summaryrefslogtreecommitdiff
path: root/sys/arch
AgeCommit message (Collapse)Author
2013-01-15Loongson: Add a stat clock for the lemote machines.Paul Irofti
The stat clock is implemented so that it also covers, at least according to the tests I've done, Torek's paper on randomized sampling. Most of the bits for randomizing the stat ticks were taken from sparc's implementation and adapted to the companion chip's MFGPT frequencies. Before this my lemote showed a 8% cpu usage for the cpuhog example from Torek's paper even though openssl speed showed differences up to 27% when ran with and without cpuhog. With the new stat clock it shows cpuhog around 18-22%, which I think is the proper value. Okay miod@.
2013-01-15implement bus_space_foo_8 for amd64. it will panic on attempts to do 8David Gwynne
byte operations on I/O space with "invalid I/O operation", but works great for mem mappings. compiled AND tested on myx(4) on amd64. ok deraadt@ kettenis@ miod@ pirofti@
2013-01-15Fix segmentation of buffers that straddle a boundary.Mark Kettenis
ok dlg@
2013-01-15Loongson: Remove some of the unused defines.Paul Irofti
Increases readability.
2013-01-15Loongson: Unbreak the tree after my last commit.Paul Irofti
Patch appended my final work on glxclk to an earlier version that I forgot was there.
2013-01-14Loongson: Replace the system clock provided by the CPU with a GLX MFGPT.Paul Irofti
CPU throttling was not possible due to the fact that the system clock was the CPU clock. So slowing down the CPU would also slow down the passing of time. This commit adds a driver for the MFGPT1 clock from the AMD companion chip found on lemote and hooks it up as the system clock. It also changes the frequency value of hz from the default, which was 100, to 128. That's because the scaling on MFGPT clocks is represented by powers of two. Okay miod@.
2013-01-12add viomb driver for virtio balloon deviceStefan Fritsch
This allows the host to request some memory back from the openbsd guest in case of a shortage. Patch by Dinar Talypov OK jasper@ OK jmc@ (man page)
2013-01-11Change boot commandline parsing to only force lowercase of the device nameMiod Vallat
part, and keep the filename unchanged. This allows a kernel which name contains uppercase letters to be bootable, instead of getting a `file not found' error with the filename being all lower case. A workaround was to prepend the filename with `/', to give an absolute path. Yet there is no reason to lowercase the filename unless there is a boot device prefix. Crank version.
2013-01-11Do not use `#' to introduce comments anymore, as(1) will no longer supportMiod Vallat
this RSN.
2013-01-11Make ENTRY() generate a .type @function declaration as well if __ELF__.Miod Vallat
Add an END() macro, to be used to match ENTRY(), which produces a .size directive.
2013-01-07Correctly signal SIGSEGV, instead of SIGBUS, for faults caused by access toMiod Vallat
pages mapped without read (or write) permissions. The existing logic would incorrectly match uvm_fault() returning EACCES with the CMMU fault logic reporting a bus error (which will still cause SIGBUS to be sent).
2013-01-07Move exec() libsa prototype to hp300 stand, now that it can no longer beMiod Vallat
found in libsa.
2013-01-07Fix typos.Kenji Aoyama
"Please commit." miod@
2013-01-06acknowlege -> acknowledgeMartynas Venckus
compatability -> compatibility OK jmc@.
2013-01-05Switch m88k ports to ELF.Miod Vallat
2013-01-02Zero out sbv_maskkey to make it boot when there's a softraid cryptoMartynas Venckus
partition. OK jsing@.
2013-01-02miod let one exec.c slideTheo de Raadt
2013-01-01Remove libsa exec.c, not used by anything anymore; all boot blocks now useMiod Vallat
loadfile or a derivative of.
2013-01-01Switch hp300 boot blocks to MI loadfile code.Miod Vallat
2013-01-01Define _STANDALONE, not STANDALONE, in the hp300 bootblocks, to follow theMiod Vallat
common (and expected) practice. This exposes a flaw, in that mkboot, a host tool, gets built with boot block settings (and won't compile if _STANDALONE is defined). Fix this by using different set of CPPFLAGS for host tools and boot blocks. While there, avoid sourcing ../Makefile.inc twice.
2013-01-01Crank NMBCLUSTERS to 4096 to match other 32-bit platforms.Miod Vallat
2013-01-01Remove unused mem_size global.Miod Vallat
2013-01-01Fix gap value when formatting 1.44MB 3"1/2 floppy disks; from NetBSDMiod Vallat
2013-01-01<machine/aout_machdep.h> is not used anywhere in the tree, move to the Attic.Miod Vallat
2013-01-01Add a "memory" clobber to the flush_pipeline() assembly construct, to make sureMiod Vallat
the compiler won't try to outsmart me.
