summaryrefslogtreecommitdiffstats
path: root/sys
Commit message (Collapse)AuthorAgeFilesLines
...
* Put DIOCWLABEL under #ifdef NO_GEOMphk2003-01-261-0/+2
|
* - Split db.desc.cmd into desc.control and desc.reqcount.simokawa2003-01-262-103/+127
| | | | - remove unncessary descriptor updates.
* - Improve IT/IR DMA queue management.simokawa2003-01-264-428/+235
| | | | | - Improve debug message for mbuf handling. - Wait 1 sec for DMA stop in fwohci_i{t,r}x_disable() before freeing buffers.
* Link /boot/boot1 to the name /boot/boot to avoid per-arch naming of thephk2003-01-261-0/+1
| | | | bootstrap code for disklabel using architectures.
* Build a file "boot" which consists of boot1 and boot2 concatenated.phk2003-01-262-2/+12
| | | | | There is little if any reason to treat the two components separately and it will simplify disklabel(8) and libdisk if we didn't.
* This file is no longer needed.phk2003-01-261-79/+0
| | | | Spotted by: tjr
* Move UPCALL related data structure out of kse, introduce a newdavidxu2003-01-2639-1510/+1663
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | data structure called kse_upcall to manage UPCALL. All KSE binding and loaning code are gone. A thread owns an upcall can collect all completed syscall contexts in its ksegrp, turn itself into UPCALL mode, and takes those contexts back to userland. Any thread without upcall structure has to export their contexts and exit at user boundary. Any thread running in user mode owns an upcall structure, when it enters kernel, if the kse mailbox's current thread pointer is not NULL, then when the thread is blocked in kernel, a new UPCALL thread is created and the upcall structure is transfered to the new UPCALL thread. if the kse mailbox's current thread pointer is NULL, then when a thread is blocked in kernel, no UPCALL thread will be created. Each upcall always has an owner thread. Userland can remove an upcall by calling kse_exit, when all upcalls in ksegrp are removed, the group is atomatically shutdown. An upcall owner thread also exits when process is in exiting state. when an owner thread exits, the upcall it owns is also removed. KSE is a pure scheduler entity. it represents a virtual cpu. when a thread is running, it always has a KSE associated with it. scheduler is free to assign a KSE to thread according thread priority, if thread priority is changed, KSE can be moved from one thread to another. When a ksegrp is created, there is always N KSEs created in the group. the N is the number of physical cpu in the current system. This makes it is possible that even an userland UTS is single CPU safe, threads in kernel still can execute on different cpu in parallel. Userland calls kse_create to add more upcall structures into ksegrp to increase concurrent in userland itself, kernel is not restricted by number of upcalls userland provides. The code hasn't been tested under SMP by author due to lack of hardware. Reviewed by: julian
* Put use of DIOCWLABEL under #ifdef NO_GEOMphk2003-01-262-1/+8
|
* - Remove a redundant scheduler option.jeff2003-01-262-2/+0
| | | | | Pointy hat to: jeff Spotted by: dillon
* - Add entries for scheduler selectionjeff2003-01-261-0/+17
|
* - Introduce the SCHED_ULE and SCHED_4BSD options for compile time selectionjeff2003-01-2612-1/+17
| | | | | of the scheduler. - Add SCHED_4BSD as the scheduler for all kernel config files in cvs.
* - Add the ule scheduler. This is intended to be a general purpose processjeff2003-01-261-0/+697
| | | | | scheduler with many SMP benefits. It is still very experimental and should be used only in test environments.
* Disable DRA to fix problems with recording.orion2003-01-261-14/+19
|
* - Call sched_sleep() instead of rolling our own in cv_waitq_add().jeff2003-01-261-2/+2
|
* Merge some code paths back together so that we only instantiate 1 copy ofjake2003-01-261-121/+97
| | | | the user tlb fault handlers.
* Allow defaulting the console to ttya when it sets to screen and keyboardjake2003-01-262-10/+16
| | | | in the prom but no keyboard is plugged in.
* Merge the best parts of maxfragpackets and maxnipq together. (Bothsilby2003-01-261-13/+5
| | | | | | | | | | | functions implemented approximately the same limits on fragment memory usage, but in different fashions.) End user visible changes: - Fragment reassembly queues are freed in a FIFO manner when maxfragpackets has been reached, rather than all reassembly stopping. MFC after: 5 days
* Add INTEL EEPRO100imp2003-01-251-0/+1
| | | | Submitted by: joerg
* Remove some unnecessary casts.silby2003-01-251-3/+3
|
* regenalfred2003-01-251-4/+4
|
* Bring shm functions closer the the opengroup standards.alfred2003-01-253-8/+13
| | | | | PR: 47469 Submitted by: Craig Rodrigues <rodrigc@attbi.com>
* regen.alfred2003-01-251-2/+2
|
* Bring semop() closer the the opengroup standards.alfred2003-01-253-5/+17
| | | | | PR: 47471 Submitted by: Craig Rodrigues <rodrigc@attbi.com>
* Add MTX_DUPOK to the initialization of system map locks.alc2003-01-251-2/+2
|
* Add ac97_patch.[ch] that provide space for ac97 codec specific patches.orion2003-01-257-78/+163
|
* Change API of FW_GDEVLST ioctl.simokawa2003-01-255-40/+50
| | | | | | - include information about itself. - define struct fw_devinfo and use it in struct fw_devlstreq. - unify EUI64 representation using struct fw_eui64.
