f5e8243c9d
Use the COMPAT_50 quotactl, not the native one.
...
Should fix build of COMPAT_AOUT_M68K
2011-03-07 19:26:35 +00:00
638f8519ad
Bump SYMTAB_SPACE.
2011-03-07 15:56:26 +00:00
0582307bb5
Bump SYMTAB_SPACE.
2011-03-07 15:49:50 +00:00
d6eff8b7ce
Bump SYMTAB_SPACE.
2011-03-07 15:37:46 +00:00
97ea5c269b
fix typo, and use the parent to determine the offset. From Dave Mills
2011-03-07 13:58:32 +00:00
f2fbc0de6d
Get improvements from FreeBSD-current (January 2011).
...
Successfully tested with WEP, WPA and WPA2 on AR9285.
Reviewed by jmcneill@
2011-03-07 11:25:41 +00:00
5c3ccd6eba
Pass down ELF Auxillary Vectors for static NetBSD binaries too.
...
Rename __libc_init to _libc_init and call it explicitly from CSU code.
This enforces the constructor run order for newly linked programs.
Keep it as constructor with run-once semantic for binary compatibility.
Implement dl_iterate_phdr for statically linked programs.
This material is based upon work partially supported by
The NetBSD Foundation under a contract with Joerg Sonnenberger.
2011-03-07 05:09:09 +00:00
974d4f3d46
Use macros clts/stts instead of directly manipulating CR0 flags.
...
Expose fpuinit to XEN build. (remove #ifdef XEN)
2011-03-07 02:24:57 +00:00
948ee160e6
Renamed "stg" to "stge" as in the kernel.
2011-03-06 20:36:29 +00:00
ac93dee5bb
Now that MACHINE is defined in user-land in either
...
mips/include/mips_param.h or powerpc/include/param.h, stop
trying to re-define it in the port-specific param.h files
when _KERNEL isn't defined.
2011-03-06 20:34:55 +00:00
fa5740cd0f
Suport for loading a kernel from memory (RAM, ROM or Flash).
...
The boot file path should look like "mem:<address>", where <address> is
the start address, in hex notation, of the file in memory.
2011-03-06 18:22:13 +00:00
063f96f3c2
merge the bouyer-quota2 branch. This adds a new on-disk format
...
to store disk quota usage and limits, integrated with ffs
metadata. Usage is checked by fsck_ffs (no more quotacheck)
and is covered by the WAPBL journal. Enabled with kernel
option QUOTA2 (added where QUOTA was enabled in kernel config files),
turned on with tunefs(8) on a per-filesystem
basis. mount_mfs(8) can also turn quotas on.
See http://mail-index.netbsd.org/tech-kern/2011/02/19/msg010025.html
for details.
2011-03-06 17:08:10 +00:00
b5c8479490
don't abuse cf_unit, requested by phone.
2011-03-06 17:00:16 +00:00
e1901f8e74
Remove trailing spaces.
2011-03-06 14:58:42 +00:00
dcaf78e5b9
Remove trailing spaces and TABs.
2011-03-06 14:54:46 +00:00
ec6990c143
Remove extra TAB.
2011-03-06 14:52:37 +00:00
094e0e9bb0
Remove trailing spaces.
2011-03-06 14:51:21 +00:00
b79587cca2
New experimental driver for SundanceIT ST1023 / IP1000+ NICs.
...
PHY initialization, media select and MAC address are working, but I found no
way to make the chip transmit any frame yet (although it clears the DONE flag).
Moved DSK_DECL to globals.h, where NIF_DECL already was.
2011-03-06 13:55:12 +00:00
7e5272d701
Bump SYMTAB_SPACE for the many other evbarm kernels that also need it increased.
2011-03-06 05:08:03 +00:00
f09748f46c
{ffs_nodealloccg,ext2fs_nodealloccg,ext2fs_mapsearch}: use XOR and ffs()
...
to find free bits in the inode and block bitmaps, instead of the loop.
Obtained from FreeBSD (changes by jhb).
