Commit Graph

3920 Commits

Author SHA1 Message Date
Alexander Kabaev
e99c09e2dc Eliminate global cons_unavailable flag and replace it by the status
bit maintained on a per-device basis. Single variable is inadequate
on machines running with multiple consoles enabled.
2004-02-05 01:56:43 +00:00
Peter Wemm
0765d4d608 Don't cast a pointer to an int that isn't big enough. 2004-02-05 00:12:08 +00:00
Peter Wemm
46d9072b4a Fix long/int printf format problems exposed by PMAP_DIAGNOSTIC 2004-02-05 00:11:05 +00:00
Peter Wemm
b132d96ea6 Checkpoint a NOTES file I had as of Nov 23rd. It doesn't quite compile
due to triggering some printf breakage in some DIAGNOSTIC printfs.
2004-02-04 22:53:50 +00:00
David E. O'Brien
02c4c7b81f Remove a device that will compile fine, isn't 64-bit clean. 2004-02-02 00:54:59 +00:00
Peter Wemm
3a0bfa76fd GRR. MFi386: white space spam 2004-01-30 03:02:45 +00:00
Peter Wemm
8a703d13b0 Merge some more changes from i386. 2004-01-30 00:24:45 +00:00
Peter Wemm
ac498460b2 Re-add debug register support.
Some other minor tweaks snuck in here, including supporting more
discontiguous memory segments and some cosmetic tweaks.
2004-01-29 00:07:29 +00:00
Peter Wemm
170a05510d Re-add user_dbreg_trap() for debug register support 2004-01-29 00:05:03 +00:00
Peter Wemm
db5272252f Take another shot at the invariants calls to __panic. They hadn't been
updated for the regparm ABI on amd64.
Context switch debug regs.
Update for fpu simplification
Don't needlessly reload %cr3, in case the cpu has the tlb flush filter
turned off.  Re-add LAZY_SWITCH stubs.
2004-01-29 00:02:54 +00:00
Peter Wemm
fdf00d6d6c deal with dbregs for fork etc
update for fpu.c simplification
Merge #include sort from i386
2004-01-28 23:58:55 +00:00
Peter Wemm
bb2fbc41af Un-stub the hardware debug register stuff. 2004-01-28 23:57:40 +00:00
Peter Wemm
eb18d248ee Export PCB_DR* symbols 2004-01-28 23:57:02 +00:00
Peter Wemm
1c89210c83 We can simplify a lot of things now that we don't have to worry about
hardware bugs on external 386 cpus and now that we can depend on SSE.
2004-01-28 23:55:58 +00:00
Peter Wemm
d957532a87 Add dbreg struct definitions for /proc/*/dbregs and a place to store the
registers in the pcb
2004-01-28 23:54:31 +00:00
Peter Wemm
1182b177ff Re-add debug register functions 2004-01-28 23:53:04 +00:00
Peter Wemm
b29fd7c4db MFi386: mp_topology(). 2004-01-28 23:51:16 +00:00
Peter Wemm
5e465ab919 MFi386: add THERMTRIP msr values 2004-01-28 23:47:22 +00:00
Peter Wemm
4612a96f08 Diff reduction with i386 2004-01-28 23:46:48 +00:00
Peter Wemm
d0f2d056fa MFi386: change an outb to a DELAY() 2004-01-28 20:46:31 +00:00
Alan Cox
11a61c16c2 MFi386 revision 1.230
- Move smp_topology to subr_smp.c so that it is defined on all architectures.
2004-01-25 06:19:37 +00:00
Jeff Roberson
048ac395be - Recruit some new ULE users by making it the default scheduler in GENERIC.
ULE will be in a probationary period to determine whether it will be left
   as the default in 5.3 which would likely mean the rest of the 5.x series.
2004-01-24 21:38:52 +00:00
Jacques Vidrine
5864cda7c6 Add PFIL_HOOKS to the GENERIC kernel configuration, primarily so
that one can load the IPFilter module (which requires PFIL_HOOKS).

Requested by:	Many, for over a year
2004-01-24 14:59:51 +00:00
Peter Wemm
ccaa41bc2c Unbreak amd64: Rename calls from panic to __panic 2004-01-23 01:04:28 +00:00
Poul-Henning Kamp
37f0ad870c remove elan_mmcr, I'm not sure I understand what it did here in the
first place.
2004-01-17 13:13:48 +00:00
Jacques Vidrine
e4dc8baa84 Provide sysarch(2) prototypes in the MD sysarch.h headers. While I'm
at it, use the ANSI C generic pointer type for the second argument,
thus matching the documentation.

Remove the now extraneous (and now conflicting) function declarations
in various libc sources.  Remove now unnecessary casts.

Reviewed by:	bde
2004-01-09 16:52:09 +00:00
Jacques Vidrine
3f6f39ff54 Remove `static' prototype from header file. 2004-01-06 20:36:21 +00:00
John Baldwin
1e4cad1384 Use i8259A register defines from shared header sys/dev/ic/i8259.h instead
of from the amd64-specific icu.h.
2004-01-06 19:05:25 +00:00
David Xu
a30ec4b99c Make sigaltstack as per-threaded, because per-process sigaltstack state
is useless for threaded programs, multiple threads can not share same
stack.
The alternative signal stack is private for thread, no lock is needed,
the orignal P_ALTSTACK is now moved into td_pflags and renamed to
TDP_ALTSTACK.
For single thread or Linux clone() based threaded program, there is no
semantic changed, because those programs only have one kernel thread
in every process.

