Bochs/bochs/cpu
Kevin Lawton 425ad824c0 I changed the TLB entry from 3 dwords to 4, and (when you compile
with GCC) align them with the GCC special alignment attribute.
Since there was then one available field, I split the protection
attributes and native host pointers into their own fields.

Before, with 3 dwords per TLB entry, some entries (about 3/8)
were spanning two processor cache lines (assuming a 32-byte
cache line).  Now, they all fit within one cache line.

Knocked about 1.4% off Win95 boot time, probably more off normal
software runs.
2002-09-10 00:01:01 +00:00
..
access.cc I changed the TLB entry from 3 dwords to 4, and (when you compile 2002-09-10 00:01:01 +00:00
apic.cc - if misaligned or wrong size write, print the address and length! 2002-07-23 15:32:14 +00:00
arith8.cc I extended the guest-to-host TLB acceleration across the 2002-09-06 21:54:58 +00:00
arith16.cc I rehashed the way the EFLAGS register was stored internally. 2002-09-08 04:08:14 +00:00
arith32.cc I rehashed the way the EFLAGS register was stored internally. 2002-09-08 04:08:14 +00:00
bcd.cc - AAM can generate an exception (divide by 0) 2002-07-06 11:02:35 +00:00
bit.cc I extended the guest-to-host TLB acceleration across the 2002-09-06 21:54:58 +00:00
cpu.cc I rehashed the way the EFLAGS register was stored internally. 2002-09-08 04:08:14 +00:00
cpu.h I changed the TLB entry from 3 dwords to 4, and (when you compile 2002-09-10 00:01:01 +00:00
ctrl_xfer8.cc - add RCS Id to header of every file. This makes it easier to know what's 2001-10-03 13:10:38 +00:00
ctrl_xfer16.cc - add RCS Id to header of every file. This makes it easier to know what's 2001-10-03 13:10:38 +00:00
ctrl_xfer32.cc Added a comment on iret flag writing. 2002-02-22 05:33:36 +00:00
ctrl_xfer_pro.cc I rehashed the way the EFLAGS register was stored internally. 2002-09-08 04:08:14 +00:00
data_xfer8.cc I extended the guest-to-host TLB acceleration across the 2002-09-06 21:54:58 +00:00
data_xfer16.cc I extended the guest-to-host TLB acceleration across the 2002-09-06 21:54:58 +00:00
data_xfer32.cc I extended the guest-to-host TLB acceleration across the 2002-09-06 21:54:58 +00:00
debugstuff.cc I rehashed the way the EFLAGS register was stored internally. 2002-09-08 04:08:14 +00:00
decode16.cc - add RCS Id to header of every file. This makes it easier to know what's 2001-10-03 13:10:38 +00:00
decode32.cc - add RCS Id to header of every file. This makes it easier to know what's 2001-10-03 13:10:38 +00:00
exception.cc I rehashed the way the EFLAGS register was stored internally. 2002-09-08 04:08:14 +00:00
fetchdecode.cc - check in Stanislav Shwartsman's patch "bochs-mmx.patch-endian-support". 2002-09-09 16:11:25 +00:00
flag_ctrl_pro.cc I rehashed the way the EFLAGS register was stored internally. 2002-09-08 04:08:14 +00:00
flag_ctrl.cc I rehashed the way the EFLAGS register was stored internally. 2002-09-08 04:08:14 +00:00
i387.h - check in Stanislav Shwartsman's patch "bochs-mmx.patch-endian-support". 2002-09-09 16:11:25 +00:00
init.cc I rehashed the way the EFLAGS register was stored internally. 2002-09-08 04:08:14 +00:00
io_pro.cc I rehashed the way the EFLAGS register was stored internally. 2002-09-08 04:08:14 +00:00
io.cc Enhanced the repeat IO accelerations (enabled by --enable-repeat-speedups) 2002-09-09 16:56:56 +00:00
lazy_flags.cc I rehashed the way the EFLAGS register was stored internally. 2002-09-08 04:08:14 +00:00
lazy_flags.h I rehashed the way the EFLAGS register was stored internally. 2002-09-08 04:08:14 +00:00
logical8.cc I extended the guest-to-host TLB acceleration across the 2002-09-06 21:54:58 +00:00
logical16.cc I extended the guest-to-host TLB acceleration across the 2002-09-06 21:54:58 +00:00
logical32.cc I extended the guest-to-host TLB acceleration across the 2002-09-06 21:54:58 +00:00
Makefile.in - check in Stanislav Shwartsman's patch "bochs-mmx.patch-endian-support". 2002-09-09 16:11:25 +00:00
mmx.cc Very small #ifdef mods so that all the static functions would 2002-09-09 17:13:13 +00:00
mult8.cc - add RCS Id to header of every file. This makes it easier to know what's 2001-10-03 13:10:38 +00:00
mult16.cc - add RCS Id to header of every file. This makes it easier to know what's 2001-10-03 13:10:38 +00:00
mult32.cc - add RCS Id to header of every file. This makes it easier to know what's 2001-10-03 13:10:38 +00:00
paging.cc I changed the TLB entry from 3 dwords to 4, and (when you compile 2002-09-10 00:01:01 +00:00
proc_ctrl.cc - check in Stanislav Shwartsman's patch "bochs-mmx.patch-endian-support". 2002-09-09 16:11:25 +00:00
protect_ctrl_pro.cc - add RCS Id to header of every file. This makes it easier to know what's 2001-10-03 13:10:38 +00:00
protect_ctrl.cc I extended the guest-to-host TLB acceleration across the 2002-09-06 21:54:58 +00:00
resolve16.cc - add RCS Id to header of every file. This makes it easier to know what's 2001-10-03 13:10:38 +00:00
resolve32.cc - add RCS Id to header of every file. This makes it easier to know what's 2001-10-03 13:10:38 +00:00
segment_ctrl_pro.cc - patch from Mike Rieker <mrieker@o3one.org> associated with this bug rpt: 2001-11-13 05:11:41 +00:00
segment_ctrl.cc - add RCS Id to header of every file. This makes it easier to know what's 2001-10-03 13:10:38 +00:00
shift8.cc I extended the guest-to-host TLB acceleration across the 2002-09-06 21:54:58 +00:00
shift16.cc I extended the guest-to-host TLB acceleration across the 2002-09-06 21:54:58 +00:00
shift32.cc I extended the guest-to-host TLB acceleration across the 2002-09-06 21:54:58 +00:00
soft_int.cc - add RCS Id to header of every file. This makes it easier to know what's 2001-10-03 13:10:38 +00:00
stack16.cc - add RCS Id to header of every file. This makes it easier to know what's 2001-10-03 13:10:38 +00:00
stack32.cc - Somebody was convinced that the enter instruction with level>0 was broken, 2002-03-05 15:50:17 +00:00
stack_pro.cc - add RCS Id to header of every file. This makes it easier to know what's 2001-10-03 13:10:38 +00:00
string.cc I rehashed the way the EFLAGS register was stored internally. 2002-09-08 04:08:14 +00:00
tasking.cc Fixed two calls to dtranslate_linear in paging.cc to use 2002-09-09 21:59:10 +00:00
vm8086.cc - included instinc's patch.stack-return-from-v86 2002-06-27 13:31:54 +00:00