summaryrefslogtreecommitdiffstats
path: root/sys/dev
Commit message (Collapse)AuthorAgeFilesLines
* 0 is not an invalid interrupt in the PCI world (just in the ia32imp2002-06-071-2/+4
| | | | | | world), do not treat it as such. This fixes the alpha boot problem. Reviewed by: drew, des
* Recognize Adaptec ANA-5910/30/40[A] boards.mdodd2002-06-073-19/+65
| | | | | | Read the MAC address from Adaptec boards correctly. Bits borrowed from sys/pci/if_en_pci.c.
* Move some code around.mdodd2002-06-074-16/+18
| | | | Minor whitespace changes.
* Hook up the ahd driver.gibbs2002-06-067-27/+11
|
* Remove one more multi-line string literal.gibbs2002-06-061-5/+5
|
* Make sc_saver_keyb_only (sceen saver interrupted by keyboard input only)alfred2002-06-061-1/+1
| | | | the default.
* Silence GCC warnings about multi-line strings.gibbs2002-06-057-48/+48
| | | | Sync Perforce IDs.
* Add some PCI-X register definitions.gibbs2002-06-051-1/+18
| | | | | PCIM_CMD_SERREN -> PCIM_CMD_SERRESPEN to be consistent with the PERR definition.
* Enter the ahd driver which supports the Adaptec AIC7902 Ultra320, PCI-Xgibbs2002-06-059-0/+19742
| | | | SCSI Controller chip.
* Add device id. for fxp chip on Intel D845EPT2L boards. This seems toambrisko2002-06-051-0/+1
| | | | | | | make the onboard NIC work. Sponsored by: Vernier Networks MFC after: 1 day
* Don't treat statistics counter wrap-overs as errors.tmm2002-06-051-10/+3
|
* Added support for 82545EM and 82546EB based adapters.pdeuskar2002-06-036-1751/+7198
| | | | | | Added Vlan support. MFC after: 1 week
* Use a per-device worker thread to avoid blocking in mdstrategy()iedowse2002-06-031-11/+36
| | | | | | | until the I/O completes. This fixes some easily reproducable deadlocks that occur when using md(4) with GEOM. Reviewed by: phk
* Quick and dirty convert to newbus. (Eventually 'eni.c' should go away.)mdodd2002-06-036-18/+670
| | | | | | | Module loads and unloads properly. Thanks to Richard Hodges <rh@matriplex.com> for donating the hardware to allow me to work on this driver.
* Fix typo in the BSD copyright: s/withough/without/schweikh2002-06-0220-20/+20
| | | | | Spotted and suggested by: des MFC after: 3 weeks
* Fix warnings.alfred2002-06-011-3/+3
| | | | | There's no need to use concatination when you have 'something.macro_arg'. Comment out comment following #endif.
* Fix a couple of places where preprocessor concatination was misused andalfred2002-06-011-2/+2
| | | | something wound up grafted to a comma.
* Silence warning.alfred2002-06-011-2/+2
| | | | | When casting a "const void *" to a "struct foo **" you want to actually cast it to "struct foo * const *" not simply "const struct foo **".
* Quick, low impact band-aide to unbreak the build. Added && 0 so weru2002-06-011-1/+1
| | | | | | | ifdef out what appears to be a bogus call to softintr using data elements that aren't in the softc. Forgotten by: imp
* Should have been part of recent commit:alfred2002-06-012-5/+3
| | | | | | | | Fix GCC warnings caused by initializing a zero length array. In the process, simply things a bit by getting rid of 'struct ng_parse_struct_info' which was useless because it only contained one field. But now I'm unbreaking compilation by adjusting these files to the recent netgraph change.
* Call destroy_dev in detach routine to cleanup properly otherwise wealfred2002-06-011-3/+8
| | | | | | | | | | | panic because of a repeat make_dev if/when the device is reattached to the system. Remove an "#if __FreeBSD__" in code that's nested under a "#if __NetBSD__" (*sigh*) Reported by: Seth Hettich <sjh@whiskey.ucf.ics.uci.edu> Tested by: Seth Hettich <sjh@whiskey.ucf.ics.uci.edu>
* Use PCI_INTERRUPT_VALI in stead of hard coded 255imp2002-06-011-5/+5
|
* More style(9) nitsimp2002-06-011-20/+23
|
* Define a PCI_INVALID_IRQ to augment PCI_INTERRUPT_VALIDimp2002-06-011-1/+2
|
* Make this file mostly conform to style(9).imp2002-06-011-87/+86
| | | | Approved by: msmith in principle before walkabout
* Quick, low impact band-aide to unbreak the build. Added && 0 so weimp2002-05-313-3/+3
| | | | | ifdef out what appears to be a bogus call to softintr using data elements that aren't in the softc.
