summaryrefslogtreecommitdiff
path: root/sys
AgeCommit message (Collapse)Author
2010-01-30Build these firmwares on loongson.Miod Vallat
2010-01-28Remove dead assignment and newly created unused variable.Charles Longeau
Found by LLVM/Clang Static Analyzer. ok krw@ henning@
2010-01-28The RA stack control bit on the loongson2 diagnostic register is not a clearMiod Vallat
but a disable bit; rename the define and keep the RA stack disabled at all times - I blame the few remaining hard freezes on it.
2010-01-28regenMiod Vallat
2010-01-28HP-UX getpgrp2() is documented as returning the same value than getpgid(),Miod Vallat
with the same error conditions. So reimplement it as the real thing. ok guenther@
2010-01-28Make sure the process tree is is loop-free by forbidding ptrace()Philip Guenthe
of a direct ancestor, closing a localhost DoS. As an exception, do permit ptrace() of pid 1 and have inferiors() stop climbing if it hits that. ok tedu@ hpux_compat suggestion from miod@
2010-01-28set differential gains on 6250 too (but use a 1.0 factor, not 1.5).Damien Bergamini
2010-01-28Make sure we also cause an iTLB and BTB flush on loongson processors duringMiod Vallat
regular tlb replacement, as well. Helps stability a lot, still not perfect.
2010-01-27change the MRR pattern from 1/1/1/1 to 2/2/2/4Damien Bergamini
2010-01-27- regenJasper Lievisse Adriaanse
2010-01-27- add a bunch of new ati and nvidia devicesJasper Lievisse Adriaanse
2010-01-27disable interrupts in rdmsr() and wrmsr() so that their operation does notMiod Vallat
risk being interrupted.
2010-01-264.7-BETA (also, lo-carb and ozone layer friendly)Miod Vallat
2010-01-26syncDamien Bergamini
2010-01-26new run(4) idsDamien Bergamini
2010-01-26backout last, it causes mem conflicts and device trouble;Otto Moerbeek
ok beck@ miod@
2010-01-26Be a bit more paranoid and disable interrupts around critical pieces of code,Miod Vallat
such as isa interrupt mask change. While there, use the same constructs to work around write posting when updating bonito interrupt masks.
2010-01-26Comment out the Geode timecounter code completely. While a nice idea, thisMiod Vallat
counter relies upon MSR access, and on the Lemote, access to them is done through the northbridge PCI configuration space, which wreaks havoc on the system stability.
2010-01-26Sprinkle some write buffer flushes, copied from powerpc; preventsMiod Vallat
lock count from becoming negative under bad timing circumstances.
2010-01-25add uthum there too. ok drahn@Matthieu Herrb
2010-01-25Provide proper ELF64_R_{SYM,TYPE,INFO} on little endian mips64, since theseMiod Vallat
do not follow the regular layout. Makes ld.so much happier on loongson. ok drahn@
2010-01-24"Fix" crash caused by FAT32 re-use of small buf from buffer cache,Kenneth R Westerback
at least until problem (starting somewhere between 4.1 and 4.2) in buffer cache is found and a general fix is in place. Simply B_INVAL small buf so it doesn't come back. Addresses PR#6290, confirmed by submitter RD Thrush. ok tedu@ "should be marked XXX" thib@
2010-01-23Give little-endian fp on mips a change to be handled correctly.Miod Vallat
2010-01-23finally, 6250 parts need crystal calibration too.Damien Bergamini
update sensitivity settings for 5000 series to workaround a performance bug in the DSP (1000 is not affected so we keep the old values for 5000.) update sensitivity settings for 6000 series. from similar changes to iwlwifi
2010-01-22Correctly return ENXIO on unhandled ioctls; ok marco@Miod Vallat
2010-01-22Move is_memory_range() from mips64 mem.c to a per-architecture location; thisMiod Vallat
allows /dev/kmem to access the 256MB alias of the low memory on loongson, which in turns makes procmap(8) happy.
