similar to: [LLVMdev] Writing a backend for the ZPU

Whoops! Forgot to CC: the list! ----- Forwarded Message ----- > From:Samuel Crow <samuraileumas at yahoo.com> > To:Øyvind Harboe <oyvind.harboe at zylin.com> > Cc: > Sent:Thursday, November 18, 2010 7:22:39 PM > Subject:Re: [LLVMdev] Writing a backend for the ZPU > > > Hello, Some > thoughts/problems: - In GCC I created > registers which were

> We mean ZPU looks like x86 fp stack. We handle it by register > allocating it like a normal register based cpu than add a x87 > stackifer pass to convert it to stack operations. Thanks! > You do not have to. Take a look at PPC and ARM branch fix up passes. > llvm backends compute exact length of instructions so it can determine > when it is necessary to convert a conditional

On Fri, 20 Jun 2008, [ISO-8859-1] ?yvind Harboe wrote: >> The ZPU has two instructions that I'd also like to use. These instructions >> can push a value from deeper down on the stack and also pop a value >> from the stack and store them deeper down on the stack. > > Sounds like the Intel X87 floating point stack, which we support. GCC does as well. Supporting floating

Hi all, Zylin has implemented the world smallest 32 bit CPU with a GCC backend. (I shall stand corrected if anyone claims & proves otherwise :-) Implementing a GCC backend for a zero operand/stack based architecture proved pretty tricky, but I'm quite pleased with the resulting code. I did make alterations to the architecture to make it fit GCC without sacrificing CPU size. I have been

> On Jun 19, 2008, at 2:30 PM, ?yvind Harboe wrote: >> My llvm.org knowledge is ... shallow ... but I'm hoping that >> someone would find the time & pity to answer my questions: >> >> Q: Is a stack based / zero operand CPU and llvm a good match? (GCC >> wasn't) > > I'm not really sure, I'm not too familiar with these architectures. >

Are JIT DLLs supported? The idea is to use llvm to put performance sensitive code into a DLL that a Windows app can then use. This would build the performance sensitive code on the target machine making it possible to exploit CPU specific x86 vector instructions. The code that calls fn's in the DLL should, ideally, be unaware that a llvm JIT is being used. -- Øyvind Harboe

> ?yvind Harboe wrote: >> Are JIT DLLs supported? >> >> The idea is to use llvm to put performance sensitive code into >> a DLL that a Windows app can then use. >> >> This would build the performance sensitive code on the target >> machine making it possible to exploit CPU specific x86 >> vector instructions. >> >> The code that calls

> For the JIT and a couple of optimizers on x86 (suitably stripped etc), > we see it taking ~2M uncompressed. Additionally, as you say, this is > very compressible for an install image. 2mByte for PC apps is "nothing" these days. The app I have in mind is 50-200mByte depending on install options. To be able to write llvm DLLs and ship intermediate code would definitely be

I recently use gbm for a binary classification problem. As expected, it gets very good results, based on Area under ROC with 7-fold cross validation. However, the application (malware detection) is cost-sensitive, getting a FP (classify a clean sample as a dirty one) is much worse than getting a FN (miss a dirty sample). I would like to tune the gbm model biased to very low FP rate. For this

An easier way would be to use a coverage tool like gcov to see what's actually *used* when the app is run normally. Then you can ask the question, what percentage of all lines of code are dead? A static analysis will not be able to see through things like virtual method calls. Reid On Sun, Jul 3, 2011 at 3:07 PM, Øyvind Harboe <oyvind.harboe at zylin.com> wrote: > Can llvm

I've tried llvm 2.8 and 2.9, but they both have problems with #include <iosfwd>. Any pointers to what this might be? /usr/include/c++/4.5/iosfwd:39:10: fatal error: 'bits/c++config.h' file not found #include <bits/c++config.h> -- Øyvind Harboe - Can Zylin Consulting help on your project? US toll free 1-866-980-3434 / International +47 51 87 40 27 http://www.zylin.com/

Can llvm generate warnings for unused C++ code using global analysis? If I could use llvm to figure out what code I can delete in a 20 year old app with millions of lines of code, this alone would defend spending time on making the app build with llvm, even if we don't actually run the code generated... -- Øyvind Harboe - Can Zylin Consulting help on your project? US toll free

On Thu, Jun 30, 2011 at 4:00 PM, Øyvind Harboe <oyvind.harboe at zylin.com> wrote: > On Thu, Jun 30, 2011 at 3:43 PM, Duncan Sands <baldrick at free.fr> wrote: >> On 30/06/11 15:13, Øyvind Harboe wrote: >>>> Sorry, I meant from which version of clang? I'm using Ubuntu 10 and >>>> I was wondering if clang/llvm 2.8 would do it. >>>

On Thu, Jun 30, 2011 at 3:43 PM, Duncan Sands <baldrick at free.fr> wrote: > On 30/06/11 15:13, Øyvind Harboe wrote: >>> Sorry, I meant from which version of clang? I'm using Ubuntu 10 and >>> I was wondering if clang/llvm 2.8 would do it. >> >> Perhaps I better try the latest, got an error when I tried 2.8: > > C++ support in clang 2.8 was poor.

>> Starting with which version? > > I think GCC last changed the C++ ABI with version 3.4. Sorry, I meant from which version of clang? I'm using Ubuntu 10 and I was wondering if clang/llvm 2.8 would do it. clang --version clang version 2.8 (branches/release_28) -- Øyvind Harboe - Can Zylin Consulting help on your project? US toll free 1-866-980-3434 / International +47 51 87

> Sorry, I meant from which version of clang? I'm using Ubuntu 10 and > I was wondering if clang/llvm 2.8 would do it. Perhaps I better try the latest, got an error when I tried 2.8: In file included from testsuite/test.cpp:8: In file included from ./testsuite/precompile.h:4: ./util/precompile.h:17:12: fatal error: 'iosfwd' file not found -- Øyvind Harboe - Can Zylin

Hi, I'm trying to find some written evidence to present to the powers that be that clang/llvm is ABI compatible with GCC 4.1 linux 64 bit. This would allow me to work with clang/llvm at my day-job and I would like nothing better! This is a pretty big app(millions of lines of code), so it would be interesting to see how llvm/clang behaves. Better warnings/errors and cutting down on build

If you have not done so, you have to configure a clang header to find your default libc++ header files. See #5 at http://clang.llvm.org/get_started.html if this is your issue. Garrison On Jun 30, 2011, at 10:33, Øyvind Harboe wrote: > I've tried llvm 2.8 and 2.9, but they both have problems with > #include <iosfwd>. > > Any pointers to what this might be? > >

On Thu, Jun 30, 2011 at 2:03 PM, Duncan Sands <baldrick at free.fr> wrote: > Hi Øyvind, > >> I'm trying to find some written evidence to present to the powers >> that be that clang/llvm is ABI compatible with GCC 4.1 linux >> 64 bit. > > both clang and gcc-4.1 are supposed to conform to the platform ABI (if they > don't then that is a bug), and thus

Hi, I'm trying to find some written evidence to present to the powers that be that clang/llvm is ABI compatible with GCC 4.1 linux 64 bit. This would allow me to work with clang/llvm at my day-job and I would like nothing better! This is a pretty big app(millions of lines of code), so it would be interesting to see how llvm/clang behaves. Better warnings/errors and cutting down on build