| Age | Commit message (Collapse) | Author | |
|---|---|---|---|
| 2006-06-30 | For 32bit sun frame buffers, tweak ri_devcmap to get the BoW palette instead | Miod Vallat | |
| of needing different WSCOL_{BLACK,WHITE} values than 8bit frame buffers. This allows us to not special case the alloc_attr() invocations depending on the color depth, and to make WSCOL_{BLACK,WHITE} constants again in the wsemul_sun land. | |||
| 2006-06-30 | Make routines which may set PSL_IOPL consistent wrt the securelevel for | Miod Vallat | |
| which this is allowed; ok deraadt@ | |||
| 2006-06-30 | Old typos bite the dust. | Miod Vallat | |
| 2006-06-30 | When allocating the default rasops attribute, it is not necessary to use | Miod Vallat | |
| non-zero parameters on a color display as long as we want white on black; this was inherited from sparc but this is not necessary on these platforms. | |||
| 2006-06-30 | Shrink internal iomap to 256KB, it's enough for all the mappings we'll need | Miod Vallat | |
| in it. | |||
| 2006-06-30 | On OSS systems (IIfx), compute addresses relative to VIA2 rather than VIA1, | Miod Vallat | |
| it's simpler. | |||
| 2006-06-30 | Some machines lack the virtual-dma property. In that case, go back | Theo de Raadt | |
| to the 128KB assumption, until we find out another way to determine the space. Damn. | |||
| 2006-06-29 | Fix PMAP_PREFER and make it use VA_ALIAS_MASK to prevent me from breaking | Mark Kettenis | |
| it again. ok miod@, jason@ | |||
| 2006-06-29 | No need to keep specific variables for the kernel messages colors on sun | Miod Vallat | |
| emulation, we can reuse the existing WSCOL_xxx codes, and WSCOL_BLACK and WSCOL_WHITE will point to variables anyway. | |||
| 2006-06-29 | If invoking mapdev() with a non-aligned pa, make sure we allocate enough pages | Miod Vallat | |
| if we cross a page boundary; really only necessary for some sun4 tricky attachments, no functional change. | |||
| 2006-06-29 | Map the correct size for the colormap registers (not that it really matters | Miod Vallat | |
| since this is rounded up one page). | |||
| 2006-06-29 | make compile w/ debug; from danielcavanagh@aanet.com.au | Michael Shalayeff | |
| 2006-06-29 | try prom mapping first, then memory mapping, then io mapping.. this works | Theo de Raadt | |
| on v210 and on netra t1 105, good sign it will work on others. mechanism pointed out by miod, tested by dlg and djm and I | |||
| 2006-06-28 | cleanup dmesg logging for iommu goo, make it follow the standard form | Theo de Raadt | |
| 2006-06-28 | look at "virtual-dma" property to decide where the iommu address space | Theo de Raadt | |
| is, instead of having iommu_init() decide | |||
| 2006-06-28 | keep track of what is a schizo and what is tomatillo | Theo de Raadt | |
| 2006-06-27 | Fix interrupt mapping. This now seems to work on all machines, even with | Mark Kettenis | |
| pci cards that have a pci-pci bridge and lack fcode. tested by many, ok jason@, dlg@ | |||
| 2006-06-27 | pull in files.gpio in case we want it | Theo de Raadt | |
| 2006-06-27 | don't bother printing "{se}bus: %s incomplete" for devices missing a reg | Jason Wright | |
| property. They're not interesting. | |||
| 2006-06-27 | Rewritten clock driver for 80219/80321, actually manages seperate stat | Dale Rahn | |
| clock and tick timer and detects missing ticks. | |||
| 2006-06-26 | Clean up some inconsistancies in definitions around DOS MBR code. No | Kenneth R Westerback | |
| functional change. ok deraadt@ | |||
| 2006-06-26 | enable mpi | David Gwynne | |
| ok deraadt@ | |||
| 2006-06-26 | atop(PAGE_SIZE) -> 1, 'tis true but it gets computed faster when PAGE_SIZE | Miod Vallat | |
