ead2ed15cf
Add few more L2 cache info entry.
2000-09-15 03:44:28 +00:00
29027825f3
When fetching cache info:
...
- Don't fall into infinite loop even if the # of iteration necessary isn't 1.
- Don't interpret lower 8bit of AL, which is # of iteration, as a descriptor.
2000-09-15 03:41:18 +00:00
3d6272fcdf
Use VNODE_OP_NOINLINE.
2000-09-13 16:26:15 +00:00
72a24b4eae
Add an align argument to uvm_map() and some callers of that
...
routine. Works similarly fto pmap_prefer(), but allows callers
to specify a minimum power-of-two alignment of the region.
How we ever got along without this for so long is beyond me.
2000-09-13 15:00:15 +00:00
d0b099a9fc
Fetch cache info via `cpuid' insn, parse it, remember it,
...
and report it at boot time.
2000-09-13 08:04:15 +00:00
90a2e46ab1
Add support for using the 586- and 686-class performance
...
counters.
2000-09-13 04:47:00 +00:00
b9b07590ec
Define some 586-class CESR MSR bits.
2000-09-13 04:44:27 +00:00
e2cc69026a
Add 686-class performance counter events.
2000-09-13 03:37:04 +00:00
cbb605647a
add (commented out) entry for WSCONS_DEFAULT_TYPE
2000-09-10 11:48:38 +00:00
62a6d0660c
move the guts of pmap_remove() to pmap_do_remove(), which has a flags arg
...
to tell it whether or not to remove wired mappings. pmap_remove() wants
to remove wired mappings, pmap_collect() doesn't. fixes PRs 10363 and 10761.
2000-09-10 03:45:58 +00:00
cb01317d62
Final fix to make this work without object directories again.
2000-09-08 19:57:05 +00:00
3220c30427
Readd at signs before commands which were removed for debugging.
2000-09-08 19:50:52 +00:00
278e796ec6
Do last change differently: "(unset MAKEOBJDIRPREFIX || true)" is a NOP,
...
use "MAKEOBJDIRPREFIX= && unset MAKEOBJDIRPREFIX" instead.
2000-09-08 19:48:00 +00:00
f5cb2786f8
Use "(unset MAKEOBJDIRPREFIX || true)" because unset return an error code
...
if the variable was not set before.
2000-09-08 19:45:45 +00:00
9c01e4e226
deal properly with MAKEOBJDIRPREFIX (which has priority over MAKEOBJDIR so
...
the MAKEOBJDIR to ${MAKE} has no effect) so we unset MAKEOBJDIRPREFIX before
invoking ${MAKE}
2000-09-08 05:37:59 +00:00
a1139ee535
Example config file for enabling large page support.
2000-09-07 18:55:30 +00:00
2fc18ed7a4
Replace the old db_memrw.c with the hp300 version, modified for i386
...
and large page support. This version is a bit more intelligent about
how it deals with kernel text pages.
Now that this is fixed, don't map the kernel text read-write if
DDB is configured.
2000-09-07 18:46:19 +00:00
a68986edfc
Add optional support for mapping the kernel with large (4MB) pages.
...
Right now, only does kernel text, and wastes a little memory, but
there is some noticeable performance improvement even with just text.
2000-09-07 17:20:58 +00:00
2204402aa1
If KERN_LDSCRIPT is specified as a makeoption, then add it to the
...
linker flags.
2000-09-07 06:39:36 +00:00
1946ce2a42
A linker script based on elf_i386.x which aligns the .data segment
...
to a 4MB boundary, thus allowing the text and read-only data to be
mapped using large pages.
2000-09-07 06:36:24 +00:00
50668be499
Deal with low addresses in pmap_k{enter_remove}. port-i386/10966.
2000-09-07 06:06:52 +00:00
fbc40806dd
vtopte() and kvtopte() now check for kernel vs. non-kernel addresses,
...
respectively.
2000-09-06 23:32:13 +00:00
b53916e196
Cast arguments to kvtopte() to vaddr_t.
2000-09-06 23:28:30 +00:00
5f98fd6c03
On second thought, back out kvtopte() change in previous. Dug, PT pages
...
are in a special VA range, not kernel VA range.
2000-09-06 22:23:46 +00:00
ba3605f2bf
Use kvtopte() for kernel addresses.
2000-09-06 22:22:27 +00:00
a7b2cf9d7a
Print CPU features, and use kvtopte() for kernel addresses.
2000-09-06 22:19:46 +00:00
e32a6a7c54
Use kvtopte() for kernel addresses.
2000-09-06 22:09:58 +00:00
7dd395bb7f
Remove some vto*() macros that won't stand a chance of working
...
with PSE or PAE, and reimplement vtphys() in terms of pmap_extract().