2011-03-06 04:46:26 +00:00
ae3b98c1b5
In a case of direct select, set only masked events, do not wakeup LWP
...
if no polled/selected events were set; also, count the correct return
value for the select.
2011-03-06 04:41:58 +00:00
0df36b27f7
Remove trailing spaces.
2011-03-06 02:22:27 +00:00
db89ee153d
Bump SYMTAB_SPACE.
2011-03-06 02:18:36 +00:00
3fb7a604bd
Fix compilation
2011-03-05 23:51:47 +00:00
f795a05e24
Disable COMPAT_IRIX and related code. The usefulness is under discussion
...
and code that dereferences pointer into userland without proper checking
needs a full audit.
2011-03-05 23:38:40 +00:00
c0fcf73bd6
Set the ENVSYS_FCHANGERFACT flag soon enough for it to make a difference.
2011-03-05 22:27:21 +00:00
405a0ee48c
Use %zu for size_t in debugging printf.
2011-03-05 21:51:17 +00:00
65734ab916
Make interrupt establishment failure be noted like most other PCI drivers do.
2011-03-05 21:43:38 +00:00
ae52eb0e4e
Correct #endif comment.
2011-03-05 21:38:34 +00:00
557eb4273a
Forward define of ps_strings.
2011-03-05 19:08:40 +00:00
8541527a4c
p_psstr -> p_psstrp
2011-03-05 19:04:58 +00:00
0b4b684637
From Dave Mills: It would appear that the two PHYs can only be
...
accessed from the registers of controller0 (mvgbec0) hence why both
PHYs are being detected on controller0. I have made the assumption
that PHY0 => controller0 (mvgbec0), PHY1 => controller1(mvgbec1)
and developed a patch accordingly. Also the irq is 15 for controller1
not 14.
2011-03-05 18:28:13 +00:00
e7a3ba5387
Regen.
2011-03-05 15:36:50 +00:00
06de72bde8
Add ATHEROS AR2427 and AR9287
2011-03-05 15:36:27 +00:00
0edd86cf19
Only call unmapiodev if va is non-zero.
2011-03-05 15:25:52 +00:00
1923f983b9
Add missing .debug lines from usr/libdata/ldscripts
...
Add missing .mdebug lines from usr/libdata/ldscripts
Fixes PR/40522
2011-03-05 15:19:44 +00:00
0ccdc9a448
Test and set xbdi->xbdi_cont at splbio(). Otherwise we could overwrite
...
xbdi->xbdi_cont and process the same request twice.
2011-03-05 15:12:16 +00:00
8d7cb782a7
Add contributed Marvell Kirkwood RTC driver from Brett Slager
...
Fixes PR 44004
2011-03-05 14:50:56 +00:00
6f6097d0cf
add defflag for AR9285.
...
Spotted by jmcneill@
2011-03-05 14:42:11 +00:00
e3bce9dd91
If _KERNEL is not defined, force MACHINE to be "m68k". Userland should be
...
using uname/sysctl to get this value.
2011-03-05 14:32:32 +00:00
a3766dd8f3
If _KERNEL is not defined, force MACHINE to be "powerpc". Userland should be
...
using uname/sysctl to get this value.
2011-03-05 14:27:48 +00:00
5152fa4e3d
If _KERNEL is not defined, force MACHINE to be "mips". Userland should be
...
using uname/sysctl to get this value.
2011-03-05 14:26:01 +00:00
a0882e35a9
Fix prototype of algor_p6032_iointr.
...
Fix buffer overrun ( PR algor/42176 ).
Constify a few things.
2011-03-05 14:04:02 +00:00
cbbbf64494
Increase SYMTAB_SPACE
2011-03-05 14:02:08 +00:00
6315c7d63e
Add __cpu_simple_lock_t. Use it, x86_read_psl(), and x86_disable_intr() to
...
disable interrupts locally and protect the access to APERF and MPERF. Also
rationalize the MD initialization sequence.
2011-03-05 09:47:19 +00:00
b3a05fed5a
If the P-state control mask is set, do a proper read-modify-write.
