GCC memory consumption increased by recent patch!
gcctest@suse.de
gcctest@suse.de
Mon Sep 27 04:12:00 GMT 2004
Hi,
Comparing memory consumption on compilation of combine.i and generate-3.4.ii I got:
comparing combine.c compilation at -O0 level:
Overall memory allocated via mmap and sbrk increased from 24121k to 24169k, overall 0.20%
Overall memory needed: 24121k -> 24169k
Peak memory use before GGC: 9296k
Peak memory use after GGC: 8607k
Maximum of released memory in single GGC run: 2870k
Garbage: 42629k
Leak: 6090k
Overhead: 5588k
GGC runs: 363
comparing combine.c compilation at -O1 level:
Overall memory allocated via mmap and sbrk increased from 25585k to 25629k, overall 0.17%
Overall memory needed: 25585k -> 25629k
Peak memory use before GGC: 9377k
Peak memory use after GGC: 8800k
Maximum of released memory in single GGC run: 2031k
Garbage: 69180k
Leak: 6674k
Overhead: 11094k
GGC runs: 575
comparing combine.c compilation at -O2 level:
Overall memory allocated via mmap and sbrk increased from 28885k to 28929k, overall 0.15%
Overall memory needed: 28885k -> 28929k
Peak memory use before GGC: 12772k
Peak memory use after GGC: 12613k
Maximum of released memory in single GGC run: 2525k
Garbage: 84273k
Leak: 6428k
Overhead: 15636k
GGC runs: 571
comparing combine.c compilation at -O3 level:
Overall memory needed: 20544k
Peak memory use before GGC: 13044k
Peak memory use after GGC: 12692k
Maximum of released memory in single GGC run: 3362k
Garbage: 112175k
Leak: 6929k
Overhead: 20684k
GGC runs: 637
comparing insn-attrtab.c compilation at -O0 level:
Overall memory needed: 116476k
Peak memory use before GGC: 77101k
Peak memory use after GGC: 45186k
Maximum of released memory in single GGC run: 42129k
Garbage: 159505k -> 159505k
Leak: 10645k
Overhead: 19805k -> 19805k
GGC runs: 310
comparing insn-attrtab.c compilation at -O1 level:
Overall memory needed: 133084k
Peak memory use before GGC: 93162k
Peak memory use after GGC: 71529k
Maximum of released memory in single GGC run: 41234k
Garbage: 453085k -> 453085k
Leak: 11078k
Overhead: 73707k -> 73707k
GGC runs: 461
comparing insn-attrtab.c compilation at -O2 level:
Overall memory needed: 207744k
Peak memory use before GGC: 108725k
Peak memory use after GGC: 87092k
Maximum of released memory in single GGC run: 34490k
Garbage: 502446k -> 502447k
Leak: 11218k
Overhead: 83073k -> 83073k
GGC runs: 384
comparing insn-attrtab.c compilation at -O3 level:
Overall memory needed: 207736k
Peak memory use before GGC: 108726k
Peak memory use after GGC: 87093k
Maximum of released memory in single GGC run: 34491k
Garbage: 503617k -> 503617k
Leak: 11256k
Overhead: 83469k -> 83469k
GGC runs: 393
comparing Gerald's testcase PR8361 compilation at -O0 level:
Overall memory needed: 115352k
Peak memory use before GGC: 92231k
Peak memory use after GGC: 90025k
Maximum of released memory in single GGC run: 20139k
Garbage: 265688k -> 265688k
Leak: 60277k
Overhead: 47198k -> 47198k
GGC runs: 378
comparing Gerald's testcase PR8361 compilation at -O1 level:
Overall memory needed: 109320k -> 109352k
Peak memory use before GGC: 95751k
Peak memory use after GGC: 88883k
Maximum of released memory in single GGC run: 19570k
Garbage: 594441k -> 594412k
Leak: 62357k
Overhead: 131029k -> 131026k
GGC runs: 605
comparing Gerald's testcase PR8361 compilation at -O2 level:
Overall memory needed: 110308k -> 110316k
Peak memory use before GGC: 95751k
Peak memory use after GGC: 88883k
Maximum of released memory in single GGC run: 19569k
Garbage: 647298k -> 647298k
Leak: 62912k
Overhead: 154283k -> 154285k
GGC runs: 638
comparing Gerald's testcase PR8361 compilation at -O3 level:
Overall memory needed: 108024k -> 108016k
Peak memory use before GGC: 91613k
Peak memory use after GGC: 90197k
Maximum of released memory in single GGC run: 20253k
Garbage: 682712k -> 682715k
Leak: 63333k
Overhead: 163435k -> 163435k
GGC runs: 626
Head of changelog is:
--- /usr/src/SpecTests/sandbox-britten-memory/x86_64/mem-result/ChangeLog 2004-09-26 18:52:33.000000000 +0000
+++ /usr/src/SpecTests/sandbox-britten-memory/gcc/gcc/ChangeLog 2004-09-27 00:57:09.000000000 +0000
@@ -1,3 +1,75 @@
+2004-09-26 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
+
+ * bitmap.h (EXECUTE_IF_SET_IN_BITMAP, EXECUTE_IF_AND_COMPL_IN_BITMAP,
+ EXECUTE_IF_AND_IN_BITMAP): Changed to iterator style.
