summaryrefslogtreecommitdiffstats
path: root/sys
Commit message (Collapse)AuthorAgeFilesLines
* Fix bug where bundle-level receive statistics were not getting updated.archie2000-08-091-0/+2
|
* MFS: sync with peter's isadma fixupcg2000-08-091-4/+6
|
* activate via82c686 audio drivercg2000-08-093-3/+5
|
* new driver: via 82c686 chipset audiocg2000-08-093-0/+770
| | | | | | | not personally tested, but i'm assured it works Submitted by: David Jones <dej@ox.org> PR: kern/20161
* warn if hwptr goes backwardscg2000-08-092-1/+8
|
* Initialize *countp to 0 in stub for softdep_flushworklist().tegge2000-08-091-0/+1
| | | | | | This allows ffs_fsync() to break out of a loop that might otherwise be infinite on kernels compiled without the SOFTUPDATES option. The observed symptom was a system hang at the first unmount attempt.
* MF4: fix style(9) whitespace breakage.peter2000-08-081-3/+3
|
* Don't attempt to actually read SAF-TE temperature objects- nobody seemsmjacob2000-08-081-19/+21
| | | | | | | | | | to be obeying the original spec as to what the numeric value means. Temperature flags are unaffected- these are still the 'pseudo-thermometers' and overtemp/undertemp warnings will be caught and translated to SES objects here. PR: 20475
* o Add POSIX1E_CAPABILITY_EXTATTR_NAME, a string constant for therwatson2000-08-081-3/+6
| | | | | | | | | | | | | extended attribute name used to store capabilities for binaries in FFS. o Uncomment CAP_SETPCAP because, unfortunately, it is useful when developing and testing, as well as in practice for a system not fully capability-enabled. o Synch up prototypes with the implementation, reducing diffs and making the patches for capabilities more accessible and less dastardly. Obtained from: TrustedBSD Project
* o Introduce vn_extattr_{get,set}, wrapper routines for VOP_GETEXTATTRrwatson2000-08-082-0/+78
| | | | | | | | | | and VOP_SETEXTATTR to simplify calling from in-kernel consumers, such as capability code. Both accept a vnode (optionally locked, with ioflg to indicate that), attribute name, and a buffer + buffer length in UIO_SYSSPACE. Both authorize the call as a kernel request, with cred set to NULL for the actual VOP_ calls. Obtained from: TrustedBSD Project
* Fix braino in the tsleep call, ata_delayed_attach might be a NULLsos2000-08-081-1/+1
| | | | pointer..
* Merged from sys/i386/conf/GENERIC revisions 1.266 and 1.267.nyan2000-08-081-5/+5
|
* Merged from sys/i386/conf/GENERIC.hints revision 1.3.nyan2000-08-081-1/+1
|
* Changed default cursor shape to non-blink mode.nyan2000-08-084-6/+12
| | | | Submitted by: Tomokazu HARADA <tkhara@osk4.3web.ne.jp>
* Fixed PC-9821 NOTE supports with LINE30 mode.nyan2000-08-084-6/+32
| | | | Submitted by: Tomokazu HARADA <tkhara@osk4.3web.ne.jp>
* Fix the comments to properly document the PQ_MEDIUMCACHE andkbyanc2000-08-082-4/+4
| | | | | | | PQ_NORMALCACHE options. PR: 20409 Submitted by: Tony Finch <dot@dotat.at>
* Typo/spelling fixes.rnordier2000-08-081-4/+4
|
* The Yamaha CDR100 doesn't seem to like multi-LUN probing.ken2000-08-081-0/+4
| | | | | PR: kern/20347 Submitted by: Andre Albsmeier <andre.albsmeier@mchp.siemens.de>
* This is an overhaul of the mode page handling in camcontrol as well askbyanc2000-08-081-1/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | related patches. These include: * Mode page editting can be scripted. This involves two things: first, if stdin is not a tty, changes are read from stdin rather than invoking $EDITOR. Second, and more importantly, not all modepage entries must be included in the change set. This means that camcontrol can now gracefully handle more intrusive editting from the $EDITOR, including removal or rearrangement of lines. It also means that you can do stuff like: # echo "WCE: 1" | camcontrol modepage da3 -m 8 -e # newfs /dev/da3 # echo "WCE: 0" | camcontrol modepage da3 -m 8 -e * Range-checking on user-supplied input values. modeedit.c now uses the field width specifiers to determine the maximum allowable value for a field. If the user enters a value larger than the maximum, it clips the value to the max and warns the user. This also involved patching cam_cmdparse.c to be more consistent with regards to the "count" parameter to arg_put (previously is was the length of strings and 1 for all integral types). The cam_cdbparse(3) man page was also updated to reflect the revised semantics. * In the process, I removed the 64 entry limit on mode pages (not that we were even close to hitting that limit). This was a nice side-effect of the other changes. * Technically, the new mode editting functionality allows editting of character array entries in mode pages (type 'c' or 'z'), however