summaryrefslogtreecommitdiffstats
path: root/sys
Commit message (Collapse)AuthorAgeFilesLines
* Fix a include gottcha in the SMP case.phk1997-12-291-2/+2
| | | | Submitted by: Simon Shapiro <Shimon@Simon-Shapiro.ORG>
* Add the vnode interlock back around vref.dyson1997-12-292-2/+18
|
* Add the vnode interlock back around vget.dyson1997-12-291-8/+2
|
* Sync with sys/i386/conf/options.i386 revision 1.66.kato1997-12-292-4/+4
|
* Sync with sys/i386/i386/microtime.s revision up to 1.39.kato1997-12-291-10/+7
|
* Sync with sys/i386/isa/clock.c revision up to 1.107.kato1997-12-293-144/+114
|
* Sync with sys/i386/isa/sio.c revision up to 1.193.kato1997-12-292-10/+50
|
* Sync with sys/i386/i386/userconfig.c revision 1.99.kato1997-12-291-29/+13
|
* Fixed style bugs in previous commit.bde1997-12-291-3/+3
|
* If available, use the device's LOGICAL blocksize as reported byjulian1997-12-291-53/+69
| | | | | | | | | | | | | the "READ_CAPACITY" command, rather than the physical blocksize reported in the physical geometry code-page. Also don't print out worrying bogus messages when probing a device that has no media. There's no point in printing out something that is unknowable. It just confuses things. Move the check for valid blocksize out of 'open()' to the subroutine that actually finds this out, thus probe/attach can also report and act on the problem.
* Fix the decl of vfs_ioopt, allow LFS to compile again, fix a minor problemdyson1997-12-296-15/+19
| | | | with the object cache removal.
* Lots of improvements, including restructring the caching and managementdyson1997-12-2922-476/+265
| | | | | | | | | | | | | | of vnodes and objects. There are some metadata performance improvements that come along with this. There are also a few prototypes added when the need is noticed. Changes include: 1) Cleaning up vref, vget. 2) Removal of the object cache. 3) Nuke vnode_pager_uncache and friends, because they aren't needed anymore. 4) Correct some missing LK_RETRY's in vn_lock. 5) Correct the page range in the code for msync. Be gentle, and please give me feedback asap.
* Add back a #include <sys/types.h> so that this header issteve1997-12-282-0/+2
| | | | | | | | self-sufficient again. This is a quick fix that should really be remedied by removing all of the gratuitous changes made in revision 1.20. PR: 5351
* More cleanup relating to our use of the TSC.phk1997-12-289-240/+164
| | | | Look in the cpu_feature (CPUID output) to see if we have it.
* wash, sort and put in order various nits from the i586_ctr -> tscphk1997-12-2816-100/+102
| | | | | | commit. Pointed out by: bde
* back out previous commitjulian1997-12-281-24/+13
| | | | got a better version.
* Merge from sys/i386/i386/microtime.s revision 1.36.kato1997-12-281-14/+11
|
* Move the sector size check to the right place,julian1997-12-281-14/+25
| | | | | also cope with some drives that return funny values. The correct fix will come in a few days.
* Fixed initialization of the divisor latch. We depended on siocnopen()bde1997-12-283-6/+63
| | | | | | initializing it for the (usual) (siocniobase != 0) case, but siocnopen() doesn't initialize it if the latch registers already have the correct values.
* YAMFsio.c (always call ttwwakeup() before returning from comstart()).bde1997-12-283-3/+6
|
* Always call ttwwakeup() before returning from comstart(). It isn'tbde1997-12-283-3/+6
| | | | | | necessary to call it when the tty layer's output state has not been changed, but siostop() sometimes changes the TS_BUSY state and then calls comstart() mainly for its side effect of calling ttwwakeup().
* Removed unnecessary (and broken) wakeup code in rpstart(). There is nobde1997-12-282-30/+0
| | | | | | | | | | | need to do it directly, since ttwwakeup() is always called just before returning from rpstart(). The brokenness was waking up the wrong address after clearing TS_SO_OLOWAT. It's not clear how processes waiting for output to drain below low water ever got woken up. Found by: when I fixed longstanding warts in output watermark handling, this was the only driver that knew too much (anything) about the watermarks
* Removed stale comment about printf's deficiencies and rewrote the codebde1997-12-281-29/+13
| | | | | | that worked around them for the CLI devtab listing. Print the `conflicts' flag in the CLI devtab listing.
* Handle "%...p" as "%#...x" instead of "0x%...x". This is a quick fixbde1997-12-281-3/+2
| | | | | | for field widths being 2 larger than specified for "%<number>p". Only printing of null pointers is "wrong" now (it is actually "right", but inconsistent with printf(3)).
* Unspammed nested include of <sys/malloc.h>. <sys/mbuf.h> hasn'tbde1997-12-281-8/+3
| | | | | | | | | really used any features from <sys/malloc.h> since it was "blasted from the past" in rev.1.15, and most .c files did not depend on the namespace pollution (others have been fixed). The mbuf allocation control values M_DONTWAIT and M_WAIT were just defined in terms of the malloc control values M_NOWAIT and M_WAITOK. The mbuf values are now defined directly.
