summaryrefslogtreecommitdiffstats
path: root/sys
Commit message (Collapse)AuthorAgeFilesLines
* Expose more of these structures to tthe user so that netstatwollman1997-01-031-3/+3
| | | | | | doesn't walk around with its KERNEL exposed. More commits to follow...
* Move the ethertypes from <netinet/if_ether.h> to <net/ethernet.h>.wollman1997-01-032-18/+20
| | | | | | Many programs need the numbers but don't need the internals of ARP. More commits to follow...
* Separate kernel-internal data structures from exposed user interfacewollman1997-01-032-272/+308
| | | | | | to interfaces. (Amazing nobody had done this!) More commits to fix up user-land to follow.
* Undo the collapse breakage (swap space usage problem.)dyson1997-01-034-16/+4
|
* Oops, delete extra push %edi and push %esi in memsize().kato1997-01-031-3/+1
| | | | Submitted by: Michio "Karl" Jinbo <karl@marcer.nagaokaut.ac.jp>
* Change IMR of master PIC from 0x7f into 0xff in isa_defaultirq().kato1997-01-031-2/+2
| | | | This is a 2.2 candidate.
* Add code to copy the LDT, if required.se1997-01-022-2/+32
| | | | | | | | | | | | | | This code was sent to me by Bruce Evans, and seems to fix some possible kernel panic in case of an execution error. It did not cause any problems on my system, but I did never observe the problem this patch is supposed to fix, anyway. This patch is a NOP, unless the kernel is built with "options USER_LDT", and doesn't affect the GENERIC kernel for this reason. I want to have it in 2.2: it fixes a bug ... Submitted by: bde
* Add Intel VX chip set specific detection and register dump code.se1997-01-021-1/+56
| | | | Submitted by: brianc@netrover.com (Brian Campbell)
* Make it possible to test kernel code in a userland harness, even if itphk1997-01-011-1/+3
| | | | uses MAKE_SET or derivatives and <sys/time.h> at the same time...
* Guess what? We left alot of the old collapse code that is not neededdyson1997-01-014-2/+15
| | | | | | anymore with the "full" collapse fix that we added about 1yr ago!!! The code has been removed by optioning it out for now, so we can put it back in ASAP if any problems are found.
* A very significant improvement in the management of process mapsdyson1996-12-312-43/+46
| | | | | | | | | | | | | | and objects. Previously, "fancy" memory management techniques such as that used by the M3 RTS would have the tendancy of chopping up processes allocated memory into lots of little objects. Alan has come up with some improvements to migtigate the sitution to the point where even the M3 RTS only has one object for bss and it's managed memory (when running CVSUP.) (There are still cases where the situation isn't improved when the system pages -- but this is much much better for the vast majority of cases.) The system will now be able to much more effectively merge map entries. Submitted by: Alan Cox <alc@cs.rice.edu>
* Fix (properly, I hope) 'panic: sillyrename dir' crash that can happenwpaul1996-12-312-4/+6
| | | | | | | | | | | | | | | | | | | | if you do: % cd /nfsdir % mkdir -p foo/foo % mv foo/foo . nfs_sillyrename() self-destructs if you try to sillyrename a directory, however nfs_rename() can be coerced into doing just that by the above sequence of commands. To avoid this, nfs_rename() now checks that v_type of the 'destination' vnode != VDIR before attempting the sillyrename. The server correctly handles this particular situation by returning ENOTEMPTY on the rename() attempt. I asked if this was the correct fix for this on -hackers but nobody ever answered. This is a 2.2 candidate.
* A couple of changes for proper mpu401 recognition, and an opl3 patchjkh1996-12-303-15/+21
| | | | | | | taken from the voxware-3.5 distribution. Also some changes to the SB and MPU IRQs to reflect more common/default settings. Submitted-By: Brian Campbell <brianc@netrover.com>
* Apply a similar fix as in gsc.c rev 1.23.joerg1996-12-301-2/+1
| | | | Pointed out by: bde
* Apparently, someone changed the gsc driver to allocate one big bufferjoerg1996-12-301-1/+0
| | | | | | | | | | | | at device attach time, instead of allocating and freeing buffers as necessary. But he or she forgot to remove the line that invalidated the buffer when the device is closed. Therefore, after using the device for the first time, the buffer was incorrectly invalidated and that caused a page fault on the second, and subsequent uses. Closes PR # kern/2319: Using Genius GS-4500 scanner... Submitted by: jmrueda@diatel.upm.es (Javier Martmn Rueda)
* Update our kernel ntp code to the latest from David Mills. The main changejhay1996-12-306-389/+777
| | | | | is the addition of the FLL code, which is used by the latest versions of xntpd. The kernel PPS code is also updated, although I can't test that yet.
