leafspark
/
Citron
mirror of https://git.citron-emu.org/citron/emu
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
18,273 Commits 1 Branch 7 Tags 74 MiB
Commit Graph

34 Commits

Author SHA1 Message Date
Morph a745d87971 general: Fix various spelling errors 2021-01-02 10:23:41 -05:00
bunnei 3d592972dc
Revert "core: Fix clang build" 2020-10-20 19:07:39 -07:00
Lioncash be1954e04c core: Fix clang build 
Recent changes to the build system that made more warnings be flagged as
errors caused building via clang to break.

Fixes #4795
 2020-10-17 19:50:39 -04:00
Lioncash 9f027b1af2 common/swap: Make use of std::endian 
Allows removing a bunch of defines in favor of a two liner.
 2020-07-14 16:26:54 -04:00
Lioncash 0d8ef2d3b9 common/swap: Improve codegen of the default swap fallbacks 
Uses arithmetic that can be identified more trivially by compilers for
optimizations. e.g. Rather than shifting the halves of the value and
then swapping and combining them, we can swap them in place.

e.g. for the original swap32 code on x86-64, clang 8.0 would generate:

    mov     ecx, edi
    rol     cx, 8
    shl     ecx, 16
    shr     edi, 16
    rol     di, 8
    movzx   eax, di
    or      eax, ecx
    ret

while GCC 8.3 would generate the ideal:

    mov     eax, edi
    bswap   eax
    ret

now both generate the same optimal output.

MSVC used to generate the following with the old code:

    mov     eax, ecx
    rol     cx, 8
    shr     eax, 16
    rol     ax, 8
    movzx   ecx, cx
    movzx   eax, ax
    shl     ecx, 16
    or      eax, ecx
    ret     0

Now MSVC also generates a similar, but equally optimal result as clang/GCC:

    bswap   ecx
    mov     eax, ecx
    ret     0

====

In the swap64 case, for the original code, clang 8.0 would generate:

    mov     eax, edi
    bswap   eax
    shl     rax, 32
    shr     rdi, 32
    bswap   edi
    or      rax, rdi
    ret

(almost there, but still missing the mark)

while, again, GCC 8.3 would generate the more ideal:

    mov     rax, rdi
    bswap   rax
    ret

now clang also generates the optimal sequence for this fallback as well.

This is a case where MSVC unfortunately falls short, despite the new
code, this one still generates a doozy of an output.

    mov     r8, rcx
    mov     r9, rcx
    mov     rax, 71776119061217280
    mov     rdx, r8
    and     r9, rax
    and     edx, 65280
    mov     rax, rcx
    shr     rax, 16
    or      r9, rax
    mov     rax, rcx
    shr     r9, 16
    mov     rcx, 280375465082880
    and     rax, rcx
    mov     rcx, 1095216660480
    or      r9, rax
    mov     rax, r8
    and     rax, rcx
    shr     r9, 16
    or      r9, rax
    mov     rcx, r8
    mov     rax, r8
    shr     r9, 8
    shl     rax, 16
    and     ecx, 16711680
    or      rdx, rax
    mov     eax, -16777216
    and     rax, r8
    shl     rdx, 16
    or      rdx, rcx
    shl     rdx, 16
    or      rax, rdx
    shl     rax, 8
    or      rax, r9
    ret     0

which is pretty unfortunate.
 2019-04-12 00:07:39 -04:00
Lioncash 66b73fd399 common/swap: Mark byte swapping free functions with [[nodiscard]] and noexcept 
Allows the compiler to inform when the result of a swap function is
being ignored (which is 100% a bug in all usage scenarios). We also mark
them noexcept to allow other functions using them to be able to be
marked as noexcept and play nicely with things that potentially inspect
"nothrowability".
 2019-04-11 20:42:44 -04:00
Lioncash 9cb4b7be40 common/swap: Simplify swap function ifdefs 
Including every OS' own built-in byte swapping functions is kind of
undesirable, since it adds yet another build path to ensure compilation
succeeds on.

Given we only support clang, GCC, and MSVC for the time being, we can
utilize their built-in functions directly instead of going through the
OS's API functions.

