summaryrefslogtreecommitdiff
path: root/sys/arch
AgeCommit message (Collapse)Author
2004-09-14eap tested to work; tybollt@solace.mh.seTheo de Raadt
2004-09-14print layers in hexMichael Shalayeff
2004-09-14poke and clean all exceptions regs and not just oneMichael Shalayeff
2004-09-14a couple of new bitsMichael Shalayeff
2004-09-14error out from btlb_insert() if there is noneMichael Shalayeff
2004-09-14always map the io spaceMichael Shalayeff
2004-09-14do not print anything about btlb if there is noneMichael Shalayeff
2004-09-14better nkpdes calculation including pdes for io space now; move ↵Michael Shalayeff
uvm_page_physload() to later where pdes are all mapped already
2004-09-14higher max kmem default and mark swap brokenMichael Shalayeff
2004-09-14a few more funcsMichael Shalayeff
2004-09-13for pcons tty and kernel console (ie. boot -a) input, replace prom ^HTheo de Raadt
with ^? as that is what we have standardized on for all delete keys. this got missed, meaning boot -a and RAMDISK console tty were gross. ok miod
2004-09-10Need to sync NLEAF() occurences too.Miod Vallat
2004-09-10typo in SAVE_CPU expansion.Miod Vallat
2004-09-10Use correct register aliases wrt the __mips_n64 regdef.Per Fogelstrom
2004-09-10This is the script used to link the kernel. A flaw in ld makesPer Fogelstrom
setting the load address using 0xffffffff80100000 impossible so a workaround using -0x7.... is used. Hopefully this file will go away and using -N and -Ttext= will work once binutils is fixed.
2004-09-10Fix LEAF usage adding new extra arg. spotted by miod.Per Fogelstrom
2004-09-09these should have gone in with the other 64 bit changesPer Fogelstrom
2004-09-09Kernel moves to 64 bit. A few more tweaks when binutils is updated.Per Fogelstrom
2004-09-09Kill mb_map and related defines, we don't need one.Miod Vallat
2004-09-09cpu_setup functions passed wrong first arg; from foleite@yahoo.com.brTheo de Raadt
ok grange markus
2004-09-07Properly put ktrsyscall (and scdebug_foo) into biglock, inspired from NetBSDNiklas Hallqvist
2004-09-06rl fits on install media againTheo de Raadt
2004-09-06Jumbo pack of fixes:Miod Vallat
- do not leak memory when polling; - bring LUN support back - Motorola documentation says LUNs are not supported, but it's a SysV/m88k limitation, not a hardware one. - honour request timeout while polling (instead of using a fixed value) - do not program the scsi command length if the hardware knows it from the scsi command group (as advised in the manual) - various minor fixes, especially better error recovery. tested by nick@ and I; ok deraadt@.
2004-08-31Buglet in cachectl(), do not use curproc anymore.Miod Vallat
2004-08-30another card tat uses system promMichael Shalayeff
2004-08-30In the interrupt handler, if there is a DMA transfer in progress, doMiod Vallat
not busy wait for the transfer to complete at the very beginning, as this interrupt could be unrelated to the transfer; busywaiting could cause the scsi command timeout to be triggered, and the code would not recover correctly. Instead, make the DMA check simply return, leaving the "busy" flag set; and perform the usual interrupt handler duties, except for the phase change. test&ok millert@
2004-08-30fix some types not agreeing with gcc 3Per Fogelstrom
2004-08-30Add the original LUNA-88K support.Kenji Aoyama
- retrieve auto-boot information from NVRAM - retrieve ethernet address information from FUSE ROM They are stored different way from LUNA-88K2. Tested by Jeff McMahill, ok by miod@
2004-08-29oops debugging printfMichael Shalayeff
2004-08-27Add wsmux to support USB keyboards on RAMDISK_CD, which has usb support,Dale Rahn
ok deraadt@
2004-08-27Support USB keyboards on install where USB support is enabled. ok deraadt@Dale Rahn
2004-08-26diskboot changesPer Fogelstrom
2004-08-26Fix a problem where the driver could get stuck in the rx int loop.Per Fogelstrom
2004-08-26sync with GENERIC for PCI NICs and MII PHY driversBrad Smith
ok deraadt@
2004-08-25Swap the sources for clock and statclock on MVME188 (the Z8536 being moreMiod Vallat
reliable), and add a large block of comments to explain the timer mess^Wsituation on MVME188.
2004-08-25Allow DMA transfers to be disabled through device flags; default is to enableMiod Vallat
DMA.
2004-08-25Set SCMD_PAD in non-integral sector DMA transfers, or the transfer neverMiod Vallat
completes; from the old driver, forgotten during the replacement here.
2004-08-25Rearm the DUART counter as soon as we enter the interrupt handler, to notMiod Vallat
drift the clock of the hardclock() processing time.
2004-08-25Minor tweaks to make things work in install.Per Fogelstrom
2004-08-25Disable ehci on macppc for now, it causes a race condition between devicesDale Rahn
attached at boot and ehci, can be enabled via UKC. ok deraadt@
2004-08-25Better DUART timer interval computation for clock.Miod Vallat
2004-08-25Use a simpler delay() routine for MVME188, from Mach via Luna88kMiod Vallat
2004-08-24The#@$#%^&}1 z8536 timers runs at PCLK/2, hence 2MHz. Compensate, sinceMiod Vallat
the code assumes 1MHz timers.
2004-08-24Sync with i386 cdboot.8:Tom Cosgrove
Better description of the path to bsd.rd, to avoid having to update it every release. Also note that cdboot's ls command doesn't work on ISO-9660 filesystems.
2004-08-24Better description of the path to bsd.rd, to avoid having to update itTom Cosgrove
every release. Mentioned by marc@, among others. Also note that cdboot's ls command doesn't work on ISO-9660 filesystems. help and ok jmc@ and deraadt@
2004-08-24Oops, unbreak dartcngetc().Miod Vallat
2004-08-24Correctly document the order in which cdboot looks for kernels.Tom Cosgrove
2004-08-24Have cdboot look for /3.6/amd64/bsd.rd first.Tom Cosgrove
"go for it" deraadt@
2004-08-24Have cdbr look for cdboot at /3.6/amd64/cdboot as well as at /cdboot.Tom Cosgrove
"go for it" deraadt@
2004-08-24Correctly document the order in which cdboot looks for kernels.Tom Cosgrove
generated by cgit v1.2.3 (git 2.25.1) at 2024-12-21 08:19:29 +0000