2000-09-06 19:09:45 +00:00
eac566737e
In pmap_extract():
...
- Return FALSE if the PTE does not have PG_V set (previously, you would
get always get true of the PDE was valid, even if the PTE was not).
- Teach it about 4MB pages.
2000-09-05 21:56:41 +00:00
7683d3f956
Define PG_LGFRAME, which is the page frame mask for large (4MB) pages.
2000-09-05 21:52:16 +00:00
ef2b361222
Fix to work with objdir.
2000-09-05 05:05:16 +00:00
3fd457414e
constify first arg to parseopts()
2000-08-31 07:21:46 +00:00
4db6fc7542
Make need_resched() take a "struct cpu_info *" argument. This
...
causes gives a primitive form of processor affinity. Its use in
roundrobin() still needs some work.
2000-08-25 01:04:06 +00:00
2ea44b1dd4
install dosboot.com in the release(7) hierarchy.
2000-08-23 21:00:46 +00:00
58e7a6954b
Add spllock(). See spl(9) for details.
2000-08-22 19:46:26 +00:00
71cabc49dd
(rbus_pccbb_parent_mem): make the rbus minimum start address
...
configurable (RBUS_MIN_START), since default value of 1GB doesn't work
well on some ThinkPads
2000-08-22 08:42:55 +00:00
df316aa06c
Change the heap limit from 40000 to 50000 so the heap is the same size as it was before the "Hack from Hell'.
2000-08-22 02:46:30 +00:00
5755dca55d
Fix typos.
...
No functional changes.
2000-08-21 07:39:42 +00:00
23a7f255d4
Make sure we provide splsched() as described in spl(9).
2000-08-21 02:06:31 +00:00
a86d1f4891
Add a lock around the scheduler, and use it as necessary, including
...
in the non-MULTIPROCESSOR case (LOCKDEBUG requires it). Scheduler
lock is held upon entry to mi_switch() and cpu_switch(), and
cpu_switch() releases the lock before returning.
Largely from Bill Sommerfeld, with some minor bug fixes and
machine-dependent code hacking from me.
2000-08-20 21:50:06 +00:00
dd87a44f4a
Initialize PV head simple locks, done slightly differently than
...
in the sommerfeld_i386mp_1 branch.
2000-08-19 19:48:53 +00:00
b0dc085df7
Move the TSS selector out of the PCB and put it into mdproc. We
...
need to access this when we have the proclist locked for reading,
and thus cannot store it in the PCB (which may be swapped out).
As part of this, call pmap_activate() from cpu_switch() to switch
to the new address space, and refresh the PCB's copy of the LDT
selector from the pmap structure (see above paragraph). We need
to do this for MP support anyhow.
Fixes a "panic: spinlock_switchcheck: CPU 0 has 1 spin locks" via
gdt_compact() reported by Nathan Williams.
2000-08-16 04:44:35 +00:00
37262919a9
Edit adv0 comment to include "SCSI"; now 'grep -v -i SCSI GENERIC' works.
2000-08-15 20:09:26 +00:00
534b45c853
Move identifycpu to a place where it gets called earlier in the process,
...
so that special setup functions needed for some CPUs will be run before
some things (like UVM) are inited.
Add vm_page_zero_enable = FALSE to the cyrix 6x86 case, as page zeroing
while idle causes problems.
2000-08-15 18:21:44 +00:00
cabafb6c60
cast to u_long before casting to u_int to appease lint.
2000-08-13 22:45:01 +00:00
de0b1dd11d
Go to splhigh() before calling power hooks and maintain this level
...
while suspended. When waking up the power hooks are again called at
splhigh() and then the level is lowered.
This prevents interrupts from reaching a device before the power hook
has reinitialized it.
2000-08-13 22:26:27 +00:00
47b371580e
make this usable as INSTALL-like kernel suitable for boot floppy
...
enable supported stuff, add also SCSI devices (only sd* and cd* uncommented)
2000-08-13 21:55:33 +00:00
40dc315f3b
#if 0 bogus mca_intr_establish() added possibly mistakely
...
by cgd in rev. 1.2
2000-08-13 10:38:38 +00:00
85dda25e94
move "options PULLDOWN_TEST" into sys/sys/mbuf.h (in #ifdef _KERNEL),
...
as it is no wthe default setting for everyone.
the reason we still use the name "PULLDOWN_TEST" while it is now default:
kame code sharing.
2000-08-13 01:31:15 +00:00
e207174eb0
Add functions to read EISA configuration data for MEM, IRQ, DMA, and IO.
...
XXX Just error stubs on the i386 right now -- someone needs to write
XXX EISA BIOS code for i386.
2000-08-11 00:43:18 +00:00
enami