since buff_encode doesn't grok them it is currently useless. * Camcontrol gained two new options related to mode pages: -l and -b. The former lists all available mode pages for a given device. The latter forces mode page display in binary format (the default when no mode page definition was found in scsi_modes). * Added support for mode page names to scsi_modes. Allows names to be displayed alongside mode numbers in the mode page listing. Updated scsi_modes to use the new functionality. This also adds the semicolon into the scsi_modes syntax as an optional mode page definition terminator. This is needed to name pages without providing a page format definition. * Updated scsi_all.h to include a structure describing mode page headers. * Added $FreeBSD$ line to scsi_modes. Inspired by: dwhite Reviewed by: ken
* Fix a bug where we were accessing already free'd memory during node shutdown.archie2000-08-071-1/+1
| | | | Detected via: 0xdeadc0de
* Add three new control messages to the ng_ether(4) netgraph node type:archie2000-08-072-4/+83
| | | | | | NGM_ETHER_GET_ENADDR: Get the device's Ethernet address NGM_ETHER_SET_PROMISC: Enable/disable promiscuous mode NGM_ETHER_SET_AUTOSRC: Enable/disable packet source address override
* Fix some warnings. Here are more part-time volatiles - ie: data thatpeter2000-08-071-3/+3
| | | | | lives in memory and is sometimes busmastered to/from the controller. I believe these are all ok.
* Make the kqueue socket read filter honor the SO_RCVLOWAT value.jlemon2000-08-071-1/+1
| | | | Spotted by: "Steve M." <stevem@redlinenetworks.com>
* Close PR 20438. Make fix for preserving LED settings conditional onwpaul2000-08-072-6/+14
| | | | presence Intel 21143 chip.
* Fix bug with timeout; previously, when attempting to poll the kqueue byjlemon2000-08-071-14/+22
| | | | | | | | | | | | | | | passing a zero-valued timeout, the code would always sleep for one tick. Change code to avoid calling tsleep if we have no intention of sleeping. Bring in bugfix from sys_select.c, r1.60 which also applies here. Modify error handling slightly; passing in an invalid fd will now result in EBADF returned in the eventlist, while an attempt to change a knote which does not exist will result in ENOENT being returned. Previously such attempts would fail silently without notification. Pointed out by: nicolas.leonard@animaths.com Rick Reed (rr@yahoo-inc.com)
* Adjust TCP checksum rather than compute it afresh.ru2000-08-071-4/+8
| | | | Submitted by: Erik Salander <erik@whistle.com>
* Back this one out until I can fix _all_ of the headers. The headers suregreen2000-08-072-2/+2
| | | | are shpxed quite nicely.
* Add workaround for livelock problem when starting APs.tegge2000-08-072-0/+8
| | | | | | | | With more than 1 AP present, an AP could fail to properly release the mp lock before waiting for smp_started to become nonzero. With early startup of APs, the BSP could fail to properly release the mp lock before waiting for smp_started to become nonzero.
* Save and restore the registers SOF and FRAMENUMBER on suspend. In somen_hibma2000-08-072-5/+15
| | | | | | | | | cases the registers are not correctly set on resume. This solves the problem of USB failing after resuming a machine. Submitted by: mike+fbsd@medianstrip.net PR: 18261
* Regen.n_hibma2000-08-062-7/+30
|
* New idsn_hibma2000-08-061-0/+4
| | | | | Submitted by: Scott Long PR: 18897
* Some new ids, from NetBSDn_hibma2000-08-061-1/+2
|
* sync CCR register definitions with creative sourcescg2000-08-061-1/+3
|
* Add experimental code for ATA100 support on:sos2000-08-063-64/+199
| | | | | | | | | | | Promise Ultra100 / Fasttrak100 HighPoint HPT370 controllers (fx Abit KA7-100 onboard ctrl, Abit HotRod 100) Intel ICH2 (Intel 815E based motherboards) So far I can read >90MB/s on the Promise and the HPT370. I can write >64MB/s on the promise and >50MB/s on the HPT370 so it seems writing is still done in ATA66 mode :( The ICH2 support is untested as of yet...
* Use tsleep instead od DELAY in probe when not in boot.sos2000-08-061-1/+4
| | | | This fixes the uptil 30s hangs on PCCARD ata device probes.
* Cleaning conflicting module tree. Use modules/3dfx instead.cokane2000-08-061-22/+0
|
* Fix clean to remove pci.hcokane2000-08-061-7/+11
| | | | | | Numerous style fixes Submitted by: Marcel Moolenaar <marcel@cup.hp.com>
* fix the staticy sound issuecg2000-08-061-154/+140
| | | | | | use timer instead of per-channel interrupts do playback like the linux driver - may fix nmi-with-ecc issue
* Remove sequence checking in the burner support.sos2000-08-062-49/+4
| | | | | This was really a leftover from the wormcontrol days, it is no longer needed when using burncd.
