[Barrelfish-users] Compile error in lib/octopus/parser/parser.c

Daniel Röhrig daniel.roehrig at mailbox.tu-berlin.de
Thu Jul 5 13:17:05 CEST 2012

Hey guys,

we encountered the same error on several machines and two different 
operating systems: Debian Unstable and Ubuntu Precise. Strangely enough 
solving these issues was achieved in different ways:
- Debian unstable machine 1: Install the package haskell-plattform
- Debian unstable machine 2: Downgrade bison from version 5 to version 4.1
- Ubuntu machine 1: New check-out, new build, no idea why it suddenly 
started working again.
- Ubuntu machine 2 & 3: No problems at all (but they use mac laptops, so 
they don't count...)

In the end, we got all machines working again, but only after an 
annoying phase of trial and error and we have no idea if what we did had 
any actual influence on the result. Maybe that helps somehow in finding 
the error.


On 05.07.2012 12:59, Patrik Eklöf wrote:
> If it helps, I reproduced these errors on a fresh Fedora 16 machine after installing the bf prerequisites:
> yum install mercurial -y
> yum install ghc -y
> yum install ghc-ghc-paths-devel -y
> yum install ghc-parsec-devel -y
> Regards,
> Patrik Eklöf
> Student at ICT, Master of Science in Engineering Education, IT Programme, KTH, Kista
> ________________________________________
> From: Gerd Zellweger [zgerd at student.ethz.ch]
> Sent: Thursday, July 05, 2012 10:40
> To: Kornilios Kourtis; Patrik Eklöf
> Cc: barrelfish-users at lists.inf.ethz.ch
> Subject: Re: [Barrelfish-users] Compile error in lib/octopus/parser/parser.c
> Hi Kornilios,
> You are correct. I'm currently trying to figure out why - for some people - hake actually tries to invoke yacc even tough I never intended this to happen. I cannot reproduce it on my Ubuntu machine so I'm planning to do a quick test with a gentoo stable setup that Georgios reported he was using. I'll keep you posted on my progress.
> @Patrick: As Kornilios mentioned, the parse.c file is generated by bison so fixing this file does not make a lot of sense. However, I declared yyerror wrong in parse.y and will fix the declaration there. Thanks for pointing this out.
> Gerd
> Kornilios Kourtis <kornilios.kourtis at inf.ethz.ch> wrote:
>> Hi Patrik,
>> On Wed, Jul 04, 2012 at 03:48:46PM +0000, Patrik Eklöf wrote:
>>> It seems there was a compile error in parser.c due to someone using bad
>>> coding practices of passing a string literal to a function expecting a
>>> char* (which really was expecting a string literal, though).  In case
>>> this error hasn't been fixed yet, I made a patch to fix the error.
>> The parser.c file is automatically generated by bison (yacc), which is why
>> lib/octopus/Hakefile silences some warnings for this particular file.
>> Based on one of your previous mails, I'm guessing that you generated
>> a new parse.c file.
>> You can check this using:
>> hg diff ./lib/octopus/parser/parse.c
>> I think the real bug was the one in your previous mail about the yacc
>> error. I can't figure out why make decided to regenerate the file and I
>> can't reproduce it here. Maybe it was some kind of implicit make rule
>> invocation.
>> cheers,
>> Kornilios.
>> --
>> Kornilios Kourtis
>> _______________________________________________
>> Barrelfish-users mailing list
>> Barrelfish-users at lists.inf.ethz.ch
>> https://lists.inf.ethz.ch/mailman/listinfo/barrelfish-users
> _______________________________________________
> Barrelfish-users mailing list
> Barrelfish-users at lists.inf.ethz.ch
> https://lists.inf.ethz.ch/mailman/listinfo/barrelfish-users

More information about the Barrelfish-users mailing list