search for: fpmathoperator

...< > llvm-dev at lists.llvm.org> wrote: > >> Hi, >> >> I need to add a flush-denormals-to-zero (FTZ) flag to FastMathFlags, >> but we've already used up the 7 bits available in >> Value::SubclassOptionalData (the "backing storage" for >> FPMathOperator::getFastMathFlags()). These are the possibilities I >> can think of: >> >> 1. Increase the size of FPMathOperator. This gives us some additional >> bits for FTZ and other fastmath flags we'd want to add in the future. >> Obvious downside is that it increases LLVM...

...eview feedback incorporated: * Changed single letter flags to short abbreviations ('S' ==> 'nsz') * Indentation fixes * Comments don't state function names -------------- next part -------------- A non-text attachment was scrubbed... Name: 0002-Fast-math-flags-added-to-FPMathOperator.patch Type: application/octet-stream Size: 4937 bytes Desc: not available URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20121115/2e3c515f/attachment.obj> -------------- next part -------------- -------------- next part -------------- A non-text attachment was scrubbed... Name: 0...

...;milseman at apple.com> wrote: > New patches with review feedback incorporated: > * Changed single letter flags to short abbreviations ('S' ==> 'nsz') > * Indentation fixes > * Comments don't state function names > > <0002-Fast-math-flags-added-to-FPMathOperator.patch> > <0003-Fast-math-interfaces-for-Instructions.patch> > <0004-Fast-math-flags-for-LLVM-IR-parsing-and-printing.patch> > <0005-Fast-math-flags-for-the-bitcode.patch> > <0006-Fast-math-test-case-for-bitcode-and-textual-reading-.patch> > <0007-Fast-ma...

Hi, I need to add a flush-denormals-to-zero (FTZ) flag to FastMathFlags, but we've already used up the 7 bits available in Value::SubclassOptionalData (the "backing storage" for FPMathOperator::getFastMathFlags()). These are the possibilities I can think of: 1. Increase the size of FPMathOperator. This gives us some additional bits for FTZ and other fastmath flags we'd want to add in the future. Obvious downside is that it increases LLVM's memory footprint. 2. Steal some low...

...v <llvm-dev at lists.llvm.org> wrote: >> >> Hi, >> >> I need to add a flush-denormals-to-zero (FTZ) flag to FastMathFlags, >> but we've already used up the 7 bits available in >> Value::SubclassOptionalData (the "backing storage" for >> FPMathOperator::getFastMathFlags()). These are the possibilities I >> can think of: >> >> 1. Increase the size of FPMathOperator. This gives us some additional >> bits for FTZ and other fastmath flags we'd want to add in the future. >> Obvious downside is that it increases LLVM...

...wrote: > >> New patches with review feedback incorporated: >> * Changed single letter flags to short abbreviations ('S' ==> 'nsz') >> * Indentation fixes >> * Comments don't state function names >> >> <0002-Fast-math-flags-added-to-FPMathOperator.patch> >> <0003-Fast-math-interfaces-for-Instructions.patch> >> <0004-Fast-math-flags-for-LLVM-IR-parsing-and-printing.patch> >> <0005-Fast-math-flags-for-the-bitcode.patch> >> <0006-Fast-math-test-case-for-bitcode-and-textual-reading-.patch> >&...

...hed are some patches for adding in an IR-level mechanism for representing fast-math flags, as discussed in my prior RFC. Patches include infrastructure, API support, textual and bitcode reader/writer support, example optimization, and test cases. >> >> <0002-Fast-math-flags-added-to-FPMathOperator.patch> >> <0003-Fast-math-interfaces-for-Instructions.patch> >> <0004-Fast-math-flags-for-LLVM-IR-parsing-and-printing.patch> >> <0005-Fast-math-flags-for-the-bitcode.patch> >> <0006-Fast-math-test-case-for-bitcode-and-textual-reading-.patch> >&...

...review feedback incorporated: >>>> * Changed single letter flags to short abbreviations ('S' ==> 'nsz') >>>> * Indentation fixes >>>> * Comments don't state function names >>>> >>>> <0002-Fast-math-flags-added-to-FPMathOperator.patch> >>>> <0003-Fast-math-interfaces-for-Instructions.patch> >>>> <0004-Fast-math-flags-for-LLVM-IR-parsing-and-printing.patch> >>>> <0005-Fast-math-flags-for-the-bitcode.patch> >>>> <0006-Fast-math-test-case-for-bitcode-and-...

...t; New patches with review feedback incorporated: >>> * Changed single letter flags to short abbreviations ('S' ==> 'nsz') >>> * Indentation fixes >>> * Comments don't state function names >>> >>> <0002-Fast-math-flags-added-to-FPMathOperator.patch> >>> <0003-Fast-math-interfaces-for-Instructions.patch> >>> <0004-Fast-math-flags-for-LLVM-IR-parsing-and-printing.patch> >>> <0005-Fast-math-flags-for-the-bitcode.patch> >>> <0006-Fast-math-test-case-for-bitcode-and-textual-reading-...

Thanks Duncan, GVN/EarlyCSE worked as suggested. Any pointers on how to optimize out: %accumulate = fadd double %6, 0.000000e+00 Using the 3.1 release and the C++ API, I can't figure out how FPMathOperator, TargetOptions, nor IRBuilder::SetDefaultFPMathTag work. I also don't see any floating point math transformation passes. I did see IRBuilder::SetFastMathFlags, do I need to update to 3.2 and use this call? -- View this message in context: http://llvm.1065342.n5.nabble.com/Which-transform-pa...

