diff --git a/CHANGELOG.md b/CHANGELOG.md --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -1,93 +1,93 @@ -# Version 0.4.5 - -- fix build with newer base - -# Version 0.4.4 - -- export some useful data types - -# Version 0.4.3 - -- fixes to compile on GHC 7.10 and on Windows - -# Version 0.4.2 - -- fix OSX compilation - -# Version 0.4.1 - -- fix a bug which caused segmentation fault if a compiled function was called multiple times - -# Version 0.4 - -- rewrite FFI -- support calling labels -- speed up long code alignments -- bugfix: auto size calculation fix - -# Version 0.3.1 - -- use multi-byte nop operations for padding -- `preBuild` operation (may speed up code generation) -- branch-predicition friendlier if-then-else -- not-condition pattern: `N` - usage example: `j (N E)` which is the same as `j NE` -- bugfix: fix if-then-else condition - -# Version 0.3 - -- simpler API for label handling - (the idea of labels as De-Bruijn indices is abandoned) - -# Version 0.2 - -- simpler API for immediate values: `Add ax, 1` -- redesigned addressing API: `Add ax (addr $ rax + 4*rdi + 3)` -- support near jumps -- support automatic decision between short and near jumps (conditional jumps too) -- support more registers and instructions: - - cmovCC instructions - - SSE registers: xmm0 - xmm7 - - SSE instructions: movd, movq, movdqa, paddb, paddw, paddd, paddq, psubb, psubw, psubd, psubq, pxor - - SSE instructions (partial support): psllw, pslld, psllq, pslldq, psrlw, psrld, psrlq, psrldq, psraw, psraq -- bugfixes - - throw an error if an immediate value does not fit - - fix show instance of shift instructions - -# Version 0.1.4 - -- OS X operating system support (Balázs Kőműves) -- better show for db -- bugfix - - save flags in traceReg - -# Version 0.1.3 - -- jmpq instruction support (George Stelle) -- support near conditional jumps -- support automatic decision between short and near conditional jumps for backward references -- support alternative condition names -- make possible to use labels as relative immediate values (not used yet) -- bugfix - - fail if a short jump is out of range - -# Version 0.1.2 - -- Windows operating system support (Balázs Kőműves) -- GHC 7.10 support -- TODO.md file added - -# Version 0.1.1.1 - -- change dependencies to reflect that the package compiles only with ghc 8 -- add 'tested-with: GHC == 8.0.1' to the cabal file - -# Version 0.1.1 - -- examples moved into the library -- more Haddock comments -- add cabal test suit -- bugfixes - - fix code generation for alignments - - smaller code is generated now for 'add rax, 100' and similar instrcutions - +# Version 0.4.5 + +- fix build with newer base + +# Version 0.4.4 + +- export some useful data types + +# Version 0.4.3 + +- fixes to compile on GHC 7.10 and on Windows + +# Version 0.4.2 + +- fix OSX compilation + +# Version 0.4.1 + +- fix a bug which caused segmentation fault if a compiled function was called multiple times + +# Version 0.4 + +- rewrite FFI +- support calling labels +- speed up long code alignments +- bugfix: auto size calculation fix + +# Version 0.3.1 + +- use multi-byte nop operations for padding +- `preBuild` operation (may speed up code generation) +- branch-predicition friendlier if-then-else +- not-condition pattern: `N` + usage example: `j (N E)` which is the same as `j NE` +- bugfix: fix if-then-else condition + +# Version 0.3 + +- simpler API for label handling + (the idea of labels as De-Bruijn indices is abandoned) + +# Version 0.2 + +- simpler API for immediate values: `Add ax, 1` +- redesigned addressing API: `Add ax (addr $ rax + 4*rdi + 3)` +- support near jumps +- support automatic decision between short and near jumps (conditional jumps too) +- support more registers and instructions: + - cmovCC instructions + - SSE registers: xmm0 - xmm7 + - SSE instructions: movd, movq, movdqa, paddb, paddw, paddd, paddq, psubb, psubw, psubd, psubq, pxor + - SSE instructions (partial support): psllw, pslld, psllq, pslldq, psrlw, psrld, psrlq, psrldq, psraw, psraq +- bugfixes + - throw an error if an immediate value does not fit + - fix show instance of shift instructions + +# Version 0.1.4 + +- OS X operating system support (Balázs Kőműves) +- better show for db +- bugfix + - save flags in traceReg + +# Version 0.1.3 + +- jmpq instruction support (George Stelle) +- support near conditional jumps +- support automatic decision between short and near conditional jumps for backward references +- support alternative condition names +- make possible to use labels as relative immediate values (not used yet) +- bugfix + - fail if a short jump is out of range + +# Version 0.1.2 + +- Windows operating system support (Balázs Kőműves) +- GHC 7.10 support +- TODO.md file added + +# Version 0.1.1.1 + +- change dependencies to reflect that the package compiles only with ghc 8 +- add 'tested-with: GHC == 8.0.1' to the cabal file + +# Version 0.1.1 + +- examples moved into the library +- more Haddock comments +- add cabal test suit +- bugfixes + - fix code generation for alignments + - smaller code is generated now for 'add rax, 100' and similar instrcutions + diff --git a/CodeGen/X86/Tests.hs b/CodeGen/X86/Tests.hs --- a/CodeGen/X86/Tests.hs +++ b/CodeGen/X86/Tests.hs @@ -242,8 +242,8 @@ And_{} -> mk (.