b999c51a1a
Code cleanup. Don't use gpr.SetRegImm where all we do is write to a scratch register
2026-02-10 19:28:53 +01:00
c3db3d5187
arm64jit: When rouding unset, use nearest.
...
The 0/default rounding mode is nearest, not toward zero.
We set hasSetRounding only when fcr31 has a non-zero rounding mode or
flush to zero set.
2023-07-25 20:30:05 -07:00
6df939034a
Core: Cleanup some sign extensions for clarity.
2023-04-05 17:16:51 -07:00
b9352354c9
Masked PSP Memory support for the AArch64 Dynarec
2019-04-15 12:07:57 +02:00
419c1fbd73
Jit: Respect flags for jit types and features.
...
Left some free space for more.
2019-02-03 14:57:08 -08:00
eb4b59b530
arm64jit: Enable breakpoints.
...
Memory breakpoints not yet really tested.
2018-06-06 17:31:56 -07:00
ab809bd19e
jit: Apply hasSetRounding at compile time.
...
Otherwise, the block will be executed with the wrong rounding mode the
first time rounding is set. This could be important if it was set for a
single operation.
This is only a problem the first time it's set.
2018-04-01 10:36:16 -07:00
5177db0f91
arm64jit: Remove unnecessary address masking.
...
We use views like on x86_64, so this isn't needed.
2017-12-28 23:58:30 -08:00
6fd17fb026
arm64jit: Use reg sum for LDR/STR.
...
Skips an add, and should be less ops anyway.
2017-12-28 10:19:55 -08:00
092f98d313
arm64jit: Fix an integer truncation warning.
2017-12-27 19:39:04 -08:00
7c2fc90def
arm64jit: Avoid MOVK elsewhere without pointerify.
2017-12-27 17:57:19 -08:00
8c3552de74
cmake: Detect features at compile time
...
Instead of relying on manually passed down flags from CMake,
we now have ppsspp_config.h file to create the platform defines for us.
This improves support for multiplatform builds (such as iOS).
2016-10-19 12:31:19 +02:00
8a3c96a413
ARM64: Don't need to avoid destroying SCRATCH1 in these functions.
2015-10-08 14:54:43 +02:00
66adc4e695
jit: Normalize CONDITIONAL_DISABLE formatting.
2015-07-02 20:31:37 -07:00
1d1c80d9cf
arm64: Use BFI for cfc1.
2015-07-02 20:31:35 -07:00
b6612edf67
arm64: Use a cached rounding func for cvt.w.s.
...
This is much faster for this particular instruction, although not all
games even use it.
2015-06-28 12:40:29 -07:00
1c163e4817
arm64: Avoid an ORR for c.ueq.
...
This is about 15% faster for this single, uncommon instruction on A57.
2015-06-28 10:52:17 -07:00
febe435946
arm64: Use FP load/stores for non-reg pointers.
2015-06-28 10:45:44 -07:00
fbd4db0fc4
arm64: Add a safemem path.
...
This is probably not optimal but at least it works.
2015-06-27 00:22:04 -07:00
b3aa6d89e9
Fix UBFX encoding (thanks SonicAdvance1.)
2015-06-26 21:27:03 -07:00
b0d291032d
armjit Avoid cfc1/mfc1 to $0.
2015-04-07 18:30:36 -07:00
0a70618f87
ARM64: Accurate floating point rounding. For some reason, FTZ doesn't seem to work though.
2015-04-06 18:13:36 +02:00
44286a2b37
ARM64: Accurate float->int conversion with rounding mode.
2015-04-06 18:13:34 +02:00
8eedcc7fb0
ARM64: Speedup fpu/vfpu load/stores too using "pointerification". Actually noticeable gain.
2015-04-06 18:13:32 +02:00
34e61ab875
ARM64: More FPU instructions (int<->float convert), minor stuff
2015-04-06 18:13:25 +02:00
25ec85551f
ARM64: Implement FP compares, misc
2015-04-06 18:13:22 +02:00
5dff3f8c89
ARM64: Implement scalar FMOV. This makes the FPU2op ops work.
2015-04-06 18:13:16 +02:00
4233921ab7
ARM64: Some more instructions, func replacements
2015-04-06 18:13:16 +02:00
2bca05c4f2
ARM64: implement shifts, movz/movn. Corresponding fixes to emitter/disasm
2015-04-06 18:13:14 +02:00
0922db6062
ARM64: Some FP work.
2015-04-06 18:13:11 +02:00
d2c746dd64
ARM64: Get the FP reg cache working, implement some basic FP arith
2015-04-06 18:13:11 +02:00
b309c83973
Initial work on ARM64, based on the ARM jit.
2015-04-06 18:13:01 +02:00
Henrik Rydgård