summaryrefslogtreecommitdiff
path: root/sys/arch
AgeCommit message (Collapse)Author
2007-04-16Crank MAXPHYS to the regular 64KB.Miod Vallat
2007-04-16macppcs can boot off usb disks, so treat "/usb" controllers in the ofw treeDavid Gwynne
as storage controllers. this lets ottos macppc root off the usb disk he booted off.
2007-04-14Crank KVM from a ridiculous pedro-sized 256MB to 1GB; needed for upcoming MIMiod Vallat
changes.
2007-04-14Correctly handle pv_flags for pages used to store ptes, so that theyMiod Vallat
are in a correct state when we uvm_pagefree() them.
2007-04-13While splitting flags and pqflags might have been a good idea in theoryArtur Grabowski
to separate locking, on most modern machines this is not enough since operations on short types touch other short types that share the same word in memory. Merge pg_flags and pqflags again and now use atomic operations to change the flags. Also bump wire_count to an int and pg_version might go int as well, just for alignment. tested by many, many. ok miod@
2007-04-13Timecounters for macppc. As clean and simple as sparc64. Uses the %tbMark Kettenis
special register. ok drahn@
2007-04-13Relax the cache flags logic in pmap_kenter_pa, to make sure that mappingsMiod Vallat
entered before vm_physmem[] are initialized will be cached. This is a temporary measure until this pmap implements pmap_steal_memory(). Help and ok drahn@
2007-04-13Update install media generation to the new semantics of the -c flag.Todd C. Miller
Also replace "newfs -O" with "newfs -O 0" now that -O takes an option.
2007-04-13When freeing PTP pages, we need to wait until TLB shootdown has beenArtur Grabowski
done until we put them on the freelist and risk them being reused. If the page is reused before being shot, there's a risk that it's still in the PDE TLB and speculative execution might use data from it to load TLBs. If the page isn't zeroed anymore we risk loading bogus tlbs from it. Inspired by a similar change in NetBSD. toby@ ok, tested by many at the hackathon.
2007-04-13Move the proc0 stack below the proc0 page tables so that in case weArtur Grabowski
overflow the stack, we run into the symbol table (not critical) instead of overwriting the page tables. From mickey, ok art@, ok toby@
2007-04-13map peeing -> mappingMiod Vallat
2007-04-13get alpha SMP into a state where it at least compiles:Martin Reindl
- add machine-dependent spinlock operations - add basic interprocessor interrupt sending and receiving code from NetBSD; ok miod@
2007-04-12Remove the lk_interlock from struct lock; Also remove the LK_INTERLOCKThordur I. Bjornsson
flag. This effectively makes the simplelock argument to lockmgr() fluff. ok miod@
2007-04-12Faster signal delivery on i386/MP.Artur Grabowski
We need to poke the other CPU so that it processes the AST immediately and doesn't wait for the next interrupt or syscall. Since IPIs really shouldn't process ASTs, we need to trigger a soft interrupt on the destination CPU to process the AST. But since we can't send soft interrupts to other CPUs, we send an IPI, that triggers a soft interrupt that in turn processes the AST. Also, this marks the beginning of moving to slightly better IPI mechanism of short and optimized IPIs instead of the large and complicated IPI infrastructure we're using now. tested by many, ok tholo@
2007-04-12untypoArtur Grabowski
2007-04-12Discard .comment sections in the final a.out file.Miod Vallat
2007-04-12__HAVE_CPUINFO for alpha, with bits from netbsdMartin Reindl
ok miod@
2007-04-12Remove __BUS_SPACE_ADDRESS_SANITY correctly, no cookie for me. Spotted byMiod Vallat
djm@
2007-04-11Instead of managing pages for intrsafe maps in special objects (aka.Artur Grabowski
kmem_object) just so that we can remove them, just use pmap_extract to get the pages to free and simplify a lot of code to not deal with the list of intrsafe maps, intrsafe objects, etc. miod@ ok
2007-04-11sili(4) works, why not enable it?David Gwynne
2007-04-10Add support for a fourth axis on wsmouse devices, e.g. on the Apple MightMiod Vallat
Mouse. Currently limited to USB mice. Adapted from a diff from Gareth <garf@loveandnature.co.za> on tech@
2007-04-10Remove long dead definitions. No functional change.Miod Vallat
2007-04-10Long forgotten stub.Miod Vallat
2007-04-10Add bbc(4) and pmc(4).Mark Kettenis
2007-04-10Driver for the BootBus Controller; sets some magic bits that makes pmc(4)Mark Kettenis
actually work.
