summaryrefslogtreecommitdiff
path: root/sys/arch/amd64
AgeCommit message (Expand)Author
2014-02-23Determine whether the cpu supports xsave (and hence AVX).dsl
2014-02-20This doesn't need fpu.h, but should include ucontext.hdsl
2014-02-20Move the amd64 and i386 pcb to the bottom of the uarea, and move thedsl
2014-02-20Since fpu.c is contained in every kernel, and it calls routines inpgoyette
2014-02-18Copy fpu.h to releasedsl
2014-02-18It seems that firefox includes machine/fpu.h on amd64.dsl
2014-02-17Add urtw (ok'd by christos).schmonz
2014-02-15Load and save the fpu registers (for copies to/from userspace) usingdsl
2014-02-15Remove all references to MDL_USEDFPU and deferred fpu initialisation.dsl
2014-02-12Change i386 to use x86/fpu.c instead of i386/isa/npx.cdsl
2014-02-12Add definitions of the default control words directly to this filedsl
2014-02-12Change the argument to fpudna() to be the trapframe.dsl
2014-02-11Move sys/arch/amd64/amd64/fpu.c and sys/arch/amd64/include/fpu.hdsl
2014-02-09Steal the table from i386/isa/npx.c than selects the highest prioritydsl
2014-02-09Simplify fputrap() considerably.dsl
2014-02-09Best if x86_stmxcsr executes stmxcsr.dsl
2014-02-09Add x86_stmxcsr for amd64.dsl
2014-02-07Convert the amd64 build to use x86/cpu_extended_state.h so that the fpudsl
2014-02-05Apparantly clang doesn't like -mno-80387 or -mno-fp-ret-in-387dsl
2014-02-05Change the compiler options to explicitly specify:dsl
2014-02-04There is no need to check for recursive calls into fpudna().dsl
2014-01-25delete VM_DEFAULT_ADDRESS; some of those should be GC'ed because they matchchristos
2014-01-25provide propert address defaults for topdown and bottomup allocationchristos
2014-01-19Remove the unused 'struct md_coredump'.dsl
2014-01-15Reduce amount of -no-integrated-as on x86 as .code16 is now supported byjoerg
2014-01-15LLVM doesn't support data32/addr32, but is smart enough to figure thejoerg
2014-01-12Try to reserve space in a way that results in the same allocation forjoerg
2014-01-11remove unused and undefined .globljakllsch
2014-01-09Make the amd64 post-halt-press-a-key code match i386, and introducedholland
2014-01-04Remove __HAVE_PROCESS_XFPREGS and add the extra parameter for the sizedsl
2014-01-01Change the type of the 'cookie' that holds the state of the core dump filedsl
2013-12-12Since regs->fxstate is actually of type 'struct fxsave64' use a structuredsl
2013-12-11Remove the fields that were used to save the i387 fp state on interrupt.dsl
2013-12-08Add some definitions for cpu 'extended state'.dsl
2013-12-01revert fpu/pcu changes until we figure out what's wrong; they cause randomchristos
2013-11-18implement the *at() syscalls.chs
2013-11-17Always define SLJIT_CACHE_FLUSH(), start include guards with '_' and use _LP6...alnsn
2013-11-11NetBSD 6.99.26: Switch i386 and amd64 to the x87 default control wordjoerg
2013-11-06- move variables inside their #ifdef usemrg
2013-11-02sprinkle __diagusedchristos
2013-10-27- fpu_state_save() on amd64: ci->ci_fpsaving is no longer used, simplify.rmind
2013-10-26Add driver for ASIX AX88178a and AX88179 Ethernet interface.nonaka
2013-10-23Use the MI "pcu" framework for bookkeeping of npx/fpu states on x86.drochner
2013-10-23Add xhci devicematt
2013-10-17add missing _christos
2013-10-17we need to return something here.christos
2013-10-17use the parameter for instruction macroschristos
2013-10-17remove unused variable, and move variable used only by XEN in the XEN ifdef.christos
2013-10-16remove unusedchristos
2013-09-12Pass PICFLAGS down to cc-as-as and use __PIC__ to decide if it is smalljoerg