summaryrefslogtreecommitdiff
path: root/sys
AgeCommit message (Collapse)Author
2006-01-28Make sure that HW VLAN stripping is not enabled and remove unused codeBrad Smith
for HW VLAN stripping.
2006-01-28remove printf's from dc_newbuf().Brad Smith
2006-01-27Oops, removed too many includes in 1.34 by mistake; spotted matthieu@Miod Vallat
2006-01-27Use letoh32 to print out configuration descriptors in debug functions. NowClaudio Jeker
big endian systems will print out correct settings.
2006-01-27add iwi(4).Brad Smith
From: Pedro la Peu via PR 4992
2006-01-26Add glenv(4).Mark Kettenis
tested by jmc@
2006-01-26Fix printf.Mark Kettenis
2006-01-26Make lm at iic detach properly, and use config_detach(9) to fully detachMark Kettenis
lm at iic if we attach lm at isa for the same chip. tested by robert@, krw@
2006-01-26make timecounter behave; grange@ okMichael Shalayeff
2006-01-26Make T1 work on the dual port version. Time slot map is different than onClaudio Jeker
the single port T1 version. Most notably T1 uses now timeslots 1-24 and not 0-23 no matter what type of card.
2006-01-26Some fixes to make the card work on big endian systems. Still not enoughClaudio Jeker
though.
2006-01-26Switch to __HAVE_DEVICE_REGISTER to find the device we're booted from.Miod Vallat
Tested by aoyama@
2006-01-26Fix character range check in local atoi() function. krw@ "looks correct"Aaron Campbell
2006-01-26regenBrad Smith
2006-01-26add Promise SATA-II chipsets and fix ordering of some existing entries.Brad Smith
2006-01-25Wow, this is clearly wrong (found in esa_set_params):Brad Smith
if (p->channels < 1 && p->channels > 2) return EINVAL; Change the 'and' to an 'or' so the check actually does something. From jmcneill NetBSD ok krw@
2006-01-25Duh! If askes to halt the input dont halt the output instead. This alsoBrad Smith
fixes the timeout on draining when closing. From reinoud NetBSD ok krw@
2006-01-25Fix problem in ioctl() handling in OSS audio emulation which causedBrad Smith
unintentional changes of the audio settings e.g. when running "kphone". From NetBSD ok jakemsr@
2006-01-25reduce stack use; pedro@ miod@ okMichael Shalayeff
2006-01-25- Count packets discarded by RX/TX MAC (cause of FIFO overflow, etc)Brad Smith
as input/output interface errors. - Keep values of rx/tx discards & tx collisions inside struct bge_softc. So we can keep statistic across ifconfig down/up runs (cause bringing bge up will reset chip). From oleg FreeBSD
2006-01-25kthread_create_deferred() do the things right, no need forAlexander Yurchenko
playing with cold global. ok miod@ drahn@
2006-01-25Prevent division by zero, and make it actually update the fan sensors.Mark Kettenis
2006-01-25Time-slot mask are u_int32_t so art_mask_tsmap() should return u_int32_t too.Claudio Jeker
While there fix a typo in a comment.
2006-01-25Check if the dma ring got cleared before accessing it. Fixes a panic thatClaudio Jeker
could be triggered in some weird cases. While there adjust a debug output because it is to verbose for level 1.
2006-01-25Add additional debug function to print framer error counters.Claudio Jeker
2006-01-25Reset and initialize part of the device early. Makes reading the ROM moreClaudio Jeker
reliable because correct bus timings were set later.
2006-01-24Initial stab at glenv(4), a driver for the GL518SM.Mark Kettenis
2006-01-24Fix NULL dereference on error condition in nfs_reply(). nmp->nm_so canAaron Campbell
become NULL if the nfs_reply() -> nfs_receive() -> nfs_reconnect() datapath fails. pedro@ ok "I think this is a good fix"
2006-01-24Match a few more NVIDIA product ids found in Linux drivers.Jonathan Gray
2006-01-24regenJonathan Gray
2006-01-24Some NVIDIA MCP04/MCP51/C51/MCP55 devices.Jonathan Gray
2006-01-24Remove unused fields from struct mac68k_machine.Miod Vallat
2006-01-23The GL518SM is a real piece of shit, and if we don't detect it bad thingsMark Kettenis
will happen. The comment in the code has all the gory details. tested by jmc@
2006-01-23support PMTU-Discovery for link2 ipsec bridges; ok claudioMarkus Friedl
2006-01-23Be more careful when mapping DAFB color registers, so that we don't preventMiod Vallat
esp to map the turbo registers on Q700/900/950.
2006-01-23Be sure to clear color capabilities from rasops at 8bpp if we do not haveMiod Vallat
a setcolor routine.
2006-01-23Don't match ITE chips matched by it(4).Mark Kettenis
tested by jsg@
2006-01-23Do not enable disabled sensors.Mark Kettenis
ok grange@, dlg@
2006-01-23update sm(4) from NetBSD, including support for more models and MII;Martin Reindl
also activate ISA attachment; ok brad@
2006-01-22account tx retry-fail errors in if_oerrors.Damien Bergamini
2006-01-22fixes miibus_{read,write}reg routines.Damien Bergamini
this resolves the ghost ukphy problem. did some cleanup while i'm here. tested by and ok jsg@
2006-01-22Colormap support for DAFB frame buffers (using information from Linux).Miod Vallat
However, color can not be enabled in emulation mode if the display is the console at this point, but 8bpp X11 works.
2006-01-22Add bus_space_vaddr() and use it instead of accessing private bus_spaceMiod Vallat
fields. No functional change.
2006-01-22Nothing uses this file anymore.Alexander Yurchenko
ok miod@
2006-01-22Fix UDMA on IDE function integrated on VIA VT8237 south bridge.Mark Kettenis
ok grange@
2006-01-22Print adb subsystem name early in attachment, and pass the proper deviceMiod Vallat
name to interrupt counter registration, rather than "adb".
2006-01-22Finally update the todclock on shutdown, if we are using a level 6 clock.Miod Vallat
2006-01-22Minimal iop code to set the serial ports to compatible mode automagically,Miod Vallat
so that it is not necessary to change settings from MacOS before booting into OpenBSD; from NetBSD.
2006-01-22Fix for ioapic irq routing.Brad Smith
From NetBSD, NetBSD PR 22728 for details ok beck@ krw@ tested by beck@ krw@ marc@ and a few others on a number of amd64 systems.
2006-01-22b_cylin -> b_cylinder; no functional change.Miod Vallat