public inbox for java@gcc.gnu.org
 help / color / mirror / Atom feed
* Re: CVS error on ppc
@ 2002-05-01  9:26 Eric Blake
  2002-05-01  9:42 ` Nic Ferrier
  2002-05-01  9:56 ` Tom Tromey
  0 siblings, 2 replies; 4+ messages in thread
From: Eric Blake @ 2002-05-01  9:26 UTC (permalink / raw)
  To: Nic Ferrier; +Cc: java, gcc

You're not the only one.  Something is wrong in the Makefile script,
because I got the same error on a RedHat box:

cvs [update aborted]: reading libjava/sysdep/i386/locks.h: Too many levels of symbolic links

Basically, something in the Makefile is linking
libjava/sysdep/<arch>/locks.h to itself, causing an infinite link
recursion. It is also affecting compilation:

...
linking ../../../gcc/libjava/sysdep/i386/locks.h to sysdep/locks.h
configure: error: ../../../gcc/libjava/sysdep/i386/locks.h: File not found
make[2]: *** [Makefile] Error 1

I'm not sure who is responsible, but would you like to file a
high-priority PR since you found it first? I'm cc'ing the gcc list, in
case anyone knows what went wrong.


> From: Nic Ferrier <nferrier at tapsellferrier dot co dot uk>
> To: <java at gcc dot gnu dot org>
> Date: 01 May 2002 15:09:39 +0100
> 
> Something strange just happened on my PPClinux box... I was just
> trying a cvs update on the gcc tree (the branch) and I got this:
> 
> cvs [update aborted]: reading libjava/sysdep/powerpc/locks.h: Too many levels of symbolic links
> nferrier@imac:~/projects/gnu/gcc/gcc-cvs$ Write failed flushing stdout buffer.
> write stdout: Broken pipe
> 
> Bizzare! It worked fine last night (though the build didn't run,
> which is why I was updating today - so I could report the fault).
> 
> 
> Nic

----
Eric Blake       ebb9@email.byu.edu

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: CVS error on ppc
  2002-05-01  9:26 CVS error on ppc Eric Blake
@ 2002-05-01  9:42 ` Nic Ferrier
  2002-05-01  9:56 ` Tom Tromey
  1 sibling, 0 replies; 4+ messages in thread
From: Nic Ferrier @ 2002-05-01  9:42 UTC (permalink / raw)
  To: Eric Blake; +Cc: java, gcc

Eric Blake <ebb9@email.byu.edu> writes:

> You're not the only one.  Something is wrong in the Makefile script, 
> because I got the same error on a RedHat box: 
>  
> cvs [update aborted]: reading libjava/sysdep/i386/locks.h: Too many levels of symbolic links 
>  
> Basically, something in the Makefile is linking 
> libjava/sysdep/<arch>/locks.h to itself, causing an infinite link 
> recursion. It is also affecting compilation: 
>  
> ... 
> linking ../../../gcc/libjava/sysdep/i386/locks.h to sysdep/locks.h 
> configure: error: ../../../gcc/libjava/sysdep/i386/locks.h: File not found 
> make[2]: *** [Makefile] Error 1 
>  
> I'm not sure who is responsible, but would you like to file a 
> high-priority PR since you found it first? I'm cc'ing the gcc list, in 
> case anyone knows what went wrong. 

I don't seem to be able to do that. Submitting with high priority
gives me an error.

Can someone else submit it please?


Nic

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: CVS error on ppc
  2002-05-01  9:26 CVS error on ppc Eric Blake
  2002-05-01  9:42 ` Nic Ferrier
@ 2002-05-01  9:56 ` Tom Tromey
  1 sibling, 0 replies; 4+ messages in thread
From: Tom Tromey @ 2002-05-01  9:56 UTC (permalink / raw)
  To: Eric Blake; +Cc: Nic Ferrier, java, gcc

>>>>> "Eric" == Eric Blake <ebb9@email.byu.edu> writes:

Eric> You're not the only one.  Something is wrong in the Makefile
Eric> script, because I got the same error on a RedHat box:

Eric> I'm not sure who is responsible, but would you like to file a
Eric> high-priority PR since you found it first? I'm cc'ing the gcc
Eric> list, in case anyone knows what went wrong.

I saw this the other day.

This happens when you rebuild an already-built tree.  A configury
change will make things mess up in just this way.

I fixed it by removing the bogus symlink from the source tree,
updating to get the missing file, and then rebuilding from scratch.
There may be something less drastic you can do.

This definitely isn't a high-priority PR thing.  The problem only
manifests when rebuilding a tree that was built before this change (I
think) went in:

    2002-04-16  DJ Delorie  <dj@redhat.com>

	    * configure.in: Allow building in $srcdir.
	    * configure: Regenerated.

It shouldn't happen for any clean build.

Tom

^ permalink raw reply	[flat|nested] 4+ messages in thread

* CVS error on ppc
@ 2002-05-01  7:09 Nic Ferrier
  0 siblings, 0 replies; 4+ messages in thread
From: Nic Ferrier @ 2002-05-01  7:09 UTC (permalink / raw)
  To: java

Something strange just happened on my PPClinux box... I was just
trying a cvs update on the gcc tree (the branch) and I got this:

cvs [update aborted]: reading libjava/sysdep/powerpc/locks.h: Too many levels of symbolic links
nferrier@imac:~/projects/gnu/gcc/gcc-cvs$ Write failed flushing stdout buffer.
write stdout: Broken pipe

Bizzare! It worked fine last night (though the build didn't run,
which is why I was updating today - so I could report the fault).


Nic

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2002-05-01 16:56 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2002-05-01  9:26 CVS error on ppc Eric Blake
2002-05-01  9:42 ` Nic Ferrier
2002-05-01  9:56 ` Tom Tromey
  -- strict thread matches above, loose matches on Subject: below --
2002-05-01  7:09 Nic Ferrier

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).