Commit Graph

154738 Commits

Author SHA1 Message Date
Alan Cox
5d4a7b7945 Eliminate an unnecessary call to pmap_remove_all(). If a page belongs to
an object whose reference count is zero, then that page cannot possibly
be mapped.
2010-04-20 04:16:39 +00:00
Nathan Whitehorn
0d526fa829 Add gpart and glabel to the release CD mfsroot. Even if sysinstall
cannot partition disks on powerpc, this will allow the user to.

PR:		powerpc/93203
Obtained from:	ia64
MFC after:	1 week
2010-04-20 01:12:23 +00:00
Rick Macklem
a318bc273d For the experimental NFS client doing an NFSv4 mount,
set the NFSCLFLAGS_RECVRINPROG while doing recovery from an expired
lease in a manner similar to r206818 for server reboot recovery.
This will prevent the function that acquires stateids for I/O
operations from acquiring out of date stateids during recovery.
Also, fix up mutex locking on the nfsc_flags field.

MFC after:	1 week
2010-04-20 01:02:39 +00:00
Attilio Rao
95335fd844 getblk lockmgr is mostly used as a msleep() and may lead too easilly to
false positives.
Whitelist it.

Reported by:	Erik Cederstrand <erik at cederstrand dot dk>
2010-04-19 23:40:46 +00:00
Attilio Rao
248bb9379f Fix a deadlock in the shutdown code:
When performing a smp_rendezvous() or more likely, on amd64 and i386,
a smp_tlb_shootdown() the caller will end up with the smp_ipi_mtx
spinlock held, busy-waiting for other CPUs to acknowledge the operation.
As long as CPUs are suspended (via cpu_reset()) between the active mask
read and IPI sending there can be a deadlock where the caller will wait
forever for a dead CPU to acknowledge the operation.
Please note that on CPU0 that is going to be someway heavier because of
the spinlocks being disabled earlier than quitting the machine.

Fix this bug by calling cpu_reset() with the smp_ipi_mtx held.
Note that it is very likely that a saner offline/online CPUs mechanism
will help heavilly in fixing similar cases as it is likely more bugs
of this type may arise in the future.

Reported by:	rwatson
Discussed with:	jhb
Tested by:	rnoland, Giovanni Trematerra
		<giovanni dot trematerra at gmail dot com>
MFC:		2 weeks

Special deciation to:	anyone who made possible to have 16-ways machines
			in Netperf
2010-04-19 23:27:54 +00:00
Pyun YongHyeon
6da6d0a9e3 With r206844, CSUM_TCP is also set for CSUM_TSO case. Modify
drivers to take into account for the change. Basically CSUM_TSO
should be checked before checking CSUM_TCP.
2010-04-19 22:10:40 +00:00
Edwin Groothuis
696dd8f41c MFV of tzdata2010i, r206865
- Marocco does have DST this year between May and August.
- Historical data for Taiwan
- Argentina / San Luis does not do DST this year.
2010-04-19 20:59:39 +00:00
Edwin Groothuis
206727f1fa Vendor import of tzdata2010i:
- Marocca will have DST this year between May and August
- Historical data for Taiwan
- No more DST for the Argentinian province/state San Luis this year.

Obtained from:	ftp://elsie.nci.nih.gov/pub/
2010-04-19 20:54:00 +00:00
Jaakko Heinonen
002d1d1c38 Fix ddb(4) "show geom addr" command when INVARIANTS is enabled. Don't
assert that the topology lock is held when g_valid_obj() is called from
debugger.

MFC after:	1 week
2010-04-19 20:07:35 +00:00
Rui Paulo
4ad53b2511 Dump the AR_PHY_TURBO register on the AR5416. This register holds 11n
configurations.

Sponsored by:	iXsystems, inc.
2010-04-19 17:16:23 +00:00
Luigi Rizzo
d55ebfbd4e fix 64-bit build
Reported by:	Robert Noland
2010-04-19 16:35:47 +00:00
Luigi Rizzo
6ba1ccc0f2 whitespace fixes (trailing whitespace, bad indentation
after a merge, etc.)
2010-04-19 16:17:30 +00:00
Kenneth D. Merry
3579cf4c4f Don't clear other flags (e.g. CSUM_TCP) when setting CSUM_TSO. This was
causing TSO to break for the Xen netfront driver.