...epresenting fast-math flags, as discussed in my prior RFC. Patches include infrastructure, API support, textual and bitcode reader/writer support, example optimization, and test cases. -------------- next part -------------- A non-text attachment was scrubbed... Name: 0002-Fast-math-flags-added-to-FPMathOperator.patch Type: application/octet-stream Size: 4940 bytes Desc: not available URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20121115/4780f05d/attachment.obj> -------------- next part -------------- -------------- next part -------------- A non-text attachment was scrubbed... Name: 0...

...some patches for adding in an IR-level mechanism for representing fast-math flags, as discussed in my prior RFC. Patches include infrastructure, API support, textual and bitcode reader/writer support, example optimization, and test cases. >>> >>> <0002-Fast-math-flags-added-to-FPMathOperator.patch> >>> <0003-Fast-math-interfaces-for-Instructions.patch> >>> <0004-Fast-math-flags-for-LLVM-IR-parsing-and-printing.patch> >>> <0005-Fast-math-flags-for-the-bitcode.patch> >>> <0006-Fast-math-test-case-for-bitcode-and-textual-reading-...

...t; Attached are some patches for adding in an IR-level mechanism for representing fast-math flags, as discussed in my prior RFC. Patches include infrastructure, API support, textual and bitcode reader/writer support, example optimization, and test cases. > > <0002-Fast-math-flags-added-to-FPMathOperator.patch> > <0003-Fast-math-interfaces-for-Instructions.patch> > <0004-Fast-math-flags-for-LLVM-IR-parsing-and-printing.patch> > <0005-Fast-math-flags-for-the-bitcode.patch> > <0006-Fast-math-test-case-for-bitcode-and-textual-reading-.patch> > <0007-Fast-ma...

...om> wrote: > Another round of improved patches, and a patch for documentation changes to LangRef. > * Make comments more up to date > * Use 'arcp' instead of 'ar' > * Use logical || > > Still based off of r168110 > > <0001-Fast-math-flags-added-to-FPMathOperator.patch> > <0002-Fast-math-interfaces-for-Instructions.patch> > <0003-Fast-math-flags-for-LLVM-IR-parsing-and-printing.patch> > <0004-Fast-math-flags-for-the-bitcode.patch> > <0005-Fast-math-test-case-for-bitcode-and-textual-reading-.patch> > <0006-Fast-ma...

...y, !fpmath !0 ... !0 = metadata !{!metadata "fast"} This indicates that accuracy loss is acceptable (just how much is unspecified) for the sake of speed. Thanks to Chandler for pushing me to do it this way! It also creates a simple way of getting and setting this information: the FPMathOperator class: you can cast appropriate instructions to this class and then use the querying/mutating methods to get/set the accuracy, whether 2.5 or "fast". The attached clang patch uses this to set the openCL 2.5 ULPs accuracy rather than doing it by hand for example. In addition it changes I...

...!{!metadata "fast"} > > This indicates that accuracy loss is acceptable (just how much is > unspecified) > for the sake of speed. Thanks to Chandler for pushing me to do it this > way! > > It also creates a simple way of getting and setting this information: the > FPMathOperator class: you can cast appropriate instructions to this class > and then use the querying/mutating methods to get/set the accuracy, whether > 2.5 or "fast". The attached clang patch uses this to set the openCL 2.5 > ULPs > accuracy rather than doing it by hand for example. >...

...adata "fast"} > > This indicates that accuracy loss is acceptable (just how much is unspecified) > for the sake of speed. Thanks to Chandler for pushing me to do it this way! > > It also creates a simple way of getting and setting this information: the > FPMathOperator class: you can cast appropriate instructions to this class > and then use the querying/mutating methods to get/set the accuracy, whether > 2.5 or "fast". The attached clang patch uses this to set the openCL 2.5 ULPs > accuracy rather than doing it by hand for example...

I've fixed this specific error in r164813. Please let me know if there are more behind it. On Thu, Sep 27, 2012 at 2:16 PM, Sid Manning <sidneym at codeaurora.org> wrote: > > I just updated my llvm sources (revision 164794.) and I see the error, > "overriding non-deleted function" when building with gcc 4.7.0 and passing > -std=c++11. > >

...etadata !{!metadata "fast"} > > This indicates that accuracy loss is acceptable (just how much is unspecified) > for the sake of speed. Thanks to Chandler for pushing me to do it this way! > > It also creates a simple way of getting and setting this information: the > FPMathOperator class: you can cast appropriate instructions to this class > and then use the querying/mutating methods to get/set the accuracy, whether > 2.5 or "fast". The attached clang patch uses this to set the openCL 2.5 ULPs > accuracy rather than doing it by hand for example. > >...

I just updated my llvm sources (revision 164794.) and I see the error, "overriding non-deleted function" when building with gcc 4.7.0 and passing -std=c++11. /usr2/sidneym/llvm/tools/install/bin/c++ -D_GNU_SOURCE -D_DEBUG -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -D__STDC_LIMIT_MACROS -std=c++11 -fPIC -fvisibility-inlines-hidden -I/local/scratch/llvm-tmp/build/lib/VMCore