+2007-06-13 Eric Christopher <echristo@apple.com>
+
+ * config/i386/i386.c (override_options): If we've specified
+ an arch then don't use TARGET_SUBTARGET_ISA_DEFAULTs.
+
2007-06-13 Bob Wilson <bob.wilson@acm.org>
-
+
* df-scan.c (df_get_entry_block_def_set): Check if STATIC_CHAIN_REGNUM
is defined.
-
+
2007-06-13 Bernd Schmidt <bernd.schmidt@analog.com>
* config/bfin/bfin.h (EH_RETURN_HANDLER_RTX): Use gen_frame_mem.
object file.
(make_gcov_file_name): Do not generate long names if input_name is
NULL.
- (output_lines): If merging results do not display graph, data and
+ (output_lines): If merging results do not display graph, data and
runs informations.
Checking source file modification is done in find_source.
Request walking the subtrees only, leaving the current is_lhs/val_only
untouched.
(convert_non_local_reference): Likewise.
-
+
2007-06-12 Nathan Sidwell <nathan@codesourcery.com>
* config/m68k/m68k-devices.def (52221, 52223, 5253): New.
-
+
2007-06-12 Richard Guenther <rguenther@suse.de>
PR tree-optimization/15353
df_process_deferred_rescans, df_notes_rescan): Fixed spelling of
word "deferred".
* df-core.c: Ditto.
-
+
2007-06-11 Daniel Berlin <dberlin@dberlin.org>
* Merge dataflow-branch into mainline (see ChangeLog.dataflow)
(__divsi3, __umodsi3, __modsi3): Likewise.
(__ashldi3, __ashrdi3, __lshrdi3): New.
* config/xtensa/t-xtensa (LIB1ASMFUNCS): Add DImode shift functions.
-
+
2007-06-08 Harsha Jagasia <harsha.jagasia@amd.com>
Tony Linthicum <tony.linthicum@amd.com>
(vectorizable_load): Call vect_model_load_cost during analysis phase.
(vectorizable_store): Call vect_model_store_cost during analysis phase.
(vectorizable_call, vectorizable_assignment, vectorizable_operation,
- vectorizable_promotion, vectorizable_demotion): Call
+ vectorizable_promotion, vectorizable_demotion): Call
vect_model_simple_cost during analysis phase.
2007-06-08 Simon Baldwin <simonb@google.com>
PR target/29953
* config/sh/sh.md (doloop_end): New pattern and splitter.
* loop-iv.c (simple_rhs_p): Check for hardware registers.
-
+
2007-06-08 Zdenek Dvorak <dvorakz@suse.cz>
PR middle-end/32209
* rtl.h (push_to_sequence2): New.
* emit-rtl.c (push_to_sequence2): New.
* function.c (assign_parm_data_all): Add new fields.
- (assign_parm_setup_block): Call push_to_sequence2 instead of
+ (assign_parm_setup_block): Call push_to_sequence2 instead of
push_to_sequence.
(assign_parm_setup_reg): Ditto.
(assign_parm_setup_stack): Ditto.
{
int i;
int ix86_tune_defaulted = 0;
+ int ix86_arch_specified = 0;
unsigned int ix86_arch_mask, ix86_tune_mask;
/* Comes from final.c -- no real reason to change it. */
if (!ix86_arch_string)
ix86_arch_string = TARGET_64BIT ? "x86-64" : "i386";
+ else
+ ix86_arch_specified = 1;
+
if (!strcmp (ix86_arch_string, "generic"))
error ("generic CPU can be used only for -mtune= switch");
if (!strncmp (ix86_arch_string, "generic", 7))
/* Enable by default the SSE and MMX builtins. Do allow the user to
explicitly disable any of these. In particular, disabling SSE and
MMX for kernel code is extremely useful. */
+ if (!ix86_arch_specified)
ix86_isa_flags
|= ((OPTION_MASK_ISA_SSE2 | OPTION_MASK_ISA_SSE | OPTION_MASK_ISA_MMX
| TARGET_SUBTARGET64_ISA_DEFAULT) & ~ix86_isa_flags_explicit);
{
target_flags |= TARGET_SUBTARGET32_DEFAULT & ~target_flags_explicit;
+ if (!ix86_arch_specified)
ix86_isa_flags
|= TARGET_SUBTARGET32_ISA_DEFAULT & ~ix86_isa_flags_explicit;