e2f2b17e31
add the new compat header.
2005-11-10 18:47:16 +00:00
60305c1ba1
fix debugging.
2005-11-10 18:45:36 +00:00
032305b8bd
fix debugging; remove some defines that are now in the header file.
2005-11-10 18:45:20 +00:00
75fdf989ea
These two really don't need the new header.
2005-11-10 18:43:32 +00:00
2a4a5534b4
Include new compat header.
2005-11-10 18:39:30 +00:00
0b0496e3d5
move the compat code out of the sys file.
2005-11-10 18:36:20 +00:00
4098e6b83d
expose some more internal implementation constants and variables used by
...
emulations.
2005-11-10 18:35:15 +00:00
65de380a95
More ipc support for linux; makes oracle work. From chuq and
...
jlrodriguez at terra dot es
2005-11-10 18:33:37 +00:00
a7ccc64926
Add an obviously missing break in a switch statement.
...
Adjusting brightness and contrast now works!
2005-11-10 15:53:05 +00:00
8fa905e040
Check possibility of virtual alias correctly in primaly_cache_is_2way case
...
(though I guess all those CPUs have L1 cache larger than 8KB).
2005-11-10 15:31:17 +00:00
2fa7cb2348
Remove redundant assignment (from Liam Foy)
2005-11-10 13:40:38 +00:00
07c827131f
regen
2005-11-10 13:31:03 +00:00
7928f127e6
Add NSC IrDA Fast Infrared Port.
2005-11-10 13:30:46 +00:00
f5602021b6
regen
2005-11-10 13:26:34 +00:00
066abf15d0
Add Synaptics PS/2-style Touchpad.
2005-11-10 13:25:59 +00:00
65ee9092eb
Abuse types a little less.
2005-11-10 12:05:01 +00:00
92fa4843f1
Give THUMB_CODE its own 'struct undefined_handler' so we don't try to
...
install the same handler structure twice (for difference coprocs).
Spotted by dbj@ using QUEUEDEBUG.
2005-11-10 11:18:55 +00:00
f55fe966f7
Apply big endian fixes submitted by Garrett D'Amore in PR kern/32032.
2005-11-10 10:00:19 +00:00
af56e26cc6
partially revert previous commit stripping arm mapping symbols
...
it broke ddb when DEBUG wasn't set to -g.
2005-11-10 09:42:42 +00:00
e75bcb9ebf
Fix build problem
2005-11-09 21:56:11 +00:00
261d102fbb
Typo in comment.
2005-11-09 17:54:12 +00:00
f98d5d856e
Improve Linux exit_group emulation by sending a SIGKILL to all other
...
processes in the group instead of letting them live. It would be better
to properly terminate them.
2005-11-09 14:56:50 +00:00
f53c4d1d1e
Prevent integer overflow in timout calculation.
...
changed DEBUG_LINUX to DEBUG_LINUX_FUTEX so that we have less debug
messages on the console
2005-11-09 14:52:18 +00:00
155c8bd666
Whitespace nit.
2005-11-09 12:47:39 +00:00
807ce4c6fc
tmpfs_reg_resize: drop swap slots on truncation.
...
fix a problem pointed by jmmv@.
2005-11-08 23:04:03 +00:00
9df1f09b91
add a function to drop all swap slots in a given range. for tmpfs.
...
XXX maybe it's better to implement true truncation.
2005-11-08 23:02:22 +00:00
4868e1f967
add missing IPL_SOFTXENEVT.
...
pointed by Hisashi T Fujinaka on current-users@.
2005-11-08 22:57:30 +00:00
67a3f47423
Add a fake entry in pms_protocols array to avoid an out-of-bounds
...
access in do_enable() if PMS_SYNAPTICS_TOUCHPAD and PMSDEBUG are both
defined.
2005-11-08 22:12:16 +00:00
26222780eb
Correctly computes futex timeout.
