similar to: [LLVMdev] Current Regressions

On Thu, 12 May 2005, John Criswell wrote: > Here is a more complete list of regressions for the platforms listed below. > Some of the regressions from the previous list I emailed a few days ago have > been fixed or were false positives. Thanks to all who've helped fix things. > > We would like to try to get as many of these fixed as possible before I > create the release

Chris Lattner wrote: > On Thu, 12 May 2005, John Criswell wrote: > >> Here is a more complete list of regressions for the platforms listed >> below. Some of the regressions from the previous list I emailed a few >> days ago have been fixed or were false positives. Thanks to all >> who've helped fix things. >> >> We would like to try to get as many

On Fri, 13 May 2005, John Criswell wrote: > These following regression tests are failing (didn't mention these before > because they're easy to spot in nightly tester emails): > > FAIL: /Users/criswell/llvm/test/Regression/CodeGen/Generic/print-arith-fp.ll: > FAIL: > /Users/criswell/llvm/test/Regression/CodeGen/X86/fast-cc-pass-in-regs.ll: > FAIL:

A big compile time regression. Any ideas? Ciao, Duncan. On 22/07/11 19:13, llvm-testresults at cs.uiuc.edu wrote: > > bwilson__llvm-gcc_PROD__i386 nightly tester results > > URL http://llvm.org/perf/db_default/simple/nts/253/ > Nickname bwilson__llvm-gcc_PROD__i386:4 > Name curlew.apple.com > > Run ID Order Start Time End Time > Current 253 0 2011-07-22 16:22:04

Perhaps you noticed that LLVM gained a new optimizing register allocator yesterday (r130568). Linear scan is going away, and RAGreedy is the new default for optimizing builds. Hopefully, you noticed because your binaries were suddenly 2% smaller and 10% faster*. Some noticed because LLVM started crashing or miscompiling their code. Greedy replaces a fairly big chunk of the code generator, so

Hi, guys, I am comparing the performance of the default scheduler (seems to be the one that minimizes register pressure) with no scheduler (-pre-RA-sched=none), and I got these numbers. The ratio is low_reg_pressure/none, that is, the lower the number, the better the performance with low register pressure: CFP2000/177.mesa/177.mesa 1.00 CFP2000/179.art/179.art

Hi all, I've started looking at the GlobalMerge pass, enabled by default on ARM and AArch64. I think we should reconsider that, at least for AArch64. As is, the pass just merges all globals together, in groups of 4KB (AArch64, 128B on ARM). At the time it was enabled, the general thinking was "it's almost free, it doesn't affect performance much, we might as well use it".

Hello; I was able to build and install llvm(3.0) under Ubuntu 11.10 (using the ./configure script found under llvm source, and then make and make install). While configuring, I gave --prefix as a directory where I would like llvm to be installed. I did not give --with-llvmgccdir and the --enable-optimized argument to configure. Because 3.0 doesn't come with llvmgcc source/binaries and I

Are these 225 compile time regressions real? It sure looks bad! Ciao, Duncan. On 01/12/11 09:39, llvm-testresults at cs.uiuc.edu wrote: > > bwilson__llvm-gcc_PROD__i386 nightly tester results > > URL http://llvm.org/perf/db_default/simple/nts/380/ > Nickname bwilson__llvm-gcc_PROD__i386:4 > Name curlew.apple.com > > Run ID Order Start Time End Time > Current 380

Hi Bob, are these performance regressions real? They look pretty serious. Ciao, Duncan. On 10/12/11 09:40, llvm-testresults at cs.uiuc.edu wrote: > > bwilson__llvm-gcc_PROD__i386 nightly tester results > > URL http://llvm.org/perf/db_default/simple/nts/332/ > Nickname bwilson__llvm-gcc_PROD__i386:4 > Name curlew.apple.com > > Run ID Order Start Time End Time >

On 01 Sep 2006, at 10:05, Kenneth Hoste wrote: > >> >>> Also, it is possible to tell make only to compile benchmark X? How >>> can I >>> enforce this? >> >> Go into the directory for that benchmark, then run 'make' or >> whatever. > I tried tom compile each of the SPEC CPU2000 benchmarks using the make command is each respective

On Jul 24, 2011, at 3:02 AM, Duncan Sands wrote: > A big compile time regression. Any ideas? > > Ciao, Duncan. False alarm. For some reason that I have not yet been able to figure out, these tests run significantly more slowly when I run them during the daytime, which I did for that run. I checked a few of the worst regressions reported here and they all recovered in subsequent

Yes, they are real. I re-ran the two tests with the biggest execution time regressions, and the results were completely reproducible. On Oct 12, 2011, at 1:24 AM, Duncan Sands wrote: > Hi Bob, are these performance regressions real? They look pretty serious. > > Ciao, Duncan. > > On 10/12/11 09:40, llvm-testresults at cs.uiuc.edu wrote: >> >>

Hello, Some problems were solved, new ones arised... Getting closer though... The fixes for the previous problems are at the bottom of this email, bug reports will be submitted when all problems are solved. +++ New/remaining problems +++ Currently, 9/26 benchmarks compile and run succesfully. One (fma3d) still has a f95 related problem (see below). The other 16 are divided into two groups:

Thanks for the perf testing. I assume that DAG legalization is equipped to handle these cases fairly well, or someone would've complained by now... FWIW (and at least some of this can be blamed on me), instcombine already does the narrowing transforms without checking shouldChangeType() for binops like and/or/xor/udiv. The justification was that narrower ops are always better for

Hello Thanks for looking into this. I can't be very confident what the knock on result of a change like that would be, especially on architectures that are not Arm. What I can do though, is run some benchmarks and look at that results. Using this patch: --- a/lib/Transforms/InstCombine/InstructionCombining.cpp +++ b/lib/Transforms/InstCombine/InstructionCombining.cpp @@ -150,6 +150,9 @@

Currently, the debug intrinsic functions each have 3 arguments: an SSA value representing either the address or Value of a local variable, a DILocalVariable, and a complex expression. If the SSA value is an Instruction, and that Instruction is at some point deleted, we attempt to salvage the SSA value by recreating the instruction within the complex expression. If the instruction cannot be

On 31 Aug 2006, at 23:46, Chris Lattner wrote: > On Thu, 31 Aug 2006, Kenneth Hoste wrote: >> Bummer. I think I'll contact the NAG support for more info on >> this. Can you >> show me the content of your Makefile.nagfortran? > > It is identical to yours. > >> Also, it is possible to tell make only to compile benchmark X? How >> can I >>

Hi, LLVM dev team: Thanks for your suggestion, I have done the experiment to compare the two pointer analysis(Andersen and Steensgaard) methods in LLVM, but the result was unexpected. In each test, I compare these two methods using same optimization; There are several tests, each with a different optimization. The benchmark is all the 11 C programs in CINT2000 of SPEC. In all the tests, I found

Can someone tell me how to build festival on a machine with gcc 3.3.2? I've searched all around and even found a reference or two that the problem exists but I'm not seeing the fix. thanks! -reed Symtoms are -- ./configure, then.... [root@telephone speech_tools]# make Check system type Remake modincludes.inc NATIVE_AUDIO ok EDITLINE