2007-04-10New driver for watchdog time on NatSemi PC87317 SuperIO chip.Mark Kettenis
2007-04-10More typos from my list.Miod Vallat
2007-04-10Remove ARM32_BUFTYPE_RAW, it's never used.Miod Vallat
2007-04-10``prob ably'' can be written in one word.Miod Vallat
2007-04-10Remove unused BUS_SPACE_ALIGNED_POINTER, __BUS_SPACE_ADDRESS_SANITY andMiod Vallat
__BUS_SPACE_ALIGNED_ADDRESS.
2007-04-10``it's'' -> ``its'' when the grammar gods require this change.Miod Vallat
2007-04-10privledges -> privilegesMiod Vallat
2007-04-10Do not compile unused dump_tlb() unless option DEBUG.Miod Vallat
2007-04-10slighly -> slightlyMiod Vallat
2007-04-10Prevent infinite loop mapping interrupts when "reg" property is missing.Mark Kettenis
ok mbalmer@, dlg@
2007-04-10pascoe@ finished sili(4). enable it so people can try it.David Gwynne
2007-04-10pascoe finished sili(4). let's enable it so people can try it.David Gwynne
2007-04-09Re-enable art(4); mglocker@'s e250 is perfectly happy with larger kernels now.Mark Kettenis
2007-04-09Unbreak the tree.Mark Kettenis
2007-04-09Seperate rtc(4) code out into its own file. Add support for catching theMark Kettenis
power button interrupts on ds1287 models. The hardware will stil power off automatically about 20 seconds after the power button is pressed, but we get a decent chance at doing a clean shutdown before that. "a good start" deraadt@
2007-04-09Fix (unused) extintr_disable_by_num behaviour, and make vmstat -i outputMiod Vallat
provide the correct irq numbers.
2007-04-08ral(4) reported to work on sparc64 by Maxim Belooussov <belooussov@gmail.com>Jonathan Gray
so enable it.
2007-04-07Remove unused variable.Mark Kettenis
2007-04-07USECPERSEC is no longer used.Mark Kettenis
2007-04-07Replace code that tries to find the boot device by pretending to walk theMark Kettenis
device tree and matching locators in the bootpath string components with code that tries to match the PROM node of the devices to bootpath components. This way we don't need a list of possible boot devices that needs to be tweaked whenever we try to support new Sun hardware (or when dlg plugs a new fancy storage controller in a sparc64 machine). Tested by many.
2007-04-07Replace code that tries to find the boot device by pretending to walk theMark Kettenis
device tree and matching locators in the bootpath string components with code that tries to match the PROM node of the devices to bootpath components. This way we don't need a list of possible boot devices that needs to be tweaked whenever we try to support new Sun hardware (or when dlg plugs a new fancy storage controller in a sparc64 machine). Tested by many.
2007-04-06Add support for PCIe. Don't map interrupts that are already mapped correctly.Mark Kettenis
Tested by deraadt@, nick@, sturm@, naddy@ and others.
2007-04-05Fix a comment. 64KB is enough for the cpu_info structure. 64MB would be aClaudio Jeker
bit lavish. OK jason@
2007-04-05Fix interrupts.Mark Kettenis
2007-04-05Disable interrupts around bit operations for propert atomicity against them.Miod Vallat
generated by cgit v1.2.3 (git 2.25.1) at 2025-01-19 14:08:31 +0000