| Commit message (Collapse) | Author | Age | Files | Lines | ||
|---|---|---|---|---|---|---|
| ... | ||||||
| * | Avoid a 'thundering herd' problem when many processes wait for free buffers. |   art | 2001-03-30 | 1 | -11/+25 | |
| | | | | | | | | Just wakeup one process (there is a possible bug here that will be fixed in the next round of cleanup). Some misc cleanup, especially in the comments. | |||||
| * | Slight cleanup. | art | 2001-03-14 | 1 | -6/+6 | |
| | | ||||||
| * | Reserve some buffers for syncer daemon. This prevent deadlock in getblk |   gluk | 2001-03-13 | 1 | -24/+122 | |
| | | | | | | | | | and getnewbuf. One process can sleep at "getnewbuf" waiting for a free buffer and it may held buffer 'A' busy. Other processes can return buffers on free lists, but they sleep on "getblk" waiting for buffer 'A'. art@ ok. | |||||
| * | Remove random trailing whitespace. | art | 2001-03-09 | 1 | -9/+9 | |
| | | | | | From gluk. | |||||
| * | art@ found a race in getnewbuf. bawrite can block so we need to restart |   csapuntz | 2001-02-27 | 1 | -2/+2 | |
| | | | | | the whole buffer allocation process | |||||
| * | Cleanup of vnode interface continues. Get rid of VHOLD/HOLDRELE. | csapuntz | 2001-02-24 | 1 | -8/+10 | |
| | | | | | | | | | | | | | Change VM/UVM to use buf_replacevnode to change the vnode associated with a buffer. Addition v_bioflag for flags written in interrupt handlers (and read at splbio, though not strictly necessary) Add vwaitforio and use it instead of a while loop of v_numoutput. Fix race conditions when manipulation vnode free list | |||||
| * | Change the B_DELWRI flag using buf_dirty and buf_undirty instead of | csapuntz | 2001-02-23 | 1 | -12/+38 | |
| | | | | | | | | | | manually twiddling it. This allows the buffer cache to more easily keep track of dirty buffers and decide when it is appropriate to speed up the syncer. Insipired by FreeBSD. Look over by art@ | |||||
| * | Latest soft updates from FreeBSD/Kirk McKusick | csapuntz | 2001-02-21 | 1 | -8/+8 | |
| | | | | | Snapshot-related code has been commented out. | |||||
| * | Use MIN, not min when counting the pages we steal. | art | 2001-02-13 | 1 | -2/+2 | |
| | | | | | | min is a function taking u_int arguments and we are counting signed longs here. | |||||
| * | remove obsolete vtrace guts; art@ |   mickey | 2000-06-23 | 1 | -2/+1 | |
| | | ||||||
| * | Collect statistics on sync and async writes. | art | 1999-12-05 | 1 | -1/+22 | |
| | | | | | From NetBSD. | |||||
| * | Indentation to make the code more readable. | art | 1999-12-02 | 1 | -6/+8 | |
| | | ||||||
| * | use clrnd(round_page(size)) instead of roundup(size, CLBYTES). | art | 1999-09-10 | 1 | -2/+2 | |
| | | | | | They do the same thing, but the former is noticeably faster on sparc | |||||
| * | vm_offset_t -> {v,p}addr_t ; vm_size_t -> {v,p}size_t | art | 1999-07-15 | 1 | -4/+4 | |
| | | ||||||
| * | zap the newhashinit hack. | art | 1999-04-28 | 1 | -2/+2 | |
| | | | | | | Add an extra flag to hashinit telling if it should wait in malloc. update all calls to hashinit. | |||||
| * | indent | art | 1998-11-29 | 1 | -2/+2 | |
| | | ||||||
| * | A couple more splbio()s in vfs_bio plus moving around a couple functions. | csapuntz | 1998-01-10 | 1 | -9/+8 | |
| | | ||||||
| * | Fixed hang on shutdown | csapuntz | 1997-11-07 | 1 | -7/+5 | |
| | | | | | Disabled vop_nolock for now. Filesystems still need to be cleaned up. | |||||
| * | Updates for VFS Lite 2 + soft update. | csapuntz | 1997-11-06 | 1 | -217/+77 | |
| | | ||||||
| * | back out vfs lite2 till after 2.2 |   deraadt | 1997-10-06 | 1 | -77/+217 | |
| | | ||||||
| * | VFS Lite2 Changes | csapuntz | 1997-10-06 | 1 | -217/+77 | |
| | | ||||||
| * | Don't look beyond the end of bdevsw[], from NetBSD PR 3748 by Michael L Hitch |   tholo | 1997-06-14 | 1 | -2/+3 | |
| | | ||||||
| * | Minor performance enhancements from NetBSD | tholo | 1997-04-14 | 1 | -2/+34 | |
| | | ||||||
| * | Remove lots of timer_state structs as they just ate memory and only a few was |   niklas | 1997-01-05 | 1 | -4/+1 | |
| | | | | | | | | ever used. Now a single state is kept for net, tty and disk events resp. Also, call the randomness from disk_unbusy instead of biodone, as biodone gets a lot of virtual events (from virtual filesystems etc), and as a bonus: feed xfer time and size into the entropy pool too. | |||||
| * | random device is permanent now. | mickey | 1996-10-19 | 1 | -8/+2 | |
| | | ||||||
| * | rnd -> random | deraadt | 1996-08-29 | 1 | -4/+4 | |
| | | ||||||
| * | Ensure we never use more than one callout table slot | tholo | 1996-07-21 | 1 | -3/+9 | |
| | | ||||||
| * | -Wall & -Wstrict-prototype fixes | niklas | 1996-07-02 | 1 | -3/+1 | |
| | | ||||||
| * | When a dirty buffer is written to, don't move it to the tail of the | tholo | 1996-06-26 | 1 | -22/+1 | |
| | | | | | | | | dirty list. Doing that can, for a block that is written to with a period of less than 30 seconds, cause the block to never be flushed to disk. Idea from Jeffrey Mogul's paper covering the same basic changes. | |||||
| * | from pk; protect some more stuff with splbio | deraadt | 1996-06-14 | 1 | -4/+8 | |
| | | ||||||
| * | Kernel-implementation of update(8) my me | tholo | 1996-06-11 | 1 | -1/+180 | |
| | | ||||||
| * | sync syscalls, no sys/cpu.h | deraadt | 1996-05-02 | 1 | -3/+4 | |
| | | ||||||
| * | NetBSD 960317 merge | niklas | 1996-04-19 | 1 | -1/+3 | |
| | | ||||||
| * | Initial commit of random source driver. | mickey | 1996-03-29 | 1 | -0/+9 | |
| | | ||||||
| * | From NetBSD: Merge with NetBSD 960217 | niklas | 1996-02-29 | 1 | -6/+17 | |
| | | ||||||
| * | initial import of NetBSD tree | deraadt | 1995-10-18 | 1 | -0/+843 | |
 |
index : wireguard-openbsd | |
| WireGuard implementation for the OpenBSD kernel | Matt Dunwoodie |
| summaryrefslogtreecommitdiffstats |