* Warnings cleanup for gcc3. Also __FUNCTION__ -> __func__anholt2002-05-3113-148/+148
| | | | Approved by: des
* Use __FreeBSD_version to test what things to useimp2002-05-312-10/+11
|
* Fix build breakage on my pccard related commit.takawata2002-05-312-2/+2
|
* Add missing unlockjulian2002-05-301-1/+3
| | | | Submitted by: dirkx@covalent.net
* Make oldcard and newcard kernel module work.takawata2002-05-309-6/+99
|
* Check for defined(__i386__) instead of just defined(i386) since the compileralfred2002-05-303-9/+9
| | | | will be updated to only define(__i386__) for ANSI cleanliness.
* Check for defined(__i386__) instead of just defined(i386) since the compileralfred2002-05-302-5/+5
| | | | will be updated to only define(__i386__) for ANSI cleanliness.
* Add one copy of crc32() and crc32_tab[] in libkern, and remove it two otherphk2002-05-291-55/+0
| | | | | | | places. Comment out crc32 related definitions in zlib.h, we don't seem to have the corresponding code in our kernel.
* NetBSD have adopted our usage of the DMAADDR macro:joe2002-05-283-12/+10
| | | | | | | | date: 2002/05/28 12:42:39; author: augustss; Change DMAADDR macro slightly. Update the $NetBSD$ tags to reflect this and make slight changes to usb_mem.h so that we're in sync with each other.
* Use %02d in track numbers, so that 1 is 01, much easier for scriptssos2002-05-281-1/+1
|
* Regen.joe2002-05-272-2/+16
|
* Add a couple of new aue ethernet adapters from NetBSD:joe2002-05-272-0/+4
| | | | | ELECOM LDUSBLTX IODATA USBETTXS
* Use aue_lookup for looking up devices.joe2002-05-262-86/+71
|
* Correct the usage of DMAADDR in a piece of '#if 0'd code. (The compilerjoe2002-05-261-1/+1
| | | | didn't pick it up.)
* MFNetBSD:joe2002-05-261-2/+2
| | | | | | | revision 1.124 date: 2002/05/26 03:10:02; author: minoura; state: Exp; lines: +3 -3 Clear done_head in the HCCA *before* acknoledging the interrupt. Driver lost some completed transfers under heavy loads.
* MFNetBSD: ohci.c (1.124), uhci.c (1.159), usbdi.c (1.100)joe2002-05-263-9/+9
| | | | | | | | | date: 2002/05/19 06:24:31; author: augustss; state: Exp; Update dma memory access API a little. NetBSD have adopted our way of using the KERNADDR macro. Update the revision tags to show that we're in sync, and remove the casts that they did in their adaptation.
* Mis-edit in last commit.phk2002-05-261-0/+1
|
* Be a bit smarter about rewriting data so we don't loose too much performance.phk2002-05-261-21/+21
| | | | Sponsored by: DARPA & NAI Labs.
* Use an umazone per unit for allocating the sectors for malloc backing.phk2002-05-261-14/+32
| | | | | | Clean up things properly when we unconfigure malloc backed units. Sponsored by: DARPA & NAI Labs.
* Give the "malloc" backing of md(4) an adaptive multilevel index tree tophk2002-05-251-64/+225
| | | | | | | | | | | remove the need for a contiguous array with pointers to all the sectors. Try to make failure to malloc(9) memory a non-hang situation. Eventually this will allow us to test the 64bit cleanness of the disk I/O patch, but more work is outstanding here and elsewhere. Sponsored by: DARPA & NAI Labs.
* Fixed printf format errors which apparently crept in while -Wformat wasbde2002-05-255-13/+10
| | | | disabled for gcc-3.
* "Fixed" printf format errors (64-bit daddr_t casualties; should use %jxbde2002-05-251-4/+4
| | | | | | | | | instead of %llx when %j is available). Changed nearby output formats from %x to %#x so that it is obvious that the numbers are in hex (vinum mostly uses 0x%x elsewhere). Didn't fix nearby format printf errors (long lines).
* Fixed printf format errors. Most of them are 64-bit daddr_t casualties.bde2002-05-251-4/+6
| | | | | | | | | | | | Printing daddr_t's using %d format was always an error, but gcc's warning about it was ignored for supported 64-bit arches and not printed for supported 32-bit arches. Hundreds if not thousands thousands of previously "fixed" daddr_t printings are now broken on 32-bit machines by casting daddr_t's to longs. daddr_t's should be printed using %jd format, but this fix uses %lld since %j is not implemented in the kernel yet. Fixed some nearby format printf errors (style bugs).
* Fully reset a gem on some error conditions; otherwise it would hang intmm2002-05-241-0/+4
| | | | | | about 1 of 10 cases. Proposed and tested by: phk
OpenPOWER on IntegriCloud