Age | Commit message (Collapse) | Author | |
---|---|---|---|
2006-10-04 | More tidying up after bad144. Remove references in comments and | Kenneth R Westerback | |
don't include dkbad.h when nothing in it is used. Missed arm/include/disklabel.h first time around. | |||
2006-08-31 | Make sure we do mbuf operations at splnet(), for some models have ze | Miod Vallat | |
interrupting at spl4 (i.e. below splnet) and this can cause pool corruption; ok deraadt@ | |||
2006-08-30 | Horrible code to work around stupid VXT2000 rom getchar() behaviour, but | Miod Vallat | |
at least this allows our boot blocks to work with glass console on these machines now. Caution: reading this code will hurt your eyes. ok deraadt@ | |||
2006-08-30 | Enable VXT2000 bits now that they are worth using; ok deraadt@ | Miod Vallat | |
2006-08-30 | Teach the SGEC driver about the Real World: | Miod Vallat | |
- check for error bits in descriptors, and count and report errors accordingly; make sure we do not pass up known-to-be-bad packets up the receive path. - check for various error conditions which would cause the transmit process to go to suspended state, and recover from them, before the timeout expires. - finally, on the VXT2000, the transmit process for some reason will reset its ring position to the beginning of the ring every time it completes a transmission burst, while we don't, so play with the head of ring register behind its back. This makes my VXT2000+ (net)boot multiuser. Tested on VXT2000 (ze@vxtbus), KA49 (ze@vsbus) and KA53 (ze@ibus); ok deraadt@ | |||
2006-08-30 | Better diagnostic messages. | Miod Vallat | |
2006-08-27 | Work-in-progress VXT2000 support (commented out in GENERIC so far). Glass | Miod Vallat | |
console works on color model, serial console works, ethernet attaches but does not work correctly yet. | |||
2006-08-27 | Extract the device-independent parts of dzms, to let lkms@non-dz exist. | Miod Vallat | |
Don't forget to rerun config(8)! | |||
2006-08-27 | Move cfdriver structs to better places, in preparation for new attachments. | Miod Vallat | |
2006-08-26 | Doh, compile before commit. | Miod Vallat | |
2006-08-26 | Enable all planes in the ramdac as well, so that color really works. | Miod Vallat | |
2006-08-24 | Add a few printfs to slow down SGEC initialization - somehow this appears | Miod Vallat | |
to be necessary on VXT2000. While large SDELAY are not enough. Someone will probably find my lack of faith in the SGEC code disturbing. | |||
2006-08-24 | Check for a selftest failure after reset in the correct register. | Miod Vallat | |
2006-08-24 | Enable the appropriate serial transmitters in VXT2000 serial routines, this | Miod Vallat | |
makes output work when the VXT2000 is not configured for serial console. | |||
2006-08-22 | More register defines. | Miod Vallat | |
2006-08-22 | Wrap dmesg | Miod Vallat | |
2006-08-17 | Check d_secpercyl in all readdisklable() functions, and have all of | Kenneth R Westerback | |
them return 'invalid geometry' when d_secpercyl == 0. While there move the check to a consistant location (after the check of d_secperunit) and use a consistant idiom (i.e. some readdisklabel()'s have no 'done' label). prodded by thib@ after a bad macppc experience. ok deraadt@ | |||
2006-08-15 | Some VS4k/90 have the uVAX bit set in confdata, do not report them as uVAXens. | Miod Vallat | |
2006-08-15 | A more reliable LCSPX probe, based on information from Blaz Antonic, tested | Miod Vallat | |
by aanriot@ | |||
2006-08-11 | Use d_secperunit rather than 0x1fffffff as the default value for | Kenneth R Westerback | |
RAW_PART's p_size. Since d_secperunit is initialized to 0x1fffffff if not specified there should be no need to use the magic number again. And if d_secperunit was specified then that value should be used instead of the magic number. This was already being done for hp300, luna88k, mac68k, mvme68k, mvme88k. Should be a no-op. | |||
2006-08-06 | Replace magic numbers with adequate #defines, support vsync disabling in the | Miod Vallat | |
burner routine, and do the colormap updates during vertical retrace, instead of using a fixed delay. | |||
2006-08-05 | Ugly hack to avoid sending the ``all keys up'' event to userland, replacing | Miod Vallat | |
