| Age | Commit message (Collapse) | Author | |
|---|---|---|---|
| 2007-11-06 | Close a race where we might save/drop the fpu state of the wrong process in | Mark Kettenis | |
| the ipi handlers. | |||
| 2007-11-06 | ramdisks should have the same behaviour as GENERIC | Theo de Raadt | |
| 2007-11-06 | Keep a pending software interrupts mask per processor, instead of having it | Miod Vallat | |
| global; and only schedule software interrupts on the currently running cpu. | |||
| 2007-11-06 | Comment out the pmap fine grained locking stuff, it is not necessary for now | Miod Vallat | |
| because of the global lock. It will get enabled again when locking work progresses. | |||
| 2007-11-06 | Be sure to pmap_deactivate() a process during context switches, so that | Miod Vallat | |
| the cpu which runs it is accounted correctly in MP kernels. | |||
| 2007-11-06 | Remove the now unused idle_u, and call the secondary processors startup | Miod Vallat | |
| stack a startup stack. | |||
| 2007-11-06 | Get rid of TC_IPL_xxx values and tc_intrlevel_t, and use IPL_xxx and int. | Miod Vallat | |
| No functional change. | |||
| 2007-11-05 | I screwed the pooch. Fix the stack offsets. | Tobias Weingartner | |
| "That's much better." kettenis@ | |||
| 2007-11-05 | Sprinkle a few holy membars around memory writes. | Miod Vallat | |
| ok kettenis@ | |||
| 2007-11-05 | fix and use DEVICE_IS_VGA_PCI, makes vgafb_pci_match() much simpler | Martin Reindl | |
| ok oga@ brad@ | |||
| 2007-11-05 | temporary ddb buffers need to be a bit larger on 64 bit systems to cope with | Miod Vallat | |
| some radix configurations. | |||
| 2007-11-05 | audio encoding conversion cleanup | Jacob Meuser | |
| - the endianness of the conversions don't depend on the endianness of machine the conversions are built on, but the endianness of the audio data itself. choose encoding conversions explicitly, instead of relying on #defines based on the endianness of the machine. - replace home-grown conversions with comparable conversions in auconv.c and mulaw.c - use the proper conversion for ulinear_be:16 -> slinear_le:16 in auixp(4) thanks ajacoutot@ and sthen@ for !x86 testing | |||
| 2007-11-04 | replace even more ctob/btoc with ptoa/atop | Martin Reindl | |
| 2007-11-03 | Fix LKM support for amd64. | Mike Belopuhov | |
| ok deraadt weingart | |||
| 2007-11-03 | Add acpi_acquire_global_lock(), and acpi_release_global_lock to | Gordon Willem Klok | |
| amd64 the not ghetto architecture. ok toby@ | |||
| 2007-11-03 | use DEVICE_IS_VGA_PCI in vga_pci_match() | Martin Reindl | |
| ok brad@ fgsch@ also tested by oga@ | |||
| 2007-11-03 | don't spam dmesg with huge, whitespace padded bios strings. | Chris Kuethe | |
| ok beck, gwk | |||
| 2007-11-03 | change what happens when acpi is enabled to include "relatively safe" | Bob Beck | |
| stuff and have acpi actually do something. note acpi is still disabled by default, but now will do more stuff when enabled with config -e/boot -c ok gwk@, weingart@, canacar@, deraadt@ | |||
| 2007-11-03 | APM switchto cleanup diff. This uses the switchto functionality to | Tobias Weingartner | |
| significantly clean up the APM idle loop things. I actually can make sense of what the APM idle loop is supposed to do. Ok gwk@, beck@, and theo says ramdisks compile. | |||
| 2007-11-03 | Add acpi_acquire_global_lock() and acpi_release_global_lock() in MD locore.s | Tobias Weingartner | |
| because it is easier to write these here in asm than it is to put them inline with gcc's __asm() construct. Also, chances are amd64's versions will need to be written in a different manner. Basically yanked from ACPI spec V3, section 5.2.10.1. Ok canacar@, beck@, gwk@ | |||
| 2007-11-02 | replace ctob/btoc with ptoa/atop (plus the usual round_page() where needed) | Martin Reindl | |
| 2007-11-02 | Use the same cache cleaning address computation as done in cpufunc_asm_xscale, | Miod Vallat | |
| for there be dragons in xscale cache and it would not be cleaned correctly, leading to wrong pcb data being restored on resume and eventually causing panics. | |||
| 2007-11-01 | Remove curproc definition. It is supposed to be in proc.h only. | Mike Belopuhov | |
| ok kettenis | |||
| 2007-10-31 | Overhaul fpu context save/restore code, making it more similar to what we do | Mark Kettenis | |
| on i386 and amd64. Don't let IPIs in when saving fpu context by disabling interrupts. | |||
