| Commit message (Expand) | Author | Age | Files | Lines |
* | Remove a six year old undocumented #ifdef : NO_B_MALLOC. | Poul-Henning Kamp | 2002-05-04 | 1 | -12/+0 |
* | Change callers of mtx_init() to pass in an appropriate lock type name. In | John Baldwin | 2002-04-04 | 1 | -1/+1 |
* | brelse() was improperly clearing B_DELWRI in the B_DELWRI|B_INVAL case | Matthew Dillon | 2002-04-03 | 1 | -5/+2 |
* | Remove __P. | Alfred Perlstein | 2002-03-19 | 1 | -2/+2 |
* | Fixed some printf format errors (hopefully all of the remaining daddr64_t | Bruce Evans | 2002-03-19 | 1 | -8/+10 |
* | Convert all pmap_kenter/pmap_kremove pairs in MI code to use pmap_qenter/ | Jake Burkholder | 2002-03-17 | 1 | -2/+2 |
* | Introduce the new 64-bit size disk block, daddr64_t. Change | Kirk McKusick | 2002-03-15 | 1 | -2/+2 |
* | Giant pushdown for read/write/pread/pwrite syscalls. | Alfred Perlstein | 2002-03-15 | 1 | -0/+2 |
* | * Move bswlist declaration and initialization from kern/vfs_bio.c to | Eivind Eklund | 2002-03-05 | 1 | -4/+15 |
* | Document all functions, global and static variables, and sysctls. | Eivind Eklund | 2002-03-05 | 1 | -69/+138 |
* | Back out all the pmap related stuff I've touched over the last few days. | Peter Wemm | 2002-02-27 | 1 | -2/+2 |
* | Jake further reduced IPI shootdowns on sparc64 in loops by using ranged | Peter Wemm | 2002-02-27 | 1 | -2/+2 |
* | GC: BIO_ORDERED, various infrastructure dealing with BIO_ORDERED. | Poul-Henning Kamp | 2002-02-22 | 1 | -2/+0 |
* | Replace bowrite() with BUF_WRITE in ufs. | Poul-Henning Kamp | 2002-02-22 | 1 | -16/+0 |
* | GC P_BUFEXHAUST leftovers, we've had a new mechanism to avoid buffer | Matthew Dillon | 2002-01-31 | 1 | -1/+0 |
* | This fixes a large number of bugs in our NFS client side code. A recent | Matthew Dillon | 2001-12-14 | 1 | -0/+12 |
* | The nbuf calculation was assuming that PAGE_SIZE = 4096 bytes, which is | Matthew Dillon | 2001-12-08 | 1 | -6/+12 |
* | Placemark an interrupt race in -current which is currently protected by | Matthew Dillon | 2001-11-08 | 1 | -0/+4 |
* | Implement IO_NOWDRAIN and B_NOWDRAIN - prevents the buffer cache from blocking | Matthew Dillon | 2001-11-05 | 1 | -4/+9 |
* | Documentation | Matthew Dillon | 2001-10-21 | 1 | -6/+3 |
* | Change the kernel's ucred API as follows: | John Baldwin | 2001-10-11 | 1 | -10/+4 |
* | Enable vmiodirenable by default. Remove incorrect comment from sysctl.conf. | Matthew Dillon | 2001-09-26 | 1 | -1/+1 |
* | KSE Milestone 2 | Julian Elischer | 2001-09-12 | 1 | -7/+7 |
* | Remove the code that limited the buffer_map to 1/2 the size of the | Matthew Dillon | 2001-08-22 | 1 | -0/+2 |
* | Move most of the kernel submap initialization code, including the | Matthew Dillon | 2001-08-22 | 1 | -7/+61 |
* | Revert previous accidental commit. FWIW, it was part of enabling | Peter Wemm | 2001-07-27 | 1 | -0/+3 |
* | Fix cut/paste blunder. Serves me right for doing a last minute tweak | Peter Wemm | 2001-07-27 | 1 | -3/+0 |
* | With Alfred's permission, remove vm_mtx in favor of a fine-grained approach | Matthew Dillon | 2001-07-04 | 1 | -77/+35 |
* | This patch implements O_DIRECT about 80% of the way. It takes a patchset | Matthew Dillon | 2001-05-24 | 1 | -2/+7 |
* | - Always call bfreekva() w/o vm_mtx held. | John Baldwin | 2001-05-23 | 1 | -28/+12 |
* | Introduce a global lock for the vm subsystem (vm_mtx). | Alfred Perlstein | 2001-05-19 | 1 | -4/+101 |
* | Revert consequences of changes to mount.h, part 2. | Greg Lehey | 2001-04-29 | 1 | -2/+0 |
* | Correct #includes to work with fixed sys/mount.h. | Greg Lehey | 2001-04-23 | 1 | -0/+2 |
* | bread() is a special case of breadn(), so don't replicate code. | Poul-Henning Kamp | 2001-04-18 | 1 | -23/+2 |
* | Write a switch statement as less obscure if statements. | Poul-Henning Kamp | 2001-04-17 | 1 | -18/+8 |
* | This patch removes the VOP_BWRITE() vector. | Poul-Henning Kamp | 2001-04-17 | 1 | -1/+8 |
* | Add debugging option to always read/write cylinder groups as full | Kirk McKusick | 2001-04-17 | 1 | -1/+5 |
* | Fix lockup for loopback NFS mounts. The pipelined I/O limitations could be | Matthew Dillon | 2001-02-28 | 1 | -3/+7 |
* | Change and clean the mutex lock interface. | Bosko Milekic | 2001-02-09 | 1 | -1/+1 |
* | This commit represents work mainly submitted by Tor and slightly modified | Matthew Dillon | 2001-02-04 | 1 | -6/+13 |
* | Use PCPU_GET, PCPU_PTR and PCPU_SET to access all per-cpu variables | Jake Burkholder | 2001-01-10 | 1 | -5/+5 |
* | This implements a better launder limiting solution. There was a solution | Matthew Dillon | 2000-12-26 | 1 | -135/+128 |
* | Stick the kthread API in a kthread_* namespace, and the specialized kproc | John Baldwin | 2000-12-15 | 1 | -2/+2 |
* | Implement a low-memory deadlock solution. | Matthew Dillon | 2000-11-18 | 1 | -80/+128 |
* | Take VBLK devices further out of their missery. | Poul-Henning Kamp | 2000-11-02 | 1 | -1/+1 |
* | Catch up to moving headers: | John Baldwin | 2000-10-20 | 1 | -2/+1 |
* | Convert lockmgr locks from using simple locks to using mutexes. | Jason Evans | 2000-10-04 | 1 | -2/+2 |
* | Add three new VOPs: VOP_CREATEVOBJECT, VOP_DESTROYVOBJECT and VOP_GETVOBJECT. | Boris Popov | 2000-09-12 | 1 | -11/+13 |
* | Major update to the way synchronization is done in the kernel. Highlights | Jason Evans | 2000-09-07 | 1 | -7/+13 |
* | Now that buffer locks can be recursive, we need to delete the panics | Kirk McKusick | 2000-07-25 | 1 | -4/+0 |