summaryrefslogtreecommitdiff
path: root/sys/arch
AgeCommit message (Collapse)Author
2004-01-03shut gcc up; ok miodPeter Valchev
2004-01-03disable TSC for Geode SC1100; Stuart Henderson, pr 3625; ok mickey, deraadtMarkus Friedl
2004-01-03put an mi wrapper around stdarg.h/varargs.h. gcc3 moved stdarg/varargs macrosMarc Espie
to built-ins, so eventually we will have one version of these files. Special adjustments for the kernel to cope: machine/stdarg.h -> sys/stdarg.h and machine/ansi.h needs to have a _BSD_VA_LIST_ for syslog* prototypes. okay millert@, drahn@, miod@.
2004-01-03backout segment register restore diff which causes reproducible hangs; ok ↵Peter Valchev
deraadt
2004-01-02Mention that, on most 187 and 197, if the SCSI controller is not the bootMiod Vallat
device and no boot from it has been attempted at the prom, it will not get detected. Unfortunately, right now I don't know how to ``warm'' the chip. This also will need to be documented in the insallation notes at some point...
2004-01-02MC88110 errata states that, for instruction faults, in case of bus errorMiod Vallat
trap class, the value of the page fault and segment fault bits are undefined; so check for bus error first.
2004-01-02Sync m197_ext_int() logic with m187_ext_int() logic.Miod Vallat
2004-01-02Better openings value computation.Miod Vallat
For now, the result is the same as the hard-coded constant it used to be, due to a pessimistic NUM_IOPB value. This will change.
2004-01-02Relaxed dependencies, and do not undef DEBUG.Miod Vallat
2004-01-02There is no such thing as ``device cpu'' for now.Miod Vallat
2004-01-02Revert 1.100 and 1.102 for now - they cause page table corruption (bloody hell!)Miod Vallat
2004-01-02The scratch pages used in pmap_zero_page() and pmap_copy_page() being specialMiod Vallat
mappings, they still need the dcache to be invalidated after use.
2004-01-02When both cmmu_dofoo() and cmmu_remote_dofoo() exist, kill the first one,Miod Vallat
and rename the second one to the first one, i.e. have the cmmu_dofoo() functions always take a cpu# parameter. No functional change, simply makes code more readable and saves a few call frames.
2004-01-02apparently my understanding of '{FLT,DBL,LDBL}_{MIN,MAX}_EXP' ahsMichael Shalayeff
been wrong and thus proven by many and there change it back to what it is meaning explained in http://www-ccs.ucsd.edu/c/float.html . also fix spmath to use it's own consts instead of (now) off by one float.h values thus to avoid any kind of binary changes.
2004-01-02attach on two more ICHsMichael Shalayeff
2004-01-02after some advice from nick, note that only com0 is currently supportedJason McIntyre
as serial; closes PR 3621 from Johan Fredin; ok nick@ krw@ millert@ deraadt@
2004-01-02PCI_PRODUCT_ALI_M1533 should use ali1543_init too. When the PCI id forTodd C. Miller
PCI_PRODUCT_ALI_M1543 was fixed machines with an M1533 stopped working. ok deraadt@
2004-01-01Crank SHMMAXPGS from 2048 to 8192. Other platforms where peopleTodd C. Miller
are using lots of shm may also wish to also increase SHMMAXPGS. OK deraadt@
2004-01-01Oops, commited from the wrong tree.Miod Vallat
2004-01-01Let this attach correctly.Miod Vallat
2004-01-01Add sram on mvme187.Miod Vallat
2004-01-01Inline pmap_copy() and pmap_phys_address().Miod Vallat
Also get rid of a spurious local initialization in pmap_remove_mapping().
2004-01-01Some typos just never die (here controler vs controller).Miod Vallat
2003-12-31workaround: during npx exception testing, delay(1). For some reasonTheo de Raadt
amd64 (in 32 bit mode) and the new transmeta cpus want this, and lock up otherwise. very odd.
2003-12-30kill a local var that is set and never used then laterMichael Shalayeff
2003-12-30Add support for all six Motorola-supported configurations of MVME376Miod Vallat
cards. The original code would only work with the first configuration, as the etherbuf addresses grow _downwards_ from 0xfd6c0000 with Motorola's jumper configuration. Reference: MVME376 Ethernet Communication Controller User's Manual (MVME376/D1) While there, do not panic() on failure to attach, but simply print an error message and return. XXX Proper care of the memory mapping still needs to be written, as the memory XXX space of the MVME376 is always outside extio.
