summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Remove duplicate const specifiers in many drivers (I hope I got all ofdim2012-11-0542-52/+52
| | | | | | | | | | | | | | | | | | | | them, please let me know if not). Most of these are of the form: static const struct bzzt_type { [...list of members...] } const bzzt_devs[] = { [...list of initializers...] }; The second const is unnecessary, as arrays cannot be modified anyway, and if the elements are const, the whole thing is const automatically (e.g. it is placed in .rodata). I have verified this does not change the binary output of a full kernel build (except for build timestamps embedded in the object files). Reviewed by: yongari, marius MFC after: 1 week
* After years of hard work by many FreeBSD and LLVM developers, makebrooks2012-11-053-2/+14
| | | | | | clang the default compiler on i386 and amd64 systems. Special thanks to: dim, ed, rdivacky
* In sys/contrib/ngatm/netnatm/msg/uni_ie.c, fix a few warnings from newerdim2012-11-051-4/+4
| | | | | | | | versions of clang 3.2, about comparing enum uni_cause values against integer constants which fall outside the enum range. No functional change. MFC after: 3 days
* Fix a few warnings from newer clang 3.2 in libpmc, about comparing enumdim2012-11-051-2/+2
| | | | | | | | pmc_event values against integer constants which fall outside the enum range. Reviewed by: fabient, sbruno MFC after: 3 days
* Fix a minor typo.mav2012-11-051-1/+1
|
* sh: Change cmdtype in tblentry from short to signed char.jilles2012-11-051-1/+1
| | | | | If this is a smaller type than int anyway, we can make it the smallest possible.
* Add a jitter buffer in the common USB serial driver code whichhselasky2012-11-053-3/+82
| | | | | | | | | | | | | temporarily stores characters if the TTY buffer is full when used a as a console. This can happen when a console is suspended. Also properly do the flow stop signalling when this happens and flow start when the condition changes back to normal again. Bump __FreeBSD_version to force external kernel modules to be recompiled. No kernel API changes. MFC after: 1 week Suggested by: ed @
* Fix build with clang: properly terminate comment even in #if 0 blocks.delphij2012-11-051-1/+1
| | | | Submitted by: dim
* A clarification to the behaviour of the active vnode list managementkib2012-11-051-0/+3
| | | | | | | regarding the vnode page cleaning. In collaboration with: pho MFC after: 1 week
* fuse_io* must be able to crunch also VDIR vnodes.attilio2012-11-051-2/+2
| | | | | | | | Update assert appropriately. Reported and Tested by: flo MFC after: 2 months X-MFC: 241519,242536
* Sync with domain and protosw data structures.kevlo2012-11-051-1/+9
| | | | Reviewed by: glebius
* Clear IFF_DRV_OACTIVE if any slots were completed.adrian2012-11-051-0/+6
| | | | This unblocks TX EDMA under high load.
* Back out r242262. The simplified window change/update logic wasn'tandre2012-11-051-47/+16
| | | | | | complete and ready for production use. PR: kern/173309
* TX EDMA debugging fixes:adrian2012-11-051-6/+7
| | | | | * Do the calculation for each ath_buf, rather than just the first * Correct the calculation in the first place.
* Revert previous commit.kevlo2012-11-051-7/+9
|
* Catch up with r242520. s/maxusers/maxbikeshedskevlo2012-11-051-9/+7
|
* Revert r228695. We use __func__ here as a format to distinguish betweenpjd2012-11-051-1/+1
| | | | | abort and assert. It would be cleaner to use NULL or "" here, but gcc complains in both cases.
* rtld: Fix fd leak with parallel dlopen and fork/exec.jilles2012-11-042-4/+4
| | | | | | | | | | | | | | | | Rtld did not set FD_CLOEXEC on its internal file descriptors; therefore, such a file descriptor may be passed to a process created by another thread running in parallel to dlopen() or fdlopen(). No other threads are expected to be running during parsing of the hints and libmap files but the file descriptors need not be passed to child processes so add O_CLOEXEC there as well. This change will break fdlopen() (as used by OpenPAM) on kernels without F_DUPFD_CLOEXEC (added in July). Note that running new userland on old kernels is not supported. Reviewed by: kib
* MFV: less v453.delphij2012-11-0421-547/+688
|
* Add support for a few more devices:eadler2012-11-041-0/+4
| | | | | | | | | | PNP0510 and FUJ02E5 for a "Wacom Tablet at FuS Lifebook T" PNP0502 and PNP0511 for some other generic devices. PR: kern/173357 Submitted by: Andrey Zakharchenko <avz@jscc.ru> Approved by: cperciva (implicit) MFC after: 1 week
* Make OK_TO_WRITE_TO_DISK an envrionment variable insteadeadler2012-11-042-46/+44
| | | | | | | | | | of a compile time option. While here, don't differ based on the existence of LOCK_EX which doesn't seem to have ever made a difference on FreeBSD. Approved by: cperciva (from discussion) MFC after: 3 days
* Prefer the use of stdbool to custom defines.eadler2012-11-041-3/+3
| | | | | Approved by: cperciva (from discussion) MFC after: 3 days
* zfs_dirlook: bailout early if directory is unlinkedavg2012-11-041-0/+14
| | | | | | | | | | Otherwise we could fail with an incorrect error if e.g. parent object id is removed too or we can even return a wrong vnode if parent object has been already re-used. Discussed with: pjd Also see: http://article.gmane.org/gmane.os.freebsd.devel.file-systems/13863 MFC after: 26 days
