Commit Graph

1018 Commits

Author SHA1 Message Date
thorpej c9d1eae045 Size the number of initial color bins based on the size and
associativity of the L1 cache.

Note, doesn't really do anything for virtually-indexed caches,
but it doesn't hurt them, either.
2001-05-02 21:23:03 +00:00
thorpej cf67ac7122 Per discussion w/ chuck and chuck, restructure the md page stuff
to use a structure called "vm_page_md", and use __HAVE_VM_PAGE_MD
and __HAVE_PMAP_PHYSSEG.
2001-05-01 02:19:13 +00:00
thorpej 2b27ac7a99 Add a VM_MDPAGE_MEMBERS macro that defines pmap-specific data for
each vm_page structure.  Add a VM_MDPAGE_INIT() macro to init this
data when pages are initialized by UVM.  These macros are mandatory,
but ports may #define them to nothing if they are not needed/used.

This deprecates struct pmap_physseg.  As a transitional measure,
allow a port to #define PMAP_PHYSSEG so that it can continue to
use it until its pmap is converted to use VM_MDPAGE_MEMBERS.

Use all this stuff to eliminate a lot of extra work in the Alpha
pmap module (it's smaller and faster now).  Changes to other pmap
modules will follow.
2001-04-29 22:44:31 +00:00
kleink fe3c9ceaf7 * Move definitions of exact-width integer types from <machine/types.h>
to <sys/types.h> and <sys/stdint.h>.
* Add a new C99 <stdint.h> header, which provides integer types of
  explicit width, related limits and integer constant macros.
* Extend <inttypes.h> to provide <stdint.h> definitions and format
  macros for printf() and scanf().
* Add C99 strtoimax() and strtoumax() functions.
* Use the latter within scanf().
* Add C99 %j, %t and %z printf()/scanf() conversions for
  intmax_t, pointer-type and size_t arguments.
2001-04-28 15:41:28 +00:00
kleink 99054d7a6c Add definitions for C99 fastest minimum-width integer types. 2001-04-26 16:25:19 +00:00
nisimura da9a00c583 Add PRiD register imp value 0x2d for Toshiba TX4900 family. 2001-04-24 08:03:44 +00:00
thorpej 1c3a62e066 Sprinkle pmap_update() calls after calls to:
- pmap_enter()
- pmap_remove()
- pmap_protect()
- pmap_kenter_pa()
- pmap_kremove()
as described in pmap(9).

These calls are relatively conservative.  It may be possible to
optimize these a little more.
2001-04-24 04:30:50 +00:00
thorpej cd31bdd2b7 Make sure virtual_end is initialized before calling uvm_pageboot_alloc(),
and garbage-collect the virtual_avail variable (it is not necessary in
this pmap implementation).
2001-04-23 17:14:17 +00:00
thorpej 55044638aa Remove pmap_kenter_pgs(). It was never really adopted by
anything, and the interface itself wasn't as flexible as
callers would have probably liked.
2001-04-22 23:42:11 +00:00
thorpej 69abdbf60c Undo a misguided previous change to the pmap_update() API. 2001-04-22 23:19:26 +00:00
thorpej 5871b80f00 Use uvm_pageboot_alloc() for early memory allocation, rather than
calling pmap_steal_memory() directly.  On these platforms, since
uvm_pageboot_alloc() is a wrapper around pmap_steal_memory(), there
is no functional change.  This is merely for API consistency.
2001-04-22 18:21:48 +00:00
thorpej cfb5c7ed9f Make pmap_virtual_space() a required pmap function, even on platforms
which have pmap_steal_memory().  This is to reduce the API differences
between pmaps that implement pmap_steal_memory() and pmaps which do
not.

Note that pmap_steal_memory() needs to adjust *vstartp and/or
*vendp only if it used addresses within the range provided to UVM
via the pmap_virtual_space() call.  I.e. it is not necessary to do
so in any current pmap_steal_memory() implementation.
2001-04-22 17:22:57 +00:00
thorpej 4738622712 Give pmap_update() an argument (a pmap_t) so that it knows which
pmap it should be updating.
2001-04-22 00:33:59 +00:00
thorpej 7f10ba88b1 #define away pmap_update() in <machine/pmap.h> so that no function
call overhead is incurred as we start sprinkling pmap_update() calls
throughout the source tree (no pmaps currently defer operations, but
we are adding the infrastructure to allow them to do so).
2001-04-21 23:51:14 +00:00
kleink f06533a1ee Add definitions of C99 integer format conversion macros.
XXX Fastest minimum-width integer types haven't been decided upon yet.
2001-04-15 17:13:04 +00:00
kleink 739cb75837 Add definitions of C99 specified-width integer type limits.
XXX Fastest minimum-width integer types haven't been decided upon yet.
2001-04-15 15:29:02 +00:00
kleink cdcf9f46c5 Add definitions of C99 integer constant macros.
Tidy Makefiles up a little.
2001-04-14 22:46:19 +00:00
kleink a7c20e5788 Add definitions of C99 integer constant macros. 2001-04-14 22:38:33 +00:00
kleink 7affdab52e Add definitions of C99 minimum-width and greatest-width integer types.
XXX Fastest minimum-width integer types haven't been decided upon yet.
2001-04-14 12:19:49 +00:00
kleink 9133b51cd8 Replace the 'unsigned __COMPILER_INT64__' construct with a new name,
__COMPILER_UINT64__, to be supplied - if such a case is made, it shouldn't
be assumed that the unsigned type-specifier may be applied to it.
2001-04-12 22:01:18 +00:00
drochner 676bacb4ea include "opt_ddb.h" explicitely 2001-03-28 18:03:49 +00:00
tsutsui 864a597f2d Define ${OBJCOPY}.
XXX: Should we use ${CROSSDIR} here or not?
2001-03-28 17:03:48 +00:00
simonb 5aa49b6404 Delete unused uvtopte() macro. 2001-03-23 01:27:59 +00:00
chs 11fe9ca446 use ubc_winshift instead of ubc_winsize in pmaps to set up kernel
virtual space.  the latter isn't initialized yet when the value is needed.
fixes PR 12440.
2001-03-21 03:16:05 +00:00
chs ac3bc537bd eliminate the KERN_* error codes in favor of the traditional E* codes.
the mapping is:

KERN_SUCCESS			0
KERN_INVALID_ADDRESS		EFAULT
KERN_PROTECTION_FAILURE		EACCES
KERN_NO_SPACE			ENOMEM
KERN_INVALID_ARGUMENT		EINVAL
KERN_FAILURE			various, mostly turn into KASSERTs
KERN_RESOURCE_SHORTAGE		ENOMEM
KERN_NOT_RECEIVER		<unused>
KERN_NO_ACCESS			<unused>
KERN_PAGES_LOCKED		<unused>
2001-03-15 06:10:32 +00:00
shin fb20946774 keep V0 value (system call number) and restore it in exception frame,
when error is ERESTART. otherwise, user process will re-issue syscall
with broken system call number and get SIGSYS signal and terminate.

patch made by Jason R Thorpe <thorpej@zembu.com>. tested by me.
2001-02-12 10:32:14 +00:00
jdolecek 8b24036797 Require the machine-dependant DDB commands to be in db_machine_command_table[]
and link it directly to db_command_table[] so that it's not necessary
to do this at runtime. Make db_machine_command_table[] const on all ports.
g/c now unneded stuff, like db_machine_commands_install(), db_machine_init()

Patch written by enami.
2001-01-22 13:56:55 +00:00
ur 4c458e5c42 Fix register name typo. 2001-01-20 07:23:21 +00:00
shin 2e0943daa2 - fix prototype of db_mips_variable_func(). 2001-01-19 01:41:36 +00:00
tv e58532a4bf No-op commit to force update to a non-"-kk" revision. 2001-01-18 17:47:58 +00:00
jdolecek d5d431e8ae make db_[e]regs[] const 2001-01-18 10:54:27 +00:00
thorpej 9f5a22b3ee New syscall entry implementation based on the Alpha version
as hacked by mycroft.
- Use syscall_intern() to give a process a plain or fancy
  syscall based on ktrace flags.
- Avoid copying from the trapframe into a local array as much
  as possible.

Yields roughly 5% improvement on a 25MHz R3000 (DECstation 5000/200)
on a simple syscall benchmark.

There's still some work that can be done using __HAVE_MINIMAL_EMUL.
2001-01-16 06:01:26 +00:00
thorpej 61473e63f9 Define __HAVE_AST_PERPROC. 2001-01-14 22:32:57 +00:00
thorpej da57125f23 Now that we won't lose signotify()'s while we're asleep, go ahead
and to signal processing in ast() again.
2001-01-14 21:22:56 +00:00
thorpej 53d2115d66 Make the astpending flag per-process. 2001-01-14 21:18:39 +00:00
thorpej a3ec356d52 splimp() -> splvm() 2001-01-14 03:22:04 +00:00
thorpej fd0ca62926 Put signal posting back in userret() for now; for it to work
properly, we need to make astpending a per-process variable.

Pointed out by mycroft.
2001-01-14 00:39:48 +00:00
thorpej 2574f338e1 Remove a couple of now-unsed variables (args that are now no longer
passed to userret()).
2001-01-14 00:35:53 +00:00
thorpej 1eb36d23fd - Make ast() loop around astpending; it's possible for a new
AST to be posted when delivering signals, or after a process
  is preempted.
- Move all signal posting to ast().  userret() is now a one-liner.
2001-01-14 00:10:28 +00:00
thorpej 777efd8cdc Check for ASTs in Syscall and UserGenException, too; AST processing
must be done on *every* return to userland.
2001-01-13 23:49:12 +00:00
thorpej fd0605fc53 Mmm, dependency problems. Add a cast to make PROC_PC() actually
work.
2001-01-11 21:08:18 +00:00
thorpej e292dc9609 Move userret() into a header file, in preparation for splitting
syscall() into plain and fancy.
2001-01-11 18:44:29 +00:00
thorpej f1152a965b Modeled after mycroft's changes to the Alpha port, add PROC_PC() to
get profiling out of userret(), and move the preemption check to ast().
2001-01-11 18:30:16 +00:00
simonb f514e805ed Move prototypes for mips_user_cachectl() and mips_user_cacheflush()
to <mips/cachectl.h>.
2001-01-07 04:13:16 +00:00
shin ce0d1ad2fe - don't ignore DEFCOPTS (Makefile.mips).
- DEFCOPTS of hpcmips is "-Os -mmemcpy" (std.hpcmips).
2001-01-04 05:13:30 +00:00
takemura c5fd828440 replace 'long long' with int64_t to compile stand alone program with
compiler other than GCC.
2001-01-03 10:08:55 +00:00
castor a0c38fa362 Fix sendsig/sigreturn for SOFTFLOAT case. 2000-12-31 19:06:40 +00:00
castor a698752050 Disable FPU in sys___sigreturn14() so floating point registers are
properly restored.
2000-12-28 09:27:09 +00:00
castor a099ff1f75 Preserve the status register process_write_regs() to avoid confusion
about the current floating point context.
2000-12-27 08:28:12 +00:00
itojun efa2b7073c make mbstate_t bigger (32 -> 128 bytes).
XXX if you have libc after citrus locale import, please recompile libc,
and your applications that use mbstate_t (rather rare).  really sorry
for the mess.
2000-12-26 10:35:28 +00:00