llvm dev - May 2014 - [LLVMdev] Is it ok to nuke fpcmp and llvm-PerfectShuffle utils?

On Wed, May 7, 2014 at 8:28 AM, Hal Finkel <hfinkel at anl.gov> wrote:
> ----- Original Message -----
> > From: "Anton Korobeynikov" <anton at
korobeynikov.info>
> > To: "Hal Finkel" <hfinkel at anl.gov>
> > Cc: "Alexey Samsonov" <samsonov at google.com>,
"LLVM Developers Mailing
> List" <llvmdev at cs.uiuc.edu>
> > Sent: Thursday, May 1, 2014 1:56:04 AM
> > Subject: Re: [LLVMdev] Is it ok to nuke fpcmp and llvm-PerfectShuffle
> utils?
> >
> > >> Ouch, indeed... Is it ok it has no build rules?
> > >
> > > I think that it should have build rules. If nothing else, can you
> > > file a bug report on this?
> > It is building using configure + make. In good old days the perfect
> > shuffle tables for PPC were generated through the normal build
> > process
> > like .td files. However, this was pretty useless - they do not change
> > and thus perfectshuffle was excluded from the build process.
>
> Agreed, but even if it is not built by default, we should have build rules
> for it; and given that more users and developers are moving to cmake-based
> builds, we should have build rules in cmake too.
>
I've added it to CMake build in r208225.

What about fpcmp binary? Does anyone still use it?

>
>  -Hal
>
> >
> > --
> > With best regards, Anton Korobeynikov
> > Faculty of Mathematics and Mechanics, Saint Petersburg State
> > University
> >
>
> --
> Hal Finkel
> Assistant Computational Scientist
> Leadership Computing Facility
> Argonne National Laboratory
>


-- 
Alexey Samsonov, Mountain View, CA
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20140507/649af136/attachment.html>

On May 7, 2014, at 10:02 AM, Alexey Samsonov <samsonov at google.com>
wrote:
> 
> On Wed, May 7, 2014 at 8:28 AM, Hal Finkel <hfinkel at anl.gov>
wrote:
> ----- Original Message -----
> > From: "Anton Korobeynikov" <anton at
korobeynikov.info>
> > To: "Hal Finkel" <hfinkel at anl.gov>
> > Cc: "Alexey Samsonov" <samsonov at google.com>,
"LLVM Developers Mailing List" <llvmdev at cs.uiuc.edu>
> > Sent: Thursday, May 1, 2014 1:56:04 AM
> > Subject: Re: [LLVMdev] Is it ok to nuke fpcmp and llvm-PerfectShuffle
utils?
> >
> > >> Ouch, indeed... Is it ok it has no build rules?
> > >
> > > I think that it should have build rules. If nothing else, can you
> > > file a bug report on this?
> > It is building using configure + make. In good old days the perfect
> > shuffle tables for PPC were generated through the normal build
> > process
> > like .td files. However, this was pretty useless - they do not change
> > and thus perfectshuffle was excluded from the build process.
> 
> Agreed, but even if it is not built by default, we should have build rules
for it; and given that more users and developers are moving to cmake-based
builds, we should have build rules in cmake too.
> 
> I've added it to CMake build in r208225.
> 
> What about fpcmp binary? Does anyone still use it?
It is still used in the test-suite.
>From Makefile.rules:
FPCMP     = $(PROJ_OBJ_ROOT)/tools/fpcmp

and from Makefile.programs:

# DIFFPROG - The program used to diff the output
DIFFPROG := $(PROGDIR)/DiffOutput.sh "$(FPCMP) $(TOLERANCEOPT)"
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20140507/18eca014/attachment.html>

On Wed, May 7, 2014 at 10:30 AM, Bob Wilson <bob.wilson at apple.com>
wrote:
>
> On May 7, 2014, at 10:02 AM, Alexey Samsonov <samsonov at google.com>
wrote:
>
>
> On Wed, May 7, 2014 at 8:28 AM, Hal Finkel <hfinkel at anl.gov>
wrote:
>
>> ----- Original Message -----
>> > From: "Anton Korobeynikov" <anton at
korobeynikov.info>
>> > To: "Hal Finkel" <hfinkel at anl.gov>
>> > Cc: "Alexey Samsonov" <samsonov at google.com>,
"LLVM Developers Mailing
>> List" <llvmdev at cs.uiuc.edu>
>> > Sent: Thursday, May 1, 2014 1:56:04 AM
>> > Subject: Re: [LLVMdev] Is it ok to nuke fpcmp and
llvm-PerfectShuffle
>> utils?
>> >
>> > >> Ouch, indeed... Is it ok it has no build rules?
>> > >
>> > > I think that it should have build rules. If nothing else, can
you
>> > > file a bug report on this?
>> > It is building using configure + make. In good old days the
perfect
>> > shuffle tables for PPC were generated through the normal build
>> > process
>> > like .td files. However, this was pretty useless - they do not
change
>> > and thus perfectshuffle was excluded from the build process.
>>
>> Agreed, but even if it is not built by default, we should have build
>> rules for it; and given that more users and developers are moving to
>> cmake-based builds, we should have build rules in cmake too.
>>
>
> I've added it to CMake build in r208225.
>
> What about fpcmp binary? Does anyone still use it?
>
>
> It is still used in the test-suite.
>
> From Makefile.rules:
>
> FPCMP     = $(PROJ_OBJ_ROOT)/tools/fpcmp
>

> and from Makefile.programs:
>
> # DIFFPROG - The program used to diff the output
> DIFFPROG := $(PROGDIR)/DiffOutput.sh "$(FPCMP) $(TOLERANCEOPT)"
>
LLVM's fpcmp I am talking about is located under utils/fpcmp.

Looks like test-suite has its own version of it in
projects/test-suite/tools/fpcmp.c.


-- 
Alexey Samsonov, Mountain View, CA
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20140507/15ff4098/attachment.html>