summaryrefslogtreecommitdiffstats
path: root/sys
Commit message (Expand)AuthorAgeFilesLines
* Remove 'device mem' from GENERIC, which markm@ mistakingly added.ssouhlal2004-08-021-1/+0
* Unbreak DEVICE_POLLING build / LINT. Sorry!mlaier2004-08-022-2/+2
* Eliminate the acquisition and release of Giant around the call toalc2004-08-021-2/+0
* Add PowerPC bridge instruction flag to the assembler to help outgrehan2004-08-022-0/+6
* G5 support: handle the case where the OpenFirmware memory array usesgrehan2004-08-023-29/+90
* Kernel traps were not being passed to trap_fatal in somegrehan2004-08-022-2/+4
* Document machdep.enable_panic_key.scottl2004-08-021-1/+1
* Fix the build by providing 'PHYS_TO_DMAP' and 'M_MEMDESC'.obrien2004-08-021-0/+6
* Bump __FreeBSD_version to designate uma_zone functions changing type.green2004-08-021-1/+1
* Add what appears to be a missing '*/' at the end of a comment.rwatson2004-08-021-0/+1
* - Fix unloading by the same way it is done in my other classes:pjd2004-08-021-3/+5
* * Add a "how" argument to uma_zone constructors and initialization functionsgreen2004-08-0213-170/+243
* Second part of ALTQ driver modifications, covering:mlaier2004-08-017-31/+56
* Comment kse_create() and make a few minor code cleanupsjulian2004-08-011-47/+121
* The watchdog callout executes with the (non-sleepable) ifnet lock heldwpaul2004-08-011-1/+1
* In ndis_alloc_bufpool() and ndis_alloc_packetpool(), the test to see ifwpaul2004-08-011-2/+2
* Typo.pjd2004-08-011-1/+1
* Another stupid error from my side. PPPOE_NONSTANDARD was first definedglebius2004-08-011-9/+9
* Big mess 'o changes:wpaul2004-08-018-146/+246
* Add the I/O device for those architectures that have it.markm2004-08-012-0/+2
* Remove extraneous ';'.markm2004-08-012-2/+2
* Implement basic support for EHCI interrupt pipes. This is unlikelyiedowse2004-08-012-11/+283
* YA oops. Remove code that was being tested locally.markm2004-08-011-21/+0
* Remove local hack that was not supposed to be committed.markm2004-08-011-1/+1
* Turn off PREEMPTION by default while it gets debugged. It's been causingscottl2004-08-013-0/+9
* MFi386: revision 1.1167nyan2004-08-011-0/+4
* MFi386: revision 1.501.nyan2004-08-011-2/+4
* Change the default to switch on DMA on ATAPI devices if they cansos2004-08-011-3/+5
* Break out the MI part of the /dev/[k]mem and /dev/io drivers intomarkm2004-08-0137-692/+833
* - Launch main provider when there are no more disks in NEW state.pjd2004-08-011-1/+4
* Add some minor changes related to PCMCIA attribute memory mappingwpaul2004-08-011-2/+18
* Specify the locking for some proc fieldsjulian2004-07-311-3/+3
* If there are no valid components after the timeout, just destroy device.pjd2004-07-311-7/+2
* Propagate size changes upwards.le2004-07-313-0/+25
* Fix a stupid error in my previous commit, which broke operationglebius2004-07-311-1/+2
* Handle spoil event in dedicated function: g_mirror_spoiled().pjd2004-07-311-2/+17
* Add copyright notices.fjoe2004-07-313-3/+78
* Turn on PCB_FULLCTX for set_mcontext, functions like kse_switchindavidxu2004-07-311-0/+1
* Remove unused field.pjd2004-07-311-1/+0
* Destroy synchronization geom immediately. This should fix unloading withoutpjd2004-07-311-1/+11
* Fix the panic (""memory modified after free") when ata_getparam() failssos2004-07-311-0/+1
* Add pmap locking to pmap_object_init_pt().alc2004-07-312-2/+12
* Bump __FreeBSD_version to reflect the removal of kmem_alloc_pageable().alc2004-07-311-1/+1
* Allow slice creation on providers from MIRROR class.pjd2004-07-311-0/+1
* Add '-p' option for 'insert' command which allows to specify prioritypjd2004-07-311-0/+7
* - Check if 'slice' argument was given.pjd2004-07-311-0/+8
* Dump correct field.pjd2004-07-311-1/+1
* Set the access counts of a subdisk correctly when attaching itle2004-07-301-2/+21
* Connect GEOM_MIRROR class to the build.pjd2004-07-304-0/+5
* Add GEOM_MIRROR class which provide RAID1 functionality and has many usefulpjd2004-07-304-0/+3567
OpenPOWER on IntegriCloud