Regression for trunk on athlon_mp-pc-linux-gnu

Autocrasher autocrasher@devphil.com
Mon Apr 28 10:52:00 GMT 2003


One or more new FAIL's have appeared for athlon_mp-pc-linux-gnu:
    libstdc++-v3.sum 22_locale/codecvt/unicode/char.cc
    libstdc++-v3.sum 22_locale/codecvt/unicode/wchar_t.cc
These tests passed the last time they were ran.

Attached are the ChangeLogs since the previous run.  Possibly a patch with
your name on it has caused the additional regression(s).


The full transcript is at http://www.devphil.com/build/

Host was specified as athlon_mp-pc-linux-gnu.  Target was specified
as athlon_mp-pc-linux-gnu.

This particular run used standard (previously installed) binutils.

This email is generated by a machine, not sent from a person.  If you think
it's been sent to you in error, please see the gcc-regression list archives
for possible followup discussion; this could be legitimate or it could be a
false positive.  Also watch for email from the other regression testers and
check the logs from the known autobuilders.


With love,
The Autocrasher (Phil's Builder Bot, version 1.2)

=========================================================================
-------------- next part --------------
diff -u1r previous/CL/ChangeLog current/CL/ChangeLog
--- previous/CL/ChangeLog	2003-04-26 04:55:01.000000000 -0400
+++ current/CL/ChangeLog	2003-04-28 04:55:01.000000000 -0400
@@ -1 +1,6 @@
+2003-04-27  Daniel Jacobowitz  <drow@mvista.com>
+
+	* Makefile.tpl: Clean $(BUILD_SUBDIR).
+	* Makefile.in: Regenerated.
+
 2003-04-25  Wolfgang Bangerth  <bangerth@dealii.org>
diff -u1r previous/CL/gcc/ChangeLog current/CL/gcc/ChangeLog
--- previous/CL/gcc/ChangeLog	2003-04-27 04:55:03.000000000 -0400
+++ current/CL/gcc/ChangeLog	2003-04-28 04:55:03.000000000 -0400
@@ -1 +1,37 @@
+2003-04-27  Zack Weinberg  <zack@codesourcery.com>
+
+	* expr.c (expand_expr <COMPLEX_CST, STRING_CST>): Always call
+	output_constant_def, use its result instead of TREE_CST_RTL (exp).
+	Can assume it has the form (mem (symbol_ref ".LCxxx")).
+	(expand_expr <COMPONENT_REF>): Can always just extract the
+	relevant field of a CONSTRUCTOR.
+	(expand_expr <ARRAY_REF, COMPONENT_REF, BIT_FIELD_REF,
+	ARRAY_RANGE_REF>): Make control flow explicit.
+	* varasm.c (output_constant_def): Can look at TREE_CST_RTL of
+	an INTEGER_CST.
+
+2003-04-27  Kazu Hirata  <kazu@cs.umass.edu>
+
+	* reload1.c (reg_set_luid): Fix a comment typo.
+
+2003-04-27  Zack Weinberg  <zack@codesourcery.com>
+
+	* varasm.c (const_str_htab_hash, const_str_htab_eq, STRHASH,
+	struct deferred_string, const_str_htab): Kill.
+	(n_deferred_strings): New static variable.
+	(build_constant_desc): Set SYMBOL_REF_DECL of the new
+	symbol_ref to point to the constant.
+	(output_constant_def): When a deferred string is forced out,
+	just clear STRING_POOL_ADDRESS_P and decrement n_deferred_strings.
+	(mark_constant): Likewise.
+	(maybe_output_constant_def_contents): When deferring a string
+	constant, just set STRING_POOL_ADDRESS_P and increment
+	n_deferred_strings.
+	(mark_constant_pool): Check n_deferred_strings, not the size