* Restored used include of <sys/malloc.h>. malloc() is not usedbde1997-12-281-21/+22
| | | | | | | | | | | | | here, but kmem_malloc() is used and it takes the same "flags" as malloc(). Use the mbuf allocation "flags" M_WAIT and M_DONTWAIT consistently. There is really only one boolean flag, M_DONTWAIT, but the "flags" were always treated as enum-like values, except in some places here where the values are tacitly converted to boolean flags. Treat them as enum-like values everywhere, except where we tacitly assume that there are only two values in order to convert them to the corresponding two kmem_malloc() "flags".
* Update comment to match updated sb0 line (conflicts keyword no longer neededjkh1997-12-283-12/+3
| | | | at IRQ 5).
* Bring back part of rev 1.44 which was commented out by rev 1.58.alex1997-12-271-7/+8
| | | | Reviewed by: nate
* Unspammed nested include of <vm/vm_zone.h>.bde1997-12-271-3/+1
|
* Back out previous commit, the so-called "unused code" was most definatelypeter1997-12-272-2/+6
| | | | | | used, and caused a reference to an uninitialised variable (state). I think I've fixed it now, but since nothing in the tree seems to use it, I'm not sure.
* Unspammed nested include of <vm/vm_zone.h>.bde1997-12-2726-29/+56
|
* #include "opt_user_ldt.h" so that the #ifdef USER_LDT checks can work, aspeter1997-12-276-6/+12
| | | | | | commented about at length in the PR audit trail. PR: 2412
* Change major number to match the one actually used (whoops!).jkh1997-12-262-2/+2
| | | | Submitted by: Steve Gericke <steveg@comtrol.com>
* Rename "i586_ctr" to "tsc" (both upper and lower case instances).phk1997-12-2622-290/+290
| | | | | | Fix a couple of printfs too. Warning: This changes the names of a couple of kernel options!
* ename "i586_ctr" to "tsc" (both upper and lower case instances).phk1997-12-265-13/+13
| | | | | | Fix a couple of printfs too. Warning: This changes the names of a couple of kernel options!
* Reorder to a more conventional if/then/else/endif structure.phk1997-12-261-14/+11
|
* Fix some breakage that prevented the Plasmon burners from being usedjoerg1997-12-261-67/+66
| | | | | | | | | | | | | | | | | | under -current. The actual preparation of the next track will now be deferred until just before the first write operation. Otherwise, opening the device with write intent will cause the execution of commands that are illegal in `limited command set mode' (i.e., after the write channel has been opened). While i was at it, cleaned up the worm_open() function a bit. Removed the volume overflow pre-check in worm_strategy(). It was time-consuming, and rather useless in many cases anyway (with the size being reported for just the entire volume only), so we can as well let the actual SCSI command fail instead, where it'll properly be reported as EIO. Partially submitted by & discussed with: jmz
* The ioopt code is still buggy, but wasn't fully disabled.dyson1997-12-251-2/+3
|
* Make kern.ncpu reports the number of detected processors when runninggpalmer1997-12-251-1/+8
| | | | with a SMP kernel.
* The spl fixes in in_setsockaddr and in_setpeeraddr that were meant todg1997-12-251-11/+17
| | | | | | | | | fix PR#3618 weren't sufficient since malloc() can block - allowing the net interrupts in and leading to the same problem mentioned in the PR (a panic). The order of operations has been changed so that this is no longer a problem. Needs to be brought into the 2.2.x branch. PR: 3618
* Support running with inadequate swap space. Additionally, the codedyson1997-12-242-8/+16
| | | | will complain with a suggestion of increasing it.
* - Add prototype for adjust_timeout_calltodo().nate1997-12-241-1/+5
| | | | Submitted/forgotten by: Ken Key <key@cs.utk.edu>
* Add a PARITITON arg to SCSI_MKFIXED, and use it tobrian1997-12-232-5/+10
| | | | | | specify RAW_PART for the suopen() in sscioctl(). Now you can `scsi -f /dev/ssc -p'.
* This patch causes the "calltodo" timer list to be decremented by the amountnate1997-12-235-7/+236
| | | | | | | | | | | | | | | | | | | | | | of time that the laptop was suspending. Thus, select() calls that might have suspended rather than firing at 1hr + "time suspended" since the timer was posted. Adding: options APM_FIXUP_CALLTODO to the kernel config enables the patch. [ This patch was slightly modified to use a consistant indent style and I removed some unused local variables. After this has been tested a few weeks we'll make the options the default, so for now I'm now documenting it in LINT. Mike can later if he wants. ] Reviewed by: Mike Smith <msmith@freebsd.org> Submitted by: Ken Key <key@cs.utk.edu>
* Document `flags' for the psm driver.yokota1997-12-233-3/+33
|
* Removed unnecessary setting of 'error' -- binding to a privileged portalex1997-12-231-2/+2
| | | | by a non-root user always returns EACCES.
* Improve my copyright.dyson1997-12-221-11/+4
|
* Improve my copyright.dyson1997-12-221-9/+2
|
* Correct my previous fix for the UPAGES problem.dyson1997-12-222-10/+6
|
* Hopefully fix the problem with the TLB not being updated correctly.dyson1997-12-222-10/+14
| | | | | Problem tracked down by bde@freebsd.org, but this is an attempted efficient fix.
OpenPOWER on IntegriCloud