2011-03-05 06:39:55 +00:00
78d39fa3b6
KNF.
2011-03-05 06:28:29 +00:00
8871ccf0f3
Fix spelling of MKZFS
2011-03-05 03:15:25 +00:00
d9f40f2741
Use deflate.sh from the right directory. Execute with HOST_SH.
2011-03-05 01:58:01 +00:00
78c213b247
Simplify copyin_psstrings a bit.
2011-03-05 01:52:18 +00:00
48717cfc00
Refactor ps_strings access. Based on PK_32, write either the normal
...
version or the 32bit compat layout in execve1. Introduce a new function
copyin_psstrings for reading it back from userland and converting it to
the native layout. Refactor procfs to share most of the code with the
kern.proc_args sysctl handler.
This material is based upon work partially supported by
The NetBSD Foundation under a contract with Joerg Sonnenberger.
2011-03-04 22:25:24 +00:00
01b117db0c
Remove bogus RF_ASSERT that must have crept in by accident.
...
Thanks to Jan-Hinrich Fessel for reporting the issue.
2011-03-04 17:45:17 +00:00
7fcafe3a34
Revert max stack size change. This is not used anymore for 32 bit binaries.
2011-03-04 13:24:19 +00:00
4e1517c4e6
Rename a badly named constant. Make it correspond with <x86/specialreg.h>.
2011-03-04 12:10:49 +00:00
a9a815071b
Only attach on the first SMT ID (as in revision 1.16).
2011-03-04 11:56:27 +00:00
d87d6f14e7
Remove redundant/wrong increment of a variable.
2011-03-04 09:28:34 +00:00
120b4f01cf
Add module build glue for odcm(4).
2011-03-04 05:02:47 +00:00
70fcc95611
Raise the return value of the match-function of est(4) and powernow(4).
...
The assigned priorities are now: 10 for acpicpu(4), 5 for est(4) and
powernow(4), and 1 for odcm(4). These are used to pick the preferred driver.
2011-03-04 04:53:28 +00:00
839fb5fa76
Move INTEL_ONDEMAND_CLOCKMOD -- or odcm(4) -- to the cpufeaturebus.
2011-03-04 04:48:39 +00:00
72233ad7fa
PR/44673: Arna Clauson: Latest MAXSSIZ bump broke netbsd32 emulation on amd64.
...
- Use MAXSSIZ32 instead of MAXSSIZ for 32 bit binaries
- Default MAXXSIZ32 to a quarter of MAXSSIZ (good enough?)
- Add debugging
XXX: Note that:
- sparc32 MAXSSIZ is larger than sparc64 MAXSSIZ
- sparc64 MAXSSIZ32 != sparc32 MAXSSIZ
2011-03-04 04:25:58 +00:00
1428cd07df
better debugging messages.
2011-03-04 04:17:12 +00:00
22701ee034
Reduce MAXSSIZ to 64MB, otherwise netbsd32 binaries crash in ld.elf_so,
...
including the trivial main(){}. Add a warning to not modify this without
testing compatibility mode.
2011-03-04 03:34:24 +00:00
af8f930405
PR/44674: Taylor R Campbell: Fix compat copyin and copyout routines which
...
were obviously broken.
2011-03-04 01:36:56 +00:00
ccc5092ed2
Do not restrict the siginfo a process sends to itself in any way, but
...
keep restrictions for foreign processes. This is needed to allow raising
of full SIGFPE siginfo from softfloat libraries for example.
2011-03-03 20:46:16 +00:00
80dcd7f76f
Add DMI quirk support via pmf_get_platform(9). If any of the listed models
...
are matched, the whole driver will be prevented from attaching. The first
entry is Supermicro PDSMi-LN4+ (a BIOS bug with bogus P-state entries).
2011-03-03 19:24:43 +00:00
d4bc3823a0
Change MIPS_CP0FL_CONFIG*
...
Add MIPS_CP0FL_HWRENA and USERLOCAL
2011-03-03 18:44:58 +00:00
1ec0684c4b
Deal with cr27 in process_{read,write}_regs
2011-03-03 09:37:21 +00:00
cd8c039974
In rename, use char[12] for new names instead of [11]. At least
...
one routine called from here (unix2dosfn) expects and uses all of
a [12].