2005-11-08 21:28:49 +00:00
96bd7a6fc4
Partially revert change in sys/arch/mips/mips/cache.c rev 1.27,
...
i.e. override mips_cache_alias_mask and mips_cache_prefer_mask
with values which match MIPS3_MAX_PCACHE_SIZE (32KB), rather than
leave them actual primary virtual indexed cache size (8KB or 16KB).
Also add comments about what the value means there.
I thought the VCE on R4000/R4400 occurred only if actual virtual alias
was detected because there was an article which mentioned that VCE
detection logic was different according to primary cache size and
it looked reasonable. But all other articles I can find later
(http://www.linux-mips.org/archives/linux-mips/1998-05/msg00084.html etc.)
claimes that VCE detection logic always verifies all 3 bits
between vaddr[14:12] and PIdx[2:0] regardless of primary cache size
(i.e. VCE could occur even if there is no actual virtual alias), and
in fact VCED still happens with the mask values adjusted for 16KB L1
but it doesn't with ones for MIPS3_MAX_PCACHE_SIZE on my R4400 news5000.
2005-11-08 15:31:10 +00:00
4dafa089d4
vput() -> vrele(). Vnode is already unlocked.
...
With much help from Pavel Cahyna.
Fixes PR 32005.
2005-11-08 11:35:51 +00:00
06cf1c7649
copy.S should belong to MD_SFILES, not MD_CFILES.
2005-11-08 01:02:39 +00:00
fe35cfd50c
Regen from pcmciadevs rev 1.213:
...
> Fix pp_cisinfo[0] of the TEAC IDECARDII entry.
> If pp_cisinfo[0] is NULL, current pcmcia_product_lookup()
> never matches the device.
2005-11-07 16:39:16 +00:00
cf09fb3893
Fix pp_cisinfo[0] of the TEAC IDECARDII entry.
...
If pp_cisinfo[0] is NULL, current pcmcia_product_lookup()
never matches the device.
2005-11-07 16:37:28 +00:00
d72135b371
Include linux_machdep.h so that LINUX_UNAME_ARCH ges defined (fixes
...
uname emulation)
2005-11-07 14:17:45 +00:00
443e88c5cf
some assym cleanup.
...
- move copyin and friends from locore.S to their own file, copy.S.
share it between i386 and xen.
- defparam KERNBASE and kill KERNBASE_LOCORE hack.
- add more symbols to assym.h and use it where appropriate.
2005-11-07 11:42:34 +00:00
83c44381ee
Xsoftnet: remove an unnecessary assignment.
2005-11-07 08:52:07 +00:00
a5d2387dbb
regen
2005-11-06 21:50:28 +00:00
80c644a6b9
Explicitly default the input baud rate (c_ispeed) to the output baud rate (c_ospeed)
...
instead of setting it to zero. Otherwise nothing ever sets it unless some code
explicitly changes the baud rate. For a serial console (in particular) we want
to use the baud rate set by the bios (or whatever) and used by theboot code.
This is the way it was before a 'new version of com driver' was added in 1997 (rev 1.99)
2005-11-06 21:34:37 +00:00
7505cb92a6
Back out sched_{set|get}affinity addition because it breaks the
...
build again on i386 now it is fixed for other archs (Tom's fix and
mine got in colision)
2005-11-06 18:16:31 +00:00
5332333501
in amap_cow_now(), handle the case where we have to sleep and some of the
...
already-copied pages are paged out. anons that have already been copied
will have refcount == 1, whereas anons that still need to be copied will
have refcount > 1. fixes PR 25392, PR 30257, PR 31924.
2005-11-06 15:57:32 +00:00
0c24d80f02
Regen with correct RCS Ids.
2005-11-06 14:25:48 +00:00
7d2d9d6fe1
Bump MAXPHYS from 32K to 64K.
...
asc.c has now its own asc_minphys(), and both osiop and oosiop at jazzio
work fine with 64K MAXPHYS.
2005-11-06 11:15:40 +00:00
5c3034f5b3
- <machine/param.h> says:
...
"XXX Maxphys temporary changed to 32K while SCSI driver is fixed."
