This is the mail archive of the
gcc-regression@gcc.gnu.org
mailing list for the GCC project.
Some aspect of GCC memory consumption increased by recent patch
- From: gcctest at suse dot de
- To: jh at suse dot cz, gcc-regression at gcc dot gnu dot org
- Date: Tue, 21 Dec 2004 01:06:32 +0000
- Subject: Some aspect of GCC memory consumption increased by recent patch
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 needed: 24585k -> 24589k
Peak memory use before GGC: 9310k
Peak memory use after GGC: 8624k
Maximum of released memory in single GGC run: 2912k
Garbage: 42437k
Leak: 6118k
Overhead: 5748k
GGC runs: 353
comparing combine.c compilation at -O1 level:
Overall memory allocated via mmap and sbrk increased from 25501k to 25545k, overall 0.17%
Overall memory needed: 25501k -> 25545k
Peak memory use before GGC: 9198k -> 9195k
Peak memory use after GGC: 8720k -> 8693k
Maximum of released memory in single GGC run: 2060k
Garbage: 68015k -> 67948k
Leak: 6514k
Overhead: 10697k -> 10667k
GGC runs: 546 -> 545
comparing combine.c compilation at -O2 level:
Overall memory needed: 29085k -> 29093k
Peak memory use before GGC: 12705k
Peak memory use after GGC: 12578k
Maximum of released memory in single GGC run: 2574k
Garbage: 82230k -> 81919k
Leak: 6331k
Overhead: 14853k -> 14545k
GGC runs: 548 -> 546
comparing combine.c compilation at -O3 level:
Amount of memory still referenced at the end of compilation increased from 6862k to 6878k, overall 0.23%
Overall memory needed: 31485k -> 31425k
Peak memory use before GGC: 12987k
Peak memory use after GGC: 12578k
Maximum of released memory in single GGC run: 3407k
Garbage: 111193k -> 110898k
Leak: 6862k -> 6878k
Overhead: 19934k -> 19547k
GGC runs: 615 -> 613
comparing insn-attrtab.c compilation at -O0 level:
Overall memory needed: 118468k
Peak memory use before GGC: 79387k
Peak memory use after GGC: 46137k
Maximum of released memory in single GGC run: 43336k
Garbage: 161992k
Leak: 10634k
Overhead: 21268k
GGC runs: 296
comparing insn-attrtab.c compilation at -O1 level:
Overall memory needed: 129456k
Peak memory use before GGC: 83952k
Peak memory use after GGC: 70040k
Maximum of released memory in single GGC run: 41104k
Garbage: 445757k -> 445755k
Leak: 10981k
Overhead: 79321k -> 79320k
GGC runs: 430
comparing insn-attrtab.c compilation at -O2 level:
Overall memory needed: 154068k -> 153936k
Peak memory use before GGC: 99964k
Peak memory use after GGC: 85453k
Maximum of released memory in single GGC run: 42091k
Garbage: 492648k -> 492647k
Leak: 10901k
Overhead: 87595k -> 87595k
GGC runs: 364
comparing insn-attrtab.c compilation at -O3 level:
Overall memory needed: 153924k -> 153928k
Peak memory use before GGC: 99966k
Peak memory use after GGC: 85455k
Maximum of released memory in single GGC run: 42091k
Garbage: 493911k -> 493910k
Leak: 10944k
Overhead: 87759k -> 87758k
GGC runs: 372
comparing Gerald's testcase PR8361 compilation at -O0 level:
Overall memory needed: 113408k
Peak memory use before GGC: 89921k
Peak memory use after GGC: 89026k
Maximum of released memory in single GGC run: 19894k
Garbage: 248878k
Leak: 57792k
Overhead: 45384k
GGC runs: 361
comparing Gerald's testcase PR8361 compilation at -O1 level:
Overall memory allocated via mmap and sbrk increased from 95920k to 106225k, overall 10.74%
