search for: ghoang84

...Giang On Fri, Oct 15, 2010 at 5:41 PM, Andrew Lenharth <andrewl at lenharth.org>wrote: > llvm-gcc doesn't not compile *on* alpha (128bit fp and int issues). I > haven't tried it as a cross compiler. > > Andrew > > On Fri, Oct 15, 2010 at 5:19 PM, Giang Hoang <ghoang84 at gmail.com> wrote: > > Hi, > > > > I wonder if anyone has been able to successfully build llvm-gcc as an > Alpha > > cross compiler? > > > > I have tried many different combinations of flags and gcc compiler, but > have > > not been able to build...

...make[1]: Leaving directory `/home/ghoang/llvm-test/llvm-gcc-2.8-build' make: *** [all] Error 2 Giang On Sun, Jan 15, 2012 at 8:06 PM, Liu <proljc at gmail.com> wrote: > try --disable-libmudflap when you config llvm-gcc FE:) > > On Mon, Jan 16, 2012 at 9:44 AM, Giang Hoang <ghoang84 at gmail.com> wrote: > > Hi, > > > > I am trying to build the front end llvm-gcc 2.8 on 32-bit Ubuntu 10.04, > > using gcc 4.2.4. I was able to configure and build llvm-2.8 in the > > directory llvm-2.8-build. Here is the command I used to configure > llvm-gcc...

Hi Giang, Given that the community deprecated the Alpha backend, I'm doubtful anyone would be able to point you in the right direction. Have you iteratively tried the difference versions of LLVM (i.e., 2.9, 2.8, 2.7 on down the line)? Chad On Feb 3, 2012, at 12:34 PM, Giang Hoang <ghoang84 at gmail.com> wrote: > Hi, > > For my work, I want to use LLVM to compile SPEC 2k for Alpha. Since Alpha support has been dropped, I tried using version 2.8, but it is quite buggy, probably because the Alpha backend has not been maintained. I was wondering if there is an earlier ve...

Hi, For my work, I want to use LLVM to compile SPEC 2k for Alpha. Since Alpha support has been dropped, I tried using version 2.8, but it is quite buggy, probably because the Alpha backend has not been maintained. I was wondering if there is an earlier version where the Alpha backend is stable enough to compile SPEC 2k? For my purpose, I do not need the most advanced optimizations, so an

llvm-gcc doesn't not compile *on* alpha (128bit fp and int issues). I haven't tried it as a cross compiler. Andrew On Fri, Oct 15, 2010 at 5:19 PM, Giang Hoang <ghoang84 at gmail.com> wrote: > Hi, > > I wonder if anyone has been able to successfully build llvm-gcc as an Alpha > cross compiler? > > I have tried many different combinations of flags and gcc compiler, but have > not been able to build successfully.  Currently, this is the comman...

Hi, I wonder if anyone has been able to successfully build llvm-gcc as an Alpha cross compiler? I have tried many different combinations of flags and gcc compiler, but have not been able to build successfully. Currently, this is the command that I used to build on Ubuntu 10.10: ../llvm-gcc-4.2-2.8.source/configure --enable-llvm=/home/ghoang/research/llvm/llvm-objects --enable-languages=c,c++

try --disable-libmudflap when you config llvm-gcc FE:) On Mon, Jan 16, 2012 at 9:44 AM, Giang Hoang <ghoang84 at gmail.com> wrote: > Hi, > > I am trying to build the front end llvm-gcc 2.8 on 32-bit Ubuntu 10.04, > using gcc 4.2.4.  I was able to configure and build llvm-2.8 in the > directory llvm-2.8-build.  Here is the command I used to configure llvm-gcc > build: > > ../llvm-...

Hi, I am trying to build the front end llvm-gcc 2.8 on 32-bit Ubuntu 10.04, using gcc 4.2.4. I was able to configure and build llvm-2.8 in the directory llvm-2.8-build. Here is the command I used to configure llvm-gcc build: ../llvm-gcc-4.2-2.8.source/configure --enable-checking --disable-bootstrap --disable-multilib --enable-llvm=/home/ghoang/llvm-test/llvm-2.8-build/ --enable-languages=c

Hi, I encountered an error while trying to use the indirectbr instruction with Alpha backend (current build). Here's part of the code sequence that I tried to compile: bb1: %1 = load i32* %i, align 4 %2 = add nsw i32 %1, 1 store i32 %2, i32* %i, align 4 indirectbr i8* blockaddress(@main, %bb1), [ label %bb1 ] br label %return This compiles correctly when I use the X86 or PPC

Hi, For my research, I need to reshape the current ARM backend to support armv2a. Zero-extend half word load (ldrh) is not supported by armv2a, so I need to make the code generation to not generate ldrh instructions. I want to replace all those instances with a 32-bit load (ldr) and then and the result with 0xffff to mask out the upper bits. These are the modifications that I have made to

Hi, I want to add an extra instruction to the Alpha ISA. This is not a new operation; it is just an addition with a lower latency, let's call it fast-add. Could you give me advice on what's an easy and clean approach to this? I am thinking of adding an extra instruction to the LLVM IR to represent this fast-add operation (I want to make the decision of using either the normal addition