Reviewed by:	gibbs, rwatson
MFC after:	7 days
2010-04-19 15:15:36 +00:00
Luigi Rizzo
37133ba702 Slightly different handling of printf/snprintf for unaligned uint64_t,
which should improve readability, and also to ease the port to
platforms that do not support %llu

MFC after:	3 days
2010-04-19 15:11:45 +00:00
Nathan Whitehorn
3560b8af1f Fix brokenness in top on big-endian 32-bit systems introduced when
changing format_k2 to take a long long. Because itoa is defined as a K&R
C function, without prototyping its arguments, format_k2 passed a 64-bit
value, but itoa() received only the first word, showing '0' in all memory
fields.
2010-04-19 14:34:44 +00:00
Michael Tuexen
307b49efef Get delayed SACK working again.
MFC after: 3 days.
2010-04-19 14:15:58 +00:00
Rui Paulo
0ff2943ecb Revert r206755. It causes some laptops to stop booting. 2010-04-19 14:07:33 +00:00
Xin LI
0e568ab25c Partially MFp4 #176265 by pjd@:
- Properly initialize and destroy system_taskq.
 - Add a dummy implementation of taskq_create_proc().

Note: We do not currently use system_taskq in ZFS so this is mostly a
no-op at this time.  Proper system_taskq initialization is required
by newer ZFS code.

Ok'ed by:	pjd
MFC after:	2 weeks
2010-04-19 09:03:36 +00:00
Juli Mallett
fe7d6b6e69 Fix MALTA64 build. 2010-04-19 09:03:34 +00:00
Juli Mallett
540247e8c1 Remove unused file. 2010-04-19 07:51:57 +00:00
Juli Mallett
745bba1ab0 o) Eliminate the "stand" frame and its use. Use CALLFRAME_* everywhere.
o) Use <machine/asm.h> macros for register-width, etc., rather than doing it
   by hand in a few more assembly files.
o) Reduce diffs between various bits of TLB refill code in exception.S and
   between interrupt processing code.
o) Use PTR_* to operate on registers that are pointers (e.g. sp).
o) Add and use a macro, CLEAR_PTE_SWBITS rather than using the
   mysteriously-named WIRED_SHIFT to select bits to truncate when loading PTEs.
o) Don't doubly disable interrupts by moving zero to the status register,
   especially since that has the nasty side-effect of taking us out of 64-bit
   mode.
o) Use CLEAR_STATUS to disable interrupts the first time.
o) Keep SR_PX set as well as SR_[KSU]X when doing exception processing.  This
   is the bit that determines whether 64-bit operations are allowed.
o) Don't enable interrupts until configure_final(), like most other ports.
2010-04-19 07:34:26 +00:00
Juli Mallett
5f3173b517 o) Fix XKPHYS physical address extraction. Also define cache coherency
attributes for XKPHYS.
o) Make coprocessor 0 accessor function macros for register+selector registers
   take the full name so that e.g. (as done in this commit), prid selector 1
   can be written through mips_wr_ebase() rather than mips_wr_prid1().
o) Allow for sign extension of 32-bit segment addresses.
o) Remove an unused MIPS-I register number.
2010-04-19 06:01:58 +00:00
Alan Cox
7b7d5b6c58 vm_thread_swapout() can safely dirty the page before rather than after
acquiring the page queues lock.
2010-04-19 00:18:14 +00:00
Juli Mallett
ca596a25f0 o) Add a VM find-space option, VMFS_TLB_ALIGNED_SPACE, which searches the
address space for an address as aligned by the new pmap_align_tlb()
   function, which is for constraints imposed by the TLB. [1]
o) Add a kmem_alloc_nofault_space() function, which acts like
   kmem_alloc_nofault() but allows the caller to specify which find-space
   option to use. [1]
o) Use kmem_alloc_nofault_space() with VMFS_TLB_ALIGNED_SPACE to allocate the
   kernel stack address on MIPS. [1]
o) Make pmap_align_tlb() on MIPS align addresses so that they do not start on
   an odd boundary within the TLB, so that they are suitable for insertion as
   wired entries and do not have to share a TLB entry with another mapping,
   assuming they are appropriately-sized.
o) Eliminate md_realstack now that the kstack will be appropriately-aligned on
   MIPS.
o) Increase the number of guard pages to 2 so that we retain the proper
   alignment of the kstack address.

