Aren/NetBSD
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
131,240 Commits 606 Branches 0 Tags 3.1 GiB
8270ea3f59
Commit Graph

131240 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
peter
8270ea3f59 Use ANSI function decls. 2004-10-25 23:37:58 +00:00
peter
7101691fa3 Remove two redundant conditionals. 2004-10-25 23:06:41 +00:00
peter
993460fbb1 Change the static array to a linked list, removes the limit of max 20 LKMs. 
Mostly from OpenBSD.
Reviewed by several people on tech-kern.
 2004-10-25 22:59:12 +00:00
enami
63a896d5a8 Fill some entry in fake wdc_ata_bustype since those are called even 
if no wd(4) device is configured in a kernel.
 2004-10-25 22:42:09 +00:00
yamt
d9a63ae94a flush stdout before switching output to stderr. 2004-10-25 20:35:40 +00:00
yamt
a2e8d4b465 short results are not fatal. 2004-10-25 20:34:50 +00:00
hannken
12c30b8690 Add speedstep support for Dothan processors. 
From Steven J. Dovich via PR#26342.

Tested on "Intel(R) Pentium(R) M processor 1.60GHz".
 2004-10-25 15:42:24 +00:00
sketch
0d451812d5 Regen (Dell Port Replicator added). 2004-10-25 11:32:02 +00:00
sketch
b06825e711 Add Dell Port Replicator. 2004-10-25 11:31:03 +00:00
martin
5c7441e83b No need to clear %o1 when returning a 64 bit integer from pseg_get if we 
are compiling for a 64 bit kernel.
 2004-10-25 10:42:48 +00:00
augustss
e2916e01dd Preserve some bits in the overlay qTD. 2004-10-25 10:29:49 +00:00
peter
4037d03620 Apply revision 1.56 from binutils CVS: 
Catch NULL bfd pointers passed to bfd_archive_filename.

Fixes a core dump in ld(8) when loading certain LKMs.

Approved by mrg.
 2004-10-25 09:45:43 +00:00
augustss
dd753df337 Add EHCI_QTD_SET_STATUS. 2004-10-25 08:53:38 +00:00
dan
813c2cdec9 Elaborate on the performance implications and suitability of 
NEW_BUFQ_STRATEGY to various hardware and system configurations.
Bump date.