&.) Sub_{} -> mk (-) Xor_{} -> mk xor - Cmp_{} -> mk_ (-) (const a) - Test_{} -> mk_ (.&.) (const a) + Cmp_{} -> mk_ (-) const + Test_{} -> mk_ (.&.) const Mov_{} -> \a b -> ((c, False), b) Shl_{} -> \a b -> let i = fromIntegral (b .&. shiftMask) diff --git a/LICENSE b/LICENSE --- a/LICENSE +++ b/LICENSE @@ -1,30 +1,30 @@ -Copyright (c) 2016, Péter Diviánszky - -All rights reserved. - -Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: - - * Redistributions of source code must retain the above copyright - notice, this list of conditions and the following disclaimer. - - * Redistributions in binary form must reproduce the above - copyright notice, this list of conditions and the following - disclaimer in the documentation and/or other materials provided - with the distribution. - - * Neither the name of Péter Diviánszky nor the names of other - contributors may be used to endorse or promote products derived - from this software without specific prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS -"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT -LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR -A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT -OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, -SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT -LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, -DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY -THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT -(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. +Copyright (c) 2016, Péter Diviánszky + +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + + * Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. + + * Redistributions in binary form must reproduce the above + copyright notice, this list of conditions and the following + disclaimer in the documentation and/or other materials provided + with the distribution. + + * Neither the name of Péter Diviánszky nor the names of other + contributors may be used to endorse or promote products derived + from this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT +OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, +SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT +LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, +DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY +THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. diff --git a/README.md b/README.md --- a/README.md +++ b/README.md @@ -1,13 +1,13 @@ -# x86-64 - -The primary goal of x86-64 is to provide a lightweight assembler for machine generated 64 bit x86 assembly instructions. - -Features: - -- The size of operands are statically checked. For example, exchanging `rax` with `eax` raises a compile time error rather than a code-generation time error. As a consequence, code generation is faster because the sizes are statically known. -- Quickcheck tests: You can quickcheck your x86 processor! Please report failures, there is a higher chance that the error is in this library rather than in your processor. -- Immediate values are automatically converted to smaller size if possible. -- Automatic calculation of short and near distances - -The package is available on HackageDB: http://hackage.haskell.org/package/x86-64bit - +# x86-64 + +The primary goal of x86-64 is to provide a lightweight assembler for machine generated 64 bit x86 assembly instructions. + +Features: + +- The size of operands are statically checked. For example, exchanging `rax` with `eax` raises a compile time error rather than a code-generation time error. As a consequence, code generation is faster because the sizes are statically known. +- Quickcheck tests: You can quickcheck your x86 processor! Please report failures, there is a higher chance that the error is in this library rather than in your processor. +- Immediate values are automatically converted to smaller size if possible. +- Automatic calculation of short and near distances + +The package is available on HackageDB: http://hackage.haskell.org/package/x86-64bit + diff --git a/x86-64bit.cabal b/x86-64bit.cabal --- a/x86-64bit.cabal +++ b/x86-64bit.cabal @@ -1,5 +1,5 @@ name: x86-64bit -version: 0.4.6 +version: 0.4.6.1 homepage: https://github.com/divipp/x86-64 synopsis: Runtime code generation for x86 64 bit machine code description: The primary goal of x86-64bit is to provide a lightweight assembler for machine generated 64 bit x86 assembly instructions. See README.md for further details. @@ -53,7 +53,7 @@ ScopedTypeVariables build-depends: - base >= 4.8 && <4.14, + base >= 4.8 && <4.15, monads-tf >= 0.1 && <0.2, tardis >= 0.4 && <0.5, vector >= 0.11 && <0.13, @@ -75,11 +75,11 @@ CodeGen.X86.Tests build-depends: - base >= 4.8 && <4.14, + base >= 4.8 && <4.15, monads-tf >= 0.1 && <0.2, tardis >= 0.4 && <0.5, vector >= 0.11 && <0.13, - QuickCheck >= 2.8 && <2.14, + QuickCheck >= 2.8 && <2.15, deepseq if os(windows)