| 2007-10-31 | Remove some XXX code that has been #ifdef'ed out for ages. | Mark Kettenis | |
| 2007-10-31 | Remove idle_u; it's been unused for a while. | Mark Kettenis | |
| 2007-10-31 | Remove some comments about 32-bit mode. | Mark Kettenis | |
| 2007-10-31 | Garbage collect ienab_bis() and ienab_bic(). | Mark Kettenis | |
| Suggested by miod@ | |||
| 2007-10-31 | for now, workaround MP timeout/splhigh/scsi race at reboot time using a | Theo de Raadt | |
| delay. will be revisited. ok art | |||
| 2007-10-31 | comment fixes. | Joel Sing | |
| ok miod@ | |||
| 2007-10-30 | avoid intermediate files with same name (ie. a.out) for parallel make | Theo de Raadt | |
| ok miod | |||
| 2007-10-29 | run depend so that SFILES depend on assym.h, which allows them to build | Marc Espie | |
| with make -j on fast SMP boxes. noticed by robert@, input by miod, kettenis, okay toby@, robert@, and tested to work if people build kernels correctly... | |||
| 2007-10-29 | UltraSPARC CPUs (and other SPARC V9 implementations) don't have a | Mark Kettenis | |
| floating-point deferred-trap queue. Remove redundant code inherited from sparc that deals with this. Also remove the code dealing with saving and restoring the FPU state from unaligned memory; we always allocate properly aligned memory for storing the FPU state. | |||
| 2007-10-29 | Make sure the dma_cachectl*() functions actually do their work on all | Miod Vallat | |
| affected processors if option MULTIPROCESSOR. It's amazing bsd.mp could boot multiuser without this. | |||
| 2007-10-29 | When a secondary cpu gets its interrupt pin stuck, be sure to savectx | Miod Vallat | |
| and put the process it was running back on the run queue (unless this was the idle proc). | |||
| 2007-10-28 | This is a horrible kluge: invoke sched_init_cpu for seconday processors | Miod Vallat | |
| before they are started (and not skipping gaps for machine setups with holes in cpu slots). Since we start secondary cpus very late in the boot process, and sched_init_cpu() has to be invoked before proc0 execve's init, I don't think there is a better way to do this. This lets MVME188 systems with more than one processor boot multiuser. | |||
| 2007-10-28 | Start secondary processors synchronously. This only wins us a nicer dmesg | Miod Vallat | |
| output, and nothing else. | |||
| 2007-10-28 | Do not flag a processor as ``alive'' until it really is ready to accept IPIs. | Miod Vallat | |
| 2007-10-28 | add new envy(4) driver for ICE1712 (aka VIA Envy24) audio devices. The | Alexandre Ratchov | |
| device uses 32bit samples, up to 96kHz sample rate, 12 input channels and 10 output channels. Currently AK4524 codecs are supported, so M-Audio Delta cards should work. Playback and capture work, but the mixer is still incomplete. ok jakemsr@ | |||
| 2007-10-28 | UltraSPARC CPUs don't have a floating-point queue, so cpu_reset_fpustate() is | Mark Kettenis | |
| redundant since there is no queue to flush. | |||
| 2007-10-28 | Match on compatible: 'pcf8584', such that we attach on the V210 again. | Mark Kettenis | |
| 2007-10-28 | get rid of btoc/ctob in favor of atop/ptoa | Martin Reindl | |
| 2007-10-28 | When handling a userland data fault occuring in kernel mode, take the kernel | Miod Vallat | |
| lock with KERNEL_LOCK, not KERNEL_PROC_LOCK. This lets bsd.mp run multiuser on a single-processor board. | |||
| 2007-10-28 | Disable interrupts around changing curproc and curpcb so these always match. | Miod Vallat | |
| 2007-10-27 | Grab kernel lock before calling interrupt handlers. | Mark Kettenis | |
| ok miod@, deraadt@ | |||
| 2007-10-27 | Grab kernel lock for EXC_DSI. | Mark Kettenis | |
| Requested by deraadt@ | |||
| 2007-10-27 | get rid of btoc/ctob in favor of atop/ptoa | Martin Reindl | |
| 2007-10-27 | Use the same assembly constraints for all inline assembler xmem constructs. | Miod Vallat | |
| 2007-10-27 | In __cpu_simple_lock() and __cpu_simple_lock_try(), use a local u_int instead | Miod Vallat | |
| of a local __cpu_simple_lock_t (which is volatile), so that the compiler can optimize it to a register, instead of using a memory location (and doing stores into it when __cpu_simple_lock() is spinning). This makes the MP code a bit smaller and a bit faster. | |||
| 2007-10-27 | No need for an explicit pipeline synchronization in invalidate_pte(), the | Miod Vallat | |
| xmem instruction does it for us. | |||
 |
index : src | |
| OpenBSD base system |
| summaryrefslogtreecommitdiff |