summaryrefslogtreecommitdiffstats
path: root/sys/amd64
Commit message (Expand)AuthorAgeFilesLines
* As discussed on -current, remove the hardcoded default maxswzone.des2012-08-141-8/+0
* Add a hackish debugging facility to provide a bit of information aboutkib2012-08-141-2/+20
* Real hardware, as opposed to QEMU, does not allow to have a call gatekib2012-08-142-6/+27
* Remove the deassert INIT IPI from the IPI startup sequence for APs.jhb2012-08-131-16/+1
* Add a 10 millisecond delay after sending the initial INIT IPI. Thisjhb2012-08-131-1/+2
* Build modules along with the XENHVM kernels.cperciva2012-08-131-2/+0
* The assertion that I added in r238889 could legitimately fail when aalc2012-08-081-1/+2
* Do not apply errata 721 workaround when under hypervisor, sincekib2012-08-071-1/+7
* Remove duplicate header inclusion of <sys/sysent.h>pluknet2012-08-071-1/+0
* Shave off a few more cycles from the average execution time of pmap_enter()alc2012-08-051-7/+2
* Add lfence().kib2012-08-011-0/+7
* Revise pmap_enter()'s handling of mapping updates that change thealc2012-08-011-22/+30
* Change (unused) prototype for stmxcsr() to match reality.kib2012-07-301-1/+1
* Shave off a few more cycles from pmap_enter()'s critical section. Inalc2012-07-291-14/+17
* Forcibly shut up clang warning about NULL pointer dereference.kib2012-07-231-0/+7
* Constently use 2-space sentence breaks.kib2012-07-211-2/+2
* Stop caching curpcb in the local variable.kib2012-07-211-16/+12
* The PT_I386_{GET,SET}XMMREGS and PT_{GET,SET}XSTATE operate on thekib2012-07-211-1/+6
* Stop clearing x87 exceptions in the #MF handler on amd64. If user codekib2012-07-211-18/+13
* Introduce curpcb magic variable, similar to curthread, which is MDkib2012-07-195-14/+30
* Don't unnecessarily set PGA_REFERENCED in pmap_enter().alc2012-07-191-7/+5
* On AMD64, provide siginfo.si_code for floating point errors when errorkib2012-07-183-15/+34
* Add stmxcsr.kib2012-07-181-0/+2
* Add support for the XSAVEOPT instruction use. Our XSAVE/XRSTOR usagekib2012-07-143-1/+73
* Wring a few cycles out of pmap_enter(). In particular, on a user-spacealc2012-07-131-75/+86
* Add a clts() wrapper around the 'clts' instruction to <machine/cpufunc.h>jhb2012-07-092-6/+13
* Partially revert r217515 so that the mem_range_softc variable is alwaysjhb2012-07-092-2/+3
* Use assembler mnemonic instead of manually assembling, contination for r238142.kib2012-07-061-6/+3
* Several fixes to the amd64 disassembler:jhb2012-07-061-22/+134
* Make pmap_enter()'s management of PV entries consistent with the other pmapalc2012-07-061-20/+15
* Now that our assembler supports the xsave family of instructions, use themjhb2012-07-052-19/+23
* Calculate the new PTE value in pmap_enter() before acquiring any locks.alc2012-07-051-32/+27
* Correct an error in r237513. The call to reserve_pv_entries() must comealc2012-07-051-8/+12
* Decode the 'xsave', 'xrstor', 'xsaveopt', 'xgetbv', 'xsetbv', andjhb2012-07-041-3/+18
* tws(4) is interfaced with CAM so move it to the same section.delphij2012-07-011-1/+1
* Optimize reserve_pv_entries() using the popcnt instruction.alc2012-06-302-3/+18
* In r237592, I forgot that pmap_enter() might already hold a PV list lockalc2012-06-291-15/+30
* Avoid some unnecessary PV list locking in pmap_enter().alc2012-06-281-1/+1
* Optimize pmap_pv_demote_pde().alc2012-06-281-10/+35
* Add new pmap layer locks to the predefined lock order. Change the namesalc2012-06-271-3/+3
* Introduce RELEASE_PV_LIST_LOCK().alc2012-06-261-8/+11
* Add PV list locking to pmap_enter(). Its execution is no longer serializedalc2012-06-261-57/+70
* Add PV chunk and list locking to pmap_change_wiring(), pmap_protect(), andalc2012-06-251-57/+109
* Introduce reserve_pv_entry() and use it in pmap_pv_demote_pde(). In orderalc2012-06-231-1/+83
* Implement mechanism to export some kernel timekeeping data tokib2012-06-221-0/+6
* Reserve AT_TIMEKEEP auxv entry for providing usermode the pointer tokib2012-06-221-0/+1
* Introduce CHANGE_PV_LIST_LOCK_TO_{PHYS,VM_PAGE}() to avoid duplication ofalc2012-06-221-24/+20
* Update the PV stats in free_pv_entry() using atomics. After which, it isalc2012-06-211-38/+14
* Add PV list locking to pmap_copy(), pmap_enter_object(), andalc2012-06-201-31/+61
* Condition the implementation of pv_entry_count on PV_STATS. On amd64,alc2012-06-191-16/+16
OpenPOWER on IntegriCloud