2012-12-31Switch mvme68k boot blocks to use the MI libsa loadfile routines to loadMiod Vallat
kernel images, instead of their own; gives us the ability to load ELF kernels for free. Fix many warnings, and generally make the code closer to the current mvme88k boot blocks to help diffability. Rewrite bugcrt and bugexec() as a short assembly shim around a compiler-independent C block; this will allow compilers other than gcc2 to be used, and removes the need for bugcrt to be compiled without optimization. Crank all bootblocks versions.
2012-12-31Allow native ELF binaries (none yet) to run.Miod Vallat
2012-12-31Remove emulation errno mapping code from platforms that no longer havePhilip Guenther
non-native emulations. ok miod@
2012-12-31Eliminate orig_errno, which could be uninitialized in one case, by doingPhilip Guenther
the emulation errno mapping directly into the register in the trapframe. Range check the value in that case to guarantee there isn't an out-of-bounds array access. Uninitialized variable issue pointed out by David Hill. Range check suggested by matthew@ ok miod@
2012-12-30Prevent a branch-with-delay-slot instruction to be the delay slot of a previousMiod Vallat
branch-with-delay-slot instruction, in the 88100 exception code path, in kernels configured without DDB. This is undefined behaviour, but we have apparently be lucky enough since RAMDISK is known to work on 88100.
2012-12-26Nuke commons.Miod Vallat
2012-12-25Add GENERIC.MPMiod Vallat
2012-12-22Attach all known U3 AGP bridgesMartin Pieuchot
2012-12-22We already zero out the trapframe, so there is no reason to set individualMark Kettenis
members to zero. Gets rid of a bogus comment as a bonus.
2012-12-17Add a list of alpha systems and their SRM footprint in the linker scriptMiod Vallat
comments, in the discussion about the kernel load address, for reference purpose. While there, admit the current value will not let the kernel boot on a GS160. If you are the happy owner of a GS160, can afford the electricity bill, and want to run OpenBSD on it, please contact alpha@
2012-12-17Unbreak bootblocks on Alpha by fixing the primary load address. AMartynas Venckus
typo crept in in rev. 1.13. "yeah! the as800 netboots" naddy@
2012-12-13iMac G3 grahic cards have a ``misc'' pci subclass that's why they weren'tMartin Pieuchot
attached to vgafb(4) since the use of the generic DEVICE_IS_VGA_PCI macro in r1.26. Correct this and document why we should attach devices with such subclass. Fix a regression introduced in august when adding the glue to attach drm(4) on macppc, reported by brynet@ and nick@ on bugs@, thanks and sorry for the delay!
2012-12-10Correct macobio functions to only read and write one byte of data insteadMartin Pieuchot
of four. This fix a kernel panic on PowerMac G5 present since r1.22 of i2s.c, issue reported by Chris Jackman, thanks!
2012-12-10Implement support for Extended Sequence Numbers for the aesni driverMike Belopuhov
exercising the same approach as in sys/crypto/cryptosoft.c rev1.67.
2012-12-10Add cdce(4).Brad Smith
ok deraadt@
2012-12-09clarify that changes to this module were not done by intelMike Belopuhov
2012-12-09get rid of awkward label names and use simple forward/backwardMike Belopuhov
notation; no binary change
2012-12-09whitespace cleanup that's been due a long timeMike Belopuhov
2012-12-09Recognize 7448 and 970MP processors.Brad Smith
ok mpi@
2012-12-08Make ldomd(8) to control the availability of virtual disks to guest domains.Mark Kettenis
This is done by opening the corresponding /dev/vdspN device file. The virtual disk will remain available until that device is closed, which happens automatically when ldomd(8) exits.
2012-12-08Fix a comment now that PowerPC 970 are recognized, make it clear thatMartin Pieuchot
changing the frequency to a quarter mode is only supported by 970MP.
2012-12-08Recognize PowerPC 970 CPUs present in early PowerMac G5 from 2003 (7,2 andMartin Pieuchot
7,3) and let OpenBSD boot on these machines, yay! The 970 are similar to the 970FX except that they are manufactured in 130nm. Fix an issue reported by Andrew Fresh and kirby@ on misc@
2012-12-08Descriptor rings can be larger than a single page, so add the offset insteadMark Kettenis
of or-ing it in.
2012-12-08Make bus_dmamem_mmap(9) honor BUS_DMA_NOCACHE.Martin Pieuchot
ok kettenis@
2012-12-07Use a timeout to do a delayed handshake if the receive queue of the LDCMark Kettenis
channel transitions into the "up" state.
generated by cgit v1.2.3 (git 2.25.1) at 2024-12-13 13:28:08 +0000