summaryrefslogtreecommitdiffstats
path: root/sys
Commit message (Collapse)AuthorAgeFilesLines
* Another bit of sx(4) removal.jhb2005-10-171-3/+0
|
* Strip the $a, $t and $d symbols if we're using DDB. There are useless andcognet2005-10-171-0/+3
| | | | confusing in a backtrace.
* Split displaying number of physical and logical cores.jkim2005-10-172-4/+6
|
* Use the clock count register as a timecounter, as it's more accurate.cognet2005-10-171-16/+13
|
* fixed a kernel crash when IPv6 PIM-SM routing is enabled and a PIM register ↵suz2005-10-171-9/+17
| | | | | | | message is received Obtained from: KAME MFC After: 3 days
* Minor commentary correctionimp2005-10-171-3/+3
|
* Unconditionally mount a CD9660 filesystem as read-only, instead ofrodrigc2005-10-172-4/+8
| | | | returning EROFS if we forget to mount it as read-only.
* Use the actual sector size of the media instead of hard-coding it to 2048.rodrigc2005-10-172-6/+24
| | | | | This eliminates KASSERTs in GEOM if we accidentally mount an audio CD as a cd9660 filesystem.
* Unconditionally mount a UDF filesystem as read-only, instead ofrodrigc2005-10-171-2/+4
| | | | returning an EROFS if we forget to mount it as read-only.
* Check against 8019 rather than 8029 for reading in the MAC addr sinceimp2005-10-171-1/+1
| | | | | | | the former is the ISA part, not the latter. MFC After 6.0 is unfrozen (this bug doesn't exist in 6.0 because I didn't MFC the rtl80x9 changes for ISA due to an error on my part)
* - Fix typo.flz2005-10-171-1/+1
| | | | | Approved by: ssouhlal MFC after: 1 week
* Update nwfs_lookup() to match the current cache_lookup() API.truckman2005-10-161-26/+11
| | | | | | | | | cache_lookup() has returned a ref'ed and locked vnode since vfs_cache.c:1.96, dated Tue Mar 29 12:59:06 2005 UTC. This change is similar to the change made to smbfs_lookup() in smbfs_vnops.c:1.58. Tested by: "Antony Mawer" ant AT mawer.org MFC after: 2 weeks
* Reflect mpsafety of the underlying filesystem in the nullfs image.kris2005-10-161-0/+1
| | | | | | | | | | I benchmarked this by simultaneously extracting 4 large tarballs (basically world images) on a 4-processor AMD64 system, in a malloc-backed md. With this patch, system time was reduced by 43%, and wall clock time by 33%. Submitted by: jeff MFC after: 1 week
* Apply the same fix to a potential race in the ISDOTDOT code intruckman2005-10-162-6/+8
| | | | | | | | cd9660_lookup() that was used to fix an actual race in ufs_lookup.c:1.78. This is not currently a hazard, but the bug would be activated by marking cd9660 as MPSAFE. Requested by: bde
* Apply the same fix to a potential race in the ISDOTDOT code intruckman2005-10-161-1/+3
| | | | | | | | | ext2_lookup() that was used to fix an actual race in ufs_lookup.c:1.78. This is not currently a hazard, but the bug would be activated by marking ext2fs as MPSAFE. Requested by: bde MFC after: 2 weeks
* Correct the type of the temporary variable used by ufs_lookup.c:1.78truckman2005-10-161-1/+1
| | | | | | | to fix the race condition in the ufs_lookup() ISDOTDOT code. Noticed by: bde MFC after: 12 days
* Use new functions to call into drivers methods.phk2005-10-161-39/+27
|
* Make ttyconsolemode() call ttsetwater() so that drivers don't have to.phk2005-10-167-6/+1
|
* Use new (inline) functions for calls into driver.phk2005-10-164-11/+7
|
* Make ttsetcompat() staticphk2005-10-162-3/+1
|
* Add a bunch of inline functions to call the drivers tty implementing methods.phk2005-10-161-0/+76
|
* we know it is called rpclose() so call it directly.phk2005-10-161-1/+1
|
* Eliminate two unused arguments to ttycreate().phk2005-10-1615-17/+19
|
* Re-implement rev.1.76 with respect to the code size.sobomax2005-10-162-30/+44
|
* Backout previous commit - for some reason it overflows space constrains onsobomax2005-10-162-64/+48
| | | | amd64. Better version will follow.
* Reduce diffs from i386/i386/machdep.cnyan2005-10-161-9/+5
|
* For AMD processors, nullify CPUID.HTT. FreeBSD has no need for theobrien2005-10-162-0/+18
| | | | | information it conveys, and it is only confusing people. This fixes incorrect output in the previous commit.