This may fix the "stack size exceeded" problem which has been
triggering in gson's test runs. (i'm not entirely sure why it
doesn't trigger in anyone else's env)
2011-03-03 08:10:45 +00:00
1caa7e3dc2
Add support boot. only support SH4 machine(mmEye-WLF and others) now.
2011-03-03 05:59:37 +00:00
b1b063f09f
check rawvp before doing ioctl or strategy.
2011-03-03 03:39:08 +00:00
5469ac9d1c
When probing the PCI bars, make sure to only probe BARS (PPBs and PCB has
...
fewer bars than type 0 devices).
2011-03-02 21:57:40 +00:00
a8c2005e67
Make this compile with PMAP_POOLPAGE_DEBUG.
...
XXX: not sure if this is ever used
2011-03-02 13:26:41 +00:00
67365afb80
We track page modified info with PG_CLEAN, so make clear_modify
...
return false. This makes rump lfs unmount work on platforms which
use the pmap stub (i.e. non-x86, which already returned false here).
Otherwise, lfs would hang itself trying to flush some buffers but
couldn't fill a segment and therefore wouldn't actually write
anything.
2011-03-02 13:11:52 +00:00
5106307ad8
Fix misplaced #endif. PR port-arc/44664
...
XXX: I'll also remove ENABLE_INT5_STATCLOCK soon since no visible improvements.
2011-03-02 10:51:04 +00:00
17bee7a3e6
Resolve build failures caused by local use of "device flash" in config.
2011-03-02 08:25:08 +00:00
3c37c8c0bc
Adjust the detection of Turbo Boost to prevent a theoretical array OOB access.
2011-03-02 06:23:17 +00:00
5bfb3ade6b
Append Intel's Turbo Boost to the debug printfs if we detect it.
2011-03-02 06:17:08 +00:00
213d0d6de1
Always return at least one parity region. Fixes PR#44239.
...
Fix from oster@, code (such as it is) by me.
2011-03-01 22:51:14 +00:00
5bf2d32443
use hardware to draw characters, while there use vcons_replay_msgbuf()
2011-03-01 21:47:13 +00:00
20a3edc3f6
Ensure that the LWP private area has a consistent value after exec.
2011-03-01 18:53:10 +00:00
da0742f9b8
Reset node's parent pointer when it's removed. Technically the
...
parent still exists, but allows us to avoid complicated g/c algorithms
if the parent *is* removed.
2011-03-01 15:14:35 +00:00
5b407ea29a
Detach evcnt(9) before the states.
2011-03-01 13:38:45 +00:00
405dec72d6
Pass accurate protection info from ubc_uiomove() to the pager.
...
Fixes nfs{,ro}_fileio tests on at least sparc64 (and probably macppc
and other fat endian machines).
The problem was that nfs was fooled to thinking read() caused a
write fault because of VM_PROT_WRITE being unconditionally set and
therefore set NMODIFIED on a r/o file system. It is absolutely
beyond me why the test worked on i386/amd64. Incidentally, I seem
to have "misplaced" a few goats.
2011-03-01 10:02:11 +00:00
1b9ef66f08
Revert the major number for satmgr(4) back to 100 and allocate it from
...
majors.powerpc, as suggested by matt@.
2011-03-01 09:17:06 +00:00
e98440f29b
Move also the evcnt(9) event counters to one place. No functional change.
2011-03-01 05:57:04 +00:00
2562de9c62
Use __arraycount(3).
2011-03-01 05:37:02 +00:00
b9f7a5accd
Simplify by moving the debug printfs to one place. No functional change.
2011-03-01 05:32:03 +00:00
83e392c16f
Remove the cross-call from the APERF/MPERF -function.
2011-03-01 05:02:16 +00:00
c9111546fe
Move the xcall(9) that does the P- and T-state transformations from the MD
...
layer to the main code. Makes the caches coherent and provides consistent
vmstat(1) output. This is still not quite right, given that most of the
cross-calls are typically unnecessary with the dependency coordination.