I (and soda) don't know what the "SCSI driver" is, but add local
asc_minphys() to limit max xfer size only for asc.c for now.
- According to old asc.c driver, DMA_START() is called before NCR_CMD_DMA
command. I'm not sure which DMAC or ASC should be started first, but
move DMA command from asc_dma_go() to asc_dma_setup().
- Disable NCR_F_DMASELECT (which allows DMA xfer for select command)
since it caused some problem on PICA.
Now gxemul-0.3.6.2 reaches mountroot() sd0 (but can't exec /sbin/init yet).
2005-11-06 11:09:17 +00:00
0f4f57561c
Add pcscp (aka DC-390) at pci, which is confirmed working.
2005-11-06 10:35:13 +00:00
830c0db95b
NCRDMA_SETUP() should be called before NCR_SET_COUNT() and NCRCMD_DMA command
...
in ncr53c9x_select(). Tested with esp on SS1+ and sun3/80, and pcscp at pci.
2005-11-06 10:31:46 +00:00
e15c104622
Add "Conexant CXT48". This should have a more meaningful name, but I
...
couldn't find it. So, I added this because it's better than printing
"unknown codec". When we find a more decent name later, we can change this.
2005-11-06 10:08:24 +00:00
35f76b0ac2
ifdef out sched_{get|set}affinity for ports that don't use it yet. That
...
should fix build.
2005-11-05 23:44:25 +00:00
df6c7b914b
nfs_read: return EISDIR rather than EPERM for !VREG files.
2005-11-05 19:21:14 +00:00
85e05fc4a2
Remove unused and incorrect MIPS_KSEG2_TO_PHYS() and MIPS_PHYS_TO_KSEG2()
...
macro.
2005-11-05 18:17:36 +00:00
dfbaaaef48
ETHERCAP_VLAN_MTU is a flag for ec_capabilities, the size of the 802.1q
...
encaptulation header is ETHER_VLAN_ENCAP_LEN. Pointed out by der Mouse on
tech-kern.
2005-11-05 16:01:52 +00:00
4daa98c618
Rename the wait channels (one was a copy&pasto, the other much too long)
2005-11-05 13:59:02 +00:00
e3fd6c6a46
Make MIPS3_PG_SHIFT a variable and initialize it accordingly
...
in mips_vector_init() if options MIPS3_4100 is specified
so that kernels which have options MIPS3_4100 also work
on other MIPS3 CPUs.
XXX: now should we rename options MIPS3_4100 to options ENABLE_MIPS_R4100,
XXX: or just make MIPS3_PG_SHIFT always a variable?
2005-11-05 11:57:25 +00:00
67e4c0b956
Check MIPS_PRID_IMPL(cpu_id) for MIPS3_4100 CPUs at runtime to
...
initialize the MIPS_COP_0_TLB_PG_MASK register with their specific value.
2005-11-05 11:08:41 +00:00
d9c04bf638
Most use of mips_paddr_to_tlbpfn() is inside if (MIPS_HAS_R4K_MMU) so
...
call mips[13]_paddr_to_tlbpfn() directly instead.
2005-11-05 10:57:49 +00:00
b3b0932a5d
mirror manu's amd64 changes to implement linux sched_{seg,get}affinity;
...
the kernel and linux lkm now compile again.
2005-11-05 10:56:48 +00:00
6305a3d092
Remove #undef MIPS3_4100 from these files since mipsX_subr.S no longer has
...
#if defined(MIPS3_4100) statements.
2005-11-05 10:39:26 +00:00
e54d906c60
Add empty <machine/wired_map.h> for the MI mips/wired_map.c.
2005-11-05 09:54:48 +00:00
fdb3b14bb4
Adapt arc port to the new MI mips3 wired map functions.
...
Tested on NEC-RD94 with several PCI and jazzio devices.
2005-11-05 09:50:50 +00:00
74f4d162fb
Add MI mips3 wired map functions based on patch in port-mips/31915
...
from Garrett D'Amore of Tadpole Computer Inc.