Reviewed by: deischen, dfr
2004-01-03 02:02:26 +00:00
Alan Cox
f79f3abcbf - Use pagezero() instead of bzero() in pmap_pinit(). (pagezero() is much
faster.)
MFi386:
 - Don't bother clearing PG_ZERO on the page table page in
   _pmap_allocpte(); it serves no purpose.
 - Don't bother clearing and setting PG_BUSY on page table directory pages.
2004-01-01 07:08:52 +00:00
Mike Silbersack
ddeb5b242e Track three new sendfile-related statistics:
- The number of times sendfile had to do disk I/O
- The number of times sfbuf allocation failed
- The number of times sfbuf allocation had to wait
2003-12-28 08:57:09 +00:00
Mike Silbersack
5caf2b00f0 Move the declaration of sfbufspeak and sfbufsused to mbuf.h,
and use imax instead of max, as sfbufspeak and sfbufsused
are signed.

Submitted by:   bde
2003-12-28 01:43:22 +00:00
Mike Silbersack
5eda9873e9 Track current and peak sfbuf usage, export the values via sysctl. 2003-12-27 07:52:47 +00:00
Peter Wemm
c15e347e22 GC the unused <machine/kse.h> file. 2003-12-24 00:51:30 +00:00
Peter Wemm
9b68618df0 Add an additional field to the elf brandinfo structure to support
quicker exec-time replacement of the elf interpreter on an emulation
environment where an entire /compat/* tree isn't really warranted.
2003-12-23 02:42:39 +00:00
Alan Cox
925692caa5 - Significantly reduce the number of preallocated pv entries in
pmap_init().  Such a large preallocation is unnecessary and wastes
   nearly eight megabytes of kernel virtual address space per gigabyte
   of managed physical memory.
 - Increase UMA_BOOT_PAGES by two.  This enables the removal of
   pmap_pv_allocf().  (Note: this function was only used during
   initialization, specifically, after pmap_init() but before
   pmap_init2().  During pmap_init2(), a new allocator is installed.)
2003-12-22 01:01:32 +00:00
Alan Cox
35833d6a20 Since we have additional kernel virtual address space, allow the buffer
cache to grow to 400M bytes.
2003-12-20 20:03:10 +00:00
Peter Wemm
d50c65bfce MFi386: remove APIC_IRQ* defines that are no longer used. 2003-12-11 02:48:25 +00:00
Peter Wemm
bf342084e5 MFi386: (jhb): Deal with MAXCPU etc correctly 2003-12-11 02:47:35 +00:00
David E. O'Brien
70005bdb38 Add just enough of i386/include/pcvt_ioctl.h to amd64/include/pcvt_ioctl.h
such that 'ispcvt' can build.  Unforunately 'ispcvt' is needed in order for
/etc/rc.d/syscons to run.  This fixes the bug where I could not get my
keymap effective at boot.
2003-12-10 02:49:17 +00:00
Nate Lawson
cac6460cfe Use the ACPI-CA definitions for the various APIC tables instead of our
own.
2003-12-09 03:04:19 +00:00
Alan Cox
4d4a286cba Increase VM_KMEM_SIZE_MAX from 200MB to 400MB.
Discussed with:	peter
2003-12-07 04:51:04 +00:00
Peter Wemm
3db6450423 Reconfigure the runq macros to use the 64 bit ffs/bsf routines instead
of doing a loop and taking two 32 bit passes at the runqueue bits.  All
the 64 bit platforms should probably do this since there are 64 run queues.

Approved by:  re (scottl)
2003-12-06 23:24:18 +00:00
Peter Wemm
176ce2b1f4 Add 64 bit bsf*/ffs* routines. Have the ffs() inline use gcc's builtin
because it uses the better cmove instructions to avoid branches.
2003-12-06 23:22:43 +00:00
Peter Wemm
21616ec307 Various whitespace and cosmetic sync-up's with i386.
Approved by:  re (scottl)
2003-12-06 23:19:47 +00:00
Peter Wemm
90be5e4e7d amd64_protection_init and the protection_codes[] array was overkill.
Inline it instead.

Approved by:  re (scottl)
2003-12-06 23:18:42 +00:00
Peter Wemm
62b689edc9 Kill the ASM versions of the mtx_lock_spin and friends. They were never
used on amd64, and were actually totally broken.  They had the wrong
calling conventions.  I believe the i386 versions are going away too.

Approved by:  re (scottl)
2003-12-06 23:17:18 +00:00
Peter Wemm
cfef943682 MFi386: put the apic disable hook in a better place.
Approved by:  re (scottl)
2003-12-06 23:14:44 +00:00
Peter Wemm
42f5377536 Revert some amd64 changes that cached curthread and converge back to the
i386 version.  The curthread special case in pcpu.h solves my complaint
about the verbose macro expansion in this case.  Note that the i386
version still has some OBE comments, I didn't re-add them back again.

Approved by:  re (scottl)
2003-12-06 23:13:22 +00:00
John Baldwin
b6c71225a9 Fix all users of mp_maxid to use the same semantics, namely:
1) mp_maxid is a valid FreeBSD CPU ID in the range 0 .. MAXCPU - 1.
2) For all active CPUs in the system, PCPU_GET(cpuid) <= mp_maxid.

Approved by:	re (scottl)
Tested on:	i386, amd64, alpha
2003-12-03 14:57:26 +00:00