summaryrefslogtreecommitdiff
path: root/sys/arch
AgeCommit message (Collapse)Author
2003-09-19Let boot -b return to debugger before invoking the kernels entry point, likeMiod Vallat
mvme88k does.
2003-09-19- some bios return a valid geometry for inexistent hd's so beforeFederico G. Schwindt
getting the geometry check the number of attached drives from the bios area at 40:75. - bump version. from toby@. we want people to test this, specially those with ghost drives showing up.
2003-09-19add debug info around #41; toby@ ok.Federico G. Schwindt
2003-09-18- move checks earlier to catch inexistent devices before testing for edd.Federico G. Schwindt
- bump version. toby@ ok.
2003-09-18- pass the correct device to bios_getdisklabel().Federico G. Schwindt
- don't blindly return but check whether bios_getdisklabel() failed and act according. this fixes (at some extent) booting from a floppy if it wasn't inserted during floppyprobe(). toby@ ok.
2003-09-17More cleaning of the exception handling code, and collateral damage:Miod Vallat
- always give C routines invoked by the assembly code some breathing room on the stack - merge error and reset exception handlers -- gets us rid of error_fault() and error_reset(). - remove all references to SR0 and "threads" inherited from Mach. In fact, we do not use SR0 at all now. - only use double load and stores instructions when we are 200% sure we are accessing a correctly aligned area. I am not fond of unaligned kernel accesses, and forcing every pgb to be aligned on an 8 byte boundary is gross.
2003-09-16Do not use | !FOO to mention undesirable flags, but rather & ~FOO - hopefullyMiod Vallat
this had no side effects...
2003-09-16Remove unused material from include files, as well as the annoying U() macroMiod Vallat
which is only used to obfuscate things. Doing this points out that the BUG memory is not at the same address on 187 and 197 (the 197 BUG is inside obio), so provide distinct constants, and treat 187 and 197 slightly differently in pmap_bootstrap(). However, we now need to map the 197 flash memory as well... While there, simplify and constify pmap_table_build() and its associated data.
2003-09-16Clean <machine/db_machdep.h> of unused material. Also, let this file beMiod Vallat
includable even without option DDB, in which case it will only define the PC_REGS(), inst_return() and inst_call() macros; this removes the need to declare them a second time in trap.c.
2003-09-16Provide more information about the CPU/CMMU configuration in the dmesgMiod Vallat
output. Also, if the board configuration parameter area is invalid, warn the user and assume safe values - this is likely caused by a dead NVRAM battery.
2003-09-16document extension support bitmap definition; requested by deraadt@.Federico G. Schwindt
toby@ ok.
2003-09-11permit ramdisks to buildTheo de Raadt
2003-09-11support new via c3 AES instruction; written by jasonTheo de Raadt
2003-09-11handle case where bios EDD support only supports a subset; from fgsch,Theo de Raadt
ok toby, and now tested for lots of machines
2003-09-09Remove DDB conditionals and always define T_KDB_*Miod Vallat
2003-09-08match size of fpreg w/ the save87 structs as figured in discussion w/ jason ↵Michael Shalayeff
two months ago and the rediscovered by grange@ and then probably by some more other people in mongolia who had no inet to report it unfortunately; this also fixes problem w/ catching sigsegv after recent bcopy(,,sizeof) fixes; deraadt@ ok
2003-09-08Revert the changes reordering the data access exception processing in theMiod Vallat
context of a regular exception. This was probably a good move, but it eventually causes issues when uptimes grows towards infinity... Of course, diagnosing such an issue at the same time that the disk drive is dying and causes problems on his own, is not exactly easy...
2003-09-07Oops, use the local MVMEPROM_CALL(), rather than the one which used to beMiod Vallat
in <machine/prom.h> -- code in stand/ does not need a specific context to invoke the BUG functions.
2003-09-07The PTE bit used to flag wiring is zeroed behind our back on 88110, soMiod Vallat
switch to another, safe, unused bit. From the MC88110UM via deraadt@
2003-09-07Fix system hangs some i386 machines experienced when booting with theKenneth R Westerback
'-a' option. Closes PR #2122 and #3437. From Thorsten Glaser. ok deraadt@ henning@ weingart@.
