summaryrefslogtreecommitdiff
path: root/sys/arch/mvme88k/include
AgeCommit message (Collapse)Author
2004-08-06Merge Luna88k and mvme88k M88200 management code. Features:Miod Vallat
- simpler structures (no more redundant or easily computable information). - split scheme configuration (for 4:1 and 8:1 designs) is only compiled in if necessary (read: only on a mvme88k kernel configured for MVME188 support), which speeds up CMMU operations on the Luna88k. - will not enable bus snopping on a monoprocessor system. Tested on Luna88k-2, MVME187 and various MVME188 by aoyama@ and I.
2004-08-03hide more inside _KERNEL; miod okTheo de Raadt
2004-08-02Fun with the BUG:Miod Vallat
- add the board's suffix to the machine description if there is one; - recognize BUG version < 5 on MVME188, which don't provide a CNFG block. in this case we'll assume 20MHz for now, until we can parse the ENV data block correctly...
2004-08-02More include files cleaning:Miod Vallat
- move MAX_CPUS constant to <machine/cpu.h> - do not include <machine/board.h> unless needed. In fact, remove this file entirely on mvme88k, and include <machine/mvme*.h> on a compiling-for-this-board basis - keep MAX_CMMUS constant private to the m8820x code
2004-08-01Move 88100 DAE code to its own file, shared between luna88k and mvme88k.Miod Vallat
2004-07-30Move evcount structures inside struct intrhand, this makes more sense andMiod Vallat
gives us more counters in the process. Also clean up intrhand structures and usage, especially move them to SLISTs.
2004-07-28Remove EH_DEBUG code.Miod Vallat
2004-07-25Merge luna88k and mvme88k pmap. The pmap will now assume that the memoryMiod Vallat
below the kernel text is reserved for the PROM, instead of using fixed (but different) values between luna88k and mvme88k. Tested on mvme88k by myself, on luna88k by aoyama@
2004-07-24intr_establish() and related defines are mvme88k-only, so move them back fromMiod Vallat
<m88k/cpu.h>, and simplify the return values while there.
2004-07-23Eventually get the MVME188 reset sequence to work, and allow it to returnMiod Vallat
to the BUG instead of spinning if our reset fails.
2004-07-20Let the boot device detection code handle secondary SCSI buses.Miod Vallat
2004-06-22Do not access [id][pl]ar registers from C code - we only care about theirMiod Vallat
values in trapframes.
2004-05-20Properly flush instruction cache for ptrace(PT_WRTIE_{DI}, ...) on powerpcMark Kettenis
and m68k. ok drahn@, millert@
2004-05-07bus_dma(9) implementation for mvme88k, mostly based upon powerpc.Miod Vallat
2004-04-30Correctly cope with unaligned memory pointers in d16_b*()Miod Vallat
2004-04-29luna88k needs a <machine/profile.h> as well, so share it with mvme88k.Miod Vallat
2004-04-26Fix RCSIDMiod Vallat
2004-04-26Start factorizing luna88k and mvme88k common parts.Miod Vallat
For now, include files only.
2004-04-25Simplify cpu_number() - adding a shortcut depending upon the board typeMiod Vallat
actually slows things down in every case.
2004-04-24Preliminary switch to an mvme88k bus_space world. Not all drivers have beenMiod Vallat
converted yet, and they rely upon some linear mappings provided by bus_space. In order to not impact performance, almost all the bus_space accesses go through macros and inline functions. This currently restricts us to D16 and D32 access modes, which are selected at compilation time. Since there are no plans to support D8 vme devices in the future, this is acceptable for now. This makes the "len" locator for vme devices go away, and allows to simplify some code which was using the pcctwo device, and will now directly access hardware which is known to exist, rather than wait until the device is attached. While there, try to enforce more interrupt vector number checks, since these changes pointed out that dart(4) had been working correctly so far by sheer luck only.
2004-04-19remove the GATEWAY junk thingy; at the same time, select a new fatTheo de Raadt
cluster value that noone will really need to crank.
2004-04-16When a VMEBus device needs to use two interrupt vectors, be nice and let itMiod Vallat
actually use two distinct vectors, rather than twice the same. Because of this, print vector last in vmeprint() so that dmesg remains pretty in the multi-vector case.
