 Rika Ichinose
|
0d5f7fa66b
Increase non-temporal i386 & x64 Fill* thresholds to 4 Mb.
|
vor 1 Jahr |
|
Rika Ichinose
|
1ec0326995
REP STOS branch for x64 Fill* (only for System V ABI for now).
|
vor 1 Jahr |
|
Rika Ichinose
|
a4c324ee23
Fill* for x64, physically sharing half of the code with FillChar.
|
vor 2 Jahren |
|
Rika Ichinose
|
b468793c63
Index/Compare refined by hand instead of mostly being GCC output.
|
vor 2 Jahren |
 florian
|
b164817e18
* check also for XGETBV support, resolves problem reported by Pierre
|
vor 1 Jahr |
|
florian
|
704ad21b23
+ centralized cpu capability detection
|
vor 1 Jahr |
|
Rika Ichinose
|
c07f36b30b
Post-modern CompareByte for x86-64/SSE2.
|
vor 2 Jahren |
|
Rika Ichinose
|
0e426db5de
x86_64.inc: shorten Interlocked*, perform macro-fused test+jz in Index* early.
|
vor 2 Jahren |
|
Rika Ichinose
|
669d41172c
Fix UTF-8 symbols in comments.
|
vor 2 Jahren |
|
Rika Ichinose
|
8d5d7b480d
Supposedly faster Move for x64.
|
vor 2 Jahren |
|
Rika Ichinose
|
f20c7b9ae9
Shorter x86_64.inc:inc/declocked.
|
vor 2 Jahren |
|
Rika Ichinose
|
b56cbad50e
Supposedly faster FillChar for x64.
|
vor 2 Jahren |
|
Rika Ichinose
|
8e884d9acd
Handle Index* / Compare* tail by directly reading last VECSIZE bytes, if there was at least one full vector.
|
vor 2 Jahren |
|
florian
|
ee16fc7b96
* patch by Rika, trivial adjustments to !373, resolves #40172
|
vor 2 Jahren |
|
Rika Ichinose
|
da12cfc867
Improved CompareWord for i386 and x86_64.
|
vor 2 Jahren |
|
florian
|
7cc94fc000
* patch by Rika: Trivial adjustments to !379, resolves #40168
|
vor 2 Jahren |
|
Rika Ichinose
|
b723178117
Even better CompareByte for x64.
|
vor 2 Jahren |
|
Rika Ichinose
|
d36e96ea74
Improved CompareDWord for i386 and x86_64.
|
vor 2 Jahren |
|
Rika Ichinose
|
eff26797ab
SSE2 IndexDWord for x64.
|
vor 2 Jahren |
|
Rika Ichinose
|
524589231f
Improved CompareByte for i386 and x86_64.
|
vor 2 Jahren |
 Jonas Maebe
|
0758aa1143
FPU exception mask: generlised system unit interface
|
vor 3 Jahren |
|
florian
|
e1698a5969
* when compiling with the main branch compiler, p2align with 3 parameters can be used now
|
vor 4 Jahren |
|
florian
|
bdc826cc18
* x86-64: do not jump in inc/declocked to return but return directly
|
vor 4 Jahren |
 svenbarth
|
0c316deab5
* readd SmallInt typecasts to SmallInt overload of SwapEndian
|
vor 5 Jahren |
|
svenbarth
|
9d86fed95b
* avoid range check error when using SwapEndian with 16-bit constants
|
vor 5 Jahren |
|
florian
|
33fa211b84
+ x86-64: assembler implementation for u128_div_u64_to_u64 for SysV ABI
|
vor 5 Jahren |
|
florian
|
62f890efa3
* improved floating point signal handling on x86/linux
|
vor 6 Jahren |
|
florian
|
5b766a1e67
* fixed CompareByte for x86-64 after r37528
|
vor 8 Jahren |
|
florian
|
0c7306e1cd
* improved CompareByte for x86-64 based on a patch by Markus Beth
|
vor 8 Jahren |
 marco
|
bc39eb29d3
* comment spelling fix (modifed->modified)
|
vor 8 Jahren |