XXX is the comment about this option being "not yet proven stable"
still relevant/true?
 2004-10-25 06:13:47 +00:00
lukem
7cc9c4ef50 fix formatting of a column 2004-10-24 23:58:58 +00:00
enami
5c26271a67 Back out part of previous. The card refered in PR/27255 needs to be 
handled differently (at least the current code doesn't handle well).
 2004-10-24 22:17:07 +00:00
augustss
d3c902a013 Add a comment. 2004-10-24 22:13:52 +00:00
augustss
2dc3c5521f Follow the spec more closely when updating the overlay qTD in the QH. 2004-10-24 22:12:24 +00:00
augustss
becc42a193 Only delay interrupt my 2 miucroframes instead of 8. This improves 
Ethernet performance.
 2004-10-24 22:11:05 +00:00
augustss
f0c6875478 No need to generate an interrupt at the data part of a control transfer, 
it's generated by the status transfer.
 2004-10-24 22:09:15 +00:00
augustss
3d0622d94b Make sure to update the data toggle on short transfers. 
This makes my Ethernet adapter much happier.
 2004-10-24 22:07:04 +00:00
cube
85c68ec88d Add a check in DEV_USES_PARTITIONS macro in case dv->dv_cfdata is NULL, 
as it is the case for md and raid. Raid case tested by Martin Husemann,
I could test the install floppies work again.
 2004-10-24 17:06:24 +00:00
wiz
2f2f733ec3 Sort options. 2004-10-24 17:01:22 +00:00
wiz
3d110d313a Sort options in usage. 2004-10-24 17:01:07 +00:00
wiz
77e2a559dc Drop empty line at EOF. 2004-10-24 16:43:21 +00:00
wiz
de2ba5e1d1 Uppercase first letter in a sentence. 2004-10-24 16:42:19 +00:00
wiz
0967fe43f4 Fix typo in Dd argument. 2004-10-24 16:40:32 +00:00
wiz
c40d738e25 Comment out Xr to non-existing hostname.if(5). 2004-10-24 16:40:01 +00:00
wiz
4506a84c31 Drop trailing whitespace. 2004-10-24 16:37:49 +00:00
lukem
feb92219d9 Protect access to all public functions via a mutex as some of the backends are 
not reentrant (such as compat getgr*_r, because it uses the non reentrant
_{dns,nis}_getgrent() backends to implement `+group').
 2004-10-24 14:52:46 +00:00
lukem
a48f586379 Protect access to all public functions via a mutex as some of the backends are 
not reentrant (such as compat, because it uses the non reentrant getnetgrent(3)
to implement `+@netgrp' & `-@netgrp').
 2004-10-24 14:46:23 +00:00
skrll
3780f859c7 Only create a KTR_USER record if it's one of the tracepoints. 
OK'd by manu.
 2004-10-24 14:42:10 +00:00
augustss
d58af06710 Regen. 2004-10-24 13:01:17 +00:00
augustss
5e26cc8165 Add 2 devices. From OpenBSD. 2004-10-24 12:58:44 +00:00
augustss
f8968198b1 Dont use lockmgr() from interrupt context. 2004-10-24 12:53:26 +00:00
augustss
122fe5d16e Whine about not being able to figure out default language if we are debugging. 2004-10-24 12:52:40 +00:00
augustss
a71f7f822d Don't make activate static. It disagrees with prototype. 2004-10-24 12:50:54 +00:00
augustss
b702a8af08 Don't allow open of low/full speed pipes types that will not work yet. 2004-10-24 08:52:26 +00:00
augustss
3f908f7bc3 Get rid of an oddly placed assert(). 2004-10-24 08:49:13 +00:00
augustss
5f05836624 Do hex conversion right. 2004-10-24 08:47:27 +00:00
skrll
22f25588af s/Id/NetBSD/ 2004-10-24 07:12:40 +00:00
skrll
e0658593c2 Fix the argument to KERNEL_PROC_LOCK/KERNEL_PROC_UNLOCK. 2004-10-24 06:58:14 +00:00
christos
cfe914cf35 Change the sc_name buffer size from 16 to 32, since we already oveflow 16 
in if_tlp_pci.c... The name code really needs to be fixed so that it does
not have a buffer overflow again:

1. the cardbus and eisa code don't initialize sc_name; it probably should?
2. tlp_attach() seems to print the name and the ethernet address, but
   the pci front end seems to do the same.
3. some of the name determination code in the pci front-end does not guarantee
   nul terminated strings.

Really, all this name code is just bloat and another place for bugs to hide.
 2004-10-24 00:35:08 +00:00
heinz
eb3298f1fb Follow the change done for port prep, revision 1.50. This makes output from 
the autoconfiguration visible again. (Ok'd by Chuck Silvers).
 2004-10-24 00:17:20 +00:00
yamt
62905a1302 uvm_pageidlezero: grab kernel_lock before uvm.fpageqlock. PR/27259. 2004-10-23 21:29:27 +00:00
yamt
c0ebe301d7 don't reference kernel_lock directly. 2004-10-23 21:27:33 +00:00
yamt
d82d2638a7 __cpu_simple_lock: loop without locking cache or asserting LOCK#. 2004-10-23 21:25:22 +00:00
yamt
fa29ea9740 to determine if an interrupt needs to grab the kernel lock or not, 
check interrupt's own ipl rather than cpu's current ipl.
 2004-10-23 21:24:05 +00:00
thorpej
40efd0eeef Check d_magic2 once, and not d_magic twice. Pointed out by tsarna. 2004-10-23 19:45:32 +00:00
christos
fbbeedbe63 PR/27286: Tom Ivar Helbekkmo: Allow RTM_GET to work with RTA_IFA|RTA_IFP set. 
Quiting Tom: The problem is the special case of an RTM_GET message
that wants interface information included in the response, and
therefore include the RTA_IFA or RTA_IFP (or both) flags in the
bitmask that says what addresses are supplied in the message.  For
the RTM_GET message, it doesn't make sense to supply addresses
other than the one you're asking about, so those two other bits
are, in that specific case, overloaded with this meaning.

There is code in sys/net/rtsock.c to handle the case, but at some
time, extra sanity checking of the received message was added, that
failed to take this possibility into account.

The patch, is needed for the Asterisk software PBX to work properly
when it has multiple interfaces active: it needs to ask the kernel
for the IP address of the interface that will be used to communicate
with a given host.
 2004-10-23 19:13:22 +00:00