Commit Graph

2386 Commits

Author SHA1 Message Date
christos
ebdc03b439 make the definitions consistent with the declarations. 2005-05-30 04:56:13 +00:00
christos
7ceb7d6ff0 another UNCONST.. 2005-05-30 04:55:35 +00:00
christos
a9cf21849d - const poisoning
- eliminate variable shadowing
2005-05-30 04:21:39 +00:00
christos
2f27be7da0 - const poisoning
- avoid shadowing
- usbd_trim_spaces was writing to const memory; give it a buffer.
2005-05-30 04:20:46 +00:00
skrll
35de2b7d10 Fix interrupt out transfers.
Reviewed by Lennart.
2005-05-28 07:44:37 +00:00
soren
5350b653c5 Sync 1284 id printing with the devinfo change, but leave it disabled for now. 2005-05-23 16:35:26 +00:00
augustss
191e9dab58 Handle DCD differently so it's not always asserted. From kern/30268 from
Wolfgang Stukenbrock.
2005-05-18 11:26:11 +00:00
augustss
36e939eaa1 Allow multiple opens of the control endpoint. This is unproblematic since
the control endpoint is in some sense always open to the device.

From wulf@ping.net.au.
2005-05-17 14:53:06 +00:00
augustss
bf68872cdb Support for the HX version of the chip.
Contributed by a donor that wishes to remain anonymous.
2005-05-11 20:25:01 +00:00
augustss
834327c847 Don't keep the devinfo string on the stack, instead use malloc/free.
This should cure some rare stack overflows.
2005-05-11 10:02:28 +00:00
augustss
cfff5f5e7a Only look at USBD_FORCE_SHORT_XFER for writes. 2005-05-08 08:12:08 +00:00
skrll
f2f62a1726 Add support for optional interrupt output pipes as described in the
HID specicification version 1.11.

Reviewed by Lennart. Thanks.
2005-05-08 06:19:10 +00:00
abs
b82eaee723 Add USB Portuguese (pt_PT) keyboard map, from Manuel Pata via
Rui Paulo in PR/30041
2005-05-05 22:27:29 +00:00
augustss
3fefd3b866 Make sure all physical addresses are filled in the qTD even when we have
a page offset.

From Dan Ellis (as was the previous fix).
2005-05-05 20:54:34 +00:00
augustss
9a2e60d4f6 Make sure to honor the USBD_FORCE_SHORT_XFER flag and generate an empty
transfer when necessary.
2005-05-05 20:40:20 +00:00
jonathan
f36cdf86c7 Fix ifdef botch in previous revision:
The `Static' forward declaration of ukbd_rawrepeat was inside #ifdef
UKBD_REPEAT but not #ifdef DISPLAY_COMPAT_RAWCKBD. The definition and
all other references are inside both #ifdefs.  The `static' function
declaration with no definition failed GCC warnings.

Pass GCC warnings by wrapping  the declaration with a check that both
CPP tokens are defined.
2005-05-03 18:37:46 +00:00
augustss
d069124a45 Don't emulate auto repeat in the keyboard driver, the wskbd driver does that.
Auto repeat emulation can be turned on by defining UKBD_REPEAT, but this
should never be necessary.
2005-05-03 13:15:21 +00:00
augustss
c2c00331d7 Use UTF8 to encode strings read from the device (instead of using '?' for
characters >=0x100).

Also add serial number string to the device information struct.
2005-05-02 15:32:18 +00:00
wiz
fbd2fc5eca Fix typo. 2005-05-01 21:47:11 +00:00
augustss
649ae4a707 Update TODO comment and copyright. 2005-05-01 19:45:40 +00:00
augustss
9c5fb57b96 Only complain when opening full speed isoc devices, other types seem to
work reasonably (even if there is no proper scheduling of the TT yet).
2005-05-01 19:35:07 +00:00
augustss
6aae573fe5 Fix the error check on completion of an xfer. Just because the qTD status
field has error bits set doesn't mean there has been an error, unless the
HALTED bit is set as well.  The old behaviour could result in a transfer
being considered failed by the software, but a success by the device.
This seems to have affected mostly mass storage devices.

Pointed out by Dan Ellis in kern/29731.
2005-05-01 19:24:39 +00:00
augustss
5adcc3d063 Timeout while polling is in ms not ticks.
From OpenBSD.
2005-05-01 14:21:27 +00:00
augustss
77e2a0a080 In eohci_waitintr, xfer->timeout is in ms, not hz.
From OpenBSD.
2005-05-01 01:14:30 +00:00
augustss
7beafdee3a Fix some spelling errors.
From OpenBSD.
2005-05-01 01:10:06 +00:00
augustss
31e5d222f7 Preserve any configuration data that may have been set by SMM/BIOS over
chip reset.

