similar to: [LLVMdev] RFC: Tailcall Improvement

For my target, I handle incoming memory arguments by creating a store to memory (in LowerCall, [1]), then creating a fixed object on the stack and loading from it (in LowerFormalArguments[2]). This approach was based on MSP430. I now have the problem that the resulting loads in my output assembly are done assuming that the call stack looks something like: ------ MemArg ------ MemArg ------

Hello Dale, this is an updated version of the tail call optimization patch for powerpc. could you have a look at it? i added code to support ppc64 (untested, will try to get access to ppc64 on a friend's machine). incorporated evan's formatting suggestions. ;) will run another round of testing (llvm-test) on my powerpc g4/800 when i get the okay to commit. testing on this machine takes

On Apr 16, 2008, at 10:07 AM, Arnold Schwaighofer wrote: > Hello Dale, > > this is an updated version of the tail call optimization patch for > powerpc. could you have a look at it? > > i added code to support ppc64 (untested, will try to get access to > ppc64 on a friend's machine). > incorporated evan's formatting suggestions. ;) > > will run another round

On Tue, Apr 22, 2008 at 12:30 AM, Evan Cheng <evan.cheng at apple.com> wrote: > More nitpicks: > ... > No need for else here. :-) Done > SPDiff = (int)CallerMinReservedArea - (int)ParamSize; > > Just change last statement to > int SPDiff = (int)... Done > > +bool > +PPCTargetLowering::IsEligibleForTailCallOptimization(SDOperand Call, > +

Comments: CheckDAGForTailCallsAndFixThem - 1. for (SelectionDAG::allnodes_iterator BE = DAG.allnodes_begin(), + BI = prior(DAG.allnodes_end()); BI != BE; BI--) { Please use pre-decrement instead of post-decrement. 2. The function is slower than it should be. You are scanning all the nodes in the DAG twice. You should just examine DAG.getRoot() to make determine whether it's a

Hi all, I changed the code that checks whether a tail call is really eligible for optimization so that it performs the check/fix in SelectionDAGISel.cpp:BuildSelectionDAG() as suggest by Evan. Also eliminated an error that caused the remaining failing test cases in the test-suite. The results look very nice (on darwin x86, r42486). The same number (46) of failing test cases on patched

Sorry that also suffered from signature mismatch. So how about more like: typedef struct vtable_t { int (*pf4)(int a, int b, int c, int d); int (*pf5)(int a, int b, int c, int d, int e); } vtable_t; int f3(int (*pf5)(int a, int b, int c, int d, int e), int a, int b, int c, int d) { return pf5(a, b, c, d, d + d); } int f4(vtable_t *vt, int a, int b, int c, int d) { return vt->pf5(a, b,

tailcall optimization stop working in jit (lli) in revision 45527. i guess this is because the jit is tailjmping to the wrong function address. the following change would reenable tailcallopt in jit. But i am pretty sure that this is not the correct fix (since i don't really understand what is going on :). maybe evan can comment on this? regards arnold Index:

On Tue, 25 Sep 2007, Evan Cheng wrote: >> the stack adjustment only fastcc was not one of them. Now that fastcc >> can cause tail call optimization i had to change the convention from >> caller pops arguments to callee pops arguments in order to allow tail >> call optimization in a general way. > > Hmmm. Ok. So this is due to X86CallingConv.td changes? Unfortunately

typedef struct vtable_t { int (*pf4)(int a, int b, int c, int d); int (*pf5)(int a, int b, int c, int d, int e); } vtable_t; int f1(int (*pf4)(int a, int b, int c, int d)) { return pf4(1, 2, 3, 4); } int f2(vtable_t *vt) { return vt->pf4(1, 2, 3, 4); } gcc and clang will not tailcall these on arm with -O3. int f3(int (*pf5)(int a, int b, int c, int d, int e)) { return pf5(1, 2, 3,

Hi there, it's me again. attached is preliminary support for tail calls on ppc 32. (once i get access to a 64bit machine that part will follow). It has passed the llvm-test with the -tailcallopt flag enabled. (after half a day on an old g4/800 :) okay to submit? probably not. :) suggestions. regards arnold -------------- next part -------------- A non-text attachment was scrubbed... Name:

I don't know how to file a PR, but I have a patch (see below), that should work regardless of abi differences, since it relies on the compiler to do the though job. void X86CompilationCallback_SSE(void) { char * SAVEBUF= (char*) alloca(64+12); // alloca is 16byte aligned asm volatile ( "movl %%eax,(%0)\n" "movl %%edx,4(%0)\n" // Save EAX/EDX/ECX

Paolo Invernizzi wrote: > There was a similar problem some time ago, and was resolved with alloca. > I think it's a better solution to use the stack instead of the heap... I tend to agree, but the constructors won't get called if it's an object array -- anyway, this particular case there was no objects, just pointers and bools so alloca should be fine. I'll leave it to

Hello, Corrado > Before you can correctly invoke a function via the Procedure Linkage > Table (plt), the ABI mandates that ebx is pointing to the GOT (Global > Offset Table) (see http://www.greyhat.ch/lab/downloads/pic.html) This is known issue, just nobody realized, that we have bunch of non- PIC-aware assembler code. :) Fixing would be not so trivial though, mostly due to ABI

On Apr 22, 2008, at 4:58 AM, Arnold Schwaighofer wrote: > On Tue, Apr 22, 2008 at 12:30 AM, Evan Cheng <evan.cheng at apple.com> > wrote: >> More nitpicks: >> ... >> No need for else here. :-) > Done >> SPDiff = (int)CallerMinReservedArea - (int)ParamSize; >> >> Just change last statement to >> int SPDiff = (int)... > Done >>

Please open a bugzilla ticket and attach your testcase. It will allow us to debug and fix the problem. Thanks - Elena From: JinGu [mailto:jingu at codeplay.com] Sent: Saturday, September 16, 2017 00:38 To: Demikhovsky, Elena <elena.demikhovsky at intel.com>; daniel_l_sanders at apple.com <daniel_l_sanders at apple.com>; Jon Chesterfield <jonathanchesterfield at

> so I think we need to use non-extending load for element size less than 8bit on "DAGTypeLegalizer::SplitVecOp_EXTRACT_VECTOR_ELT" like this roughly. > if (N->getOperand(0).getValueType().getVectorElementType().getSizeInBits() < 8) { > return DAG.getLoad(N->getValueType(0), dl, Store, StackPtr, MachinePointerInfo()); > } else { > return

Hello. I found that the X86CompilationCallback_SSE wrapper for X86CompilationCallback2 is not setting up properly for the PIC invocation. Before you can correctly invoke a function via the Procedure Linkage Table (plt), the ABI mandates that ebx is pointing to the GOT (Global Offset Table) (see http://www.greyhat.ch/lab/downloads/pic.html) Dump of assembler code for function

> extends the elements to 8bit and stores them on stack. Store is responsible for zero-extend. This is the policy... - Elena -----Original Message----- From: jingu at codeplay.com [mailto:jingu at codeplay.com] Sent: Friday, September 15, 2017 17:45 To: llvm-dev at lists.llvm.org; Demikhovsky, Elena <elena.demikhovsky at intel.com>; daniel_l_sanders at apple.com Subject: Re: Question

Hi Evan, I incoporated the changes you request but to the following i have got a question: > Also, moving the option > there will allow us to change fastcc ABI (callee popping arguments) > only when this option is on. See Chris' email: I am not to sure on that. because that would make modules compiled with the flag on incompatible with ones compiled without the flag off as