+ (bitmap_iterator): New type.
+ (bmp_iter_common_next_1, bmp_iter_single_next_1, bmp_iter_single_init,
+ bmp_iter_end_p, bmp_iter_single_next, bmp_iter_and_not_next_1,
+ bmp_iter_and_not_init, bmp_iter_and_not_next, bmp_iter_and_next_1,
+ bmp_iter_and_init, bmp_iter_and_next): New functions.
+ * basic-block.h (EXECUTE_IF_SET_IN_REG_SET,
+ EXECUTE_IF_AND_COMPL_IN_REG_SET, EXECUTE_IF_AND_IN_REG_SET): Changed to
+ use iterator-style EXECUTE_IF_IN_BITMAP macros.
+ * bitmap.c (bitmap_print): Ditto.
+ * bt-load.c (clear_btr_from_live_range, add_btr_to_live_range,
+ btr_def_live_range): Ditto.
+ * cfganal.c (compute_dominance_frontiers_1) Ditto.
+ * cgraphunit.c (convert_UIDs_in_bitmap, cgraph_characterize_statics):
+ Ditto.
+ * ddg.c (build_inter_loop_deps): Ditto.
+ * df.c (FOR_EACH_BB_IN_BITMAP, df_bb_reg_info_compute, df_refs_update):
+ Ditto.
+ * except.c (remove_eh_handler): Ditto.
+ * flow.c (reg_set_to_hard_reg_set): Ditto.
+ * gcse.c (clear_modify_mem_tables): Ditto.
+ * global.c (build_insn_chain): Ditto.
+ * ifcvt.c (dead_or_predicable): Ditto.
+ * loop-invariant.c (get_inv_cost, set_move_mark, move_invariant_reg):
+ Ditto.
+ * ra-build.c (livethrough_conflicts_bb, conflicts_between_webs): Ditto.
+ * ra-rewrite.c (reloads_to_loads, rewrite_program2,
+ detect_web_parts_to_rebuild, delete_useless_defs, actual_spill): Ditto.
+ * tree-cfg.c (allocate_ssa_names, tree_duplicate_sese_region,
+ tree_purge_all_dead_eh_edges): Ditto.
+ * tree-into-ssa.c (compute_global_livein, insert_phi_nodes,
+ insert_phi_nodes_for, debug_def_blocks_r, invalidate_name_tags,
+ rewrite_ssa_into_ssa): Ditto.
+ * tree-outof-ssa.c (find_replaceable_exprs): Ditto.
+ * tree-sra.c (scan_function, decide_instantiations, scalarize_parms):
+ Ditto.
+ * tree-ssa-alias.c (init_alias_info, compute_points_to_and_addr_escape,
+ compute_flow_sensitive_aliasing, maybe_create_global_var,
+ dump_points_to_info_for): Ditto.
+ * tree-ssa-dce.c (EXECUTE_IF_CONTROL_DEPENDENT): Ditto.
+ * tree-ssa-dse.c (dse_finalize_block): Ditto.
+ * tree-ssa-live.c (live_worklist, calculate_live_on_entry,
+ calculate_live_on_exit, build_tree_conflict_graph, dump_live_info):
+ Ditto.
+ * tree-ssa-loop-ivopts.c (find_induction_variables,
+ find_interesting_uses, add_old_ivs_candidates, alloc_use_cost_map,
+ determine_use_iv_costs, determine_set_costs, find_best_candidate,
+ set_cost_up_to, create_new_ivs, remove_unused_ivs, free_loop_data):
+ Ditto.
+ * tree-ssa-loop-manip.c (add_exit_phis_var, add_exit_phis): Ditto.
+ * tree-ssa-operands.c (get_asm_expr_operands, add_call_clobber_ops,
+ add_call_read_ops): Ditto.
+ * tree-ssa-pre.c (bitmap_print_value_set, insert_aux): Ditto.
+
+2004-09-26 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
+
+ * pa.c (print_operand): Use non-trapping completers for UNLE, UNLT,
+ UNGE, UNGT, UNEQ, UNORDERED and ORDERED comparisons.
+
+2004-09-26 Ulrich Weigand <uweigand@de.ibm.com>
+
+ * builtins.c (expand_builtin_memcmp): Adjust MEM_SIZE to
+ reflect size of memory regions being compared.
+
+2004-09-26 Ulrich Weigand <uweigand@de.ibm.com>
+
+ * builtins.c (get_memory_rtx): Set mem attributes for non-ADDR_EXPR
+ expressions. Always clear MEM_SIZE and alias set.
+
2004-09-26 Roger Sayle <roger@eyesopen.com>
Giovanni Bajo <giovannibajo@gcc.gnu.org>
I am friendly script caring about memory consumption in GCC. Please contact
jh@suse.cz if something is going wrong.
The results can be reproduced by building compiler with
--enable-gather-detailed-mem-stats targetting x86-64 and compiling preprocessed
combine.c or testcase from PR8632 with:
-fmem-report --param=ggc-min-heapsize=1024 --param=ggc-min-expand=1 -Ox -Q
The memory consumption summary appears in the dump after detailed listing of
the places they are allocated in. Peak memory consumption is actually computed
by looking for maximal value in {GC XXXX -> YYYY} report.
Yours testing script.
More information about the Gcc-regression
mailing list