similar to: [LLVMdev] Some questions about building an LLVM frontend

I've been working some more on my front-end, incorporating the advice from your previous email. I've had some success, in that I can generate a Module object, populated it with functions and instructions, and print out the resulting LLVM IR. But I'm still a little puzzled on how to connect the front end with LLVM's backend code generator, I can't seem to figure out how to

In the language I am working on, there are cases where the call flow graph can affect the type of a variable. A simple example is a "nullable" type (a pointer which is allowed to be NULL), which can be converted into a non-nullable type via an if-test. So for example if x is nullable, and I test x != NULL, then inside the conditional block the type of x is no longer nullable.

On May 24, 2008, at 02:53, Talin wrote: > In the language I am working on, there are cases where the call flow > graph can affect the type of a variable. A simple example is a > "nullable" type (a pointer which is allowed to be NULL), which can > be converted into a non-nullable type via an if-test. So for example > if x is nullable, and I test x != NULL, then

On May 23, 2008, at 11:53 PM, Talin wrote: > In the language I am working on, there are cases where the call flow > graph can affect the type of a variable. Ok. > A simple example is a > "nullable" type (a pointer which is allowed to be NULL), which can be > converted into a non-nullable type via an if-test. So for example if x > is nullable, and I test x != NULL,

Thanks (everyone) for the suggestions, they were helpful. What I decided to do was to essentially "wrap" LLVM's BasicBlock structure with my own block class that contains additional high-level information. Up to this point I was generating LLVM basic blocks on the fly as I walked the statement tree; Now I build my own BB graph, and then use that to generate the BB graph for

On Apr 7, 2011, at 12:14 PM, Talin wrote: > > OK I've been checking this out some more, and the DIEs don't look valid to me. Take a look at this output from dwarfdump -v: > > 0x000000c7: TAG_subprogram [3] > 0x000000c8: AT_name( .debug_str[0x000001bd] = "construct" ) > 0x000000cc: AT_MIPS_linkage_name( .debug_str[0x000001c7] =

Hi, after some experience with Rails on MySQL databases I gave it a try on one of our larger Oracle database. I was facing a problem when Rails tried to detect the columns for a model/table. The appropriate statement ran about 1 min which led to a timeout. Our Oracle guru told me to analyze the SYSTEM schema. After that hadn''t helped, he said the only remaining chance is to use a

On Tue, Nov 9, 2010 at 9:04 AM, Devang Patel <dpatel at apple.com> wrote: > > > On Nov 8, 2010, at 10:52 PM, Talin <viridia at gmail.com> wrote: > > On Mon, Nov 8, 2010 at 9:56 AM, Devang Patel < <dpatel at apple.com> > dpatel at apple.com> wrote: > >> >> On Nov 6, 2010, at 7:35 PM, Talin wrote: >> >> After to speaking to Devang

Hi, Long time ago (llvm-svn june 2008) I asked here about a way to output the assembly code of my JIT generated code to a string, so I could use it to read it on the screen. I came up with this solution: std::string Err; const llvm::TargetMachineRegistry::entry* _arch = llvm::TargetMachineRegistry::getClosestTargetForJIT(Err); std::string FeaturesStr; llvm::TargetMachine*

Patches 1 & 2 are fine, ACK. I have a separate comment about patch 4 coming up. In libguestfs we handle optional args by putting them in a separate list (so instead of { args; ret } we have we have { args; optargs; ret }). This gets translated into a language like Python in the natural way. However for C it's translated into a varargs list with a rather complex system of flags, see:

Dieter Menne pointed out that the (small) xml attachment didn't make it. Here is an in-line version (see end of message). Let's hope it works this time. I'm struggling with interpreting XML files created by ADODB as data.frames and I'm looking for advice. Note: This xlm contains a result set which comes from a rectangular data array. I've been trying to play with

On Sat, Oct 15, 2011 at 9:49 PM, Chandler Carruth <chandlerc at google.com>wrote: > On Sat, Oct 15, 2011 at 9:20 PM, Talin <viridia at gmail.com> wrote: > >> I recently updated my version of LLVM from revision 140108 to 142082, and >> several things broke, most of which were easily fixed. However, I'm now >> getting a "pure virtual method called"

So far I'm really liking the new type system -- I've been able to simplify my code generator in a number of areas. And the IR is now vastly more readable, both in the debugger (using dump()) and when printing modules via llvm-dis. It's a tremendous improvement. I do have a few comments / questions: -- I think I may be misunderstanding how named structs are supposed to be combined in

Interestingly, I also get a similar error in a different executable (my unittest): pure virtual method called terminate called without an active exception 0 tartc 0x00000001010a8265 PrintStackTrace(void*) + 53 1 tartc 0x00000001010a88cc SignalHandler(int) + 364 2 libSystem.B.dylib 0x00007fff831341ba _sigtramp + 26 3 libSystem.B.dylib 0x7261742e65637365 _sigtramp +

I have a question about regarding the use of ''has_one'' in this scenario: Schema: User ---- id first_name last_name UserMessage ---- user_id message_id Message ---- id Assuming business rules dictate that a Message can have at most ONE User (let''s assume a message can be created without a user associated to it), these are my questions: 1) How would I use has_one

Here's a more specific question about compiling native code: When I use llc to generate a .s file, and then I try to assemble it using 'as', I get tons of errors - it appears to be choking on lines like the following: movl L"_tart.core.Iterable<tart.core.String>:type"$non_lazy_ptr, %eax The error I get is:

I recently updated my version of LLVM from revision 140108 to 142082, and several things broke, most of which were easily fixed. However, I'm now getting a "pure virtual method called" exception (__cxa_pure_virtual) which I wasn't getting before. This is happening in the destructor of a statically-initialized object. (More precisely, it's blowing up in a BumpPtrAllocator,

On Sat, Oct 15, 2011 at 9:20 PM, Talin <viridia at gmail.com> wrote: > I recently updated my version of LLVM from revision 140108 to 142082, and > several things broke, most of which were easily fixed. However, I'm now > getting a "pure virtual method called" exception (__cxa_pure_virtual) which > I wasn't getting before. This is happening in the destructor of

On Mon, Oct 24, 2011 at 9:17 AM, Devang Patel <dpatel at apple.com> wrote: > > On Oct 23, 2011, at 12:03 AM, Talin wrote: > > Just a follow up on this - I am still having problems, I never did figure > out a solution. (I've been running with debug off for the last month so that > I could get work done.) > > Here's what I am seeing: I am definitely calling

On Mon, Aug 23, 2010 at 5:58 PM, Devang Patel <devang.patel at gmail.com>wrote: > > On Sun, Aug 22, 2010 at 12:50 PM, Talin <viridia at gmail.com> wrote: > >> I recently started getting this error when I try to debug my LLVM-compiled >> program in GDB: >> >> Dwarf Error: Cannot find DIE at 0x16769 referenced from DIE at 0x1713c >> [in module