From OpenBSD.
2005-05-01 01:03:11 +00:00
augustss
e556a8eca4 If the SMM driver had enabled ownership change interrupts, re-enable them
temporarily before we attempt to take control.

From OpenBSD.
2005-05-01 00:45:55 +00:00
augustss
5283bc3d82 Root hubs don't have transaction translators, so don't print anything
about them for depth 0.  From FreeBSD.
2005-04-30 20:54:13 +00:00
augustss
5b3acf742e Fix a race condition in xfer abort. Derived from a FreeBSD patch.
An xfer could be aborted twice (which means that the second abort might
access deallocated memory).  This happened when an xfer timed out and
the timeout started an abort.  While that abort was taking place the
xfer could be cancelled (usually by closing the pipe), causing a second
abort to begin.
This is now handled by having flags indicating the abort state of an xfer.

Hopefully this will fix the occasional crashes when printing.
2005-04-30 14:38:40 +00:00
skrll
e71eebcad0 Complete the interrupt output transfer support copy from FreeBSD, but
remove the incorrect access to the bulk part of the union.

Reviewed and OK'ed by Lennart.
2005-04-29 19:15:13 +00:00
augustss
91c4478386 Check if we have any SCSI busses so this file doesn't cause link errors
if used without SCSI.
2005-04-29 17:52:46 +00:00
augustss
71f12914f1 If UKBD_LAYOUT isn't defined, but PCKBD_LAYOUT is, use the latter as the
default layout.  Fixes half of kern/28668.
2005-04-29 17:02:06 +00:00
augustss
17e2d8f05a Allow 32 chars in the saved vendor string. Fixes kern/29760. 2005-04-29 15:04:29 +00:00
augustss
abcf879dd7 Get bit reversal table right this time. 2005-04-27 23:39:54 +00:00
augustss
35f6423e4f Add define for EHCI_MAX_POLLRATE. 2005-04-27 21:23:41 +00:00
augustss
c3aad684bd Interrupt polling was spread very unevenly. Use the same bit reversal
trick as in ohci.c to correct this.
2005-04-27 16:42:26 +00:00
augustss
d514cae370 With the interrupt fix in the previous commit low and full speed devices
have a chance to work, so allow them (with a warning).
Make a small change to the transaction translator scheduling to make it
at least hobble along (again from Dan Ellis).
2005-04-27 16:20:26 +00:00
skrll
5597345e50 KNF and s/a an/an/ 2005-04-27 07:47:25 +00:00
augustss
6490a3e8e5 Don't null the qh_link point in the interrupt tree after it has been
initialized.  From Dan Ellis in kern/30003.
2005-04-27 02:12:20 +00:00
drochner
1e5f8afdd2 remove debugging stuff which caused conflicts with USB_DEBUG kernels 2005-04-18 18:45:02 +00:00
toshii
2db3ae3497 Correct status value check for OHCI isoc transfer;
the spec says that "not accessed" is 111x, not 1111.
2005-04-17 14:46:49 +00:00
itohy
0ee992857f ukyopon(4): Kyocera AIR-EDGE PHONE driver
Close NetBSD PR #25954 in a different way.
2005-04-15 17:18:18 +00:00
itohy
15a9291a5d Fix function name in debug messages.
White space police.
2005-04-15 15:08:34 +00:00
itohy
8bd72ab6a4 If CM descriptor is found, use the data interface in it;
if CM descriptor is NOT found but UNION descriotor is found,
use the data interface in it.
Do not require CM or ACM descriptors as long as
the data interface is found.

Should fix NetBSD PR #29754 by rivo nurges.
2005-04-15 14:43:05 +00:00
itohy
70d7fffcc5 Split common modem part of umodem.c to umodem_common.c and umodemvar.h,
to be shared with other umodem-like drivers.
2005-04-15 14:14:09 +00:00
itohy
43d5a75df2 Allow multiple modem interfaces per configuration. 2005-04-12 13:12:10 +00:00
itohy
1bf2f3ae6e Add usb_find_desc_if(), which is similar to usb_find_desc(),
but searches only in the specified interface rather than
in whole configuration.
2005-04-12 13:10:14 +00:00
xtraeme
6b34c93b70 regen 2005-03-15 17:54:17 +00:00
xtraeme
8223efec1b Add entry for "SAMSUNG Mighty drive" provided by Thilo Manske in PR
kern/29550.
2005-03-15 17:52:18 +00:00
skrll
41889bdcaa Remove an item - uhidev(4) now exists.
OK'd by Lennart.
2005-03-13 08:21:50 +00:00