summaryrefslogtreecommitdiffstats
path: root/sys
Commit message (Collapse)AuthorAgeFilesLines
* We don't need to generate vnode_if.h anymoreimp2004-12-2924-27/+23
|
* No need to generate vnode_if.h anymoreimp2004-12-295-5/+5
|
* Move WINBOND 4 port hub to its proper location in the file. It belongsimp2004-12-291-1/+3
| | | | | in a section of its own, not arbitrarily grouped in the hawking section just because hawking sells it...
* Eliminate (now) unnecessary acquisition and release of the global pagealc2004-12-291-4/+0
| | | | queues lock.
* Access to the page's busy field is (now) synchronized by the containingalc2004-12-291-1/+0
| | | | | object's lock. Therefore, the assertion that the page queues lock is held can be removed from vm_page_io_start().
* Add the Winbond made Hawking UH104 4-port hub.obrien2004-12-291-0/+6
| | | | Add a note where to find the offical USB assigned numbers.
* Rewrite ehci_abort_xfer() to use the method hinted at in the EHCI spec.julian2004-12-292-45/+108
| | | | | | | | | | | | | to remove a transaction from the async schedule. The previous method didn't work well and led to the hardware writing to free'd buffers etc, as it didn't always know that the transaction had been aborted. Written after consultation with David Brownell who wrote the Linux EHCI driver. As part of this give the sqh structure a "previous" pointer. MFC after: 1 week
* Pass an explicit pointer to the bus structure in the 'usb' device ivarjb2004-12-283-3/+3
| | | | | | | | rather than a softc pointer (with the bus structure at the start). This is a non-functional change. It just helps when reading the code to know that the ehci, ohci and uhci drivers share the bus structure, not the entire softc.
* Remove debug code.pjd2004-12-281-10/+2
|
* - Up the WITNESS_COUNT macro from 200 to 1024 to support the growing numberjhb2004-12-281-2/+1
| | | | | | | | | of lock types in the kernel. This results in an increase of witness data usage from ~145k to ~280k on i386 for kernels with 'options WITNESS'. - Remove the unused witness malloc bucket. Submitted by: Michal Mertl mime at traveller dot cz (1)
* - Plug a memory leak in ng_netflow_cache_init().glebius2004-12-281-5/+10
| | | | | | | - Initialize error to 0 in ng_netflow_flow_add() (a nop change). - Update cache statistics holding workqueue mutex. MFC after: 3 days
* Don't probe for a disk unless explicitly enabled by a tunable.grehan2004-12-281-0/+6
| | | | | | | | | This allows boot to proceed on a real system until the issue of calling back into certain OpenFirmware calls (e.g. finddevice) in thread context is understood. (this commit only affects psim users, of which I think I am the only one...)
* Add FR support to sppp (MFCronyx).rik2004-12-285-15/+688
| | | | | | | | | | Silence on: net@, current@, hackers@. No objections: joerg Requested by: by many (mostly Cronyx) users for a long long time. MFC after: 10 days PR: kern/21771, kern/66348
* nos->now typo reported by njlimp2004-12-271-1/+1
|
* Add note about where major numbers are now assignedimp2004-12-271-0/+3
|
* Formatting nitsimp2004-12-271-6/+6
|
* Rewrite piece of code which I committed some time ago that allows topjd2004-12-272-12/+10
| | | | | | | | | show file name for 'mdconfig -l -u <x>' command. This allows to preserve API/ABI compatibility with version 0 (that's why I changed version number back to 0) and will allow to merge this change to RELENG_5. MFC after: 5 days
* Fix mbuf leak.pjd2004-12-271-1/+3
| | | | | Submitted by: Johnny Eriksson <bygg@cafax.se> MFC after: 5 days
* Attempt to slightly refine the print out from "show alllocks" -- listrwatson2004-12-271-2/+2
| | | | | the process and thread numbers/names on the same line rather than on separate lines, and print the thread pointer not just the tid.