This shrinks the overall code down to just

if (msvc)
  use msvc's functions
else if (clang or gcc)
  use clang/gcc's builtins
else
  use the slow path
 2019-04-11 20:36:19 -04:00
Lioncash 598954436f common/swap: Remove 32-bit ARM path 
We don't plan to support host 32-bit ARM execution environments, so this
is essentially dead code.
 2019-04-11 20:15:47 -04:00
bunnei 3e930304fe
Merge pull request #2090 from FearlessTobi/port-4599 
Port citra-emu/citra#4244 and citra-emu/citra#4599: Changes to BitField
 2019-03-20 23:44:20 -04:00
tgsm 030814b1cb Remove GCC version checks 
Citra can't be compiled using GCC <7 because of required C++17 support, so these version checks don't need to exist anymore.
 2019-02-24 15:24:06 +01:00
Weiyi Wang 71530781f3 common/swap: remove default value for swap type internal storage 
This is compromise for swap type being used in union. A union has deleted default constructor if it has at least one variant member with non-trivial default constructor, and no variant member of T has a default member initializer. In the use case of Bitfield, all variant members will be the swap type on endianness mismatch, which would all have non-trivial default constructor if default value is specified, and non of them can have member initializer
 2019-02-06 17:24:27 +01:00
Weiyi Wang 6734c64976 common/swap: use template and tag for LE/BE specification 
The tag can be useful for other type-generic templates like BitFields to forward the endianness specification
 2019-02-06 17:24:13 +01:00
Weiyi Wang 94bc48dd78 common/swap: add swap template for enum 2019-02-06 17:21:15 +01:00
Cameron Cawley 1670c4421f Port #3732 from Citra: "common: Fix compilation on ARM" 2018-07-29 15:51:31 +02:00
Lioncash 5c47ea1a4e common/swap: Remove unnecessary const on return value of swap() 2018-07-19 09:35:54 -04:00
Lioncash 0a868641fa common/swap: Use static_cast where applicable 2018-07-19 09:35:13 -04:00
Lioncash 1edf4dd7ef common/swap: Use using aliases where applicable 2018-07-19 09:32:13 -04:00
fearlessTobi 70a6691e3b Port #3579 from Citra 2018-07-07 13:59:18 +02:00
Daniel Lim Wee Soong a66204eb5c common: fix swap functions on Bitrig and OpenBSD 
swap{16,32,64} are defined as macros on the two, but client code
tries to invoke them as Common::swap{16,32,64}, which naturally
doesn't work. This hack redefines the macros as inline functions
in the Common namespace: the bodies of the functions are the
same as the original macros, but relying on OS-specific
implementation details like this is of course brittle.
 2018-04-03 02:25:11 +08:00
Jan Beich 94d23b480e common: use system bswap* functions on more BSDs 2016-10-27 23:28:30 +00:00
Emmanuel Gil Peyrot ebdae19fd2 Remove empty newlines in #include blocks. 
This makes clang-format useful on those.

Also add a bunch of forgotten transitive includes, which otherwise
prevented compilation.
 2016-09-21 11:15:47 +09:00
Yuri Kunde Schlesner 396a8d91a4 Manually tweak source formatting and then re-run clang-format 2016-09-18 21:14:25 -07:00
Emmanuel Gil Peyrot dc8479928c Sources: Run clang-format on everything. 2016-09-18 09:38:01 +09:00
Lioncash d5b983a8c0 swap: Get rid of pointer casting for swapping structs 
These shouldn't haphazardly convert types
 2016-05-08 23:33:52 -04:00
Lioncash 47ca79ba4b swap: Get rid of undefined behavior in swapf and swapd 
This isn't well-defined in C++.
 2016-05-08 23:21:47 -04:00
Lioncash aef4630102 swap: Remove unused methods 
Also gets rid of pointer data variants as this prevents the use of
the regular swapping routines as unary predicates in std lib functions.

They also cast to stricter alignment types, which is undefined behavior.
 2016-05-08 23:12:04 -04:00
Lioncash bc886a7a03 common: Get rid of a cast in swap.h 2015-09-11 08:54:33 -04:00
Emmanuel Gil Peyrot 2d044a67c9 Common: Cleanup memory and misc includes. 2015-06-28 00:36:54 +01:00
Emmanuel Gil Peyrot b1503b2020 Remove every trailing whitespace from the project (but externals). 2015-05-29 21:59:29 +01:00
archshift e011acaa84 Removed swap code redundancy and moved common swap code to swap.h 2015-03-05 22:46:45 -08:00
Emmanuel Gil Peyrot 6b411c63c9 Common: Remove dead platform #ifdefs to make the code more readable. 
Symbian, Xbox, Blackberry and iOS got removed.
FreeBSD and Android kept due to them potentially being able to run Citra in the future.
The iOS specific part also got removed from PPSSPP in order to fix a bug there.
 2015-01-06 18:57:28 +00:00
Yuri Kunde Schlesner da564d3fe0 Fix compile errors in Clang 2014-10-26 16:18:04 -02:00
bunnei 7a136b8a84 fixes to build on linux 2014-04-22 19:42:29 -07:00
bunnei 63e46abdb8 got rid of 'src' folders in each sub-project 2014-04-08 19:25:03 -04:00