2011-03-01 04:35:48 +00:00
01a744c41e
Add the trapframe pointer to the syscall information.
2011-02-28 21:24:25 +00:00
3c6d72d744
__HAVE___LWP_GETPRIVATE_FAST works now.
2011-02-28 21:23:33 +00:00
641cef61dc
Don't use %tr3 (aka cr27) in emu trap. cr27 is now used for lwp private
...
info.
2011-02-28 21:22:35 +00:00
0dad921873
Remove old/incorrect comment.
2011-02-28 21:20:50 +00:00
0437d56b01
Make error checking consistent, possibly fixes PR/44652.
2011-02-28 18:28:20 +00:00
44bf8be11e
Update firmware sets for the (rare) 2322, 2400 and 2500. In particular the
...
latter two sets have boatloads of bug fixes.
2011-02-28 17:19:43 +00:00
29e9eb7163
Update isp driver to be in sync with other platforms. Mostly very minor changes
...
with effectively zero impact on NetBSD.
2011-02-28 17:17:55 +00:00
a110815e0b
Use le16enc()/be16enc() to encode LE/BE values back into the audio stream.
...
This fixes the tests/dev/audio test on my macppc box.
While I'm here, use le16dec()/be16dec() directly instead of rewriting them.
2011-02-28 16:56:39 +00:00
8281676ab4
lom_refresh():
...
Update only the sensor status specified by the edata as noted in
sysmon_envsys(9).
lom_sysctl_alarm():
Update alarm status before reading via sysctl to make it usable at
a boot time.
2011-02-28 14:29:29 +00:00
ad600d617b
use ARM32_MMAP_WRITECOMBINE when mmap()ing video memory for improved X11 speed
2011-02-28 10:05:04 +00:00
7b3a5e8aef
implement arm32_pmap_flags() to allow mappings with write buffering enabled,
...
mostly for video memory
Tested on shark
2011-02-28 10:03:49 +00:00
15e33f3bf4
Revert previous, which doesn't cover all the cases if F_OK isn't 0,
...
and just CTASSERT that it is, as that's not remotely likely to change.
Per source-changes-d; ok by Christos.
2011-02-28 03:23:44 +00:00
6f2f866802
don't depend on F_OK being 0.
2011-02-28 00:12:15 +00:00
45b74603dd
Check for bogus flags to access() up front. Otherwise we end up
...
calling VOP_ACCESS with flags 0 and something asserts deep in the
bowels of kauth. PR 44648 from Taylor Campbell. (I moved the check
earlier relative to the suggested patch.)
Pullup candidate.
2011-02-27 23:06:40 +00:00
cd966b244a
Provide MD wrappers for match and attach.
2011-02-27 18:32:53 +00:00
c5d6c4cef0
follow the lead of genfb_pci and call pci_mapreg_probe before pci_mapreg_info
2011-02-27 18:10:25 +00:00
50f3ce9fa2
revert previous; instead of commenting out the error message how about fixing
...
the code that causes it to be printed in the first place
2011-02-27 18:01:28 +00:00
cf0085d7f5
Remove the following aprint_debug(9): "pci_mem_find: void region". I have
...
a system where this is printed about a hundred times. No functional change.
2011-02-27 17:31:08 +00:00
adec1f4f3c
Claim to support the dependency coordination during the _PDC/_OSC query.
...
(Although we do not actually support it.) Only after these bits are set,
many Intel-based BIOSes are willing to relinquish the necessary information.
2011-02-27 17:27:28 +00:00
494badeb16
Move acpicpu(4) from "acpinodebus" to "cpufeaturebus".
2011-02-27 17:10:33 +00:00
9bd36d4186
tmpfs has two layers of uvm objects (vnode->uobj and the anon object
...
in tmpfs_node), so when playing with pages make sure we lock the
uvm object the pages belong to instead of the vnode's uvm object.
per test from Nicolas Joly (which I'm sure he will commit soon ;)
2011-02-27 13:37:39 +00:00
355c56ede4
Missing semicolon.