Please refer discussion filed in the PR for details.
2005-11-05 09:46:07 +00:00
2ae48df4fc
regen
2005-11-05 08:11:30 +00:00
0cf4b35fe6
Implement more of Linux futex(2)
2005-11-05 08:07:44 +00:00
41455b2ed0
Implement Linux sched_{set|get}affinity
2005-11-05 08:06:58 +00:00
67fcd6abf5
Fix build when LINUX_NPTL is not defined
2005-11-05 07:26:45 +00:00
2cac35470c
Fix a messed up patch
2005-11-05 00:58:57 +00:00
88a5eb33ed
Instead of ifdef'ing __amd64__ all the Linux NPTL stuff, introduce an
...
ifdef LINUX_NPTL.
Also implement SETTLS flag to Linux clone()
2005-11-05 00:47:26 +00:00
fe1de4d6e3
PR/32003: Brian Buhrow: msdosfs doesn't properly zero out high cluster data
...
on non-FAT32 msdos filesystems.
2005-11-04 21:04:20 +00:00
a0a8d930a4
nfs_bioread: push delayed truncation and tweak loop accordingly.
...
PR/31926 from Jed Davis.
2005-11-04 19:39:30 +00:00
df018c387b
make this compile on m68k by including sys/device.h explicitly.
...
on other platforms it's apparently included implicitly, probably through cpu.h.
2005-11-04 17:00:43 +00:00
b72e0d5382
Fix machine name returned by uname on Linux
2005-11-04 16:58:14 +00:00
78ad642708
regen
2005-11-04 16:57:03 +00:00
e7609e6478
Implement Linux futex ang gettid
2005-11-04 16:54:11 +00:00
3b8100b96a
register_t is not an int on amd64
2005-11-04 16:52:51 +00:00
e1c0c1c9dd
Implement Linux futex and gettid system calls for amd64
2005-11-04 16:51:56 +00:00
0a3dbc01d5
build fix
2005-11-04 16:49:55 +00:00
78422d0423
Check MIPS3_CONFIG_CS and adjust csizebase at runtime on MIPS_R4100 CPUs,
...
and remove "XXXCDC: THIS MIPS3_4100 SPECIAL CASE SHOULD GO AWAY" part
from cpuregs.h. Tested on gxemul.
BTW, cache.c doesn't have MIPS_RC32364 config which was added
in mips_machdep.c rev 1.101?
2005-11-04 16:19:31 +00:00
5936836bb1
physio:
...
- update uio_resid properly.
- don't miss i/o errors near the end of partition.
2005-11-04 08:39:33 +00:00
3de46f226c
PR 31996: fix the comments for MPBIOS and MPACPI. By Damon Brodie.
2005-11-03 23:12:16 +00:00
ff50b97f19
fix the SYSV semop number of operations per request to be limited by
...
seminfo.semopm (which is tunable via the SEMOPM config variable)
rather than an arbitrary constant.
2005-11-03 16:57:03 +00:00
72d7d84d98
regen
2005-11-03 16:49:09 +00:00
59fddeceaa
turn off DEBUG and DIAGNOSTIC in GENERIC kernels so that they
...
are compatible with the LKMs that are part of the release.
2005-11-03 16:48:29 +00:00
06085e8f52
- use sys/spl.h.
...
- add some IPL_ definitions.
2005-11-03 13:06:06 +00:00
e9b064c703
a helper to implement splxxx() using splraiseipl().
2005-11-03 12:31:32 +00:00
2de7c6cd0d
Adjust signature of softdep_freefile (dummy stub which always panics
...
if called) to match ffs_extern.h so that kernels w/o softdep can compile.
2005-11-02 22:10:41 +00:00
e81a08ca3f
pool_printit: don't keep a lock when printing info.
...
we can't clean it up if ddb pager is quitted.
2005-11-02 14:32:54 +00:00
94809e83ec
don't pass -t option to xargs unless MAKEVERBOSE is 2.
