| Age | Commit message (Collapse) | Author | |
|---|---|---|---|
| 1996-11-04 | Q900, etc. have 2nd scsi adapter at offset of 0x402, not 0x400 from the 1st. | briggs | |
| 1996-11-04 | Improve performance some. Still pretty bad. | briggs | |
| 1996-11-02 | added $OpenBSD$ tag | kstailey | |
| 1996-11-02 | mem* routines for sun3 private libsa | kstailey | |
| 1996-11-02 | Improve reliability on my Q700, at least. Was getting the occasional | briggs | |
| glitch--an extra byte on occasion. | |||
| 1996-10-31 | Correct mmmap's 2nd arg type | Niklas Hallqvist | |
| 1996-10-30 | Merge to NetBSD 961020. Retained our kernel APIs where NetBSD has changed. | Niklas Hallqvist | |
| -Wall -Wstrict-prototypes -Wmissing-prototypes too. | |||
| 1996-10-30 | Add an esp1. | briggs | |
| 1996-10-30 | Make a guess that scsi-b's IRQ is on bit 0 of via2. | briggs | |
| 1996-10-30 | Go ahead and try dual scsi even though we'll probably need to make another | briggs | |
| pass at it. Macs with scsi offsets of 0x10000 have a SCSI CLK of 16.5MHz. All other (68k-based) appear to use 25MHz. | |||
| 1996-10-30 | Fix spelling error in panic. Double since of ResHndls[]. Use I/O kludge on | briggs | |
| all systems, not just CLASSII--A bunch of Quadras apparently have hardcoded I/O addresses in the ROMs, too. | |||
| 1996-10-29 | bring the source tree a bit up to date with my local tree | Thomas Graichen | |
| 1996-10-29 | removed by accident | Michael Shalayeff | |
| 1996-10-29 | not int new libsa, or moved | Michael Shalayeff | |
| 1996-10-28 | SCSI driver, functional but slow on the Quadra 700. | briggs | |
| 1996-10-28 | Remove a blank line. | briggs | |
| 1996-10-28 | Include macinfo.h. | briggs | |
| 1996-10-28 | Add a stub for mac68k_register_scsi_b_irq() to be filled in when we figure out | briggs | |
| which interrupt Apple decided to use. | |||
| 1996-10-28 | Add prototype for function to install SCSI-B IRQ. | briggs | |
| 1996-10-28 | Include macinfo.h. | briggs | |
| 1996-10-28 | Mac model info out of cpu.h and into new macinfo.h. | briggs | |
| 1996-10-28 | Rename ncr96scsi/scsi96 to esp. | briggs | |
| 1996-10-28 | Fix d_done prototype | Niklas Hallqvist | |
| 1996-10-28 | Obsolete. | briggs | |
| 1996-10-28 | Most 040 models have scsi regs at base + 0x10000, not base + 0xf000 | briggs | |
| like the Q700. | |||
| 1996-10-28 | Enable quotas | Thorsten Lockert | |
| 1996-10-27 | Check for __OpenBSD__. From Matthieu Herrb <Mathieu.Herrb@mipnet.fr> | Todd C. Miller | |
| 1996-10-26 | ep at same places at generic | Theo de Raadt | |
| 1996-10-26 | floppy driver is now in dev/isa. Not in files.isa yet, though, since arc | Jason Downs | |
| changes are still needed. | |||
| 1996-10-25 | grow kvm space; fix an over-agressive pmap optimization | Theo de Raadt | |
| 1996-10-25 | be slightly more verbose for my sake | Jason Downs | |
| 1996-10-25 | sync | Jason Downs | |
| 1996-10-24 | Discourage MATH_EMULATE. | Jason Downs | |
| 1996-10-24 | kernel is 'bsd' | Michael Shalayeff | |
| 1996-10-24 | add 3 another build directories | Michael Shalayeff | |
| 1996-10-23 | correct disklabel writing; from price | Theo de Raadt | |
| 1996-10-23 | avoid long lines | Theo de Raadt | |
| 1996-10-23 | An empty arglist is called __P((void)) not __P(()) | Niklas Hallqvist | |
| 1996-10-23 | Get memory configuration from boot blocks or /boot when it becomes | etheisen | |
| available. Use MEM_COMPUTE option to use old behavior. Use EXTMEM_SIZE to override extended memory size. | |||
| 1996-10-23 | Patches from Dave Huang <khym@bga.com> to get something in and out of the | briggs | |
| Quadra SCSI. Patched by me. VERY slow (~50K/s), but working. | |||
| 1996-10-23 | Sync up with NetBSD and include patch from Dave Huang <khym@bga.com> | briggs | |
| that puts AV I/O in the right place. | |||
| 1996-10-22 | reality check | Jason Downs | |
| 1996-10-22 | patch com speed initialization | Jason Downs | |
| 1996-10-21 | -Wall happiness | Michael Shalayeff | |
| 1996-10-21 | random is stock now | Michael Shalayeff | |
| 1996-10-21 | Added 'S' for Stack tracebacks. There are usually 3 items on the | imp | |
| stack before the routine that caused you to get into the debugger (due to the calls to stacktrace, actually). However, other than that garbage on the top of the stack, this has proven to be useful for me in tracking down strange things. While it might be better to get mdbstacktrace working (since it looks like it wants to be more complete), this gets functionality right away. Hope people find it useful. | |||
| 1996-10-21 | Fix two bugs noticed by Taras Ivanenko <ivanenko@ctpa03.mit.edu> | briggs | |
| via.c: don't allow spl to remain at splhigh on error. mainbus.c: Make sure that variable is initialized. | |||
| 1996-10-20 | Out of NetBSD (scottr): | briggs | |
| Define MACH_CLASSQ2, used by Performa 575. Remove susword() because of prototype conflict with <sys/systm.h>. | |||
| 1996-10-20 | Spaces -> tabs that were my fault | imp | |
| 1996-10-20 | Truncate rdpmc result to 40 bits. Make pctr.c build with -Wall. | dm | |
 |
index : src | |
| OpenBSD base system |
| summaryrefslogtreecommitdiff |