* The BCM5401 dspcode load on media changes also appliesgrehan2005-10-161-0/+2
| | | | | | | | to the 100/1000 BCM5400 phy. This fixes the problem with the GEM port not syncing up on Sawtooth G4's. Obtained from: NetBSD Reported by: Ben Rosengart <ben + freebsd org at narcissus net>
* Mempage should be 1024 not 4096. This fixes the multiple 16-bit cardsimp2005-10-161-1/+1
| | | | | | | | failing to work in, eg, Soekris boxes. This is a critcal fix for 6.0. MFC After: ASAP
* Add new option `q', which makes second stage loader quiet unless autobootsobomax2005-10-162-48/+64
| | | | | | is disabled or fails. MFC after: 1 week
* Fix compiling.davidxu2005-10-151-1/+1
|
* added a missing unlocksuz2005-10-151-1/+3
| | | | | Submitted by: JINMEI Tatuya MFC After: 1 day
* regen after recvmsg, recvfrom, sendmsgps2005-10-154-12/+34
|
* Implement the 32bit versions of recvmsg, recvfrom, sendmsgps2005-10-154-8/+188
| | | | Partially obtained from: jhb
* regen for clock_gettime, clock_settime, clock_getresps2005-10-154-13/+28
|
* Implement 32bit wrappers for clock_gettime, clock_settime, andps2005-10-154-25/+116
| | | | clock_getres.
* regenps2005-10-154-6/+6
|
* Correct the prototype for freebsd32_nanosleep and use the properps2005-10-152-4/+4
| | | | size when copying struct timespec32 in and out.
* Fix compiling.davidxu2005-10-151-1/+1
|
* Correct few MSR addresses.jkim2005-10-151-8/+8
| | | | | PR: amd64/85852 Submitted by: Nate Eldredge <nge at cs dot hmc dot edu>
* mpsafevm has been stable and defaulted to 1 on sparc64 for over 6 months,kris2005-10-141-1/+1
| | | | | | | | so we are ready for mpsafevfs=1 by default on sparc64 too. I have been running this on all my sparc64 machines for over 6 months, and have not encountered MD problems. MFC after: 1 week
* Let modules use the kernel's opt_*.h files if built along withyar2005-10-1466-72/+155
| | | | | | | | | | | | | | the kernel by wrapping all targets for fake opt_*.h files in .if defined(KERNBUILDDIR). Thus, such fake files won't be created at all if modules are built with the kernel. Some modules undergo cleanup like removing unused or unneeded options or .h files, without which they wouldn't build this way or the other. Reviewed by: ru Tested by: no binary changes in modules built alone Tested on: i386 sparc64 amd64
* In preparation for making the modules actually use opt_*.h filesyar2005-10-143-8/+6
| | | | | | | | | | | | | | | | | provided in the kernel build directory, fix modules that were failing to build this way due to not quite correct kernel option usage. In particular: ng_mppc.c uses two complementary options, both of which are listed in sys/conf/files. Ideally, there should be a separate option for including ng_mppc.c in kernel build, but now only NETGRAPH_MPPC_ENCRYPTION is usable anyway, the other one requires proprietary files. nwfs and smbfs were trying to ensure they were built with proper network components, but the check was rather questionable. Discussed with: ru
* - Print number of physical/logical cores and more CPUID info.jkim2005-10-148-23/+196
| | | | | | | | | - Add newer CPUID definitions for future use. Many thanks to Mike Tancsa <mike at sentex dot net> for providing test cases for Intel Pentium D and AMD Athlon 64 X2. Approved by: anholt (mentor)
* Close a race in the ufs_lookup() code that handles the ISDOTDOTtruckman2005-10-141-1/+3
| | | | | | | | | | | | | | | | | | | | | | case by saving the value of dp->i_ino before unlocking the vnode for the current directory and passing the saved value to VFS_VGET(). Without this change, another thread can overwrite dp->i_ino after the current directory is unlocked, causing ufs_lookup() to lock and return the wrong vnode in place of the vnode for its parent directory. A deadlock can occur if dp->i_ino was changed to a subdirectory of the current directory because the root to leaf vnode lock ordering will be violated. A vnode lock can be leaked if dp->i_ino was changed to point to the current directory, which causes the current vnode lock for the current directory to be recursed, which confuses lookup() into calling vrele() when it should be calling vput(). The probability of this bug being triggered seems to be quite low unless the sysctl variable debug.vfscache is set to 0. Reviewed by: jhb MFC after: 2 weeks
* Merge over the remaining changes from i386 of the ksiginfo_t changes sojhb2005-10-141-43/+21
| | | | | | that this compiles. Pointy hat to: davidxu
* Make four more functions static that were missed in the last commit.thompsa2005-10-141-5/+5
|
* Add a default value for VM_BCACHE_SIZE_MAX of 400MB. This is copied fromkris2005-10-141-0/+8
| | | | | | | | amd64, and is a factor of 3 less than the value previously auto-sized on a 12GB machine, which would cause an overflow in calculations involving the maxbcache int, causing bufinit() to loop forever at boot. Reviewed by: mlaier, peter
* The signal code is now an int rather than a long, so update debug printfs.jhb2005-10-142-4/+4
|
* Partially revert revision 1.66, which contained a change that did notkris2005-10-141-4/+4
| | | | | | | | | | | | | | | | | | correspond to the commit log. It changed the maxswzone and maxbcache parameters from int to long, without changing the extern definitions in <sys/buf.h>. In fact it's a good thing it did not, because other parts of the system are not yet ready for this, and on large-memory sparc machines it causes severe filesystem damage if you try. The worst effect of the change was that the tunables controlling the above variables stopped working. These were necessary to allow such large sparc64 machines (with >12GB RAM) to boot, since sparc64 did not set a hard-coded upper limit on these parameters and they ended up overflowing an int, causing an infinite loop at boot in bufinit(). Reviewed by: mlaier
OpenPOWER on IntegriCloud