1
0
mirror of https://github.com/openbsd/src.git synced 2025-01-10 06:47:55 -08:00
Commit Graph

111293 Commits

Author SHA1 Message Date
damien
72ad83af2c remove two printfs that slipped through my last commit. 2008-11-08 17:15:54 +00:00
jmc
c69e92d259 sort; 2008-11-08 15:46:49 +00:00
damien
d6aeb078b3 initialize k to NULL in wpi_tx_data() 2008-11-08 14:09:03 +00:00
dlg
92df2d4e6f fix macros up so they use the do { } while (/* CONSTCOND */ 0) idiom
ok deraadt@ otto@
2008-11-08 12:54:58 +00:00
damien
5c37862f33 do not call if_start() from tx_done() unless there is a reason to do so
(IFF_OACTIVE was set.)
2008-11-08 12:27:57 +00:00
damien
e21489260e do not allocate cmd memory for rings 5 and 6 (HCCA) which are unused. 2008-11-08 12:21:36 +00:00
tobias
b88e6b7b8a Support for retrieving revisions from branches in branches was missing.
Reported by chl@, ok joris@
2008-11-08 11:48:49 +00:00
damien
4cf47a4e3e rework iwn_start(). 2008-11-08 11:05:36 +00:00
damien
9d41d52fad new firmware & HW CCMP crypto 2008-11-08 11:02:59 +00:00
damien
9f114821c7 major wpi(4) overhaul.
wpi(4) and iwn(4) used to be in sync but with the latest iwn(4)
changes, this was no longer the case.  that commit repair this.
it brings HW CCMP encryption/decryption among other things.

requires an update of the wpi-firmware package.
please report any regression you might see.
2008-11-08 11:02:09 +00:00
damien
2fb01dd557 major wpi(4) overhaul.
wpi(4) and iwn(4) used to be in sync but with the latest iwn(4)
changes, this was no longer the case.  that commit repair this.
it brings HW CCMP encryption/decryption among other things.

requires an update of the wpi-firmware package.
please report any regression you might see.
2008-11-08 10:54:29 +00:00
ratchov
96b931e2c8 set the default device to /dev/audio for legacy mode.
From Thomas Pfaff <tpfaff _at_ agderlink.no> and tweaks from me
2008-11-08 10:40:52 +00:00
ratchov
3b0b3dc5cf when dev_attach()ing play-only or record-only streams, don't
use both play and record parameters (one of them might be
NULL, causing aucat to segfault).
2008-11-08 10:01:43 +00:00
jmc
abe48976b2 from reed@netbsd:
Be consistent: document the birthtime field of struct stat for
the "B" field specifier.
2008-11-08 08:24:10 +00:00
deraadt
8013a2019f brad, noone said you could go making changes like that (doubling the
amount of memory and interrupt time each bge uses) without discussing
it with other people.  lots of people are complaining about the lack
of communication in your process, and you are going to have to change
that.
2008-11-08 07:37:07 +00:00
deraadt
8adc6e4d84 When initializing the standard RX ring, do not allocate mbufs clusters for the
whole ring.
2008-11-08 07:13:02 +00:00
brad
5cdc842f54 Clean up some of the comments. 2008-11-08 07:10:13 +00:00
brad
079481096b When initializing the standard RX ring, allocate mbufs clusters for the
whole ring.
2008-11-08 07:00:52 +00:00
deraadt
71d7aa9e2f sync 2008-11-08 06:55:06 +00:00
brad
475df7c727 Reduce the delay's a bit in the miibus read/write routines.
Based on the Linux r8169 driver.

Tested on PCI/CardBus and PCIe adapters.
2008-11-08 06:52:49 +00:00
canacar
b284ba03b5 Add malloc and buckets views that display kernel malloc statistics similar
to vmstat -m
2008-11-08 06:38:27 +00:00
brad
cab1516596 Reduce the number of TX descriptors to 512, such a large number is not
necessary without TSO.
2008-11-08 03:49:27 +00:00
brad
7bcf0f6fab Add a softc field mii_oui to store the OUI. 2008-11-08 03:11:16 +00:00
brad
9121da7cbe Add additional link status checking for for the BCM5906 PHY.
Based on info from the Linux tg3 driver.
2008-11-08 03:03:50 +00:00
brad
95b1f0bed5 Only read the Gig status register if Gig media status is detected. 2008-11-08 01:50:48 +00:00
krw
920fcdb10c Some clients don't parse a DHO_DHCP_OPTION_OVERLOAD option the way
I expected. They object if there are no bits set in the option
value. So just use DHO_PAD in the reserved space unless at least
one of the bits is set.