* zfsctl_snapdir_lookup: obtain a snapname in the remount caseavg2012-11-041-0/+1
| | | | | | | ... which is triggered if somebody did regular umount on a snapshot mount. Reviewed by: Matthew Ahrens <mahrens@delphix.com> MFC after: 20 days
* zfs: set MNTK_EXTENDED_SHARED flagavg2012-11-041-0/+1
| | | | | Discussed with: kib MFC after: 20 days
* opensolaris compat: clear VI_MOUNT before returning if mount_snapshot failsavg2012-11-041-0/+3
| | | | | | | To do: investigate if it would be possible to use normal vfs_domount here. Reviewed by: kib MFC after: 19 days
* zfs_vnode_forget: dispose of larvae vnode using public vfs api (mostly)avg2012-11-041-5/+4
| | | | | Reviewed by: kib MFC after: 19 days
* zfs_umount: no need to set MNTK_UNMOUNTF here, dounmount handles thatavg2012-11-041-4/+0
| | | | | Reviewed by: kib MFC after: 19 days
* opensolaris_lookup: use vfs_busy in traverse before calling VFS_ROOTavg2012-11-041-0/+4
| | | | | | | ... to ensure that we have a valid mountpoint during the call. Reviewed by: kib MFC after: 19 days
* zfs_vnode_lock: no need to double-guess caller's intentions hereavg2012-11-041-8/+0
| | | | | | | | vn_lock should do the right thing with respect to given vnode lock flags. If a caller doesn't mind a doomed vnode, then zfs should deliver. Reviewed by: kib MFC after: 19 days
* zfs_mount: drop vfs.zfs.rootpool.prefer_cached_config tunableavg2012-11-041-14/+2
| | | | | | | | | It turned out to be not that useful, because its default value may lead to a problem when a root pool is present in zpool.cache, but its on-disk status is 'exported'. This may happen if the pool was imported in a different environment with -f flag and then exported. MFC after: 12 days
* zfs_freebsd_close: call zfs_close with count=1 instead of count=0avg2012-11-041-1/+1
| | | | | | | Otherwise we may be leaking z_sync_cnt, which may lead to unnecessary ZIL sync-ing. MFC after: 12 days
* cpuctl_do_cpuid: explicitly use ecx=0 for cpuid callavg2012-11-041-1/+1
| | | | | | | | | ... instead of whatever random value may happen to be in the register. ecx is important to some cpuid leaves. To do: extend cpuctl interface to provide for ecx value parameter. MFC after: 5 days
* Add decoding of the missed MNT_KERN_ flags to ddb "show mount" command.kib2012-11-041-0/+5
| | | | MFC after: 3 weeks
* Add decoding of the missed VI_ and VV_ flags to ddb "show vnode" command.kib2012-11-041-3/+9
| | | | MFC after: 3 days
* Order the enumeration of the MNT_ flags to be the same as the order ofkib2012-11-041-2/+2
| | | | | | their definitions. MFC after: 3 days
* New sentence, new line.joel2012-11-041-4/+4
| | | | Submitted by: brueffer
* Add a few basic examples.joel2012-11-041-1/+19
|
* - Add my mentor relationships to committers-ports.dotgblach2012-11-042-0/+4
| | | | | | - Add myself to calendar.freebsd Approved by: tabthorpe (mentor)
* Add an example showing the use of gnop(8) to skip over header data.wblock2012-11-041-1/+13
| | | | | | PR: kern/145999 Reviewed by: mjg MFC after: 1 week
* Change default prompt to show ~ again for the home directorybapt2012-11-042-2/+2
| | | | | Submitted by: flo Approved by: eadler
* Oops - conditionalise that.adrian2012-11-041-0/+2
|
* Fix a bug where operations was carried on even if not implemented,attilio2012-11-031-6/+4
| | | | | | | | leading to handling of an invalid fdip object. Reported and tested by: flo MFC after: 2 months X-MFC: 241519
* Replace all uses of the page queues lock by a R/W lock that is privatealc2012-11-031-34/+40
| | | | | | | | to this pmap. Eliminate two redundant #include's. Tested by: marcel
* Rework the known rwlock to benefit about staying on their ownattilio2012-11-037-63/+7
| | | | | | | cache line in order to avoid manual frobbing but using struct rwlock_padalign. Reviewed by: alc, jimharris
* EDMA TX tweaks:adrian2012-11-031-7/+28
| | | | | | | | | | * don't poke ath_hal_txstart() if nothing was pushed into the FIFO during the refill process; * shuffle around the TX debugging output a little so it's logged at TX hardware enqueue; * Add logging of the TX status processing.
* Merge the FDT versions of initarm.andrew2012-11-037-2487/+491
| | | | | | | | | The copies of initarm used on platforms with FDT support were almost identical. The differences were pulled out into separate functions that were called by initarm. This change merges the, now identical, copies of initarm and a few of it's support functions. This is a step towards a common kernel on ARMv6.
* sh: Use C99 flexible array instead of accessing array beyond bounds.jilles2012-11-031-3/+2
| | | | | | | | Although sufficient memory is available for a longer string in cmdname, this is undefined behaviour anyway. Side effect: for alignment reasons, an additional byte of memory is allocated per hashed command.
* Add tty_set_winsize().ed2012-11-033-37/+24
| | | | | This removes some of the signalling magic from the Syscons driver and puts it in the TTY layer, where it belongs.
* For AR9380 NICs - the non-enterprise versions don't support RTS protectionadrian2012-11-031-2/+10
| | | | | | | | | | of small (< 256 byte) aggregate frames. This needs to be done or 11n aggregation TX just simply doesn't work on these NICs. Whilst here, extend some debug printing; I was using this whilst debugging the TX power setup in the TX descriptor(s) on the AR9380.
OpenPOWER on IntegriCloud