summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2009-05-16udl(4) and the wsmux(4) work on armishTheo de Raadt
2009-05-16must pull in the wscons gooTheo de Raadt
2009-05-15Oops, forgot to add this file, part of the dsrtc DS174x changes.Miod Vallat
2009-05-15TOD clock support for IOC3 flavours with DS174x chips (which are rebadgedMiod Vallat
MK48Txx). Entangled with preliminary changes which will hopefully eventually lead to power(4) attaching on IP30 (but not finished yet).
2009-05-15Don't forget to make sure IMR writes hit the hardware (by reading backMiod Vallat
from the Hub) before continuing (this was missing in one place). Also exit early from the interrupt mask handling loop if all sources have been serviced.
2009-05-15Revert 1.18, the memory bank allocation logic was actually correct, and IMiod Vallat
must have been on drugs when thinking otherwise.
2009-05-15well, PATCH_SITES has been dead for years...Marc Espie
2009-05-15On the Intel 82801BAM Hub-to-PCI, pass the main PCI io and mmio extents ifMark Kettenis
there are no windows enabled. Restores CardBus functionality on some machines with Intel chipsets.
2009-05-15Always subregion the main PCI I/O extents. This will handle things properlyMark Kettenis
when we pass the main PCI I/O extents to the secondary bus of subtractive decode brigdes.
2009-05-15syncTheo de Raadt
2009-05-15syncTheo de Raadt
2009-05-15Stop this silliness with machine dependent man pagesTheo de Raadt
2009-05-15Enable udl(4) and ukbd(4) because this lets a landisk actually have aTheo de Raadt
non-serial "console"; it is not really a console since we probe for it so late. But enabling it in ttys(5) works with getty...
2009-05-15A few stanzas from GENERIC-IP32 I forgot to uncomment or carry out.Miod Vallat
2009-05-15Forgot to remove that now unused function in previous commit.Miod Vallat
2009-05-15install wsconsctl.conf on more machinesTheo de Raadt
2009-05-15do not fake out some config-supplied foo.h / NFOO filesTheo de Raadt
2009-05-15pull in the right wscons/rasops partsTheo de Raadt
2009-05-15second (annoying) part of the run-rt2870 firmware file split.Damien Bergamini
2009-05-15properly identify the RT3052 radio (dual-band 2T2R) in dmesg.Damien Bergamini
2009-05-15Do not ignore the route metric for "redistribute default".Marco Pfatschbacher
Got broken in -r 1.62. Pointed out by bluhm@ OK bluhm@, claudio@
2009-05-15update run(4) microcode.Damien Bergamini
split it into two separate files while i'm at it since there are actually two different firmwares (one for RT2870/RT3070 and one for >=RT3071).
2009-05-15Add the "Mvix Nubbin MS-811N" adapter to the list ofDamien Bergamini
run(4) devices. The Mvix Nubbin MS-811N is the new world's smallest USB 802.11n adapter. It beats the Planex GW-USMicroN by a few millimeters (30.0*14.0*7.0 vs 33.5*16.0*8.0).
2009-05-15attach to more devices; we now have 113 usb ids.Damien Bergamini
2009-05-15regenDamien Bergamini
2009-05-15more run(4) usb ids.Damien Bergamini
one new vendor id for "Amigo Technology".
2009-05-15cleanup forgotten calls to exit(3)Alexandre Ratchov
2009-05-15replace s/sa/sio/, left from when libsa was renamed to libsndioAlexandre Ratchov
2009-05-15in headers, macro definitions come first, prototypes come nextAlexandre Ratchov
2009-05-15if DEBUG is defined, use a global debug level ratherAlexandre Ratchov
than a per-stream debug level. No behaviour change.
2009-05-15nicer (unique) wmesg for the tsleep(9) in nfs_reconnect();Thordur I. Bjornsson
2009-05-15describe sio_getcap(), sio_pollfd() and sio_setvol() retrurn valuesAlexandre Ratchov
spotted by Philipp Schafft
2009-05-15Update to match new kernel directories.Miod Vallat
2009-05-15Do not explicitely clear interrupt sources in the interrupt handler, butMiod Vallat
ask the Bridge to do this for us; this makes the initial interrupt deadlock workaround functional again.
2009-05-15Since _defdsk is now only set to "", just nuke it.Kenneth R Westerback
ok deraadt@
2009-05-15Validate the main MBR minimally; if it does not have the AA55 thingy we mustTheo de Raadt
not read garbage values as partitions... which we then put into the spoofed label... and which would lead disklabel -A to make surprising decisions. earlier versions which did too much validation tested by many
2009-05-15Zap tedious verbiage announcing disk configuration will start soon.Kenneth R Westerback
ok deraadt@
2009-05-15If we have more than one disk, we don't know what the user wants but doneTheo de Raadt
is probably not the right default answer. ok krw
2009-05-14handle the new require-order default of "no" in existing tests, and addStuart Henderson
a new test to check it's working correctly. ok deraadt@
2009-05-14actually change the require-order default to No; I missed a part withStuart Henderson
my last commit. ok deraadt@
2009-05-14Add configuration files for IP27 (includes some IP35 support) and IP30Miod Vallat
systems. Not connected to the builds yet.
2009-05-14replace the $OUTPUT idiom by the new next_part function from daily(8)Ingo Schwarze
shortening the script by 100 lines and making it easier to understand no functional change intended feedback and ok sthen@ ajacoutot@ okan@
2009-05-14Rename sgi GENERIC and RAMDISK with -IP32 suffixes, and have the buildMiod Vallat
machinery minimally cope with this. Discussed with and reluctantely accepted by deraadt@; we both dislike losing the ``naturally comes out of your fingers'' kernel names and the ``one size fits all'' logic; but for now I don't see any easy way to get a single kernel binary able to run on multiple IP## flavours.
2009-05-14More interrupt deadlock tomfoolery; turns out the non-XBridge workaroundMiod Vallat
doesn't appear to trigger the expected interrupt, so use a fugly nanotimeout instead. This makes Origin 200 disks not stall as soon as some serious I/O is intended. They now run multiuser.
2009-05-14When walking cpu configuration, report fpu type as the cpu type, sinceMiod Vallat
various R14000 (which use R14010 FPC) report different values there. That's only to get a prettier cpu0 line in dmesg.
2009-05-14Update to match the new bootblocks filenames.Miod Vallat
2009-05-14In addition to the existing IP32 boot block, build a real 64 bit boot blockMiod Vallat
as a relocatable object, suitable for IP27/IP30/IP35. Boot blocks are now found is /usr/mdec/boot{32,64}, with boot-IP## being hard links to them for easy boot block pick.
2009-05-14Remove unused local.Miod Vallat
2009-05-14Add two hooks allowing MD code get a MD header to be included if needed,Miod Vallat
and to override the way HEAP_START and HEAP_LIMIT are defined.
2009-05-14udl(4) works on macppc, too. Disabled in GENERIC by default.Marcus Glocker
generated by cgit v1.2.3 (git 2.25.1) at 2024-12-28 08:31:49 +0000