[m-rev.] diff: don't try to make analysis files in parallel
Peter Wang
novalazy at gmail.com
Tue Jul 1 11:39:08 AEST 2008
Branches: main
compiler/make.program_target.m:
Don't try to build `.analysis' files in parallel with `mmc --make'.
It doesn't and can't work without more extensive changes.
Index: compiler/make.program_target.m
===================================================================
RCS file: /home/mercury/mercury1/repository/mercury/compiler/make.program_target.m,v
retrieving revision 1.84
diff -u -p -r1.84 make.program_target.m
--- compiler/make.program_target.m 29 May 2008 04:41:54 -0000 1.84
+++ compiler/make.program_target.m 1 Jul 2008 01:33:48 -0000
@@ -748,7 +748,7 @@ build_analysis_files(MainModuleName, All
% Ensure all interface files are present before continuing. This
% prevents a problem when two parallel branches try to generate the
% same missing interface file later.
- %
+ % (Although we can't actually build analysis files in parallel yet.)
make_all_interface_files(AllModules, Succeeded1, !Info, !IO),
( Succeeded1 = no, KeepGoing = no ->
Succeeded = no
@@ -787,7 +787,7 @@ build_analysis_files_1(MainModuleName, A
build_analysis_files_2(MainModuleName, TargetModules, LocalModulesOpts,
Succeeded0, Succeeded, !Info, !IO) :-
globals.io_lookup_bool_option(keep_going, KeepGoing, !IO),
- foldl2_maybe_stop_at_error_maybe_parallel(KeepGoing,
+ foldl2_maybe_stop_at_error(KeepGoing,
make_module_target_extra_options(LocalModulesOpts),
make_dependency_list(TargetModules, module_target_analysis_registry),
Succeeded1, !Info, !IO),
--------------------------------------------------------------------------
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