d/freebsd-dev
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
80,740 Commits 72 Branches 135 Tags 3.2 GiB
fe74650816
Commit Graph

35191 Commits

Author SHA1 Message Date
Josef Karthauser
fe74650816 In rev 1.51 of usb_port.h I switched over to using the USB_USE_SOFTINTR 
code path to fix a bug in the non USB_USE_SOFTINTR path that caused
the usb bus to hang and generally misbehave when devices were unplugged.
In the process though it also reduced the throughput of usb devices because
of a less than optimal implementation under FreeBSD.

This commit fixes the non USB_USE_SOFTINTR code in uhci and ohci
so that it works again, and switches back to using this code path.

The uhci code has been tested, but the ohci code hasn't.  It's
essentially the same anyway and so I don't envisage any difficulties.

Code for uhci submitted by:	Maksim Yevmenkin <myevmenk@exodus.net>
 2002-09-30 17:50:18 +00:00
Yoshihiro Takahashi
a92b303976 Call bus_set_resource() to set the ioport resource. 2002-09-30 16:41:47 +00:00
David E. O'Brien
dbe57134ec Turn back on the "SMP: AP CPU #N Launched!" message on normal boots. 
Peter's rev 1.189 should fix the lost console on SCSI-based systems due
to this message.
 2002-09-30 15:39:57 +00:00
Poul-Henning Kamp
c91a63aa2c Add support for DIOCGMEDIASIZE and DIOCGSECTORSIZE. 
Remove all traces of disklabel.

Sponsored by:	DARPA & NAI Labs.
 2002-09-30 13:53:22 +00:00
Poul-Henning Kamp
150b2bdfb2 If GEOM is in the kernel, take these three out. I have no way of 
testing any modifications to them, they shouldn't even bother with
disklabels in the first place and they are just plain obsolete old
hardware which should be axed entirely before 5.0-R IMO.

Sponsored by:	DARPA & NAI Labs.
 2002-09-30 13:49:20 +00:00
Poul-Henning Kamp
419f39ce0f Prefix private BIO commands with "FD" so people get a hint that they 
are in fact private.

Sponsored by:	DARPA & NAI Labs.
 2002-09-30 13:42:06 +00:00
Guido van Rooij
467f686e93 Add quirk for Apacer HandyDrive 
MFC after:	1 week
 2002-09-30 10:11:34 +00:00
Poul-Henning Kamp
3a24c28f37 Don the asbestos underwear and add the code which lets DIOCWDINFO 
write modified disklabels back to disk.

Sponsored by:	DARPA & NAI Labs.
 2002-09-30 08:59:59 +00:00
Poul-Henning Kamp
72840432e0 Retire g_io_fail() and let g_io_deliver() take an error argument instead. 
Sponsored by:	DARPA & NAI Labs.
 2002-09-30 08:54:46 +00:00
Poul-Henning Kamp
90b1cd5615 Introduce g_write_data() function. 
Sponsored by:	DARPA & NAI Labs
 2002-09-30 08:50:47 +00:00
Poul-Henning Kamp
5b3317e9e6 Add missing g_enc_le2(). 
Sponsored by:	DARPA & NAI Labs.
 2002-09-30 08:47:46 +00:00
Poul-Henning Kamp
5dcf28b202 Disable the g_sanity() check unless people ask for it in the debugflags. 
Sponsored by:	DARPA & NAI Labs.
 2002-09-30 08:46:29 +00:00
Poul-Henning Kamp
cd4f50fbb8 Make sure we don't loose our topology lock in a call_me() handler. 
Sponsored by:	DARPA & NAI Labs.
 2002-09-30 08:27:29 +00:00
Poul-Henning Kamp
f46708e081 Don't leak memory in case device_add_child_ordered() returns NULL. 
Found by:	FlexeLint
 2002-09-30 07:56:12 +00:00
David E. O'Brien
469541bfde Only print out the "SMP: AP CPU #N Launched!" message on verbose boots. 
The kernel printf() isn't race-free
 2002-09-30 07:03:16 +00:00
David E. O'Brien
e4544528f6 Save the FP state in the PCB as that is compatable with releng4 binaries. 
This is a band-aid until the KSE pthread committers get back on the ground
and have their machines setup.

Submitted by:	eischen
 2002-09-30 07:02:22 +00:00
Matthew N. Dodd
0e0e9411f2 HARP driver for the IDT77201/211 NICStAR ATM Adapter (Including Fore LE155). 
Obtained from:	 Richard Hodges <rh@matriplex.com>
 2002-09-30 05:12:39 +00:00
David E. O'Brien
940f2da74e Use fcntl.h from inside /sys. 
Reviewd by:	scottl
 2002-09-30 02:47:23 +00:00
Warner Losh
0d9ad27e7e mbuf leak in the error case has been fixed. When we have 
an error, go ahead and m_freem the buffer.

