summaryrefslogtreecommitdiff
path: root/sys/arch
AgeCommit message (Collapse)Author
2014-11-20When removing the identity mapping in low memory used during bootstrap,Mike Larkin
there is no reason to keep the NX bit around on null PTEs (PTEs that have been removed).
2014-11-20Don't attempt to clear/disable %tick_cmpr. This register is no longer presentMark Kettenis
on newer sun4v implementations (possibly SPARC T3 and later, definitely not there on SPARC T5) and it should not be necessary on older systems. Tested by dlg@
2014-11-20Disentagle bcopy/memcpy/memmove.Theo de Raadt
The situation was: memcpy swaps registers and drops into bcopy, which does the overlap check, before getting around to business. But memcpy is not supposed to handle overlapped arguments special, so we don't need all this complex register swapping and tests up front -- when memcpy is the most common operation. Refactor all these upside downupside down. Survived multiple make builds, so probably good, and faster. ok dlg
2014-11-20Move previous PTE permission fixup code into locore, and fixup some moreMike Larkin
ranges while we're there. ok deraadt@, tested by many and in snaps
2014-11-20Revert the previous changes that enabled various PTE permissions fixupsMike Larkin
in pmap.c - equivalent code will shortly be committed to locore.S that accomplishes the same thing. ok deraadt@
2014-11-19Only keep {recv,send}udp() in the pxe-specific code, the remainder of whatMiod Vallat
used to be local net.c (renamed to pxe_udp.c) can now be obtained from libsa's net.c rather than being duplicated. No functional change.
2014-11-19Remove some unused i386 pmap functions. Also fix two typos in commentsMike Larkin
while I had the hood open. ok deraadt@, mpi@
2014-11-19Add netudp.c here too for consistency, even though this code ends up not beingMiod Vallat
referenced by the bootblocks.
2014-11-19Move {recv,send}udp() out of net.c into their own file; to be used shortly.Miod Vallat
Add the new files to the few md libsa builds which require it.
2014-11-19No boot block code shall include <string.h>.Miod Vallat
2014-11-18Nuke some obvious #include duplications.Kenneth R Westerback
ok espie@ deraadt@ millert@ tedu@
2014-11-18make pmap_zero_page MP-safe, by using the directmapTheo de Raadt
mpi will investigate speedups after this. ok mpi kettenis
2014-11-18drop "pl" from the pool name and use pool_setlowatMike Belopuhov
2014-11-18(except for the arm architectures...) unify the way the ld.script isTheo de Raadt
used and depended, mimicking new changes by guenther to amd64.
2014-11-18rename ld.script to same name as other architecturesTheo de Raadt
2014-11-17Pull in the .openbsd.randomdata sections, assigned to both the data andPhilip Guenther
openbsd_randomize segments. While here, make the 'bsd' target depend on the ldscript. problem noted by and ok deraadt@
2014-11-17split the int and long implementations to avoid compiler warnings.David Gwynne
tested by and ok jsg@
2014-11-17Two additional POSIX_MADV_RANDOM conversionsTheo de Raadt
2014-11-16Replace a plethora of historical protection options with justTheo de Raadt
PROT_NONE, PROT_READ, PROT_WRITE, and PROT_EXEC from mman.h. PROT_MASK is introduced as the one true way of extracting those bits. Remove UVM_ADV_* wrapper, using the standard names. ok doug guenther kettenis
2014-11-16Add cas(4).Brad Smith
ok deraadt@
2014-11-14enable trunk.Henning Brauer
made possible by the install/upgrade changes by krw and the ifconfig adjustments by me. ok krw reyk
2014-11-14implement the atomic_foo things on arm.David Gwynne
testing and ok jsg@
2014-11-11Do not held the kernel lock when calling hardclock() and statclock().Martin Pieuchot
This is not necessary, as confirmed by tedu@ and kettenis@. Solve the major contention problem seen on my Dual G5. Reduce the build time of a kernel with two jobs from 5h+ to 18min and expose more easily pmap/memroy corruptions 8) ok kettenis@
2014-11-11Local APIC page doesn't need X permissions.Mike Larkin
ok deraadt, guenther
2014-11-09remove commented entries for tedu'd bluetooth driversJonathan Gray
2014-11-08Delete commented out 'romtty' device configuration.Kenji Aoyama
This had been used in the early years of poring OpenBSD to LUNA-88K, but no longer used.