Reviewed by:	[1] alc
X-MFC-after:	Making sure alc has not come up with a better interface.
2010-04-18 22:32:07 +00:00
Rick Macklem
7ea710b3b1 Avoid extraneous recovery cycles in the experimental NFS client
when an NFSv4 server reboots, by doing two things.
1 - Make the function that acquires a stateid for I/O operations
    block until recovery is complete, so that it doesn't acquire
    out of date stateids.
2 - Only allow a recovery once every 1/2 of a lease duration, since
    the NFSv4 server must provide a recovery grace period of at
    least a lease duration. This should avoid recoveries caused
    by an out of date stateid that was acquired for an I/O op.
    just before a recovery cycle started.

MFC after:	1 week
2010-04-18 22:21:23 +00:00
Jilles Tjoelker
348c81a032 sh: Add testcases for double-quotes within quoted ${var+-...} (non-POSIX).
POSIX leaves things like "${var+"word"}" undefined.
We follow traditional ash behaviour here.
Hence, these testcases also work on stable/8.
2010-04-18 22:13:45 +00:00
Alan Cox
b28889a2fc Remove a nonsensical test from vm_pageout_clean(). A page can't be in the
inactive queue and have a non-zero wire count.

Reviewed by:	kib
MFC after:	3 weeks
2010-04-18 21:29:28 +00:00
Rui Paulo
685b0bb814 Delete svn:executable prop. 2010-04-18 18:43:36 +00:00
Konstantin Belousov
8d03319930 Revert r206649.
Simplify the presented declaration of struct sigaction, noting the
caveat in the text. Real layout of the structure and exposed
implementation namespace only obfuscates the usage.

Submitted by:	bde
MFC after:	3 days
2010-04-18 18:23:11 +00:00
Alan Cox
4b9dd5d537 There is no justification for vm_object_split() setting PG_REFERENCED on a
page that it is going to sleep on.  Eliminate it.

MFC after:	3 weeks
2010-04-18 17:50:09 +00:00
Pawel Jakub Dawidek
bd7226a572 Restore previous order. 2010-04-18 12:43:33 +00:00
Pawel Jakub Dawidek
224329fb6b Style fixes. 2010-04-18 12:36:53 +00:00
Pawel Jakub Dawidek
eb998be67d Add missing list and lock destruction. 2010-04-18 12:27:07 +00:00
Pawel Jakub Dawidek
ad3cb80827 Extend locks scope to match OpenSolaris. 2010-04-18 12:25:40 +00:00
Pawel Jakub Dawidek
57a81a8bbc Remove racy assertion.
Obtained from:	OpenSolaris
2010-04-18 12:21:52 +00:00
Pawel Jakub Dawidek
5195ca2307 Set ARC_L2_WRITING on L2ARC header creation.
Obtained from:	OpenSolaris
2010-04-18 12:20:33 +00:00
Jilles Tjoelker
5d66b54e27 ln: Refuse deleting a directory entry by hardlinking it to itself.
Two pathnames refer to the same directory entry iff the directories match
and the final components' names match.

Example: (assuming file1 is an existing file)
  ln -f file1 file1
This now fails while leaving file1 intact. It used to delete file1 and then
complain it cannot be linked because it is gone.

With -i, this error is detected before the question is asked.

MFC after:	2 weeks
2010-04-17 22:39:53 +00:00
Rui Paulo
9eb448a7e6 Use ubthidhci_enable="NO" to avoid the bootup warning.
Submitted by: 	Jilles Tjoelker <jilles@stack.nl>
MFC after:	3 days
2010-04-17 21:31:42 +00:00
Alan Cox
b11b56b55b In vm_object_madvise() setting PG_REFERENCED on a page before sleeping on
that page only makes sense if the advice is MADV_WILLNEED.  In that case,
the intention is to activate the page, so discouraging the page daemon
from reclaiming the page makes sense.  In contrast, in the other cases,
MADV_DONTNEED and MADV_FREE, it makes no sense whatsoever to discourage
the page daemon from reclaiming the page by setting PG_REFERENCED.

