This is the mail archive of the gcc-regression@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

A recent patch decreased GCC's memory consumption.


Hi,

I am a friendly script caring about memory consumption in GCC.  Please
contact jh@suse.cz if something is going wrong.

Comparing memory consumption on compilation of combine.i, insn-attrtab.i,
and generate-3.4.ii I got:


comparing combine.c compilation at -O0 level:
    Overall memory needed: 24641k -> 24637k
    Peak memory use before GGC: 9586k
    Peak memory use after GGC: 8934k
    Maximum of released memory in single GGC run: 2754k
    Garbage: 41358k
    Leak: 6696k
    Overhead: 5818k
    GGC runs: 321

comparing combine.c compilation at -O1 level:
    Overall memory needed: 26808k
    Peak memory use before GGC: 17359k
    Peak memory use after GGC: 17170k
    Maximum of released memory in single GGC run: 2402k
    Garbage: 66308k
    Leak: 7186k
    Overhead: 7737k
    GGC runs: 415

comparing combine.c compilation at -O2 level:
    Overall memory needed: 26808k
    Peak memory use before GGC: 17360k
    Peak memory use after GGC: 17170k
    Maximum of released memory in single GGC run: 2503k
    Garbage: 84750k
    Leak: 7426k
    Overhead: 10298k
    GGC runs: 482

comparing combine.c compilation at -O3 level:
  Ovarall memory allocated via mmap and sbrk decreased from 27916k to 26808k, overall -4.13%
  Peak amount of GGC memory allocated before garbage collecting run decreased from 18805k to 17609k, overall -6.79%
  Amount of produced GGC garbage decreased from 127918k to 115642k, overall -10.62%
  Amount of memory still referenced at the end of compilation decreased from 7803k to 7496k, overall -4.09%
    Overall memory needed: 27916k -> 26808k
    Peak memory use before GGC: 18805k -> 17609k
    Peak memory use after GGC: 17170k
    Maximum of released memory in single GGC run: 5454k -> 3332k
    Garbage: 127918k -> 115642k
    Leak: 7803k -> 7496k
    Overhead: 14863k -> 13503k
    GGC runs: 535 -> 534

comparing insn-attrtab.c compilation at -O0 level:
    Overall memory needed: 81640k
    Peak memory use before GGC: 70259k
    Peak memory use after GGC: 44999k
    Maximum of released memory in single GGC run: 37046k
    Garbage: 149918k
    Leak: 11332k
    Overhead: 19106k
    GGC runs: 254

comparing insn-attrtab.c compilation at -O1 level:
    Overall memory needed: 114904k -> 114504k
    Peak memory use before GGC: 96260k
    Peak memory use after GGC: 85581k
    Maximum of released memory in single GGC run: 32777k
    Garbage: 295377k
    Leak: 11556k
    Overhead: 36053k
    GGC runs: 251

comparing insn-attrtab.c compilation at -O2 level:
    Overall memory needed: 125220k
    Peak memory use before GGC: 115308k
    Peak memory use after GGC: 85547k
    Maximum of released memory in single GGC run: 32461k
    Garbage: 381553k
    Leak: 11670k
    Overhead: 47864k
    GGC runs: 280

comparing insn-attrtab.c compilation at -O3 level:
    Overall memory needed: 125236k
    Peak memory use before GGC: 115331k -> 115333k
    Peak memory use after GGC: 85570k -> 85572k
    Maximum of released memory in single GGC run: 32787k -> 32774k
    Garbage: 382209k -> 382072k
    Leak: 11692k -> 11686k
    Overhead: 47930k -> 47917k
    GGC runs: 283 -> 282

comparing Gerald's testcase PR8361 compilation at -O0 level:
    Overall memory needed: 118624k
    Peak memory use before GGC: 95333k
    Peak memory use after GGC: 94386k
    Maximum of released memory in single GGC run: 20160k
    Garbage: 224282k
    Leak: 49424k
    Overhead: 36953k
    GGC runs: 341

comparing Gerald's testcase PR8361 compilation at -O1 level:
    Overall memory needed: 106072k
    Peak memory use before GGC: 94843k
    Peak memory use after GGC: 93892k
    Maximum of released memory in single GGC run: 19796k
    Garbage: 597266k
    Leak: 55674k
    Overhead: 69424k
    GGC runs: 521

comparing Gerald's testcase PR8361 compilation at -O2 level:
    Overall memory needed: 106064k
    Peak memory use before GGC: 94843k
    Peak memory use after GGC: 93893k
    Maximum of released memory in single GGC run: 19796k
    Garbage: 679910k
    Leak: 56611k
    Overhead: 80631k
    GGC runs: 584

comparing Gerald's testcase PR8361 compilation at -O3 level:
    Overall memory needed: 108464k
    Peak memory use before GGC: 95603k
    Peak memory use after GGC: 94656k
    Maximum of released memory in single GGC run: 19694k
    Garbage: 726345k -> 726348k
    Leak: 56846k -> 56843k
    Overhead: 84354k -> 84354k
    GGC runs: 592

Head of the ChangeLog is:

--- /usr/src/SpecTests/sandbox-britten-memory/x86_64/mem-result/ChangeLog	2005-07-19 14:55:06.000000000 +0000
+++ /usr/src/SpecTests/sandbox-britten-memory/gcc/gcc/ChangeLog	2005-07-19 19:28:59.000000000 +0000
@@ -1,3 +1,15 @@
+2005-07-19  Danny Berlin <dberlin@dberlin.org>
+	    Kenneth Zadeck <zadeck@naturalbridge.com>
+
+	* Makefile.in: Removed tree-promote-statics.c
+	* tree-promote-statics.c: Removed.
+	* common.opt: Removed flag-promote-statics.
+	* opts.c: Ditto.
+	* passes.c: Removed tree-promote-statics pass. 
+	* tree-pass.h: Ditto.
+	* timevar.def: Removed TV_PROMOTE_STATICS.
+
+
 2005-07-19  Gerald Pfeifer  <gerald@pfeifer.com>
 
 	* config.gcc: Add support for *-*-freebsd7, *-*-freebsd8,
@@ -24,6 +36,11 @@
 	* config/i386/i386.md (lrint<mode>2): Use temporary
 	instead of clobbering non-existent memory.
 
+2005-07-19  Nick Clifton  <nickc@redhat.com>
+
+	* config/avr/avr.c (legitimate_address_p): Fix debugging print
+	statement to avoid displaying ASCII control characters.
+
 2005-07-19  Ben Elliston  <bje@au.ibm.com>
 
 	* bt-load.c (link_btr_uses): Fix uninitialised warnings.


The results can be reproduced by building a 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.

Your testing script.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]