2005-11-02 14:29:52 +00:00
60df7649c9
bump to 3.99.11. yamt-vop merge.
2005-11-02 12:42:58 +00:00
f3602c2466
regen.
2005-11-02 12:41:50 +00:00
a748ea88dd
merge yamt-vop branch. remove following VOPs.
...
VOP_BLKATOFF
VOP_VALLOC
VOP_BALLOC
VOP_REALLOCBLKS
VOP_VFREE
VOP_TRUNCATE
VOP_UPDATE
2005-11-02 12:38:58 +00:00
ba91bb71cf
Bug fix: correct the sense of an if-condition in tmpfs_spec_close:
...
because VOP_UPDATE() usually succeeded, spec_close() was not usually
called. Only skip the spec_close() step if VOP_UPDATE() returns
an error result. Now /dev/watchdog works as expected when /dev/
is a tmpfs; previously, it was impossible to disarm a user-tickled
watchdog.
2005-11-02 05:41:50 +00:00
b441f8100f
Bug fix: change tmpfs_spec_bwrite from tmpfs_bwrite to vn_bwrite,
...
per yamt's suggestion. Previously, if /dev/ was mounted on a tmpfs,
block device buffers were never flushed to disk. Trying to unmount
a dirty filesystem (umount /dev/wd0e, say) caused an endless stream
of vflushbuf warnings, because tmpfs_bwrite was not flushing buffers.
The fix told to me by yamt solves the problem.
2005-11-02 05:39:16 +00:00
24f32d9b06
Put the "cache chip bug" message back on a line of its own after the CPU type
...
is announced, which is where it was before 1997. This finally stops the
CPU attachment message overflowing a line on an ELC.
2005-11-01 22:49:17 +00:00
8380617e54
Regen after adding Radeon 9000 PCI (0x4966)
2005-11-01 22:01:44 +00:00
f643837128
Add Radeon 9000 PCI (0x4966)
2005-11-01 22:00:56 +00:00
8481673c7a
Don't decrement the ttl, until we are sure that we can forward this packet.
...
Before if there was no route, we would call icmp_error with a datagram
packet that has an incorrect checksum. (From Liam Foy)
2005-11-01 21:21:09 +00:00
58654bce56
Fix second "no drives attached" message (as seen on sun4c) so as not to
...
include the device name, since this message is printed on the same line
as the attach message anyway.
2005-11-01 21:00:56 +00:00
b1134bf887
Rework the funny "user_strat: No ioctl" error messages - the functions
...
have not been called user_strat for quite some time and the message is
confusing if you do not know the code.
2005-11-01 20:44:04 +00:00
f1e915cc01
Split the linux syscall out to make the code easier to read and maintain.
2005-11-01 16:28:28 +00:00
1421f4d030
Don't claim there is a data in or out phase if the datalen is 0 (userland
...
shouldn't claim it either, but a buggy software shouldn't be able to crash
the kernel anyway). Should fix port-sparc64/31925 by Johan A.van Zanten
(which should really be kern/31925).
Analysed and patch tested by Martin Husemann.
2005-11-01 15:40:00 +00:00
e326d4a34d
Linux does not seems to set rdx on syscall exit.
2005-11-01 09:13:48 +00:00
8ab7495c2d
make scheduler work better when a system has many runnable processes
...
by making p_estcpu fixpt_t. PR/31542.
1. schedcpu() decreases p_estcpu of all processes
every seconds, by at least 1 regardless of load average.
2. schedclock() increases p_estcpu of curproc by 1,
at about 16 hz.
in the consequence, if a system has >16 processes
with runnable lwps, their p_estcpu are not likely increased.
by making p_estcpu fixpt_t, we can decay it more slowly
when loadavg is high. (ie. solve #1.)
i left kinfo_proc2::p_estcpu (ie. ps -O cpu) scaled because i have
no idea about its absolute value's usage other than debugging,
for which raw values are more valuable.