2014-11-08No need to keep the temporary mappings for the MMU pages for the resumeMike Larkin
time page table after we are done creating it in the resuming kernel. Note the resume time stack has to remain mapped as it is used during the initial phase of the hibernate unpack while the original page table is still being used.
2014-11-08Kill #if 0'ed code for boot args that we don't care about on amd64.Philip Guenther
ok deraadt@ mlarkin@ kettenis@
2014-11-07Wrong comment - NX is handled later (for now), not in locore. No functionalMike Larkin
change. noticed by deraadt@
2014-11-07Enable xhci(4). Most of the features are here, USB 1.x devices only workMartin Pieuchot
if they are connected to the root hub and isochronous transfers are not supported for the moment. Let me know if your controller/device do not work. In this case attach a dmesg of a kernel build with XHCI_DEBUG. ok deraadt@
2014-11-07Map .rodata and the KVA range corresponding to the ISA hole as NX on amd64Mike Larkin
processors that support NX. Tested on amd64 machines with and without NX capability. Additional NX ranges can be added as desired to the table defined by this diff. ok deraadt@
2014-11-05remove uneeded bpf includesJonathan Gray
ok miod@
2014-11-05Map .rodata RO after boot on amd64. Makefile.amd64 changes from deraadt.Mike Larkin
ok deraadt@
2014-11-04Uncomment code that does voltage capabilities initialisation. WithoutJonathan Gray
this using an sd card with my rev C BeagleBone Black won't work. Should fix similiar problems reported by Gerke Preussner and Andrew Hills on arm@ ok syl@
2014-11-04Add rgephy(4) for the RTL8211E phy in the LeMaker Banana Pi and Banana Pro.Jonathan Gray
2014-11-04Add atphy(4) for the Atheros AR8031 phys in the AM335x starter kit.Jonathan Gray
Tested by Bernard Merindol.
2014-11-02Only mark segment 0 as executable on 64-bit systems. There it is harmless asMark Kettenis
we have a proper X bit in the page tables. On 32-bit systems kernel .text is handled by an IBAT, so we don't need page table entries that are executable in the kernel pmap. ok mpi@
2014-11-01Now the armv7 ramdisks are compiled with SMALL_KERNELJonathan Gray
the omap and sunxi ramdisks won't build because they call wdog_register() and kern_watchdog.c is !small_kernel in config. Leaving the watchdog devices out of the ramdisks won't work either as armv7_machdep.c uses platform_watchdog_reset() to reboot. Deal with this by only calling wdog_register() when not compiled with SMALL_KERNEL.
2014-10-31The config logic with "rl needs-flag" gotJonathan Gray
removed back in 2011. Remove code under #if NRL > 0. ok miod@
2014-10-31remove a duplicate pckbdvar.h include under NGSCKBD > 0Jonathan Gray
gsckbd was removed back in 2008 so this was never defined.
2014-10-31#if NSWDISPLAY > 0 -> #if NWSDISPLAY > 0Jonathan Gray
now wsdisplay_suspend() will run on suspend
2014-10-31Fix a missing include in amd64 pmap.c that resulted in an erroneous memoryMike Larkin
map entry being entered in uniprocessor (UP) kernels. Multiprocessor (MP) kernels not affected. ok guenther, deraadt
2014-10-27Remove the spilling code for IBAT register and set up the first IBAT registerMark Kettenis
to cover the first 8 MB of memory such that it covers kernel .text and not much else. This is a first step towards W^X in the kernel for machines with G4 and older processors. ok mpi@
2014-10-27Remove execute permission from most pages in the kernel pmap. This is a firstMark Kettenis
step towards W^X in the kernel, even though it is only effective on machines with a G5 processor. ok mpi@
2014-10-27use #ifdef DDB for Debugger()Jonathan Gray
2014-10-26uvm_km_valloc -> km_allocMark Kettenis
2014-10-26- add board type of edgerouter proJasper Lievisse Adriaanse
- don't attach octrtc on the edgerouter pro either openbsd boots fine on this system, but as there's no ethernet support yet there's no storage at all right now. ok pirofti@
2014-10-26define octeon model cn61xx pass 1.1Jasper Lievisse Adriaanse
ok pirofti@
2014-10-26Don't forget to free() the malloc()s.Miod Vallat
2014-10-25Don't let qlw attach to non-sun4m yet, as this driver currently depends uponMiod Vallat
sun4m's iommu for bus_dma. (Guess what task just hit my todolist...)
generated by cgit v1.2.3 (git 2.25.1) at 2024-12-04 00:08:22 +0000