2004-04-15Since *ramread and *ramwrite are the same code, really, factorize them a laMiod Vallat
memread/memwrite.
2004-04-14After enough checks on real hardware, it turns out the documentation is wrong,Miod Vallat
for once - SCON in GLOBAL1 on MVME188 is an SCON*, thus low if the board is the system controller.
2004-04-14Nuke getdevunit(), nothing uses it.Miod Vallat
2004-04-14When trying to program tho 88410, be sure to unmask both flash banksMiod Vallat
(as advertized in the comments...)
2004-04-14Simplify pmap_bootstrap() interface.Miod Vallat
2004-04-12Minor tweaks to appease gcc 3.Miod Vallat
2004-01-26sign mismatch correction, from miodTheo de Raadt
2004-01-26A few more BUG syscall definitions and structures. Fix (not used yet)Miod Vallat
mvmeprom_netctrl definition while there.
2004-01-23Get rid of a few (mostly DDB related) unused defines and macros.Miod Vallat
2004-01-20Use shifts and logical arithmetic, rather than multiplications and integerMiod Vallat
arithmetic, when computing cmmu numbers.
2004-01-19Get rid of a few useless defines, as well as a few duplicated ones; andMiod Vallat
move 8820x-only definitions to the adequate header. Also crank MAXPHYSMEM to a more realistic value.
2004-01-19``3 pages of u-area ought to be enough for everybody''Miod Vallat
(deraadt wanted 2, I wanted 4, actually...)
2004-01-14Some MC8820x operation cleaning:Miod Vallat
- define more constants and macros for readability. Especially cache initialization suddently becomes more readable. - after every flush operation, wait for the operation to complete by accessing the status register, before returning.
2004-01-14Do not expose kernel prototypes to userland.Miod Vallat
2004-01-13Get rid of more unused useless broken invalid constant laughable values.Miod Vallat
Also shrink iomap to a more reasonable value for now.
2004-01-13No need to set [di]sar for complete tlb invalidation.Miod Vallat
2004-01-13Get rid of old MID binary compatibility. This has been a lifesaver back whenMiod Vallat
the toolchain was in limbo, but now that we have a solid enough in-tree toolchain, this makes no sense. Plus the libc back then was full of bugs...
2004-01-13Less pedantic types; helps when userland includes <signal.h> withoutMiod Vallat
<sys/types.h>; spotted by deraadt@
2004-01-13During signal delivery, force the sigframe to be on an 8 byte boundary.Miod Vallat
This removes the need for extra padding in struct sigcontext.
2004-01-12Use a struct reg in struct sigcontext, rather than an unsorted pot-pourriMiod Vallat
of registers. This also makes signal and sigreturn much simpler.
2004-01-12Shrink the reg structure, for it to only contain registers (eh), and makeMiod Vallat
the remaining fielhs only available in struct trapframe. This has the additional benefit of preventing smarty-pants from altering the frame a bit too easily from ddb, at the expense of a few ugly casts in ddb.
2004-01-12Get rid of that ugly m88100_saved_state structure, use trapframe everywhereMiod Vallat
instead. Allow struct reg and struct trapframe to live different lives and grow separately. Righty now they are still the same, and code expects a trapframe to always start with a struct reg. This may change...
2004-01-11Mark more code as dependent upon option PTRACE, and slight cleaning while there.Miod Vallat
2004-01-09Do not define new m8820x specific constants for what are common cache andMiod Vallat
apr validity constants, really. And get rid of the old cache_policy debug trick.
2004-01-08Some splfoo() have raise-if-lower semantics; so introduce raiseipl() whichMiod Vallat
never lowers current spl value, and use it where appropriate.
2004-01-07Remove duplicate definitions with MI ddb headers, and fix db_stack_trace_printMiod Vallat
prototype in the process.
2004-01-04Working kluge to ``enable'' the on-board SCSI controller on 187 and 197, ifMiod Vallat
it is not the boot device and has not been expliictely ``enabled'' in the BUG. Will definitely help netboot installations...
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@.
generated by cgit v1.2.3 (git 2.25.1) at 2024-12-22 06:31:45 +0000