similar to: Not able to use PGO with LLVM+Clang built from source

I was trying to build llvm + clang with cmake cache PGO.cmake and ninja stage2. I used the 4.0.0 final tag vfrom svn. This seems to work for me. I added libcxx, libcxxabi, llld etc. And now I get the following cmake error. -- Performing Test LIBCXX_SUPPORTS_STD_EQ_CXX11_FLAG -- Performing Test LIBCXX_SUPPORTS_STD_EQ_CXX11_FLAG - Failed CMake Error at projects/libcxx/CMakeLists.txt:396

Hello Mentors, I did analyze assembly files generated for IPRA + PGO. (1) I observed that I did not considered the scope of the optimization so changing callee saved register set for non local function is bad because IPRA can not pass this information to other modules. (2) applying this change to indirect function also has no effect because for such case IPRA is currently not able to propagate

On Sat, Jul 9, 2016 at 12:18 AM, Mehdi Amini <mehdi.amini at apple.com> wrote: > > On Jul 8, 2016, at 11:41 AM, vivek pandya <vivekvpandya at gmail.com> wrote: > > > > On Fri, Jul 8, 2016 at 11:46 PM, Mehdi Amini <mehdi.amini at apple.com> > wrote: > >> >> On Jul 8, 2016, at 11:12 AM, vivek pandya <vivekvpandya at gmail.com> wrote:

Dear Community, Sorry for being late for weekly status report but I was traveling back to my college. Please consider this week's summary as follows: Implementation: ============ This week I tried to get experimental PGO driven IPRA work. The idea is to save all register in prolog and restore it in epilog for cold function so that IPRA can propagate some free register to upper region of

The code in X86TargetLowering::IsEligibleForTailCallOptimization() has this part: // The callee has to preserve all registers the caller needs to preserve. const X86RegisterInfo *TRI = Subtarget.getRegisterInfo(); const uint32_t *CallerPreserved = TRI->getCallPreservedMask(MF, CallerCC); if (!CCMatch) { const uint32_t *CalleePreserved = TRI->getCallPreservedMask(MF, CalleeCC);

----- Original Message ----- > From: "vivek pandya" <vivekvpandya at gmail.com> > To: "Mehdi Amini" <mehdi.amini at apple.com> > Cc: "llvm-dev" <llvm-dev at lists.llvm.org>, "Hal Finkel" <hfinkel at anl.gov>, "Quentin Colombet" <qcolombet at apple.com> > Sent: Thursday, July 28, 2016 2:59:02 PM >

I have been working on PGO driven IPRA and I want to measure if this help to reduce execution time. So as mentioned earlier the idea is to make cold function register usage free i.e saving and restoring all used register by such cold function so caller of that function will have more free registers. So here I am changing standard callee saved registers set to a set which will be decided

On Fri, Jul 29, 2016 at 9:01 AM, Hal Finkel <hfinkel at anl.gov> wrote: > ----- Original Message ----- > > From: "vivek pandya" <vivekvpandya at gmail.com> > > To: "Mehdi Amini" <mehdi.amini at apple.com> > > Cc: "llvm-dev" <llvm-dev at lists.llvm.org>, "Hal Finkel" <hfinkel at anl.gov>, > "Quentin

----- Original Message ----- > From: "vivek pandya" <vivekvpandya at gmail.com> > To: "Hal Finkel" <hfinkel at anl.gov> > Cc: "llvm-dev" <llvm-dev at lists.llvm.org>, "Quentin Colombet" > <qcolombet at apple.com>, "Mehdi Amini" <mehdi.amini at apple.com> > Sent: Friday, July 29, 2016 5:02:44 AM >

Hello , I have build llvm from release_38 branch ( only llvm and clang ) and install it. My DYLD_LIBRARY_PATH points to installation-directory/lib. I am compiling example files for Kaleidoscope with following command : clang++ -g toy.cpp -std=c++11 `llvm-config --cxxflags --ldflags --system-libs --libs core mcjit native` -O3 -o toy but it fails with following error: Undefined symbols for

