similar to: [LLVMdev] Mark Hill's seminar

On Nov 19, 2004, at 9:25 PM, Mohd-Hanafiah Abdullah wrote: > Dear Dr Vikram: > > I need to know how to invoke llvmg++ so that it generates C code. Here's how you can do a single file (i.e., an unlinked executable): llvmg++ -c infile.cpp -o - | llc -march=c -o infile.cbe.c Reid might be able to tell you how to do it with a single command using the llvmc tool. In future, please

Yiping, If you are doing architectural synthesis, I do agree that you likely need to capture some operations in the instruction set, especially if you want to perform analyses and optimizations on those operations. Some specific comments: On Oct 20, 2004, at 6:04 PM, Yiping Fan wrote: > Vikram, >     I also agree with you. I understand that target-independent > representation is

>> 3. Is LLVM able to support advanced runtime features as continuations, >> garbage collection and resuming exception handling. Would there be >> anything in LLVM that would prevent these sorts of features? Is there >> anything in the LLVM runtime that is assumed (dynamic typing, etc). Or >> is LLVM as it's title sugests: Low-Level, so it won't get in your

Sorry for that last message, it was intended for llvm at cs instead of llvmdev at cs. --Vikram http://www.cs.uiuc.edu/~vadve > -----Original Message----- > From: Vikram S. Adve [mailto:vadve at cs.uiuc.edu] > Sent: Thursday, October 23, 2003 1:55 PM > To: 'Nate Fyie' > Cc: 'LLVM Developers List' > Subject: RE: Ticket #7559: FW: Bradfields/PCJF-10959 > >

LLVM uses structural equivalence to distinguish types, so there's no direct way to retain multiple types with identical contents. The C++ front-end generates a single struct for two such classes, and simply initializes their vtables differently to get the right function pointers. You would need some hack like you said to create distinct structs, which may be ok if you use it only for

On Nov 29, 2004, at 12:46 AM, Zhang Qiuyu wrote: > Hi, >   > Under directory, /llvm/test/programs/external/spec, It seems there are > several makefiles for SPEC. I am  just wondering could I get the SPEC > code or bytecode. Thanks in advance. Because of the SPEC licensing restrictions, we cannot distribute SPEC to non-UIUC users. I expect UCLA would have a site-wide license so

Great! Could you please add a /localhome directory to each of them? Since there are 2 disks per machine, perhaps the best thing to do would to make /localhome a real directory and make separate local directories on each disk, so that we can each do something like this: /localhome/vadve -> /mounts/seraph/disks/0/localhome/vadve /localhome/lattner ->

Rahul, Were you talking about tracing in the sense of "profiling" (which Anand did), or tracing of values as they are computed for debugging generated code (which I did)? For tracing of values for debugging, the answer is yes: we insert code in each BB and print out the values computed in that BB that are live at the end of the BB (so we don't print out temporaries computed and

I too am disappointed that not all talk slides are posted. To me it speaks poorly of the whole meeting since there were parallel sessions and not everyone could attend in person all the sessions they wanted to. The slides are, I thought, meant to be like the "proceedings" and If I recall some of the Apple slides are present from previous years but not this year. Vinod On Sat, Oct 17,

When you create a parser via ANTLR you specify the output language of the resulting recursive descent parser, at the moment there exists no C++ output template to my knowledge, thus you would have to generate the parser as C code for which a template exists. The runtime support should be there, at least partially but it won't use things like exceptions, nor will it have a very modular design

Our department outreach coordinator spotted this and forwarded it to me. It's a lot of speculation but I thought many of you would be interested. --Vikram http://www.cs.uiuc.edu/~vadve http://llvm.org Begin forwarded message: > From: "Jennifer C La Montagne" <jsandone at uiuc.edu> > Date: June 8, 2007 12:53:00 PM CDT > To: "Vikram Sadanand Adve"

On Apr 26, 2006, at 10:45 AM, Tom Tromey wrote: >>>>>> "Vikram" == Vikram Adve <vadve at cs.uiuc.edu> writes: > > Vikram> Either way, one issue that you will have to deal with is > preserving > Vikram> the behavior of Java exceptions (assuming you care about > that). LLVM > Vikram> does not preserve the order of potentially

On Sep 5, 2005, at 10:21 AM, Andrew Lenharth wrote: > On Mon, 2005-09-05 at 14:45 +0800, Tzu-Chien Chiu wrote: > >> why there is no general dependency analysis for the "machin code"? >> perhaps it's because the instruction scheduling is only implemented >> for sparcv9? >> > > Most backends use the SelectionDAG infastructure to do this kind of >

Reid, I have one substantial change to suggest to this. I think the distinction between module-level and cross-module optimization is artificial and unnecessary in LLVM because transparent link-time optimization makes intra-module and cross-module optimizations indistiguishable. It *is* important to distinguish between fast and slow optimizations. Because of this, I would suggest a

Dear Prof. Adve, Now I can use llvmgcc to compile a .c file into .bc file. But I still have trouble simply run the .bc code. Below is the sequence I got when I tried. I really don't know what's going on here. Please let me know how can I fix it. Thanks, xiaodong xli3|csil-suna48|~/cs426|[13]% llvmgcc scalarize.c -o scalarize xli3|csil-suna48|~/cs426|[14]% scalarize Cannot load value of

That sounds like a problem. Just so I understand, do you mean there isn't the run-time support etc. to write back ends for the C++ language, or that the compiler IR is also somehow insufficient to write a code generator? --Vikram Associate Professor, Computer Science University of Illinois at Urbana-Champaign http://llvm.org/~vadve On Jul 11, 2009, at 3:00 PM, Granville Barnett

On Oct 15, 2009, at 10:45 AM, Chris Lattner wrote: > Unfortunately, we found out at the last minute that Apple has a rule > which prevents its engineers from giving video taped talks or > distributing slides. We will hold onto the video and slide assets in > case this rule changes in the future. > > -Chris Chris, I hope you can pass my message along to the people at Apple

> On Jan 31, 2017, at 7:53 PM, Tian, Xinmin <xinmin.tian at intel.com> wrote: > > In this case, inliner is educated to add all local variables to the tag of enclosing parallel region, if there is enclosing parallel region. So isn’t it a good example that shows that your intrinsic *cannot* be opaque and that IR passes need to be modified to handle not only the IR-region

Hi Terry, Reid is exactly right about the benefits of static (link-time) optimization for whole programs. When all libraries are available, it could alllow significantly better optimization without run-time overhead. But it is increasingly common today for libraries to be dynamically linked. In these cases, you could get the benefits of LLVM optimization in two ways, *if* you compile the

>>>>> "Vikram" == Vikram Adve <vadve at cs.uiuc.edu> writes: Vikram> Either way, one issue that you will have to deal with is preserving Vikram> the behavior of Java exceptions (assuming you care about that). LLVM Vikram> does not preserve the order of potentially excepting instructions Vikram> (e.g., a divide or a load). This would have to be handled