PR: 32666
Submitted by: Chi-Fung Fan
 2002-09-30 00:18:12 +00:00
Warner Losh
559e54225b Make beep not depend on HZ setting. 
PR: 25201
Submitted by: Akio Marita-san
# This is the last part of the PR uncommitted.
 2002-09-29 23:41:25 +00:00
Warner Losh
521c179f7d SMIENB not needed, but maybe PCI_CLOCK is, so put it in #if 0'd out 2002-09-29 23:37:07 +00:00
Warner Losh
253b72896c Parens considered good. 2002-09-29 23:36:17 +00:00
Julian Elischer
2735483034 uh, commit all of the patch 2002-09-29 23:28:58 +00:00
Julian Elischer
e081731767 commit the version I actually tested.. 
Submitted by:	davidxu
 2002-09-29 23:23:25 +00:00
Julian Elischer
9eb1fdea37 Implement basic KSE loaning. This stops a hread that is blocked in BOUND mode 
from stopping another thread from completing a syscall, and this allows it to
release its resources etc. Probably more related commits to follow (at least
one I know of)

Initial concept by: julian, dillon
Submitted by:	davidxu
 2002-09-29 23:04:34 +00:00
Warner Losh
d909f3e20c Fix comment 2002-09-29 18:42:14 +00:00
Warner Losh
7f2907fd70 Limit the TX key to a valid range 
PR: 39960, 39961 (patches here pointed out problem, but didn't quite fix it)
 2002-09-29 18:40:35 +00:00
Warner Losh
214c0b3da4 Don't leak the bar list for each thing we allocate. 
# This code really needs a rewrite

Spotted by the eagle eyes of: phk
 2002-09-29 18:07:29 +00:00
Yoshihiro Takahashi
7698537b29 Added some buggy PC-98 PnP cards support. 2002-09-29 13:31:26 +00:00
Bruce Evans
a7f6243009 Fixed some of the namespace pollution in rev.1.33. <sys/systm.h> was 
included here because it was once a prerequisite of <sys/mutex.h>
although that bug was fixed long ago.
 2002-09-29 12:09:08 +00:00
Bruce Evans
65c66e222d Include <sys/systm.h> instead of depending on namespace pollution in 
<net/if_var.h>.  But depend on the standard pollution in <sys/param.h>.
Removed unused includes.
 2002-09-29 12:01:36 +00:00
Bruce Evans
6364993cd3 Include <sys/systm.h> instead of depending on namespace pollution in 
<net/if_var.h>.  But depend on the standard pollution in <sys/param.h>.
 2002-09-29 11:59:53 +00:00
Yoshihiro Takahashi
a9efc762e3 Merged from sys/isa/fd.c revision 1.238. 2002-09-29 11:43:14 +00:00
Yoshihiro Takahashi
b47483f6b8 Merged from sys/dev/syscons/syscons.c revision 1.389. 2002-09-29 11:42:36 +00:00
Yoshihiro Takahashi
8bb39f7153 Merged from sys/dev/sio/sio.c revisions 1.380 and 1.381. 2002-09-29 11:41:48 +00:00
David E. O'Brien
21b68415cd Fix style nit where conditionally compiled code was unconditionalized, 
but style(9) was consulted.

Submitted by:	bde
 2002-09-29 04:47:41 +00:00
Julian Elischer
0cd3964f6d lock proc while calling psignal 
(plus related cleanups)

Submitted by:	davidxu
 2002-09-29 02:48:37 +00:00
Peter Wemm
8eb8107b44 Deal with some SMP races by doing the entire copyin at once rather 
than doing the checks piecemeal and then doing a second copyin later.

PR:		38021
Submitted by:	davidx (I've tweaked the patch a bit)
 2002-09-28 22:44:45 +00:00
Poul-Henning Kamp
61ddb5c9d7 Rename "struct device" to "struct midway_device" to avoid clashing 
with newbus.
 2002-09-28 21:59:59 +00:00
Poul-Henning Kamp
651dde1b81 Move includ of <sys/bus_priate.h> later to get semantic identity of 
device_t the same throughout kernel.

This is a very fine point of C which fortunatly does not make any
difference in normal circumstances but which due to the pervasiveness
of device_t in the kernel can make a lint barf a lot.
 2002-09-28 21:38:35 +00:00
Poul-Henning Kamp
2f9752e95e Change a return to a break so the local buffers get properly freeed. 
Spotte by:	FlexeLint

Reviewed by:	rwatson
 2002-09-28 21:34:31 +00:00
Poul-Henning Kamp
19ebba326b s/struct dev_t */dev_t */ 2002-09-28 21:21:01 +00:00
Poul-Henning Kamp
b9b7876d2b This driver doesn't need to include <sys/bus_private.h> either. 
Spotted by:	FlexeLint
 2002-09-28 20:59:59 +00:00
Poul-Henning Kamp
a3ae595729 Disabuse this driver of the notion that it needs to include 
<sys/bus_private.h> to find the unit from a device_t.

Spotted by:	FlexeLint
 2002-09-28 20:56:32 +00:00
Peter Wemm
feea0ab05b There is no need for start/num to be signed in i386_ldt_args. 2002-09-28 20:42:04 +00:00
Poul-Henning Kamp
079a527a3d Zero the local-variable mutexes before we call mtx_init() on them, 
failing to do this may lead mtx_init() to belive they have already
been initialized.

Detected by:	Marc Recht <marc@informatik.uni-bremen.de>
 2002-09-28 19:38:29 +00:00
Peter Wemm
8b223ac209 Repair range checking for reading the ldt list. 
PR:		38016
Submitted by:	davidx
 2002-09-28 19:37:54 +00:00
Juli Mallett
85de3147ea When spamming me with a printf(9), under DIAGNOSTIC, at least be nice enough 
to include a newline.

MFC after:	4 days
Sponsored by:	Bright Path Solutions
 2002-09-28 19:04:49 +00:00
Poul-Henning Kamp
d94c4c67a2 Make P_MAGIC fit in p_magic. 2002-09-28 18:48:22 +00:00
Mitsuru IWASAKI
01598b8d4a Add support for the BCM5702x chips. 
Reviewed by:	jdp
MFC after:	1 week
 2002-09-28 18:04:12 +00:00