[m-rev.] trivial diff: workaround source distribution build problem

Julien Fischer juliensf at csse.unimelb.edu.au
Mon Jan 26 22:23:14 AEDT 2009

I'll look into what the actual problem here is tomorrow.


 	Workaround a problem that has been preventing the source
 	distribution from building on jupiter.


Index: Mercury.options
RCS file: /home/mercury/mercury1/repository/mercury/compiler/Mercury.options,v
retrieving revision 1.39
diff -u -r1.39 Mercury.options
--- Mercury.options	25 Sep 2008 13:44:48 -0000	1.39
+++ Mercury.options	26 Jan 2009 10:41:54 -0000
@@ -70,6 +70,12 @@
  # Work around a problem in the HiPE compiler (as of Erlang R11B5).
  MCFLAGS-libs.options += --erlang-switch-on-strings-as-atoms

+# Work around a problem first observed in rotd-2009-01-26.  Compiling
+# revision 1.135 of the module below results in an abort in the LLDS
+# code generator at -O5 or above.  The problem occurs when 
+# --inline-compound-threshold is set to 100 (as it is at -O5). 
+MCFLAGS-backend_libs.compile_target_code = --inline-compound-threshold 20
  # The c_code in the module gcc.m needs the header files from the GNU C
  # distribution.  Note that we need to compile these with
  # -DMR_NO_BACKWARDS_COMPAT, because otherwise there are name

mercury-reviews mailing list
Post messages to:       mercury-reviews at csse.unimelb.edu.au
Administrative Queries: owner-mercury-reviews at csse.unimelb.edu.au
Subscriptions:          mercury-reviews-request at csse.unimelb.edu.au

More information about the reviews mailing list