* Be a bit more restrictive in using multisector transfers.sos2000-08-061-4/+6
|
* - Fixed missing initialization of current device number.kato2000-08-062-1/+2
| | | | | | - Fixed comment. Pointed out by: nyan
* Change the behavior of isa_nmi to log an error message instead ofps2000-08-067-99/+217
| | | | | | | | | panicing and return a status so that we can decide whether to drop into DDB or panic. If the status from isa_nmi is true, panic the kernel based on machdep.panic_on_nmi, otherwise if DDB is enabled, drop to DDB based on machdep.ddb_on_nmi. Reviewed by: peter, phk
* None of select.h needs to be exposed to !_KERNEL.green2000-08-062-2/+2
|
* Be more verbose when changing APIC ID on an IO APIC.tegge2000-08-0611-14/+1008
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | Don't allow cpu entries in the MP table to contain APIC IDs out of range. Don't write outside array boundaries if an IO APIC entry in the MP table contains an APIC ID out of range. Assign APIC IDs for all IO APICs according to section 3.6.6 in the Intel MP spec: - If the current APIC ID on an IO APIC doesn't conflict with other IO APICs or CPUs, that APIC ID should be used. The copy of the MP table must be updated if the corresponding APIC ID in the MP table is different. - If the current APIC ID was in conflict with other units, the corresponding APIC ID specified in the MP table is checked for conflict. - If a conflict is still found then fall back to using a new unique ID. The copy of the MP table must be updated. - IDs out of range is considered to be in conflict. During these operations, the IO_TO_ID array cannot be used, since any conflict would have caused information loss. The array is then corrected, since all APIC ID conflicts should have been resolved. PR: 20312, 18919
* Fix misspelling.archie2000-08-051-1/+1
|
* Argh! Fix a brainfart of mine. In the old boot0, we relocated ourselfjhb2000-08-041-1/+1
| | | | | | | | | | | | | | | | to 0x600 via a 'rep movsw'. Once that was done, %cx was zero, so we could simply use 'movb' to update the lower byte of %cx in preparation for zeroing out the fake partition entry used to boot to other drives via F5. Well, in the new boot0, we don't actually relocate ourselves, instead it is easier to create the fake partition entry first and then just use it to get the BIOS to load all of boot0 into memory at 0x600. However, since we aren't doing the relocate code anymore, we don't know that %cx == 0 when we hit the 'movb' to setup %cx for clearning the fake partition entry. Thus, if %ch != 0 when the BIOS started boot0, then it would end up zeroing a lot more memory than just 8 words. The solution is to do a word move of $8 into %cx. Debugging help from: David Wolfskill <dhw@whistle.com>
* Cleanup debug messages and Add some enhancements from linux oniwasaki2000-08-043-126/+105
| | | | | | | | | | | | | | display control by apm -d. - Remove APM_DEBUG to avoid re-build kernel with such a unspported optioin. - Introduce new denug flag `debug.apm_debug' which can be controlled by sysctl interface and loader by setting like "debug.apm_debug=1", you will get debug messages from APM driver. - Add some enhancements from linux on display control by apm -d. I'm expecting that we can see some improvements on some laptops where apm -d doesn't work correctly so far. Reviewed by: -mobile and -current folks (no objection) Suggested by: Susumu WAKABAYASHI <susumu@wakabaya.net>
* Finally make the module dependencies work. kern_linker.c now allowsn_hibma2000-08-042-25/+20
| | | | | | | | modules to depend on modules in the same file (uhub depends on usb) or even on themselves (usb on usb, makes the define in usb_port.h a lot less convoluted). Use ANSI prototypes.
* *smack* #if 0, not #ifdef 0.wpaul2000-08-041-1/+1
|
* The check_extport() function appears to have some new code in it that checkswpaul2000-08-041-0/+7
| | | | | | | | | | | | | | the scratch RAM for data normally found in the SEEPROM (presumably in the event that the SEEPROM is unavailable or can't be read). This code causes a spontaneous reboot on monster.osd.bsdi.com, which has an embedded aic7880 controller. The problem appears to happen either when it writes to the SCBPTR port and then reads from the SCB_CONTROL port. Somewhere during the inb/outb operations, the system has a heart attack and restarts. This code looks very suspicious, particularly since it has unconditionalized debug mesages such as "Got here!" and "And it even worked!". With this block #ifdef'ed out, the machine boots and runs properly. I stronly suggest that it stay #ifdef'ed out until it's properly tested.
OpenPOWER on IntegriCloud