summaryrefslogtreecommitdiff
path: root/sys/arch
AgeCommit message (Expand)Author
2009-01-31Don't cast pointer to int, use register_t here.Alexander Yurchenko
2009-01-31Rename loadpoint to loadaddr and add to usage().Alexander Yurchenko
2009-01-30Set mbuf watermarks.Mark Kettenis
2009-01-30Deal with running out of receive buffers.Mark Kettenis
2009-01-30Properly clean out our rings when we down the interface.Mark Kettenis
2009-01-30- add missing trailing 't' of the mos(4) entry's 'Etherne'Jasper Lievisse Adriaanse
2009-01-29Add a few cache operations after I/O completes.Miod Vallat
2009-01-29The ``flush before inval'' 88410 workaround needs to be done inMiod Vallat
2009-01-29Some of the latest macppc laptops with no adb subsystem at all, spontaneouslyMiod Vallat
2009-01-29Only start using the rtc for statclock after we've received the firstMark Kettenis
2009-01-29Make multicast actually work this time.Mark Kettenis
2009-01-29Switch tsec(4) over to using MCLGETI when allocating mbufs for the rx ring.Mark Kettenis
2009-01-28Implement proper multicast support.Mark Kettenis
2009-01-28Make bus_mem_add_mapping() return ENOMEM if we fail to allocate kva insteadMark Kettenis
2009-01-28Add missing braces in trunc_page() call from vmapbuf().Alexander Yurchenko
2009-01-27Get rid of the last traces of uvm.pager_[se]vaMiod Vallat
2009-01-27Switch the amd64 PCI_CONF_LOCK from being a simplelock and splhigh to aOwain Ainsworth
2009-01-25Remove /dev/drum and related code.Miod Vallat
2009-01-25In hpibid(), when computing timeouts from the cpuspeed variable, account forMiod Vallat
2009-01-25If hdattach() aborts early because the disk is not in good shape, alwaysMiod Vallat
2009-01-25Behave correctly in interrupt handlers if no tty has been allocated yetMiod Vallat
2009-01-23Add schsio(4), a driver for SMSC SCH311x LPC Super I/O ICs providingMichael Knudsen
2009-01-23Make write_user_windows() do what the name suggests: flush user windows insteadMark Kettenis
2009-01-22Make bus_mem_add_mapping() return ENOMEM if we fail to allocate kva insteadMark Kettenis
2009-01-22Call checkdisklabel() for labels built from native info. This willKenneth R Westerback
2009-01-21Extend the acceleration code to handle jfb (XVR-600, XVR-1200) models, theMiod Vallat
2009-01-20add softraidTodd T. Fries
2009-01-20I can't figure out why flushing all user windows before writing out windowsMark Kettenis
2009-01-20Install ACPI S3 resume trampoline code in a lowmem page. First partMike Larkin
2009-01-20Add a few more inline functions to read priviliged registers.Mark Kettenis
2009-01-19The acceleration code would not work on XVR-600 and XVR-1200, so if it turnsMiod Vallat
2009-01-19Clear RI_CLEARMARGINS in the ri_bits == NULL ifb(4) kluge, the new windowMiod Vallat
2009-01-18Define HEAP_START so that we can use space below it for ethernet buffers;Miod Vallat
2009-01-18Recompute file pointer position when seeking backwards, as it gets reinitializedMiod Vallat
2009-01-18enable atphy/ageKevin Lo
2009-01-17gdt_compact() was removed years ago, so kill the code and commentsPhilip Guenthe
2009-01-17If the hypervisor call that fetches randomness returns EWOULDBLOCK,Mark Kettenis
2009-01-17Clean things up a bit, and be a bit more fussy about the proper sequence inMark Kettenis
2009-01-17Return NO_CCB if we run out of descriptors.Mark Kettenis
2009-01-17Sync vio_sendmsg() with equivalent code in vdsk(4).Mark Kettenis
2009-01-17Enable vrng(4).Mark Kettenis
2009-01-17Make this work in guest domains.Mark Kettenis
2009-01-17Recognize environmental change machine checks on ES40, and report whatMiod Vallat
2009-01-17Implement SYNCHRONIZE CACHE and make SCSI_POLL work while we're there.Mark Kettenis
2009-01-16Set the number of openings to the number of ring descriptors minus one. ForMark Kettenis
2009-01-16Make it possible to boot from vdsk(4).Mark Kettenis
2009-01-16Make transfers bigger than PAGE_SIZE work.Mark Kettenis
2009-01-16Remove bogus KASSERT.Mark Kettenis
2009-01-16Call start callback unconditionally.Mark Kettenis
2009-01-16Improve VIO state machine, and initiate VIO handshake ourselves. Makes thisMark Kettenis