| Commit message (Expand) | Author | Age | Files | Lines |
|---|
| * | spelling |   jsg | 2021-03-10 | 1 | -3/+3 |
| * | Add dt(4) TRACEPOINTs for pool_get() and pool_put(), this is simmilar to the |   claudio | 2021-01-06 | 1 | -1/+6 |
| * | pool(9): remove ticks |   cheloha | 2021-01-02 | 1 | -11/+26 |
| * | pool(9): replace custom TAILQ concatenation loops with TAILQ_CONCAT(3) | cheloha | 2020-01-24 | 1 | -11/+3 |
| * | pool(9): pl_sleep(): drop unused timeout argument | cheloha | 2020-01-23 | 1 | -9/+9 |
| * | After the kernel has reached the sysclt kern.maxclusters limit, |   bluhm | 2019-07-19 | 1 | -1/+7 |
| * | Remove file name and line number output from witness(4) |   visa | 2019-04-23 | 1 | -29/+22 |
| * | revert revert revert. there are many other archs that use custom allocs. |   tedu | 2019-02-10 | 1 | -22/+24 |
| * | if waitok flag is set, have the interrupt multipage allocator redirect | tedu | 2019-02-10 | 1 | -1/+9 |
| * | make it possible to reduce kmem pressure by letting some pools use a more | tedu | 2019-02-10 | 1 | -24/+14 |
| * | Constipate all the struct lock_type's so they go into .rodata |   guenther | 2018-06-08 | 1 | -5/+5 |
| * | slightly randomize the order that new pages populate their item lists in. |   dlg | 2018-02-06 | 1 | -2/+13 |
| * | While booting it does not make sense to wait for memory, there is | bluhm | 2018-01-18 | 1 | -1/+7 |
| * | New flag PR_RWLOCK for pool_init(9) makes the pool use rwlocks instead | guenther | 2017-08-13 | 1 | -69/+248 |
| * | Compute the level of contention only once. | visa | 2017-07-12 | 1 | -4/+5 |
| * | When there is no contention on a pool cache lock, lower the number | visa | 2017-07-12 | 1 | -1/+4 |
| * | set the alignment of the per cpu cache structures to CACHELINESIZE. | dlg | 2017-06-23 | 1 | -3/+3 |
| * | change the semantic for calculating when to grow the size of a cache list. | dlg | 2017-06-23 | 1 | -14/+8 |
| * | dynamically scale the size of the per cpu cache lists. | dlg | 2017-06-19 | 1 | -1/+22 |
| * | add garbage collection of unused lists percpu cached items. | dlg | 2017-06-16 | 1 | -2/+40 |
| * | split returning an item to the pool pages out of pool_put as pool_do_put. | dlg | 2017-06-16 | 1 | -24/+36 |
| * | report contention on caches global data to userland. | dlg | 2017-06-15 | 1 | -1/+2 |
| * | white space tweaks. no functional change. | dlg | 2017-06-15 | 1 | -4/+4 |
| * | implement the backend of the sysctls that report pool cache info. | dlg | 2017-06-15 | 1 | -17/+126 |
| * | when enabling cpu caches, check the item size against the right thing | dlg | 2017-06-13 | 1 | -2/+3 |
| * | Tweak lock inits to make the system runnable with witness(4) | visa | 2017-04-20 | 1 | -3/+3 |
| * | revert 1.206 because it allows deadlocks. | dlg | 2017-02-20 | 1 | -1/+4 |
| * | the splvm() in pool_gc_pages is unecessary now. | dlg | 2017-02-08 | 1 | -4/+1 |
| * | Force a context switch for every pool_get(9) with the PR_WAITOK flag |   mpi | 2017-01-24 | 1 | -2/+2 |
| * | let pool page allocators advertise what sizes they can provide. | dlg | 2016-11-21 | 1 | -14/+42 |
| * | rename some types and functions to make the code easier to read. | dlg | 2016-11-07 | 1 | -143/+149 |
| * | poison the TAILQ_ENTRY in items in the per cpu pool cache. | dlg | 2016-11-02 | 1 | -12/+52 |
| * | add poisoning of items on the per cpu caches. | dlg | 2016-11-02 | 1 | -4/+38 |
| * | use a TAILQ to maintain the list of item lists used by the percpu code. | dlg | 2016-11-02 | 1 | -10/+8 |
| * | add per cpu caches for free pool items. | dlg | 2016-11-02 | 1 | -1/+316 |
| * | all pools have their ipl set via pool_setipl, so fold it into pool_init. | dlg | 2016-09-15 | 1 | -28/+13 |
| * | move pools to using the subr_tree version of rb trees | dlg | 2016-09-15 | 1 | -9/+11 |
| * | revert moving pools from tree.h to subr_tree.c rb trees. | dlg | 2016-09-05 | 1 | -11/+9 |
| * | move pool red-black trees from tree.h code to subr_tree.c code | dlg | 2016-09-05 | 1 | -9/+11 |
| * | add a "show socket" command to ddb | dlg | 2016-01-15 | 1 | -4/+6 |
| * | Now that interrupt-safe uvm maps are porperly locked, the interrupt-safe |   kettenis | 2015-09-11 | 1 | -5/+1 |
| * | Give the pool page allocator backends more sensible names. We now have: | kettenis | 2015-09-08 | 1 | -20/+19 |
| * | Now that msleep(9) no longer requires the kernel lock (as long as PCATCH | kettenis | 2015-09-08 | 1 | -15/+2 |
| * | We no longer need to grab the kernel lock for allocating and freeing pages | kettenis | 2015-09-06 | 1 | -5/+11 |
| * | Push down the KERNEL_LOCK/KERNEL_UNLOCK calls into the back-end allocator | kettenis | 2015-09-01 | 1 | -7/+21 |
| * | re-enable *8. | dlg | 2015-08-21 | 1 | -2/+2 |
| * | remove the POOL_NEEDS_CATCHUP macro, it isnt used. | dlg | 2015-07-23 | 1 | -4/+1 |
| * | Move `ticks' declaration to sys/kernel.h. |   uebayasi | 2015-07-20 | 1 | -3/+1 |
| * | disable *8 again for now. incoherent archs arent having much fun with it. | dlg | 2015-04-21 | 1 | -2/+2 |
| * | nothing uses pool_sleep, so get rid of it | dlg | 2015-04-07 | 1 | -3/+1 |