2011-02-27 09:01:56 +00:00
f99cc07e58
use pa_intrpin instead of pa_intrline for irq#, since drivers treat irq=0
...
as "no irq" and this matches pa_intrpin encoding
2011-02-27 01:02:39 +00:00
ed7e15e7c6
Can't pass a void.
2011-02-26 22:33:45 +00:00
efac3e47ba
Build altboot.img, which fakes a Linux kernel module, loadable with "bootm",
...
for extremely stripped U-Boot firmware. Arguments are passed through the
"bootargs" environment variable, which is detected automatically when
using bootm.
The startup code also fixes a bug in bootm, which doesn't flush the cache
before running the image.
2011-02-26 20:11:24 +00:00
a6c55fbbe4
Enable LIBSA_PRINTF_LONGLONG_SUPPORT.
...
Useful in any of the cases where we already print a (64-bit) daddr_t.
2011-02-26 18:22:58 +00:00
bdf5f7401d
Add quad srcs if building x86_64 32-bit libkern.
2011-02-26 18:17:55 +00:00
2b6ee22130
Import the Flash and NAND subsytem code contributed by the University
...
of Szeged, Hungary.
The commit includes:
- Flash layer, which gives a common API to access flash devices
- NAND controller subsystem for the flash layer
- An example OMAP driver which is used on BeagleBoard or alike ARM boards
2011-02-26 18:07:13 +00:00
7fe5f9cb4c
Add quad srcs if using MIPS O32 ABI
2011-02-26 16:28:10 +00:00
953a720806
Make sure that sgimips64 can build the O32 bootloaders
2011-02-26 16:26:58 +00:00
3d7fbda8d9
Use mips_fpexcept() instead of fpemul_trapsignal() to deliver SIGFPE,
...
and remove now unused fpemul_trapsignal() introduced for PR port-mips/26410.
Fixes PR port-mips/35326 and now t_except unmasked tests in
/usr/tests/lib/libc/ieeefp pass.
Note t_subnormal double test still fails as mentioned in PR port-mips/44639.
2011-02-26 15:41:32 +00:00
ad8a961da7
Use proper CALLFRAME_FRAME and CALLFRAME_CAUSE macro.
...
XXX: is it safe to use (CALLFRAME_SIZ + 3*SZREG(sp)) to save FSR?
2011-02-26 15:01:31 +00:00
4939599599
Use config_defer(9) for cpu_rescan() in cpu_attach().
...
Also mark few local functions as static.
2011-02-26 14:43:18 +00:00
c267011572
- clear MIPS_FPU_EXCEPTION_BITS in MIPS_FPU_CSR in SIGILL case
...
as noted in commit log of rev 1.158
- update comment to reflect changes in rev 1.109
2011-02-26 13:58:34 +00:00
08938afcbd
Mechanically adapt to new interrupt/spl framework. Untested.
2011-02-26 12:54:41 +00:00
45a7a81b2b
Explicitly include <mips/locore.h> for softint_process().
2011-02-26 12:09:34 +00:00
b543d3edb5
Explicitly include <mips/locore.h> for mips_locoresw.
2011-02-26 12:08:30 +00:00
69d0ae1d3d
Explicitly include <mips/locore.h> for wbflush().
2011-02-26 12:07:45 +00:00
beb830523e
No need to include <mips/locore.h> here.
2011-02-26 12:04:25 +00:00
0a73905435
Use mips_fpuillinst() instead of fpemul_trapsignal() to deliver SIGILL.
2011-02-26 12:02:01 +00:00
dbe74b603b
Um, it's mips_fpuillinst(), not fpemul_fpuillinst().
2011-02-26 11:27:59 +00:00
191091a034
Use fpemul_fpuillinst() instead of fpemul_trapsignal() to deliver SIGILL.
2011-02-26 11:16:12 +00:00
d26a5ea969
mips_fpu_trap() no longer passes pc to mips_fpuillinst().
...
Use _R_PC value in trapframe instead for ksi_addr of siginfo.