| is not a constant. | |||
| 2006-06-26 | Always skip the offending instruction for fpu faults, even if it is | Miod Vallat | |
| theoretically recoverable, as there is no way we can rely on the signal handler to do this properly; consistent with other platforms, and lets a few autoconf tests in ports pass again. spotted by sturm@; reluctantly ok kettenis@ | |||
| 2006-06-26 | Enable wide targets on FAS366 esp; note that this changes the way esp | Miod Vallat | |
| user-controllable device flags are parsed, be sure to rerun config(8) on your next kernel compile. | |||
| 2006-06-26 | Use MI clock conversion code for "old" clock. | Mark Kettenis | |
| ok miod@ | |||
| 2006-06-26 | add commented out entries for bnx. | Brad Smith | |
| 2006-06-26 | Kludge around the fact that redboot uses ELF's vaddr, however maps | Dale Rahn | |
| memory at 0, not where memory is or where we want the kernel loaded. | |||
| 2006-06-26 | enable ne*; ok mpf who tested it, we all feel bad for you | Theo de Raadt | |
| 2006-06-25 | fix a typo in a comment and add the comment to the sparc hme driver. | Brad Smith | |
| 2006-06-25 | reorganize the code to be more sane. | Brad Smith | |
| ok toby@ | |||
| 2006-06-24 | Configure network devices at ipl 3, not ipl 1. | Miod Vallat | |
| 2006-06-24 | Replace physaccess() usage with pmap_kenter_cache(). | Miod Vallat | |
| 2006-06-24 | Use pmap_enter_cache() instead of physacc() in bus_mem_add_mapping(), and let | Miod Vallat | |
| physacc() die. As a bonus, kvtop() dies too. | |||
| 2006-06-24 | Allocate ethernet buffers as uvm pages, and map them with pmap_enter_cache(), | Miod Vallat | |
| instead of crossing fingers and expecting malloc() to return aligned and contiguous memory. | |||
| 2006-06-24 | Add pmap_enter_cache(), similar to pmap_kenter_cache() but for managed | Miod Vallat | |
| pages, and implement pmap_enter() as a particular case of it. | |||
| 2006-06-24 | Use pmap_kenter_cache() for device memory mapping, instead of physacc(0 | Miod Vallat | |
| which bites the dust. | |||
| 2006-06-23 | Use MI clock conversion code for "new" clock. | Mark Kettenis | |
| ok deraadt@, miod@ | |||
| 2006-06-23 | we choose more drivers/firmwares over ext2fs | Theo de Raadt | |
| 2006-06-23 | match tomatillo varient (not yet fully working) | Theo de Raadt | |
| 2006-06-23 | consistantly count context switches on exit; miod@ ok | Michael Shalayeff | |
| 2006-06-23 | In a drivers activate() entry point, if on DVACT_DEACTIVATE it does | Miod Vallat | |
| something, then it MUST return zero on DVACT_ACTIVATE, not EOPNOTSUPP; this very popular bug has been cut and pasted a lot of times... ok deraadt@ mickey@ | |||
| 2006-06-22 | Enable sk(4). | Brad Smith | |
| ok reyk@ | |||
| 2006-06-22 | BBC-based machines swap the two registers (good going Sun -- that is why we | Theo de Raadt | |
| don't get documentation, right?), and the clock fed into the chip can vary from model to model; help & testing from dlg | |||
| 2006-06-22 | mpi works here too. | David Gwynne | |
| ok deraadt@ | |||
| 2006-06-22 | do not enable the gpio parent either | Theo de Raadt | |
| 2006-06-21 | if machdep.kbdreset==1 and the front panel button interrupt fires, tell init | Jason Wright | |
| that it's time to go sleepy-bye. | |||
| 2006-06-21 | Only attach timer to the first counter-timer found. | Miod Vallat | |
| 2006-06-21 | this gpio should not be enabled | Theo de Raadt | |
| 2006-06-21 | add the machdep.kbdreset logic to sparc64 (stuff to use it is coming) | Jason Wright | |
 |
index : src | |
| OpenBSD base system |
| summaryrefslogtreecommitdiff |