2003-09-07Narrow a conditional in non-88110 area...Miod Vallat
2003-09-07Get rid of PREP2's SSBR parameter, since it does not have sense for 88110.Miod Vallat
2003-09-06Fix boot -a, add protected cnpollc(). ok deraadt@Dale Rahn
2003-09-06Large cleaning of exception handling:Miod Vallat
- faster code paths in eh.S, especially for 88110 cpus - do not service data access exceptions more than one time; also, be sure that interrupts are disabled while servicing them. - cleanup the 88110 trap handler in the same way the 88100 flavour has been cleaned recently. This spotted two bugs, which are fixed there as well: - add systrace support to the 88110 syscall code - add emulation support to the 88100 syscall code
2003-09-05Never enable interrupts just before processing data exceptions, but ratherMiod Vallat
afterwards... and only if this is a wise thing to do!
2003-09-05Do not drop the user into DDB without a really, really good reason.Miod Vallat
2003-09-04What I blamed on a specific CMMU fault pecularity turned out to be, in fact,Miod Vallat
a nice bug preventing signals to be delivered to process faulting in some conditions...
2003-09-04a better cpu identification; testing by miodMichael Shalayeff
2003-09-04correct sizeof for bzero/bcopyAnil Madhavapeddy
ok tedu krw fgsch deraadt
2003-09-03I'm walking on very, very thin ice here. Still trying to get the ``special''Miod Vallat
DAE to pass; the first fix was too brutal, the second one too shy; this third one is apparently just in the middle to make everything happy so far...
2003-09-03Use symbolic constants for pbus fault codes.Miod Vallat
2003-09-03Be more cautious in the previous fix - we don't want to skip valid data accessMiod Vallat
exceptions by mistake.
2003-09-02User mode page faults for page zero require special treatment to preventMiod Vallat
a recursive DAE. Scientists say this is proof that there's a ``strange attractor'' in the m88k processors. What is certain, though, is that the processors' documentation contains black holes.
2003-09-02indent audio comment properlyDavid Krause
ok deraadt@
2003-09-02There is a weird problems related to interrupts - we need to enable them tooMiod Vallat
early for the console to be reliable, so revert to the bug console for boot -a.
2003-09-02Revert the last change since both Pentium 4 and Mobile Pentium 4Alexander Yurchenko
can be with model = 2. Better cpu ident will be after 3.4. Problem noted by Wouter Clarie <rimshot@pandora.be>. ok deraadt@
2003-09-02just like mvme68k; miod okTheo de Raadt
2003-09-02Our extent implementation relies on the fact that extents all have distinctMiod Vallat
names. However, the tsp code uses one global extent, which both device instances want to reinitialize and use for themselves (whether this is a genuine bug or not is left to the reader). Solve the problem by putting the extents into the tsp softc, and change the pci_bwx_bus code to handle variable extent names. This makes DS20 systems with tsp0 and tsp1 work again. ok deraadt@
2003-09-02zap more driversTheo de Raadt
2003-09-02Cleanup and simplify data and instruction fault trap handling for 88100.Miod Vallat
2003-09-01The most harmful bugs are always so blatant it takes hours to spot them...Miod Vallat
Correctly setup the vector table for T_SIGSYS and T_SIGTRAP traps, and handle T_SIGSYS userland traps. This passes a complete usertrap regression test now.
2003-09-01Be sure to always provide a nonzero number in PREP() and PREP2() for softwareMiod Vallat
traps not tied to a particular vector, in order to correctly process data access faults whenever necessary, before servicing the trap.
2003-09-01Prevent userland from invoking BUG system calls, by saving the initial BUGMiod Vallat
trap vector contents, putting the bugtrap() stub in place, and switching back and forth between them when the kernel needs to invoke BUG functions.
2003-09-01Do not let userland programs enter DDB at will - it's a BAD thing.Miod Vallat
Also, fix a logic error in m88100_syscall() which prevented ERESTART and EJUSTRETURN to be returned (gasp!)
2003-09-01Correctly map the message buffer.Miod Vallat
2003-09-01remove unused sc_echobug code; ok fgs, mickeyMarkus Friedl
2003-08-31Upon returning from m18*_ext_int(), do not invoke data_access_emulation()Miod Vallat
if trap(T_DATAFLT) has been able to correct the issue. This is, in fact, really a can't-happen situation, anyways.
2003-08-31pdc_model needs to be visible from cpu.cMiod Vallat
2003-08-30Compile kernels with -mno-check-zero-division. Saves two pages, and thingsMiod Vallat
will go wrong anyway, should the kernel divide by zero...
2003-08-28from NetBSD via Tim Weiss, tim at zetaflops dot net: deal with FS block ↵Jason Wright
sizes larger than 8K testing: krw
generated by cgit v1.2.3 (git 2.25.1) at 2025-01-22 23:15:30 +0000