2011-02-26 11:05:54 +00:00
40434da6a3
- #if DEBUG -> #ifdef FPEMUL_DEBUG
...
- use __func__ to print function name
- add debug printf()s in mips_fpuexcept() and mips_fpuillinst() too
2011-02-26 10:56:56 +00:00
a1b66a73cd
Changed satmgr(4) device major number from 100 to 144, which is reserved
...
for local/vendor use according to src/sys/conf/majors. This prevents
problems when the shared PowerPC device majors list gets another entry.
2011-02-26 09:52:16 +00:00
f40a2de6c2
Fix comment about mips_emul_fp() call.
2011-02-26 09:47:24 +00:00
5a6623c3f0
Whitespace.
2011-02-26 07:54:48 +00:00
08ab7fd006
Add a comment.
2011-02-26 07:53:47 +00:00
80322835aa
(Attempt to) bundle some stw/ldws and remove redundant CR_TLS save.
2011-02-26 07:52:32 +00:00
a64b689a43
Include <uvm/uvm_extern.h>.
...
s/tf->/tf->tf_/.
2011-02-26 05:25:27 +00:00
4ad49e742f
s/tf_tf_/tf_/.
2011-02-26 05:23:22 +00:00
d42a19b052
Don't rely on having a mips64eb toolchain.
2011-02-25 23:37:32 +00:00
1879cdcf5a
mi_userret: disable an assertion for __HAVE_PREEMPTION case as it's racy.
2011-02-25 22:37:12 +00:00
04a6913f3d
tprof_start: don't forget to restore refcount when failed to start backend.
2011-02-25 22:35:38 +00:00
0db13a33ab
Comment the coordination types.
2011-02-25 20:59:37 +00:00
b4da53f1e6
make the use of SHA2-HMAC by FAST_IPSEC compliant to current standards:
...
-RFC2104 says that the block size of the hash algorithm must be used
for key/ipad/opad calculations. While formerly all ciphers used a block
length of 64, SHA384 and SHA512 use 128 bytes. So we can't use the
HMAC_BLOCK_LEN constant anymore. Add a new field to "struct auth_hash"
for the per-cipher blocksize.
-Due to this, there can't be a single "CRYPTO_SHA2_HMAC" external name
anymore. Replace this by 3 for the 3 different keysizes.
This was done by Open/FreeBSD before.
-Also fix the number of authenticator bits used tor ESP and AH to
conform to RFC4868, and remove uses of AH_HMAC_HASHLEN which did
assume a fixed authenticator size of 12 bytes.
FAST_IPSEC will not interoperate with KAME IPSEC anymore if sha2 is used,
because the latter doesn't implement these standards. It should
interoperate with at least modern Free/OpenBSD now.
(I've only tested with NetBSD-current/FAST_IPSEC on both ends.)
2011-02-25 20:13:10 +00:00
a080ec141b
Start to derive the percpu(9) (or per-domain) state coordination
...
mechanisms by parsing the _CSD, _PSD, and _TSD objects by default.
2011-02-25 19:55:06 +00:00
1177b3daa4
Fix LOCKDEBUG spin lock held panic on MIPS_CACHE_VIRTUAL_ALIAS CPUs.
...
Now cobalt boots multiuser, but many programs complains
"clntudp_create: RPC: Program not registered"
as well as R5000 sgimips O2. Smells virtual cache alias issue around pool...
2011-02-25 19:32:51 +00:00
7559b861a8
Use AUDIO_DEVICE instead of 0 as the minor number for /dev/audio since
...
0 is incorrect. While I'm here, add /dev/sound, audioctl, and mixer too.
ok pooka@
2011-02-25 19:27:05 +00:00
4d59265f95
Don't autogenerate a large number of unnecessary device nodes, just
...
slows bootstrap.
2011-02-25 18:56:20 +00:00
0f6b94e451
Fix an oversight; the APERF and MPERF counters are per-CPU, so also reset
...
these by broadcasting to all CPUs with x86_msr_xcall(9).
2011-02-25 17:23:34 +00:00
bouyer