it with as many key release events as necessary. Necessary for X until it knows how to handle this event properly. | |||
2006-08-05 | Colormap ioctls, at last. | Miod Vallat | |
2006-08-05 | If a dz child driver has taken over a line, it will never be accessible | Miod Vallat | |
as /dev/tty0X, period. So return ENXIO instead of EBUSY in dzopen(). | |||
2006-08-05 | Make sure smg will not pretend being a console on smg+gpx hardware with | Miod Vallat | |
smg forced to attach. | |||
2006-08-05 | A potpourri of unrelated changes: | Miod Vallat | |
- remove the 11/750 pcs code, it is never called. - remove the list of kernels to try to boot in order; it is a good idea only if you can interrupt the process, which you can't due to quirks on some models, so enter interactive boot immediately if booting /bsd fails. - work around some PROM bugs in graphics mode which cause \h and \t not to behave as expected. Crank version to 1.10. | |||
2006-08-05 | Use faster {copy,erase}cols routines for smg, with ad hoc vax assembly, | Miod Vallat | |
adapted from X11 via hp300. | |||
2006-08-03 | LK-401 french layout, tested on actual keyboard (I even have two of them | Miod Vallat | |
in the closet, does anyone want them?) | |||
2006-08-03 | A more reliable way to find out that the card is not there. | Miod Vallat | |
2006-08-03 | Enable all planes, so that color has a real chance to work; still untested. | Miod Vallat | |
2006-08-03 | Since we will only attempt to configure a keyboard on line 0 on machines | Miod Vallat | |
where line 0 is a keyboard connector, which is never used for serial console, remove an always-true test. | |||
2006-08-03 | No need to force a 8-bit font anymore, welcome to the 12x22 font. | Miod Vallat | |
2006-08-02 | Provide the usual 8x16 font, instead of the qvss 8x15 font, for low-res vax | Miod Vallat | |
frame buffers (and smg until it can use non-8bit wide fonts), as the latter has too many symbols in the 8-bit range stubbed out. | |||
2006-08-01 | Avoid a few NULL dereferences, and boot -c works on vax with glass console. | Miod Vallat | |
2006-08-01 | Support for font widths from 9 to 16 bits, and use the 12x22 font by | Miod Vallat | |
default (since it fits). | |||
2006-08-01 | If there is VRAM at the lcg address, but we can't make sense of the | Miod Vallat | |
configuration register, match anyway and bail out during attach, printing the configuration register value, so that this gets noticed and the code can be made aware of this particular model. | |||
2006-08-01 | make this more like a real cylon mode (no pauses at the end!) | Theo de Raadt | |
2006-07-31 | Since we now know which keyboard type we are connected to: | Miod Vallat | |
- report the proper type to WSKBDIO_GTYPE. - merge the LK201 and LK401 keymaps. - if the keyboard is LK201, handle the compose key as the left alt key, unless shifted, so that you can have both an alt and a compose key. | |||
2006-07-31 | Underline support. | Miod Vallat | |
2006-07-31 | More dead structures and defines... | Miod Vallat | |
2006-07-31 | Allow more time for the keyboard to initialize itself, helps on slower VS3100. | Miod Vallat | |
2006-07-31 | Move the relevant mouse defines from lk201.h to dzms.c and get rid of it; | Miod Vallat | |
it is too confusing to keep lk201.h and lk201reg.h. | |||
2006-07-31 | Get my raster ops correct for the cursor, fixes the display glitches. | Miod Vallat | |
2006-07-31 | On startup, and when a keyboard change is detected, probe for the keyboard | Miod Vallat | |
type. For now it is only reported to the user, but we'll use this very soon to alter keycodes depending upon the keyboard type (so that we do not have the compose and alt keys swapped on LK401 keyboards). | |||
2006-07-31 | Do not mess with pv_tables for I/O mappings, as these are never managed pages; | Miod Vallat | |
from NetBSD | |||
2006-07-30 | Always invoke the lk201 input handler, even when the wskbd device is disabled, | Miod Vallat | |
so that the keyboard gets configured correctly if plugged late. | |||
2006-07-30 | Disable keyclick by default. | Miod Vallat | |
2006-07-30 | Report MicroVAX 3100 m{10,20}{,e} as such instead of bare ``MicroVAX 3100''; | Miod Vallat | |
ok hugh@ | |||
2006-07-30 | Use DZ_CSR_xxx symbolic constants. | Miod Vallat | |
2006-07-30 | Add wscons stuff to the installation media as well. | Miod Vallat | |