Hello LLVM Developers, Please feel free to send any ideas that you can think to improve current IPRA. I will work on it and if possible I will implement that. Please consider summary of work done during this week. Implementation: ============ The reviews requests has been updated to reflect the reviews. Testing: ===== To get more benefit from IPRA I experimented it with LTO and results

> On Jul 28, 2016, at 12:59 PM, vivek pandya <vivekvpandya at gmail.com> wrote: > > I have been working on PGO driven IPRA and I want to measure if this help to reduce execution time. So as mentioned earlier the idea is to make cold function register usage free i.e saving and restoring all used register by such cold function so caller of that function will have more free registers.

> On May 24, 2016, at 10:08 PM, vivek pandya <vivekvpandya at gmail.com> wrote: > > > > On Wed, May 25, 2016 at 10:08 AM, Mehdi Amini <mehdi.amini at apple.com <mailto:mehdi.amini at apple.com>> wrote: > >> On May 24, 2016, at 9:17 PM, vivek pandya <vivekvpandya at gmail.com <mailto:vivekvpandya at gmail.com>> wrote: >> >> Dear

Sent from my iPhone > On May 24, 2016, at 11:04 PM, vivek pandya <vivekvpandya at gmail.com> wrote: > > > >> On Wed, May 25, 2016 at 10:46 AM, Mehdi Amini <mehdi.amini at apple.com> wrote: >> >>> On May 24, 2016, at 10:08 PM, vivek pandya <vivekvpandya at gmail.com> wrote: >>> >>> >>> >>> On Wed, May 25,

> On May 24, 2016, at 9:17 PM, vivek pandya <vivekvpandya at gmail.com> wrote: > > Dear Mentors, > > Please help me to understand our plan to implement Interprocedural Register allocator by propogating register usage info. While writing this mail I am considering all previous discussion over llvm-dev and IRC. > > 1) A MachineFunction pass to be executed POST-RA to

On Saturday, March 4, 2017, Ryan Taylor <ryta1203 at gmail.com> wrote: > Why you can't still expand it through MUL with a Custom lowering? Or am I > missing something? > > Yes we can but problem occurs when we know that it is shift with constant value than if we return ISD::MUL with constant imm operand than LLVM will convert it to SHL again because the constant will be

> On May 27, 2016, at 6:55 PM, vivek pandya via llvm-dev <llvm-dev at lists.llvm.org> wrote: > > > > On Sat, May 28, 2016 at 12:23 AM, vivek pandya <vivekvpandya at gmail.com <mailto:vivekvpandya at gmail.com>> wrote: > > > On Sat, May 28, 2016 at 12:21 AM, Mehdi Amini <mehdi.amini at apple.com <mailto:mehdi.amini at apple.com>> wrote:

On Fri, Jul 8, 2016 at 11:46 PM, Mehdi Amini <mehdi.amini at apple.com> wrote: > > On Jul 8, 2016, at 11:12 AM, vivek pandya <vivekvpandya at gmail.com> wrote: > > Hello LLVM Developers, > > I have a thought to improve IPRA and I would like summaries discussion on > IRC regarding that so we can develop an idea out of that if it really helps. > > So idea is

Mehdi, I’m compiling embedded applications which are small enough to do whole-program-compilation. There’s no advantage in breaking them up into separate compilation pieces and linking them, even though in source form they are composed of a couple of separate source files. So for me the compilation unit is always the entire program (and includes main()) Except for some

static void setXXX(MachineInstr &MI, ...) { for (MachineOperand &MO : MI.operands()) { if (MO.isRegMask()) MO.setRegMask(...); } } > On May 27, 2016, at 7:02 PM, vivek pandya via llvm-dev <llvm-dev at lists.llvm.org> wrote: > > > > On Sat, May 28, 2016 at 7:29 AM, Matthias Braun <matze at braunis.de <mailto:matze at braunis.de>> wrote: >