summaryrefslogtreecommitdiff
path: root/sys/arch/sparc64
AgeCommit message (Collapse)Author
2008-04-08Split the cdev makro for bpftun into two seperate definitions. tun(4) andClaudio Jeker
bpf(4) are different enough so that the split makes sense -- this is necessary to make bpf(4) cloneable. requested deraadt@, OK thib@
2008-04-07Don't grab the kernel biglock for syscalls marked SY_NOLOCK;Thordur I. Bjornsson
make the amd64 code a bit nicer and make the sparc64 match that. tested by a few; OK toby@, dlg@, kettenis@ (for the sparc64 part).
2008-04-03Use atomic operations to invalidate TSB entries.Mark Kettenis
2008-04-03Make ebus(4) work on sun4v.Mark Kettenis
2008-04-03Some cosmetic changes.Mark Kettenis
2008-04-02Add nsgphy(4).Mark Kettenis
2008-04-02For some reason using the ASI_SCRATCHPAD register at offset 0x08 makes theMark Kettenis
t1k freak out upon reboot/halt/powerdown. Use the register at offset 0x10 instead.
2008-03-31Make MULTIPROCESSOR kernels work on sun4v. Won't gracefully halt, powerdownMark Kettenis
or reboot yet, but that will (hopefully) be fixed in the near future.
2008-03-30Remove for #if 0'ed out code I left behind with the previous commit.Mark Kettenis
2008-03-30More sun4v support. GENERIC and RAMDISK kernels will now boot on bothMark Kettenis
sun4u and sun4v. GENERIC.MP won't work yet though.
2008-03-29Don't double count clock and prof interrupts.Mark Kettenis
2008-03-26Disable some code that reads some sun4u-specific registers for diagnosticMark Kettenis
reasons, but isn't otherwise used.
2008-03-24Remove KASSERT that's causing grieve on sun4v, and I've never seen triggered.Mark Kettenis
2008-03-24Use correct cpuid on sun4v.Mark Kettenis
2008-03-23Add code to initialize CPUs on sun4v.Mark Kettenis
2008-03-23Sigh! The sun4v TTEs have a different layout than sun4u TTEs. Rename theMark Kettenis
existing sun4u defines and add sun4v. For now, decide which set to use at compile time. Change the sun4u-specific code in locore.s to use the sun4u defines.
2008-03-23typo; deraadtMiod Vallat
2008-03-23Not-functional-enough driver for the Sun Expert3D cards, put in tree as anMiod Vallat
incentive for further tinkering only; not for the faint of heart.
2008-03-23Use new defines from pcidevs to match Expert3D.Miod Vallat
2008-03-23Fix ovbcopy() operation when copying shorts backwards, similar to sparcMiod Vallat
locore.s r1.76
2008-03-22Add support for DIOCRLDINFO, DIOCGPDINFO, DIOCGPART to the fdKenneth R Westerback
driver(s). Adapted from sd driver. ok marco@
2008-03-22Reintroduce the cputyp variable, and use it to distinguish between sun4u andMark Kettenis
sun4v.
2008-03-22Simplify cpu_switchto() such that it doesn't need to know the number ofMark Kettenis
register windows provided by the hardware; this number is not readily available on sun4v. This removes the optimization that skips flushing register windows if a process exits. We can add that back later if it turns out to make a significant impact.
2008-03-22Remove TRAPS_USE_IG code.Mark Kettenis
2008-03-22Split out the code that sets the TSB registers and call it fromMark Kettenis
pmap_bootstrap_cpu().
2008-03-22Use ASI_BLK_P instead of ASI_BLK_COMMIT_P when saving floating pointMark Kettenis
registers. UltraSPARC T1 doesn't support ASI_BLK_COMMIT_P, and I can't see why this code needs commit force.
2008-03-22Switch to our initial stack after switching over to our own trap table. ThisMark Kettenis
removes the need to lock the stack (and cpuinfo) into the TLB on sun4v.
2008-03-21Do not attach vgafb if not the console device, since the PROM will not haveMiod Vallat
initialized the device in this case.
2008-03-20Introduce GET_CPUINFO_VA(), GET_CPCB() and GET_CURPROC() macros to getMark Kettenis
some important members of 'struct cpuinfo'. Preparation for sun4v.
2008-03-20Fix whitespace.Mark Kettenis
2008-03-20Bring isa/fd.c into line with sparc/sparc64 fd code and prepare allKenneth R Westerback
for the addition of missing disklabel related ioctl's. ok deraadt@
2008-03-19Establish per-cpu locked mappings for `struct cpuinfo' through the PROM asMark Kettenis
well.
2008-03-19Use PROM calls to enter locked kernel text and data mappings into the TLB.Mark Kettenis
Gets rid of a big chunk of nasty asm code and makes us boot on the e10k with multi-systemboard domains.
2008-03-19Add prom_itlb_load() and prom_dtlb_load().Mark Kettenis
2008-03-18Get rid of some dead wood.Mark Kettenis
2008-03-17Remove KGDB code. It was never converted to 64-bit, and just makes locore.sMark Kettenis
even more unreadable.
2008-03-16Include <sparc64/dev/starfire.h> instead of having our own definitions.Mark Kettenis
2008-03-16Add some e10k support code.Mark Kettenis
2008-03-16Set up interrupt translation for e10k.Mark Kettenis
2008-03-16Add some e10k support code.Mark Kettenis
2008-03-15Make GENERIC.MP work on the e10k. The e10k is a bit funky since UPA onlyMark Kettenis
supports 32 ports, and a machine with up to 64 CPUs obviously needs more. So the machine has a special ASIC that does port translation, and because of that we need to distinguish between port ID's and interrupt target ID's.
2008-03-14Don't even try to attach pcons(4) if we have a real console.Mark Kettenis
2008-03-13Make this catch up with some changed structs. Fix indentation while I'm there.Mark Kettenis
2008-03-13Add prtc(4).Mark Kettenis
2008-03-13Attach prtc(4) if no (hardware) real-time clock was found.Mark Kettenis
2008-03-13Add prtc(4).Mark Kettenis
2008-03-13First shot at a driver to get the time-of-day on the e10k.Mark Kettenis
2008-03-13Remove code to set mmu context to 0. It should already be set to 0 at thatMark Kettenis
point; this code is probably a leftover from some code that tried to find out the number of available context dynamically. tested by ckuethe@, jsg@, sthen@
2008-03-12Introduce a per-handler interrupt acknowledgement function.Mark Kettenis
2008-03-09Add files for vpci(4).Mark Kettenis
generated by cgit v1.2.3 (git 2.25.1) at 2025-01-24 10:56:57 +0000