* Add sysctl kern.timecounter.nsetclock which indicates the number ofphk2003-01-251-0/+2
| | | | | | | | | | potential discontinuities in our UTC timescale. Applications can monitor this variable if they want to be informed about steps in the timescale. Slews (ntp and adjtime(2)) and frequency adjustments (ntp) will not increment this counter, only operations which set the clock. No attempt is made to classify size or direction of the step.
* Moved some (gas) macros up so they can be used in more places.jake2003-01-241-25/+25
|
* Remove extraneous FILEDESC_LOCKs around atomic reads.hsu2003-01-241-4/+0
| | | | Reviewed by: jhb
* Remove FW_SSTDV ioctl. It is not used anymore.simokawa2003-01-245-10/+35
|
* Merge little and big endian case.simokawa2003-01-242-251/+127
|
* Fix pmap_extract so that it doesn't panic if the user typesdfr2003-01-241-3/+10
| | | | | | 'cat /proc/pid/map' Submitted by: Arun Sharma <arun.sharma@intel.com>
* Remove the now unused sys/malloc.h header.mux2003-01-241-1/+0
|
* - add support for IPX (tested with mount -t nwfs and mars_nwe),fjoe2003-01-245-93/+173
| | | | | | | | | | | IP fast forwarding, SIOCGIFADDR, setting hardware address (not currently enabled in cm driver), multicasts (experimental) - add ARC_MAX_DATA, use IF_HANDOFF, remove arc_sprintf() and some unused variables - if_simloop logic is made more similar to ethernet - drop not ours packets early (if we are not in promiscous mode) Submitted by: mark tinguely (partially)
* - The 'aux' field isn't around any longer.mdodd2003-01-241-7/+5
| | | | - Add missing splnet()/splx()
* add forgotten IFA_LOCK_INITfjoe2003-01-231-0/+1
|
* More useful announce message containing current speed of CPUnjl2003-01-231-3/+3
|
* Remove redundant printf from targbhasync() since all places that return annjl2003-01-231-4/+0
| | | | error do their own, more useful printf.
* - Add sysctl knob for bus manager. (hw.firewire.try_bmr)simokawa2003-01-232-18/+29
| | | | | - Check invalid SID length. - Add some debug messages.
* Clean up some junk defines, and GC the TPR options.peter2003-01-231-23/+0
|
* Nuke CHEAP_TPR stuff, including LOPRIO_LEVEL (bogus) and ALLHWI_LEVELpeter2003-01-233-51/+0
| | | | | (which we never used). There is no need to tweak the TPR anymore and only causes problems.
* Hack around a bug in the 2200 and 2120 controllers that connot DMAscottl2003-01-232-4/+18
| | | | | | | | commands from below the first 8K of physical memory. A better fix is to modify the busdma api to allow either inclusion ranges or multiple exclusion ranges, but that debate is for another day. MFC After: 2 days
* Now that TPR isn't bogusly raised at boot, there is no need to clearpeter2003-01-233-24/+0
| | | | it at context switch.
* Dont raise the TPR register at initialization time. It only causespeter2003-01-231-2/+0
| | | | problems and we only ever clear it.
* Implement SIOCGIFMEDIA for vlan devices by passing the request to thefenner2003-01-221-0/+23
| | | | | | | | parent device, if there is a parent configured. Modify the result returned by the parent to indicate that the only supported media is the currently configured one. Reviewed by: brooks
* YAM: This id is found in a Sony Vaio GRX-670. When will this end?marcel2003-01-222-0/+2
| | | | Submitted by: Chuck McCrobie <mccrobie2000@yahoo.com>
* Revert AHD_TIMER_MAX_US to its original definition now that thegibbs2003-01-221-2/+1
| | | | ahc and ahd modules have correct dependencies on the assembler.
* Update GENSRCS and aicasm options correctly depending on whether registergibbs2003-01-222-26/+28
| | | | | | | | | pretty printing is enabled Add a dependency on the source files for aicasm so that it will be rebuilt if out of date. Simplify.
* Provide a sysctl to allow defaulting of the connectionless (-c) featuredillon2003-01-221-0/+15
| | | | | | | | to mount_nfs. The sysctl defaults to 1 (paranoid mode). Setting it to 0 will allow an NFS client to receive replies on a different IP then they were sent to by default. Submitted by: Sean Eric Fagan <sef@kithrup.com>
* - Move enable_sse()'s prototype to machine/md_var.h.jhb2003-01-2210-30/+18
| | | | | | | | - Sort definition of cpu_* variables appropriately. - Move cpu_fxsr out of the magic non-BSS set of variables and stick it in the BSS along with hw_instruction_sse (make the latter static as well). Submitted by: bde (partially)
OpenPOWER on IntegriCloud