2005-11-01 09:07:53 +00:00
b7ba0ebc16
Fix pipe linux emulation on amd64
2005-10-31 18:00:30 +00:00
5d2a593dcd
build fix
2005-10-31 17:58:07 +00:00
8217506e75
fix ioctl problems after the recent physio changes
...
in some drivers including wd and scsi.
- physio: if a caller provided a buf, stick to use it
because some drivers use it as an identifier.
- sprinkle simple_locks.
- scsistrategy: rather than issueing an async request and
waiting for its completion, simply issue a sync request.
the way to wait for the completion had an assumption that
B_CALL is never used. it isn't the case after the recent
physio() changes.
pointed/analyzed/tested by Martin Husemann.
2005-10-31 14:36:41 +00:00
f243f4debb
Fix paste-o in the NFSV3SATTRTIME_TOSERVER case of mtime handing (need
...
to set va_mtime, not va_atime).
2005-10-31 14:21:35 +00:00
fae23817af
back out the lwp portion of the previous commit which is not ready yet to
...
be committed.
2005-10-31 04:39:41 +00:00
cd2a437026
Don't clear unconditionally P_SUGID when we exec. Clear it only when the
...
real and effective user and group ids match. From Tavis Ormandy.
2005-10-31 04:31:58 +00:00
ad6779af8d
Don't install <sys/systm.h>.
2005-10-30 23:35:17 +00:00
ad33b0d825
We don't need <sys/systm.h> here.
2005-10-30 23:34:34 +00:00
1d1300cd80
Only include <sys/systm.h> if _KERNEL is defined.
2005-10-30 23:34:04 +00:00
a9894b0f08
- localize some definitions.
...
- use PPQ macro where appropriate.
2005-10-30 20:28:56 +00:00
d17f6e14bc
add support for the linux PROT_GROWS{DOWN,UP} mprotect() flags.
...
fixes PR 30008.
2005-10-30 16:25:50 +00:00
df1287ee8b
fix crash dumps on m68k platforms, cpu_kcore_hdr_t is larger than
...
512 bytes now so we need a bigger buffer. PR 26958.
2005-10-30 15:49:39 +00:00
94ce3d822f
don't include uvm_*_i.h unless needed,
...
to reduce bogus header dependencies.
2005-10-30 11:56:51 +00:00
eae67f894f
physio: comment out IPL_BIO because it's not available for some ports.
...
(it isn't used in the current implementation anyway.)
2005-10-30 09:17:02 +00:00
5797e5013d
DELAY() is done (though the implementation is MIPS3 dependent).
2005-10-30 07:55:46 +00:00
bdf356473e
Use new delay(9) for DELAY(9).
2005-10-30 07:53:16 +00:00
bfa9ff2e95
sparc needs mostly the same compilation flags for standalone build as does
...
sparc64. Move them out of the sparc64 conditional.
Okayed by martin@
2005-10-30 07:50:31 +00:00
bc7a643d4f
Use mips3_cp0_count_read() for precise microtime(9).
...
XXX microtime(9) function should be moved from machdep.c to timer.c?
2005-10-30 05:27:14 +00:00
2f451ddebd
Use #define<space> for consistency.
2005-10-30 04:40:43 +00:00
48c586e2b6
According to the new mips3_cp0_count_read() based delay(9),
...
NEC-JC94 seems to have 200MHz clock, not 150MHz.
2005-10-30 04:35:49 +00:00
49b8c2626d
- use mips3_cp0_count_read() for delay(9).
...
- delay(9) should take unsigned int, not int.
2005-10-30 04:34:03 +00:00
c0caafab2b
Set up ci_cpu_freq, ci_cycles_per_hz and ci_divisor_delay in curcpu()
...
from platform->clock.
2005-10-30 04:31:22 +00:00
14eb3fca0f
Don't drop to ddb in the default case if ddb.onpanic is 0.
...
OK'd by martin@.
2005-10-29 21:18:28 +00:00
4b8daa06de
Wow, this is clearly wrong (found in esa_set_params):
...
if (p->channels < 1 && p->channels > 2)
return EINVAL;
Change the 'and' to an 'or' so the check actually does something.