* Add NetBSD/OpenBSD compatable timeradd()/timersub() user-space macros.peter1996-12-302-2/+44
| | | | | | | These are deliberately not visible to the kernel since we have timevaladd() and timevalsub() functions there. Obtained from: NetBSD/OpenBSD
* Add INADDR_LOOPBACK, moved from <rpc/rpc.h>peter1996-12-301-1/+2
|
* First commit of a series of cleanups for the libc rpc code which has beenpeter1996-12-301-7/+3
| | | | | | | | | | | suffering a bad case neglect for the last few years. - Add full prototypes, including to function pointers. - Make the wire protocols 64-bit type safe, eg: 32 bit quantities are int32_t, not long. The orginal rpc code was implemented when an int could be 16 bits. Obtained from: a diff of FreeBSD vs. OpenBSD/NetBSD rpc code.
* Let the VM system know that on certain arch's that VM_PROT_READdyson1996-12-303-3/+19
| | | | | | | | | | | | | | also implies VM_PROT_EXEC. We support it that way for now, since the break system call by default gives VM_PROT_ALL. Now we have a better chance of coalesing map entries when mixing mmap/break type operations. This was contributing to excessive numbers of map entries on the modula-3 runtime system. The problem is still not "solved", but the situation makes more sense. Eventually, when we work on architectures where VM_PROT_READ is orthogonal to VM_PROT_EXEC, we will have to visit this issue carefully (esp. regarding security issues.)
* EEEK!!! useracc and kernacc didn't lock their respectivedyson1996-12-301-2/+16
| | | | | | | maps. Additionally, eliminate the map->hint distortion associated with useracc. That may/may-not be the "right" thing to do -- but time will tell. Submitted by: Partially by Alan Cox <alc@cs.rice.edu>
* Fixed keeping track of interrupt nesting level across ASTs. It sometimesbde1996-12-291-1/+3
| | | | | became -1, and this recently became fatal if an address error occurs in copyin/out/etc.
* Addphk1996-12-291-1/+7
| | | | | | | | SLIST_EMPTY(head) SLIST_FIRST(head) SLIST_NEXT(elm, field) Which do the obvious things while hiding implementation details.
* Reserve 4 malloc types for generic geometry handling.phk1996-12-291-2/+10
|
* Delete bogus inconsistency check that could cause a gratuitous panic. Idg1996-12-291-8/+1
| | | | | had added this years ago when I didn't understand all the subtilties of the flock code.
* Synchronize with sys/i386/isa/npx.c revision 1.35 (disablekato1996-12-291-1/+3
| | | | i586-optimized copyin and copyout).
* Synchronize with sys/i386/i386/machdep.c revision 1.223 (clean-up ofkato1996-12-292-6/+6
| | | | useracc call).
* This commit is the embodiment of some VFS read clustering improvements.dyson1996-12-2913-160/+256
| | | | | | | | | | | | Firstly, now our read-ahead clustering is on a file descriptor basis and not on a per-vnode basis. This will allow multiple processes reading the same file to take advantage of read-ahead clustering. Secondly, there previously was a problem with large reads still using the ramp-up algorithm. Of course, that was bogus, and now we read the entire "chunk" off of the disk in one operation. The read-ahead clustering algorithm should use less CPU than the previous also (I hope :-)). NOTE: THAT LKMS MUST BE REBUILT!!!
* This commit is the embodiment of some VFS read clustering improvements.dyson1996-12-292-2/+4
| | | | | | | | | | Firstly, now our read-ahead clustering is on a file descriptor basis and not on a per-vnode basis. This will allow multiple processes reading the same file to take advantage of read-ahead clustering. Secondly, there previously was a problem with large reads still using the ramp-up algorithm. Of course, that was bogus, and now we read the entire "chunk" off of the disk in one operation. The read-ahead clustering algorithm should use less CPU than the previous also (I hope :-)).
* Superficial cleanup of comment.dyson1996-12-291-2/+2
|
* Superficial clean-up of useracc calls. (The useracc usage ofdyson1996-12-292-6/+6
| | | | | B_READ/B_WRITE is bogus anyway.) Might as well make the call prettier anyway.