Amount of produced GGC garbage decreased from 552582k to 483979k, overall -14.17%
Overall memory needed: 95920k -> 106225k
Peak memory use before GGC: 88923k
Peak memory use after GGC: 87937k
Maximum of released memory in single GGC run: 19408k
Garbage: 552582k -> 483979k
Leak: 59824k -> 59823k
Overhead: 115280k -> 71065k
GGC runs: 610 -> 550
comparing Gerald's testcase PR8361 compilation at -O2 level:
Overall memory allocated via mmap and sbrk increased from 95920k to 106285k, overall 10.81%
Amount of produced GGC garbage decreased from 601773k to 518760k, overall -16.00%
Overall memory needed: 95920k -> 106285k
Peak memory use before GGC: 88923k
Peak memory use after GGC: 87938k
Maximum of released memory in single GGC run: 19408k
Garbage: 601773k -> 518760k
Leak: 60405k -> 60405k
Overhead: 137310k -> 80693k
GGC runs: 651 -> 590
comparing Gerald's testcase PR8361 compilation at -O3 level:
Overall memory allocated via mmap and sbrk increased from 98572k to 107933k, overall 9.50%
Amount of produced GGC garbage decreased from 641712k to 539487k, overall -18.95%
Overall memory needed: 98572k -> 107933k
Peak memory use before GGC: 90319k
Peak memory use after GGC: 88770k
Maximum of released memory in single GGC run: 20096k
Garbage: 641712k -> 539487k
Leak: 60736k -> 60737k
Overhead: 148688k -> 82417k
GGC runs: 645 -> 575
Head of changelog is:
--- /usr/src/SpecTests/sandbox-britten-memory/x86_64/mem-result/ChangeLog 2004-12-20 12:59:20.000000000 +0000
+++ /usr/src/SpecTests/sandbox-britten-memory/gcc/gcc/ChangeLog 2004-12-21 00:05:16.000000000 +0000
@@ -1,3 +1,37 @@
+2004-12-20 Eric Botcazou <ebotcazou@libertysurf.fr>
+
+ PR middle-end/18776
+ * expr.c (write_complex_part): Use a subreg if the original object
+ is a hard reg that spans an even number of regs or a MEM.
+ (read_complex_part): Likewise.
+
+2004-12-20 Mark Mitchell <mark@codesourcery.com>
+
+ * Makefile.in (install-common): Do not install the specs file
+ unless SPECS is defined.
+
+2004-12-20 Zdenek Dvorak <dvorakz@suse.cz>
+
+ PR rtl-optimization/18942
+ * simplify-rtx.c (simplify_relational_operation_1): Simplify
+ x + cst1 == cst2 to x == cst2 - cst1. Made static.
+
+2004-12-20 Matt Austern <austern@apple.com>
+
+ PR c++/19044
+ * c-common.c (set_builtin_user_assembler_name): New.
+ * c-common.h (set_builtin_user_assembler_name): Declare.
+ * c-decl.c (finish_decl): Use set_builtin_user_assembler_name
+
+2004-12-20 Diego Novillo <dnovillo@redhat.com>
+
+ PR tree-optimization/19080
+
+ Merge from tree-cleanup-branch
+
+ * tree-flow-inline.h (clear_call_clobbered): New function.
+ * tree-ssa-alias.c (init_alias_info): Call it.
+
2004-12-20 Richard Henderson <rth@redhat.com>
* config/i386/i386.md (sse2_movsd): Remove ix86_binary_operator_ok
--- /usr/src/SpecTests/sandbox-britten-memory/x86_64/mem-result/ChangeLog.cp 2004-12-20 02:27:51.000000000 +0000
+++ /usr/src/SpecTests/sandbox-britten-memory/gcc/gcc/cp/ChangeLog 2004-12-21 00:05:18.000000000 +0000
@@ -1,3 +1,8 @@
+2004-12-20 Matt Austern <austern@apple.com>
+
+ PR c++/19044
+ * decl.c (make_rtl_for_nonlocal_decl): Use set_builtin_user_assembler_name
+
2004-12-19 Mark Mitchell <mark@codesourcery.com>
* cp-tree.h (note_decl_for_pch): New function.
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.