* Remove duplicate CHPRODUCTS entry.imp2004-12-271-3/+2
| | | | Use tab religiously in the vendor section.
* MFi386: Restore cpu_reset proxy code to enable reset from ddb on an AP.njl2004-12-271-4/+39
|
* Reduce diffs to i386.njl2004-12-271-16/+11
|
* Restore the cpu_reset proxy code. It is needed if you want to reset thenjl2004-12-271-1/+41
| | | | | | | system from an AP at runtime (i.e., calling cpu_reset from ddb). Someday, if we move to an NMI for stopping cpus instead, we can do away with this. Requested by: jhb
* Minor cleanup: Reduce stack usage in probe method. Outdent probe. Usenjl2004-12-271-41/+38
| | | | TRUE/FALSE instead of 1/0 for booleans. Remove trailing and extra whitespace.
* Remove trailing whitespace.njl2004-12-276-10/+10
|
* Note that access to the page's busy count is synchronized by the containingalc2004-12-271-1/+1
| | | | object's lock.
* Do not vput(9) unlocked vnode and do not VREF it with the sole purposekan2004-12-271-2/+0
| | | | | | of vputting it back immediately. Complained by: DEBUG_VFS_LOCKS
* - Unintentionally checked in a debugging panic. Remove that.jeff2004-12-261-4/+0
|
* - Remove a 4BSD specific hack since this will work on ULE too.jeff2004-12-261-4/+0
|
* - Fix a long standing problem where an ithread would not honor sched_pin().jeff2004-12-261-127/+140
| | | | | | | | | | | | | | | | | | | | | | | | - Remove the sched_add wrapper that used sched_add_internal() as a backend. Its only purpose was to interpret one flag and turn it into an int. Do the right thing and interpret the flag in sched_add() instead. - Pass the flag argument to sched_add() to kseq_runq_add() so that we can get the SRQ_PREEMPT optimization too. - Add a KEF_INTERNAL flag. If KEF_INTERNAL is set we don't adjust the SLOT counts, otherwise the slot counts are adjusted as soon as we enter sched_add() or sched_rem() rather than when the thread is actually placed on the run queue. This greatly simplifies the handling of slots. - Remove the explicit prevention of migration for ithreads on non-x86 platforms. This was never shown to have any real benefit. - Remove the unused class argument to KSE_CAN_MIGRATE(). - Add ktr points for thread migration events. - Fix a long standing bug on platforms which don't initialize the cpu topology. The ksg_maxid variable was never correctly set on these platforms which caused the long term load balancer to never inspect more than the first group or processor. - Fix another bug which prevented the long term load balancer from working properly. If stathz != hz we can't expect sched_clock() to be called on the exact tick count that we're anticipating. - Rearrange sched_switch() a bit to reduce indentation levels.
* Add "show alllocks" command to DDB, which dumps a list of processesrwatson2004-12-261-0/+42
| | | | | | | | | | | | and threads currently holding sleep mutexes (and spin mutexes for curthread). This can be quite useful in looking for a lock condition summary for a system, as it avoids manually iterating through threads and processes to find all the interesting locks. NB: "alllocks" is up there with "lockedvnods" for a bad argument for show. MFC after: 2 weeks
* Assert that the vm object is locked on entry to vm_page_sleep_if_busy();alc2004-12-261-8/+3
| | | | remove some unneeded code.
* tsleep => msleeprik2004-12-261-2/+2
| | | | | | | | | read sys/kern/kern_synch.c, msleep (): KASSERT(timo != 0 || mtx_owned(&Giant) || mtx != NULL, ("sleeping without a mutex")); MFC after: 3 days
* LCA is 21066 and 21068. Add EV7 (bloody optimist.. :)wilko2004-12-261-2/+3
|
* * Remove some code that's in a #ifndef FreeBSD that's no longer used.darrenr2004-12-261-5/+0
|
* * The #ifdef's to cause mutex's for freebsd to be declared were meant to pickdarrenr2004-12-262-19/+4
| | | | | up on USE_MUTEX being defined, but this patch * Remove some code that's in a #ifndef FreeBSD that's no longer used.