* Allow pmap_protect to increase permissions. This mod can eliminatedyson1996-12-292-10/+14
| | | | | the need for unnecessary vm_faults. Submitted by: Alan Cox <alc@cs.rice.edu>
* Eliminate the redundancy due to the similarity between the routinesdyson1996-12-283-84/+18
| | | | | | | vm_map_simplify and vm_map_simplify_entry. Make vm_map_simplify_entry handle wired maps so that we can get rid of vm_map_simplify. Modify the callers of vm_map_simplify to properly use vm_map_simplify_entry. Submitted by: Alan Cox <alc@cs.rice.edu>
* The code unnecessarily created an object with no handle up-front, whichdyson1996-12-281-5/+10
| | | | | | has the negative effect of disabling some map optimizations. This patch defers the creation of the object until it needs to be at fault time. Submitted by: Alan Cox <alc@cs.rice.edu>
* Disabled i586-optimized copyin and copyout again. The fault handlerbde1996-12-283-3/+9
| | | | | | | is still broken - it doesn't restore the floating point state. 2.2-BETA users should disable it using npx0 flags 0x04 the same as 2.2-ALPHA users should have.
* Turn off qcam_debug flag by defaultache1996-12-281-1/+1
| | | | | Should go in 2.2 Reviewed by: pst
* Staticize dmapageport and isa_dmarangecheck.kato1996-12-271-4/+3
|
* Put INCLUDE_CONFIG_FILE into opt_config.h rather than the CFLAGS linepeter1996-12-261-1/+4
|
* Don't synchronously update the directory entry at the end of everybde1996-12-252-4/+4
| | | | | | | | successful write. Only do it for the IO_SYNC case (like ufs). On one of my systems, this speeds up `iozone 24 512' from 32K/sec (1/128 as fast as ufs) to 2.8MB/sec (7/10 as fast as ufs). Obtained from: partly from NetBSD
* Synchronize with sys/i386/isa/sio.c revision 1.155 (use breakpoint()kato1996-12-242-6/+4
| | | | instead of Debugger()).
* Synchronize with sys/i386/i386/userconfig.c revision 1.79 (cosmetickato1996-12-241-2/+4
| | | | change).
* Slightly re-order the sequence of commands issued, so try to sendjoerg1996-12-241-10/+7
| | | | | | | | | | | | | the START UNIT command before testing whether the device is ready. Maybe it should be done even earlier, i'm not 100 % sure. Again, CD changers will most likely benefit from it. While i was at it, also made the debugging case a little more verbose about why the cdopen() yielded an ENXIO. (Only in effect when SCSIDEBUG is specified.) Should eventually also go into 2.2.
* Do something Peter Dufault long since intended to do: make a currentjoerg1996-12-241-4/+18
| | | | | | | | | | | | error code with ASC/ASCQ 4/1 (``Logical unit is in the process of becoming ready'') non-fatal. Retry the operation until it will eventually either yield a real error condition, or finally succeed. Devices like CD changers or tape drives with a freshly inserted cartridge should benefit from this. Should go into 2.2 after some testing in -current. I'd like to see this in the release if possible.
* Use breakpoint() instead of Debugger() in siointr1(). Debugger() doesn'tbde1996-12-233-9/+6
| | | | | work in fast interrupt handlers because it calls db_printf() which uses %es for string stuff and %es isn't initialized.
* Don't redefine SCSI_NCR_DFLT_TAGS even in if FAILSAFE is defined.bde1996-12-231-4/+6
|
* Added undocumented SCSI_DELAY and SCSI_NCR_* options. SCSI_DELAY getsbde1996-12-233-3/+21
| | | | | tested a lot in GENERIC, but the others weren't in any config file and some of them were broken.
* Sorted options.bde1996-12-231-30/+30
|
* Fixed spelling of SCSI_NCR_MAX*.bde1996-12-231-3/+3
|
* Fixed quoting of MAXDSIZ and DFLTDSIZ. The quoting rules changed whenbde1996-12-233-9/+9
| | | | | | they were put in an options header. Should be in 2.2.
* Cosmetic (wrt. the screen display) change: when re-enabling a device,joerg1996-12-231-2/+4
| | | | | | | make sure it won't go into the PCI section. Disabling and re-enabling ed0 made it to the wrong section. Submitted by: msmith
OpenPOWER on IntegriCloud