mirror of
https://github.com/xenia-project/xenia.git
synced 2026-04-04 14:17:24 +00:00
Adding CONTRIBUTING.md so that github shows it on issues.
This commit is contained in:
Ben Vanik
parent
45816d801b
commit
a79ef87889
10 changed files with 120 additions and 64 deletions
|
|
@ -1,45 +0,0 @@
|
|||
# Building
|
||||
|
||||
You must have a 64-bit machine for building and running the project. Always
|
||||
run your system updater before building and make sure you have the latest
|
||||
video drivers for your card.
|
||||
|
||||
## Setup
|
||||
|
||||
### Windows
|
||||
|
||||
* Windows 8 or 8.1
|
||||
* Visual Studio 2015
|
||||
* [Python 2.7](http://www.python.org/download/releases/2.7.6/)
|
||||
* If you are on Windows 8, you will also need the [Windows 8.1 SDK](http://msdn.microsoft.com/en-us/windows/desktop/bg162891)
|
||||
|
||||
Ensure Python is in your PATH (`C:\Python27\`).
|
||||
|
||||
I recommend using [Cmder](http://bliker.github.io/cmder/) for git and command
|
||||
line usage.
|
||||
|
||||
#### Debugging
|
||||
|
||||
VS behaves oddly with the debug paths. Open the xenia project properties
|
||||
and set the 'Command' to `$(SolutionDir)$(TargetPath)` and the
|
||||
'Working Directory' to `$(SolutionDir)..\..`. You can specify flags and
|
||||
the file to run in the 'Command Arguments' field (or use `--flagfile=flags.txt`).
|
||||
|
||||
To redirect output, use the following command flags:
|
||||
`--flagfile=$(SolutionDir)scratch\flags.txt 2>&1 1>$(SolutionDir)scratch\stdout.txt`
|
||||
|
||||
### Linux
|
||||
|
||||
Linux support is extremely experimental and incomplete.
|
||||
|
||||
Only tested with GCC 4.9 on Ubuntu 14. [CodeLite](http://codelite.org) is the
|
||||
IDE of choice and `xb premake` will spit out files for that. Make also works via
|
||||
`xb build`.
|
||||
|
||||
Currently building requires that CC == CXX == g++. If you know a way around this
|
||||
(to force .c files to be built with g++) let me know.
|
||||
|
||||
## Running
|
||||
|
||||
To make life easier you can use `--flagfile=myflags.txt` to specify all
|
||||
arguments, including using `--target=my.xex` to pick an executable.
|
||||
|
|
@ -1,9 +1,5 @@
|
|||
# CPU Documentation
|
||||
|
||||
## Alloy
|
||||
|
||||
TODO
|
||||
|
||||
## Memory Management
|
||||
|
||||
TODO
|
||||
|
|
|
|||
|
|
@ -16,15 +16,6 @@ will relocate the x64 when placing it in memory, which will be at a different
|
|||
location each time. Instead it would be nice to have the xbyak `calcJmpAddress`
|
||||
that performs the relocations use the address of our choosing.
|
||||
|
||||
### Stack Walking
|
||||
|
||||
Currently the Windows/VC++ dbghelp stack walking is relied on, however this is
|
||||
not portable, is slow, and cannot resolve JIT'ed symbols properly. Having our
|
||||
own stack walking code that could fall back to dbghelp (via some pluggable
|
||||
system) for host symbols would let us quickly get stacks through host and guest
|
||||
code and make things like sampling profilers, kernel callstack tracing, and
|
||||
other features possible.