Wrap a nearby line.

Discussed with:	kib
MFC after:	3 weeks
2010-04-17 21:14:37 +00:00
Doug Barton
1a58736816 In case a user wants to configure only an IPv6 link-local address
add an example that shows how to do it.
2010-04-17 18:48:18 +00:00
Alan Cox
aefea7f519 In vm_object_backing_scan(), setting PG_REFERENCED on a page before
sleeping on that page is nonsensical.  Doing so reduces the likelihood
that the page daemon will reclaim the page before the thread waiting in
vm_object_backing_scan() is reawakened.  However, it does not guarantee
that the page is not reclaimed, so vm_object_backing_scan() restarts
after reawakening.  More importantly, this muddles the meaning of
PG_REFERENCED.  There is no reason to believe that the caller of
vm_object_backing_scan() is going to use (i.e., access) the contents of
the page.  There is especially no reason to believe that an access is
more likely because vm_object_backing_scan() had to sleep on the page.

Discussed with:	kib
MFC after:	3 weeks
2010-04-17 18:35:07 +00:00
Bernhard Schmidt
82ff07c047 Remove IPW_LOCK_DECL and fix various LORs.
Approved by:	rpaulo (mentor)
2010-04-17 18:18:46 +00:00
Bernhard Schmidt
6e4c30c9cd Use iv_appie_wpa, with this commit WPA works again.
Approved by:	rpaulo (mentor)
2010-04-17 18:17:25 +00:00
Bernhard Schmidt
557c25b48f - Make ipw usable again by moving directly into ASSOC state.
- No need to manually switch to RUN state, assoc response takes care
  of that.

Approved by:	rpaulo (mentor)
2010-04-17 18:16:14 +00:00
Bernhard Schmidt
892d2e9d9b Pass correct RSSI to ieee80211_input*().
Approved by:	rpaulo (mentor)
2010-04-17 18:14:49 +00:00
Bernhard Schmidt
d02ae91a4c Fix comment about ipw_assoc and remove some whitespaces; no functional
changes.

Approved by:	rpaulo (mentor)
2010-04-17 18:13:52 +00:00
Alan Cox
0b6ace4743 Setting PG_REFERENCED on the requested page in swap_pager_getpages() is
either redundant or harmful, depending on the caller.  For example, when
called by vm_fault(), it is redundant.  However, when called by
vm_thread_swapin(), it is harmful.  Specifically, if the thread is later
swapped out, having PG_REFERENCED set on its stack pages leads the page
daemon to reactivate these stack pages and delay their reclamation.

Reviewed by:	kib
MFC after:	3 weeks
2010-04-17 17:02:17 +00:00
Jilles Tjoelker
702f62fb48 getcwd(3): Clarify that EACCES may or may not be checked.
POSIX permits but does not require checking access on the current and parent
directories.

Because various programs do not like it if getcwd(3) fails, it seems best
to avoid checking access as much as possible. There are various reports in
GNATS about this (search for getcwd).

Our getcwd(3) implementation first queries the kernel for the pathname
directly, which does not check any permissions but sometimes fails, and then
falls back to reading all parent directories for the names.

PR:		standards/44425
MFC after:	2 weeks
2010-04-17 15:52:50 +00:00
Jilles Tjoelker
8eac1f9477 sh: On startup of the shell, use PWD from the environment if it is valid.
Unset PWD if it is incorrect and no value for it can be determined.
This preserves the logical current directory across shell invocations.

Example (assuming /home is a symlink):
$ cd
$ pwd
/home/foo
$ sh
$ pwd
/home/foo

Formerly the second pwd would show the physical path (symlinks resolved).
2010-04-17 14:35:46 +00:00
Michael Tuexen
37f144eb5d Fix a bug where SACKs are not sent when they should.
Move some protection code to INVARIANTS.
Cleanups.

MFC after: 3 days.
2010-04-17 12:22:44 +00:00