2010-01-22Fix previous to pick the right year as the timebase (mips64 routines expectMiod Vallat
the year to be relative to 1900, not zero).
2010-01-21Do not assume mac address and bss id are stored at 32-bit aligned addresses;Miod Vallat
inspired by other wireless drivers, makes urtw work on strict alignment architectures. ok jsg@ martynas@
2010-01-21*Write* to the magic Loongson2 register to invalidate iTLB and branchMiod Vallat
prediction caches, instead of *reading* it. Ahem. Blush. Yeelong lemote now run single user.
2010-01-21Use db_printf() instead of printf() in db_dump_tlb_cmd() and trapDump(), soMiod Vallat
that the output can be paused by ddb if it spans too many lines.
2010-01-21Pick the correct year when reading clock.Miod Vallat
2010-01-21Use ahci_nvidia_mcp_attach() for nvidia MCP79 ("Ion"), without whichStuart Henderson
the driver fails to detect disks. From Wouter Coene, ok jsg@
2010-01-20One \ too many in pfvar.h. From Daniel Dickman <didickman@gmail.com>Ryan Thomas McBride
ok otto
2010-01-20Fix collider offsets to correct LBA.Jordan Hargrave
Thanks to Niklas ok @marco
2010-01-20Remove horrible hack in ahc/ahd that made some commands retryKenneth R Westerback
endlessly. Use the shiny new XS_NO_CCB instead to retry commands that can't start. Neither dlg@ nor I can figure out why 4.6 and earlier worked. Problem noted, debugged with and fix tested by Jeff Ross and his new external enclosure. Thanks! ok dlg@
2010-01-19Compute the right value for hw.ncpusfound sysctl on IP30 regardless of theMiod Vallat
value of MAXCPUS.
2010-01-19IP checksum is still broken with fragments, turn it off for now.Reyk Floeter
ok dlg@
2010-01-19A couple #ifdefs were missed when the final name of the DDB_STRUCTPhilip Guenthe
option was settled on. Fix those last couple so that the new commands are actually accessible. "Oops" miod@
2010-01-18Convert pf debug logging to using log()/addlog(), a single standardisedRyan Thomas McBride
definition of DPFPRINTF(), and log priorities from syslog.h. Old debug levels will still work for now, but will eventually be phased out. discussed with henning, ok dlg
2010-01-18regen, even though there is no code change, such that the RCS IDs match,Ingo Schwarze
as suggested by art@
2010-01-18fix a spelling error in a comment, no code change; from Brad TilleyIngo Schwarze
ok miod@ art@
2010-01-18In stacktrace(), when aborting a traceback because of nonsensical pc or spMiod Vallat
value, don't print a final bogus traceback line.
2010-01-18Kernel configuration for multiprocessor Octane machines. There are a fewMiod Vallat
rough edges left to be polished, but the kernel is otherwise stable.
2010-01-18Do not call splx() from cpu_switchto(), we are at IPL_SCHED anyway and ipl willMiod Vallat
be lowered later, by the caller. This was supposed to be harmless, but newly created processes inherit a bogus level from proc0, which in turn causes a race on MULTIPROCESSOR kernel.
2010-01-18In the interrupt handler, only attempt to acquire kernel_lock if coming fromMiod Vallat
an ipl lower than IPL_SCHED, not IPL_IPI.
2010-01-18dosoftint() being invoked at IPL_SOFTINT, don't bother checking the ipl beforeMiod Vallat
grabbing the kernel lock.
2010-01-18Define IPL_SCHED as IPL_CLOCK, not IPL_HIGH.Miod Vallat
2010-01-18In proc_trampoline(), don't spl0() if option MULTIPROCESSOR; this is alreadyMiod Vallat
done for us by proc_trampoline_mp().
2010-01-18Make trapdebug code MP-safe.Miod Vallat
2010-01-17Remove dead assignment and newly created unused variable.Charles Longeau
Found by LLVM/Clang Static Analyzer. ok otto@ krw@ tedu@
generated by cgit v1.2.3 (git 2.25.1) at 2025-01-24 21:12:40 +0000