Various versions tested by Tobias Ulmer on OpenSolaris, matthieu@
on busybox's DHCP client, and Uwe Dippel on Solaris. All of which
failed before.
2008-11-08 01:42:24 +00:00
chl
1ca17e461a remove dead stores and newly created unused variables.
Found by LLVM/Clang Static Analyzer.

ok krw@
2008-11-08 01:32:06 +00:00
chl
8194b5ee61 fix format string
ok otto@
2008-11-08 01:29:10 +00:00
chl
99b3a16025 fix potential null dereference.
Found by LLVM/Clang Static Analyzer.

ok krw@ pedro@
2008-11-08 01:27:51 +00:00
ckuethe
6e12971179 Document the ddb.trigger sysctl.
feedback and ok mpf@, deraadt@
2008-11-08 01:17:52 +00:00
mpf
ad129c4224 Add a new sysctl ``ddb.trigger''
Writing to it, will bring the system into the kernel debugger.
ddb.console=1 is required and the sysctl command has to be executed
from the actual console tty.
If the securelevel is < 1, the tty check is ignored.
Feedback and OK ckuethe@, deraadt@.
2008-11-08 01:14:51 +00:00
mpf
14bbaec125 Take into account that our ether_input() already strips the
ethernet header. This lets us actually process the incoming
LACP-Packets.  It should now work with a lot more switches.
At least a Catalyst 3500 seems happy.
OK brad@
2008-11-08 01:00:01 +00:00
dtucker
8cdc75f4a5 Move JPAKE define to make life easier for portable. ok djm@ 2008-11-07 23:34:48 +00:00
brad
c4538420db When initializing the MAC put the PHY into ready state for BCM5906
chipsets.

From the Linux tg3 driver via FreeBSD.

Tested by Jordi Creix <jbcreix dot mail at gmail dot com>
2008-11-07 22:52:15 +00:00
brad
61cf33e06e - Set default RX coal packets to 64, so RX coal timeout takes control during
bulk data operations.  This greately reduces the interrupt rate while keeps
  bulk data performance.
- Set default TX coal packets to 64 and TX coal timeout to 65535us (maximum
  allowable value).  This greately reduce the interrupt rate while greately
  increases (almost doubles) small packet TX performance.

From DragonFly
2008-11-07 22:41:17 +00:00
ratchov
d0e01a5c0c expose the block size in the sndio API by making par->round writable
and thus remove the ugly rate <-> block-size table from
sio_setpar(3). Handle the block size negociation in aucat(1), since
it has few constrains the code is overally simpler.
ok jakemsr@, major crank suggested by deraadt@
2008-11-07 21:01:15 +00:00
deraadt
7b7b8a270d more models, muhahahaha 2008-11-07 20:53:41 +00:00
deraadt
40ee43b790 same umass dance required for Huawei E169g/E618; Stefan Sperling; ok jsg 2008-11-07 20:49:21 +00:00
espie
367ea1a779 clean-up regexps, replacing some . (any char) with \. (actual dot),
adding grouping parentheses, compiling whatever's compilable with /o,
restricting words with word boundaries. /b

okay grunk@, jmc@ (with help from jmc@)

#10000 !

Oddly fitting, as my first commit was concerned with manpages as well...
2008-11-07 20:20:33 +00:00
todd
29097c8601 different fix for pr5883 than proposed by the submitter, found when glancing
at apple code
go ahead drahn@
2008-11-07 19:53:20 +00:00
deraadt
1ecaadf246 use do { } while (/* CONSTCOND */ 0) construct for all multi-line functional macros; ok otto 2008-11-07 19:09:03 +00:00
stevesk
ba017f0388 add space to some log/debug messages for readability; ok djm@ markus@ 2008-11-07 18:50:18 +00:00
brad
a297daa0ba Try recollecting RX/TX descriptors if we are going to switch back to TX/RX
interrupts.  There seems to be a race between turning on TX/RX interrupt and
asserting TX/RX interrupt by the hardware.

From Sephe@DragonFly
2008-11-07 18:12:22 +00:00
brad
0d741c3ad6 Simplify the interface flag handling in the ioctl handler. 2008-11-07 18:03:52 +00:00
brad
bd4f2bb3f4 Remove storing the Ethernet header in the receve interrupt handler, the
disabled code that used to use this was removed 7 years ago.
2008-11-07 18:00:31 +00:00
brad
4ed9a83d2a Add more initialisation steps (read reset register; pause; set burst size
in config register).

From NetBSD
2008-11-07 17:44:14 +00:00
brad
457ff5e62e - According to the Apple GMAC driver, the GEM ASIC specification and
the OpenSolaris eri(7D) the TX FIFO threshold has to be set to 0x4ff
  for the Gigabit variants and 0x100 for the ERI in order to avoid TX
  underruns.
- Turn on workarounds for silicon bugs in the Apple GMAC variants.
  This was based on information obtained from the Darwin GMAC and Linux GEM
  drivers.
- Turn on "infinite" (i.e. maximum 31 * 64 bytes in length) DMA bursts.

From FreeBSD

Tested by a few users with Apple GMAC's and Sun ERI.
2008-11-07 17:37:59 +00:00
deraadt
cabc23fb40 use do { } while (/* CONSTCOND */ 0) loops for all multi-line macros
ok dlg otto claudio
2008-11-07 17:31:24 +00:00
deraadt
e007d5f515 give this some /* CONSTCOND */ love 2008-11-07 05:50:33 +00:00
deraadt
14b5d89bbe wrap do/while construct around the function macro DISPATCH(); ok dlg 2008-11-07 02:23:04 +00:00