[PATCH] Improve and consolidate sparc PIC assembler.

[Torbjorn Granlund]

David Miller <davem at davemloft.net> writes:

  From: Torbjorn Granlund <tg at gmplib.org>
  Date: Thu, 11 Apr 2013 10:50:40 +0200
  
  > There are syntax errors for swift.nada.kth.se, a Solaris system.  See
  > http://gmplib.org/devel/tm-date.html.
  > 
  > The offending lines:
  > 
  > swift (ABI=64)
  > 99:     sethi   %gdop_hix22(ctz_table), %i5
  > swift-32 (ABI=32)
  > 99:     sethi   %gdop_hix22(.Lnoll), %l0
  > 
  > We need things to work on Solaris, *BSD.
  
  The code I use in the acinclude.m4 test is:
  
          sethi   %gdop_hix22(symbol), %g1
          or      %g1, %gdop_lox10(symbol), %g1
  
  The sethi syntax used in both cases is the same.
  
  Maybe the final CC isn't set at the time that the test is run, so
  a different assembler is used for the tests compared to the actual
  build?
  
  I saw you futzing around with the location of the CC setting on
  sparc in the config files recently, specifically:
  
          * config.guess (sparc*): Invoke set_cc_for_build to get $dummy.
  
  Might be related?
  
I seriously doubt that.  Local settings of config.guess cannot possibly
leak out into configure.

The new got feature test passes.  Do you have a Slowlaris system where
you could debug this?

-- 
Torbjörn