2005-10-29 17:44:34 +00:00
2b6672222b
regen
2005-10-29 17:18:40 +00:00
b2880bdd35
Add Creative Labs's Notebook Optical Mouse.
2005-10-29 17:18:08 +00:00
44545da4e2
Fix more debug const fallout from Hauke Fath.
2005-10-29 16:34:19 +00:00
9a01903be3
correct SADB_X_MIGRATE. pointed by Francis Dupont.
2005-10-29 15:05:13 +00:00
fd96f3e9ea
pull splraiseipl() from newlock branch.
2005-10-29 14:38:51 +00:00
0c66e9687c
fix kmem grovellers.
2005-10-29 14:16:45 +00:00
b9269a5f4f
don't reset interface on rx overrun because it makes
...
the situation worse, if the overrun is merely a symptom of heavy load.
increament if_ierrors instead. PR/29831.
2005-10-29 13:24:41 +00:00
dd80ddaf14
just use ltsleep rather than lockmgr + PCATCH with horrible timeout dance.
2005-10-29 12:31:07 +00:00
c14bdacc17
MALLOC/FREE -> malloc/free where appropriate.
2005-10-29 12:26:37 +00:00
9877d180a5
Re-arrange a bitfield according to host byte-order.
2005-10-29 12:03:44 +00:00
59246ac4e1
fix compilation problem in the previous when !defined(DIAGNOSTIC).
...
pointed by xtraeme@.
2005-10-29 11:49:01 +00:00
428a59477a
physio: make requests with large buffers faster,
...
by queueing more i/o simultaneously.
2005-10-29 11:23:19 +00:00
8d3549ebb0
add something like pthread_once().
2005-10-29 11:13:21 +00:00
54f971f435
add a simple "do it in thread context" framework.
2005-10-29 11:10:37 +00:00
0416d5be24
add splraiseipl().
2005-10-29 11:07:45 +00:00
aee1af1ef4
Expunge an extraneous bus_dmamap_sync() call in iwi_frame_intr(). The entire
...
frame has already been sync'd by iwi_rx_intr().
2005-10-29 11:00:17 +00:00
b6845ed55f
- Use the DMA map size when calling bus_dmamap_sync(), instead of
...
blindly assuming MCLBYTES will DTRT.
- Use bus_dmamap_load_mbuf() instead of bus_dmamap_load() where
appropriate.
- If we have to coalesce a Tx mbuf chain comprised of more than IWI_MAX_NSEG
segments, allocate a cluster iff the payload won't fit in the header.
2005-10-29 10:48:02 +00:00
22049753c1
Don't byte swap unnecessarily in iwi_read_prom_word. Fix the user instead.
2005-10-29 08:44:28 +00:00
842139fe29
Support big-endian hosts:
...
- Always byte-swap data read from EEPROM.
- Byte-swap firmware image when host is big-endian.
In iwi_tx_start():
- Fix descriptor initialisation bogon (only problematic on big-endian hosts).
- Fix bus_dmamap_sync() 'len' parameter..
2005-10-29 08:10:38 +00:00
57e242fa1e
PR/31942: Hauke Fath: -current DEBUG kernel build fails because of gcc whining
2005-10-28 21:54:52 +00:00
daf2726a5c
- delete ugly debugging crap.
...
- remove bogus multicast handling [pointed out by thorpej]
and don't reset the chip on ENETRESET; ENETRESET is a sign
that only the multicast filter needs changing.
2005-10-28 20:56:13 +00:00
e30fd207ff
use ap->a_flags instead of fl->f_type; pointed out by: Devon H. O'Dell
2005-10-28 15:37:23 +00:00
17f31a8395
forgot to commit this.
2005-10-28 14:36:15 +00:00
ffd31cc9b5
According to thorpej, we don't need to stop when we attach. But then again
...
we don't detach yet...
2005-10-28 13:20:55 +00:00
54007a40c0
remove duplicated spllpt().
2005-10-28 10:34:18 +00:00
christos