2003-12-30provide accounting for the intr time properly from fredette@Michael Shalayeff
2003-12-30In pmap_copy_page() and pmap_zero_page(), it is not enough to flush theMiod Vallat
cache for the affected pages; force the ptes in write through instead. This gets rid of the corrupted svnd issues encountered randomly (but frequently) on mvme88k.
2003-12-29disable ie1@gsc as not really supported and a commented out bha@pciMichael Shalayeff
2003-12-29Since pmap_activate() does a complete user tlb flush, it is not necessaryMiod Vallat
to do another one in process.S...
2003-12-29Properly recognize SiS CPU family;Alexander Yurchenko
tested by Ian Zagorskih <ianzag@megasignal.com>. ok mickey@
2003-12-29correct probe console printf; miod okTheo de Raadt
2003-12-28Crank default data and stack limits, align with m68k values.Miod Vallat
2003-12-28Having vmel and vmes keep another pointer to their parent in their softc isMiod Vallat
just plain silly!
2003-12-28dma_cachectl() takes a va and computes the pa from it, no need to invoke itMiod Vallat
a second time with the pa...
2003-12-28More optimizations borrowed from the m68k pmap:Miod Vallat
- in loops over va space, do the empty segment test only once per segment - do not flush tlb for wiring-only changes While there: - in pmap_remove_all(), do not treat wired pages special. - move more paranoid tests from DIAGNOSTIC to DEBUG.
2003-12-27Replace the MVME376 driver from a homemade and dusty if_ve driver toMiod Vallat
a ``regular'' if_le driver, sharing the common am7990 code.
2003-12-27Handle odd sizes in d16_bcopy() and d16_bzero().Miod Vallat
2003-12-27Do not print vaddr in vme attachments.Miod Vallat
2003-12-27Basic cleaning and KNF; no functional change.Miod Vallat
2003-12-26A simple compile-time micro-optimization in pmap_protect() forMiod Vallat
non-hp300 flavours.
2003-12-26Merge the attachment part and the real work part of the SCSI drivers inMiod Vallat
one file per driver.
2003-12-25Enable vx(4) again - while it still awaits more tests, it can attach withoutMiod Vallat
affecting a multiuser boot.
2003-12-25Provide common D16 vmespace block access functions, instead of vs and vx eachMiod Vallat
rolling their own. Use them more cleverly in vx, in order to get the driver to at least attach and frob chips. Not tested besides multiuser boot (hence ttyflags -a), and checking cu(1) connects. More testing to come once I remember where I have hidden the 332XT transition module...
2003-12-25Remove duplicate ident strings.Miod Vallat
2003-12-25Disable vx* for now.Miod Vallat
2003-12-25Floating-point status register exception bits are supposed to be maintainedMiod Vallat
by software, so be sure to set them both the precise _and imprecise_ floating point exception handlers, whatever the state of the HANDLER define is (which is anyway, soon to be hitting the dust in a cvs tree near you...) This allows userland software to really trust fpgetsticky() results.
2003-12-24Sync ahc with NetBSD, which was in turn updated from FreeBSD by PascalKenneth R Westerback
Renauld of Network Storage Solutions, Inc. Many fixes, wider device support. In particular, the notorious 'Target 0' problem seems to be fixed. Does *not* include any updates to isa or eisa code beyond what was necessary to compile. Known issues: 1) Tagged Queuing is probably not optimal. 2) PPR negotiation may not be fully functional. 3) No support yet for freezing devices or channels. 4) The mechanism for preventing 'A' and 'B' channel confusion during probe can fail if scsibus > 254 found. 5) Requeuing I/O's not working. A workaround will be committed almost immediately. At the moment timeouts, SCSI message rejects, aborting SCB's and trying to freeze a device may cause incomplete i/o's to be reported as complete. 6) Verbosity and probe messages need work. 7) Last disk on bus seems to go through an extra re-negotiation. 8) >16 devices on an adapter will trigger the usual problems of total openings exceeding available SCB's under heavy load. Tested by deraadt@, beck@, miod@, naddy@, drahn@, marc@ amoung others. ok deraadt@.
2003-12-24Murphy's law applied to BSD:Miod Vallat
Every non-KNF file in the source tree contains at least one severe bug. KNF and a timid start at cleaning, but not very far, so there is no functional change at the moment. This code is still pathetic for now, but it sorta works and the price was right.
2003-12-24Since we have real bus error faults, and handle them as such, always treatMiod Vallat
uvm_fault() returning EACCES as a segmentation fault rather than a bus error, whatever address the fault is at. As a result, this correctly delivers SIGSEGV, rather than SIGBUS, when attempting to write to a page with only PROT_READ permissions.
generated by cgit v1.2.3 (git 2.25.1) at 2025-01-01 14:34:05 +0000