Re: [Tinycc-devel] Sugggestion: Add lib type *.c fortcc_add_library() on windows ('-l' option parse)

2014-06-18 Thread lifenjoiner
Hi, I work on win platform. As tcc can't use gcc's library (*.a) file, tcc can use *.def file (ONLY) for dynamic link avoid making the library files. For many libraries small or simple, I think tcc also should can use *.c file (ONLY) as library. I don't know Windows very well

Re: [Tinycc-devel] Sugggestion: Add lib type *.c fortcc_add_library() on windows ('-l' option parse)

2014-06-18 Thread grischka
lifenjoi...@163.com wrote: With this patch, we won't need to make the libname.a library, and also can use -l option. It will search for the *.def file, if succeed it will do an extra search for the *.c file. And what if dllname.c is the source of dllname.dll as it much more likely happens in

Re: [Tinycc-devel] Sugggestion: Add lib type *.c fortcc_add_library() on windows ('-l' option parse)

2014-06-18 Thread lifenjoiner
With this patch, we won't need to make the libname.a library, and also can use -l option. It will search for the *.def file, if succeed it will do an extra search for the *.c file. And what if dllname.c is the source of dllname.dll as it much more likely happens in peoples' projects?

Re: [Tinycc-devel] Request code review, and merge

2014-06-18 Thread David Mertens
Hello jiang, You're request for a code review is a bit premature. You suggested adding uninitialized variable warnings to tcc about a week ago, and managed to solicit some useful feedback. The general consensus seemed to be opposed to adding these warnings. Your next step in this whole process

Re: [Tinycc-devel] State of the tcc project

2014-06-18 Thread David Mertens
And now lifenjoiner has jumped in, adding patches without discussion. Can we forcibly revert tcc's mob branch at http://repo.or.cz/ back to grishka's commit from 4-17 and cherry pick grishka's and Thomas' commits that were made thereafter? This will cause confusion for anybody who has pulled

[Tinycc-devel] 回复: Tinycc-devel Digest, Vol 134, Issue 11

2014-06-18 Thread jiang
I will soon as possible to write a patch, to tell you my ideas. jiang 30155...@qq.com编写: Thank you wrote back. two weeks before I am more busy, my English not very good, so there is no what to say. you to see the World Cup, and I do not bother. thanks David jiang

Re: [Tinycc-devel] Sugggestion: Add lib type *.c for tcc_add_library() on windows ('-l' option parse)

2014-06-18 Thread lifenjoiner
Hi there, Here is my patch for libtcc.c for the minimum change I can do the best maybe. Attachments also content a small program that can be used as a test. I use: tcc %CFLAGS% -vvv WinHash.c -ladvapi32 -lntdll -lgetopt -lshlwapi -DHashMain -lshell32 If you still think this feature is useless,

[Tinycc-devel] A few omissions from the docs

2014-06-18 Thread Wendell P
I found a few things missing from the documentation. 1. In addition to -vv, there is also -vvv, which behaves a little differently. 2. The -E switch to preprocess only. I'm not sure, but -xc also seems to be recognized. 3. In the file tcc-win32.txt, the syntax given for windres is windres -O

Re: [Tinycc-devel] State of the tcc project

2014-06-18 Thread jiang
jiangint main() http://repo.or.cz/w/tinycc.git/commitdiff/e5e7f488e22190f893152c0b2f73e9ba499c4169 - causes regression. Test case: struct { unsigned a:9, b:7, c:5; } s; s.a = s.b = s.c = 3; printf(%d / %d / %d\n, s.a, s.b, s.c);