* More minor diff reduction against netbsd:imp2004-12-261-4/+5
| | | | | | | ADVANCELOGIC->AVANCELOGIC (nothing in the tree uses it, so safe to do) sort HAGIWARA vendor entry sort ACTIONTAR vendor entry Minor change to SYSTEMTALKS vendor entry.
* - Run sched_userret() after thread_userret(). Before, sched_userret() wouldjeff2004-12-261-5/+4
| | | | | | | | | lower the priority of the returning thread to a user priority before calling into thread_userret() which would call wakeup() which in turn would cause the returning thread to eventually context switch rather than completing its slice. Allowing this thread to complete its slice first yields a 15% performance improvement in super-smack on my dual opteron with 4BSD.
* Minor merges from NetBSD (up through 1.388):imp2004-12-261-27/+36
| | | | | | | | | | | | | | | | Add $NetBSD$ in a comment at the top Update copyright dates Update header comment Add some of the entries not present in FreeBSD's usbdevs file Harmonize some descriptions with NetBSD where NetBSD's were shorter More work needs to happen here, as there's many conflicting vendor names. There's also more harmonization that can happen before that problem is tackled. This was inspired by recent discussions, but none of the patches posted were consulted to produce this commit. Other, similar ones will follow.
* Add my copyright and update Jeff's copyright on UMA source files,bmilekic2004-12-265-10/+20
| | | | | | as per his request. Discussed with: Jeffrey Roberson
* - Wrap the thread count adjustment in sched_load_add() and sched_load_rem()jeff2004-12-261-6/+30
| | | | | | so that we may place some ktr entries nearby. - Define other KTR_SCHED tracepoints so that we may graph the operation of the scheduler.
* - Remove earlier KTR_ULE tracepoints.jeff2004-12-261-32/+14
| | | | | - Define new KTR_SCHED points so that we can graph the operation of the scheduler.
* - Define KTR points for KTR_SCHED.jeff2004-12-263-0/+22
|
* - Add KTR_SCHED to produce ktr data for sys/tools/sched/schedgraph.py.jeff2004-12-261-2/+3
|
* Remove an errant blank line apparently introduced inrwatson2004-12-251-1/+0
| | | | ip_output.c:1.194.
* In the dropafterack case of tcp_input(), it's OK to release the TCPrwatson2004-12-252-2/+2
| | | | | pcbinfo lock before calling tcp_output(), as holding just the inpcb lock is sufficient to prevent garbage collection.
* Revert parts of tcp_input.c:1.255 associated with the header predictedrwatson2004-12-252-4/+14
| | | | | | | | | | | | | cases for tcp_input(): While it is true that the pcbinfo lock provides a pseudo-reference to inpcbs, both the inpcb and pcbinfo locks are required to free an un-referenced inpcb. As such, we can release the pcbinfo lock as long as the inpcb remains locked with the confidence that it will not be garbage-collected. This leads to a less conservative locking strategy that should reduce contention on the TCP pcbinfo lock. Discussed with: sam
* fix commentphk2004-12-251-1/+1
|
* - Add genid field to the metadata which will allow to improve reliability a bit.pjd2004-12-253-61/+206
| | | | | | | | | | | | | | | After this change, when component is disconnected because of an I/O error, it will not be connected and synchronized automatically, it will be logged as broken and skipped. Autosynchronization can occur, when component is disconnected (on orphan event) and connected again - there were no I/O error, so there is no need to not connected the component, but when there were writes while it wasn't connected, it will be synchronized. This fix cases, when component is disconnected because of I/O error and can be connected again and again. - Bump version number. - Implement backward compatibility mechanism. After this change when metadata in old version is detected, it is automatically upgraded to the new (current) version.
* Make _umtx_op() as more general interface, the final parameter needn't bedavidxu2004-12-258-13/+12
| | | | timespec pointer, every parameter will be interpreted by its opcode.
OpenPOWER on IntegriCloud