+	of const_str_htab.
+	(init_varasm_once): No need to create const_str_htab.
+
+	* rtl.def, rtl.h, doc/rtl.texi: Document possibility that
+	SYMBOL_REF_DECL points to a constant.
+
 2003-04-26  Zack Weinberg  <zack@codesourcery.com>
diff -u1r previous/CL/gcc/cp/ChangeLog current/CL/gcc/cp/ChangeLog
--- previous/CL/gcc/cp/ChangeLog	2003-04-26 17:04:51.000000000 -0400
+++ current/CL/gcc/cp/ChangeLog	2003-04-28 04:55:17.000000000 -0400
@@ -1 +1,18 @@
+2003-04-27  Mark Mitchell  <mark@codesourcery.com>
+
+	* decl.c (maybe_commonize_var): Fix thinko in last patch.
+
+2003-04-27  Mark Mitchell  <mark@codesourcery.com>
+
+	PR c++/10506
+	* method.c (use_thunk): Decrement immediate_size_expand.
+
+	PR c++/10503
+	* cp-tree.h (DECL_VAR_MARKED_P): New macro.
+	(DECL_MAYBE_TEMPLATE): Remove.
+	* class.c (fixed_type_or_null): Avoid infinite recursion.
+
+	* decl.c (maybe_commonize_var): Make the code match the comments.
+	* pt.c (instantiate_decl): Move call to import_export_decl.
+
 2003-04-26  Mark Mitchell  <mark@codesourcery.com>
diff -u1r previous/CL/gcc/testsuite/ChangeLog current/CL/gcc/testsuite/ChangeLog
--- previous/CL/gcc/testsuite/ChangeLog	2003-04-27 04:55:09.000000000 -0400
+++ current/CL/gcc/testsuite/ChangeLog	2003-04-28 04:55:21.000000000 -0400
@@ -1 +1,9 @@
+2003-04-27  Mark Mitchell  <mark@codesourcery.com>
+
+	PR c++/10506
+	* g++.dg/init/new6.C: New test.
+
+	PR c++/10503
+	* g++.dg/init/ref6.C: New test.
+
 2003-04-26  David Edelsohn  <edelsohn@gnu.org>
diff -u1r previous/CL/libstdc++-v3/ChangeLog current/CL/libstdc++-v3/ChangeLog
--- previous/CL/libstdc++-v3/ChangeLog	2003-04-27 16:40:23.000000000 -0400
+++ current/CL/libstdc++-v3/ChangeLog	2003-04-28 04:55:35.000000000 -0400
@@ -1 +1,35 @@
+2003-04-27  Benjamin Kosnik  <bkoz@redhat.com>
+
+	* include/std/std_fstream.h (basic_filebuf): _M_pback_destroy to
+	_M_destroy_pback. _M_pback_create to
+	_M_create_pback. _M_underflow_common to
+	_M_underflow. _M_really_overflow to _M_overflow.
+	* include/bits/fstream.tcc: Same.
+	* src/fstream.cc: Same.
+	* include/std/std_streambuf.h (basic_streambuf): _M_in_cur_move to
+	_M_move_in_cur.  _M_out_cur_move to _M_move_out_cur.
+	* include/bits/streambuf.tcc: Same.
+	* include/bits/fstream.tcc: Same.
+	* include/bits/sstream.tcc: Same.
+	
+2003-04-27  Benjamin Kosnik  <bkoz@redhat.com>
+
+	* include/bits/locale_classes.h (locale::_Impl): Change _M_names
+	from fixed size array.
+	(locale): Change _S_categories as well.
+	Formatting tweaks.
+	* include/bits/locale_facets.tcc: Tweak.
+	* config/locale/gnu/c_locale.cc: Assign _S_categories.
+	* config/locale/generic/c_locale.cc: Same.
+	* src/locale.cc: Tweak.
+	* src/globals.cc: Change facet_name to name_vec, add names_c.
+	* src/localename.cc: Use them.
+	(locale::_Impl::~_Impl): Destroy _M_names.
+	(locale::_Impl::_Impl): Create _M_names.
+	
+2003-04-27  Andreas Schwab  <schwab@suse.de>
+
+	* config/locale/ieee_1003.1-2001/codecvt_specializations.h
+	(__enc_traits): Use __ibom and __ebom instead of ignoring them.
+
 2003-04-27  Nathan Myers  <ncm@cantrip.org>


More information about the Gcc-regression mailing list