|
||||
|
||||
### Sampling Profiler
|
||||
|
||||
Once we have stack walking it'd be nice to take something like
|
||||
|
|
|
|||
|
|
@ -1,37 +0,0 @@
|
|||
# Debugger
|
||||
|
||||
## Protocol
|
||||
|
||||
Framed messages:
|
||||
|
||||
```
|
||||
[4b type] (0=request, 1=notification, etc)
|
||||
[4b content source id]
|
||||
[4b request/response id]
|
||||
[4b size]
|
||||
[payload]
|
||||
```
|
||||
|
||||
## Content Sources
|
||||
|
||||
### `xe::dbg::sources::MemorySource`
|
||||
|
||||
[ ] Paged view into memory
|
||||
[ ] Search operations
|
||||
[ ] Live streaming updates
|
||||
[ ] Writes
|
||||
|
||||
### `xe::dbg::sources::ProcessorSource`
|
||||
|
||||
[ ] Thread list
|
||||
[ ] State read/write (per thread)
|
||||
[ ] Modules
|
||||
[ ] Statistics
|
||||
[ ] Basic control (pause/resume)
|
||||
[ ] Breakpoints/checkpoints/traces
|
||||
[ ] Trace stream
|
||||
|
||||
### `xe::dbg::sources::ModuleSource`
|
||||
|
||||
[ ] Paged view into all symbols
|
||||
[ ] Get function contents (data, disasm, llvm, x86, etc)
|
||||
|
|
@ -1,63 +0,0 @@
|
|||
# Style Guide
|
||||
|
||||
The style guide can be summed up as 'clang-format with the Google style set'.
|
||||
In addition, the [Google Style Guide](http://google-styleguide.googlecode.com/svn/trunk/cppguide.xml)
|
||||
is followed and cpplint is the source of truth.
|
||||
|
||||
Base rules:
|
||||
|
||||
* 80 column line length max
|
||||
* LF (Unix-style) line endings
|
||||
* 2-space soft tabs, no TABs!
|
||||
* [Google Style Guide](http://google-styleguide.googlecode.com/svn/trunk/cppguide.xml) for naming/casing/etc
|
||||
|
||||
Code that really breaks from the formatting rules will not be accepted, as then
|
||||
no one else can use clang-format on the code without also touching all your
|
||||
lines.
|
||||
|
||||
The buildbot runs `xb lint --all` on the master branch, and will run
|
||||
`xb lint --origin` on pull requests. Run `xb format` before you commit each
|
||||
local change so that you are consistently clean, otherwise you may have to
|
||||
rebase. If you forget, run `xb format --origin` and rebase your changes (so you
|
||||
don't end up with 5 changes and then a 6th 'whoops' one - that's nasty).
|
||||
|
||||
The buildbot is running LLVM 3.6.1. If you are noticing style differences
|
||||
between your local lint/format and the buildbot, ensure you are running that
|
||||
version.
|
||||
|
||||
## Tools
|
||||
|
||||
### clang-format
|
||||
|
||||
clang-format with the Google style is used to format all files. I recommend
|
||||
installing/wiring it up to your editor of choice so that you don't even have to
|
||||
think about tabs and wrapping and such.
|
||||
|
||||
#### Command Line
|
||||
|
||||
To use the `xb format` auto-formatter, you need to have a `clang-format` on your
|
||||
PATH. If you're on Windows you can do this by installing an LLVM binary package
|
||||
from [the LLVM downloads page](http://llvm.org/releases/download.html). If you
|
||||
install it to the default location the `xb format` command will find it
|
||||
automatically even if you don't choose to put all of LLVM onto your PATH.
|
||||
|
||||
#### Visual Studio
|
||||
|
||||
Grab the official [experimental Visual Studio plugin](http://llvm.org/builds/).
|
||||
To switch to the Google style go Tools -> Options -> LLVM/Clang -> ClangFormat
|
||||
and set Style to Google. Then use ctrl-r/ctrl-f to trigger the formatting.
|
||||
Unfortunately it only does the cursor by default, so you'll have to select the
|
||||
whole doc and invoke it to get it all done.
|
||||
|
||||
If you have a better option, let me know!
|
||||
|
||||
#### Xcode
|
||||
|
||||
Install [Alcatraz](http://alcatraz.io/) to get the [ClangFormat](https://github.com/travisjeffery/ClangFormat-Xcode)
|
||||
package. Set it to use the Google style and format on save. Never think about
|
||||
tabs or linefeeds or whatever again.
|
||||
|
||||
### cpplint
|
||||
|
||||
TODO(benvanik): write a cool script to do this/editor plugins.
|
||||
In the future, the linter will run as a git commit hook and on travis.
|
||||
Loading…
Add table
Add a link
Reference in a new issue