SMIs (System Management Interrupts). Always set 
the PIRQD bit.

This fixes the problem of uhub0 hanging forever during boot when USB
keyboard support is switched on in the BIOS on motherboards with Intel
chipsets (UHCI).

Approved by:	The Sheep
 2000-02-06 14:52:27 +00:00
n_hibma
ffc4c1858f Commit missing bits, forgotten when committing support for suspend/resume 
for USB.

Call uhci_power when suspending and resuming.

Approved by: The One.
 2000-02-04 10:18:37 +00:00
n_hibma
576210bc48 bus_release_resource is spellt like bus_release_resource not like 
bus_delete_resource.

Fixes a problem when the probe succeeded, but the attach failed. The
release of the resources was done inproperly.

Approved by:	jkh
 2000-01-31 14:05:21 +00:00
n_hibma
c403f310c9 Get the KASSERT right. 2000-01-29 18:53:42 +00:00
n_hibma
7a76a6d4d2 Fix debugging output. The TDs are linked through the dnext field, not 
nexttd.
 2000-01-29 18:01:15 +00:00
n_hibma
5b4848a6e2 Remove a warning from LINT 2000-01-29 16:25:55 +00:00
n_hibma
66f371b6b8 Unbreak LINT. 
Pointed out by:   Peter Wemm
 2000-01-29 15:55:03 +00:00
n_hibma
46a154cb23 First part of the fix for ohci_hash_find_td panic. 
Some controllers submit bogus pointers to the Done queue.
ohci_hash_find_td fails to find these in its hash and panics. Instead of
panicing we now assume the whole done queue is lost and let the timeout
code to clean up the mess after us.
 2000-01-29 14:53:47 +00:00
n_hibma
c63e4d9590 Add comments and debugging info. 2000-01-29 14:41:48 +00:00
n_hibma
0701def723 The toggle carry bit is stored in the headp not the tailp. 2000-01-29 14:18:31 +00:00
n_hibma
904db0db42 Mask off the last two bits before comparing. It might just be that some 
hardware might leave those bits in the wrong state.
 2000-01-29 11:59:31 +00:00
n_hibma
720b7903cb Tripmine for bad hardware. 2000-01-29 11:50:44 +00:00
wpaul
e81c093824 Minor tweak: the D-Link 10/100 USB ethernet adapter is apparently using 
the same design as the LinkSys adapter and needs the same special handling
to enable its PHY.
 2000-01-29 02:16:47 +00:00
n_hibma
76324b32bd Correct the list of error messages. It was incomplete. 2000-01-28 13:05:26 +00:00
n_hibma
f9bed91ed9 Regen. 2000-01-28 10:31:12 +00:00
n_hibma
0ea7dddb88 Correct the entry for the Kodak DC290. 
Submitted By:	Oliver Fromme <olli@dorifer.heim3.tu-clausthal.de>
 2000-01-28 10:26:29 +00:00
wpaul
10d6a2ddbb Fix a bug in the uhci driver that breaks large bulk IN transfers. The 
uhci_check_intr() routine needs to be more careful about deciding when
the end of a transfer has been detected.

This allows me to remove the nasty workaround code from if_aue and if_cue.
Receive performance is now much better for these adapters (500KB/sec
vs. 350KB/sec).

Also removed unused KUE_CUTOFF define from if_kuereg.h.

Submitted by: Lennart Augustsson
Reviewed by: n_hibma
 2000-01-28 02:15:31 +00:00
n_hibma
d9cd3d7688 The values for OHCI_HALTED and OHCI_TOGGLECARRY were reversed. 2000-01-28 00:03:45 +00:00
n_hibma
5674e93381 Move the defines to the right location. 
Add a DIAGNOSTIC when closing interrupt pipes.

Insert splx(s) which were left out with iso pipes (non-functional yet) and
in a DIAGNOSTIC.
 2000-01-27 23:25:58 +00:00
n_hibma
64633582ac Add umass.c 2000-01-27 23:19:14 +00:00
n_hibma
7e6dc45292 Properly teardown the allocated and initialised stuff when an error 
occurs (OHCIwas already done for UHCI).

Get rid of the usbus variable. It is confusing.

Align uhci_pci.c and ohci_pci.c again.
 2000-01-26 10:52:27 +00:00
n_hibma
64cda4e365 Don't set ivars until we checked the return value from device_add_child. 2000-01-26 10:27:12 +00:00
n_hibma
4a29dd1cc9 Add comments 2000-01-26 08:46:41 +00:00
n_hibma
97f54a6227 Add usbd_clear_endpoint_toggle(pipe) function. 
This is needed to implement a asynchronous callback for clear endpoint
stall, needed for the umass driver.
 2000-01-25 18:40:39 +00:00
n_hibma
54e504ae5d Correct subroutine names in DPRINTF 2000-01-25 18:37:58 +00:00
n_hibma
5bef12b54e The previous fix for the USB ugen device set configuration ioctl() 
has a slight error, and returns EIO way too often.

PR:		16315
Submitted by:  Louis Mamakos <louie@TransSys.COM>
 2000-01-24 10:23:57 +00:00
peter
fb5d4fa843 Fix some warnings. 2000-01-23 19:27:11 +00:00
peter
6588369b69 remove "extern int cold" from SPLUSBCHECK - it causes a heap of warnings. 2000-01-23 19:26:16 +00:00
peter
71c7f2956d Fix a warning in DIAGNOSTIC code. 2000-01-23 19:25:29 +00:00
n_hibma
744b2db779 Check whether we can change config at all. If any of the endpoints 
has been opened, we should not allow configuration changes as endpoint
descriptors might disappear.

PR:	16256, 16168
 2000-01-23 15:48:29 +00:00
n_hibma
46612cf2e7 Zap the vnodes for the control endpoint as well. 2000-01-23 15:42:08 +00:00
n_hibma
154c09625c Add ID for the AMD-756 OHCI controller 2000-01-21 10:46:29 +00:00