summaryrefslogtreecommitdiffstats
path: root/sys/sparc64
Commit message (Expand)AuthorAgeFilesLines
* Add upa.h, which I had previously forgotten, to unbreak the sparc64tmm2002-01-081-0/+34
* Catch up to the latest and greatest.jake2002-01-081-1/+26
* Add initial smp support. This gets as far as allowing the secondaryjake2002-01-087-0/+1093
* Setup the normal global pcb register as well on entry from user land.jake2002-01-082-92/+116
* Adapt the vectored interrupt handler for receiving ipis. If the secondjake2002-01-082-94/+106
* Use the per-cpu panic stack in the case of a fault with a bad kerneljake2002-01-082-4/+10
* Remove ATOMIC_INC_INT macro which has moved elsewhere.jake2002-01-082-22/+0
* Catch up to change in compile time assertion interface.jake2002-01-082-3/+3
* Make this code more robust in the event of stray interrupts. Handlejake2002-01-081-12/+22
* Use cpufunc macros instead of using inline asm directly.jake2002-01-081-1/+1
* Set the normal global pcb register when context switching.jake2002-01-082-0/+4
* Add a macro for getting the tlbs (itlb and/or dtlb) which the givenjake2002-01-081-0/+1
* Prototype pmap_map_tsb().jake2002-01-081-0/+2
* Remove PANIC_STACK_PAGES which is no longer used.jake2002-01-081-5/+3
* Add declarations needed by last commit.jake2002-01-081-0/+8
* Update comments about _start, the kernel entry point, to reflect newjake2002-01-084-101/+321
* Fix qsort callouts used for sorting memory regions during boot. vm_offset_tjake2002-01-081-4/+22
* Add a md field to pcpu for the upa module id.jake2002-01-081-3/+6
* Define CKLF_PC in terms of TRAPF_PC.jake2002-01-081-1/+1
* Add a mov() macro, which is used in conjunction with the register definesjake2002-01-081-0/+4
* Update comments and defines to reflect that normal and alternate g6 pointjake2002-01-081-22/+52
* Add asis for the upa config reg, which contains the hardware cpu id, andjake2002-01-081-0/+2
* Convert a bunch of 1 << PCPU_GET(cpuid) to PCPU_GET(cpumask).peter2002-01-051-1/+1
* 1. Implement an optimization for pmap_remove() and pmap_protect(): if atmm2002-01-024-48/+143
* Correct the defintion of struct ofw_upa_regs, and use it instead oftmm2002-01-026-61/+82
* Close a window of time during early boot in which an interrupt wouldtmm2002-01-021-18/+18
* Correctly identify the cpu in certain ultra 1s.jake2002-01-011-0/+1
* Define __ASM__ so that libc will know not to define C things.jake2002-01-011-0/+2
* Add a define for the fp restore soft trap type.jake2002-01-011-1/+2
* Add constants needed by user trap code.jake2002-01-011-0/+6
* Enable virtual caching for kernel pages. When we enabled virtual cachingjake2002-01-011-3/+3
* Add some more info to traces.jake2002-01-012-38/+26
* Ensure that the syscall trap vector is properly aligned.jake2002-01-012-0/+2
* Implement sysarch(SPARC_UTRAP_INSTALL).jake2002-01-011-0/+69
* Implement user trap delivery as specified by the sparc abi. This providesjake2002-01-016-4/+189
* Add a panic stack, which is used as a known good stack when there isjake2002-01-016-9/+244
* Add a soft trap for restoring the fpu registers from the pcb.jake2002-01-012-0/+24
* Fix long lines in the trap table due to the abi specificied trap typesjake2002-01-012-258/+260
* Do not include pcib.h, which only existed in my development tree, and dotmm2001-12-301-2/+1
* Add bus_common.h, which contains some definiton that apply to both PCItmm2001-12-301-0/+69
* Make these compile.jake2001-12-293-24/+18
* Remove local change that crept in.jake2001-12-291-1/+0
* Add a header for user trap types required by the sparc abi.jake2001-12-291-0/+84
* Adapt for used by upcoming fp emulation code.jake2001-12-291-62/+41
* Print the correct v9 opcodes.jake2001-12-291-18/+18
* Update to new constants.jake2001-12-291-34/+34
* Make cont in ddb work.jake2001-12-291-0/+1
* Prototype dcache_inval_phys.jake2001-12-291-0/+1
* Add comments as to why VM_MAXUSER_ADDRESS is magic (abitrary).jake2001-12-291-6/+21
* Make tte bit constants explicitly unsigned and long.jake2001-12-291-46/+33
OpenPOWER on IntegriCloud