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 increased 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 empty function compilation at -O0 level:
  Overall memory allocated via mmap and sbrk increased from 7499k to 8332k, overall 11.11%
  Peak amount of GGC memory allocated before garbage collecting increased from 2354k to 3323k, overall 41.16%
  Peak amount of GGC memory still allocated after garbage collecting increased from 2019k to 2988k, overall 47.99%
  Amount of memory still referenced at the end of compilation increased from 2282k to 3720k, overall 63.00%
    Overall memory needed: 7499k -> 8332k
    Peak memory use before GGC: 2354k -> 3323k
    Peak memory use after GGC: 2019k -> 2988k
    Maximum of released memory in single GGC run: 335k
    Garbage: 490k
    Leak: 2282k -> 3720k
    Overhead: 404k -> 874k
    GGC runs: 3

comparing empty function compilation at -O0 -g level:
  Overall memory allocated via mmap and sbrk increased from 7515k to 8348k, overall 11.08%
  Peak amount of GGC memory allocated before garbage collecting increased from 2382k to 3351k, overall 40.68%
  Peak amount of GGC memory still allocated after garbage collecting increased from 2047k to 3015k, overall 47.29%
  Amount of memory still referenced at the end of compilation increased from 2315k to 3752k, overall 62.10%
    Overall memory needed: 7515k -> 8348k
    Peak memory use before GGC: 2382k -> 3351k
    Peak memory use after GGC: 2047k -> 3015k
    Maximum of released memory in single GGC run: 335k -> 336k
    Garbage: 492k
    Leak: 2315k -> 3752k
    Overhead: 409k -> 878k
    GGC runs: 3

comparing empty function compilation at -O1 level:
  Overall memory allocated via mmap and sbrk increased from 7551k to 8384k, overall 11.03%
  Peak amount of GGC memory allocated before garbage collecting increased from 2354k to 3323k, overall 41.16%
  Peak amount of GGC memory still allocated after garbage collecting increased from 2019k to 2988k, overall 47.99%
  Amount of memory still referenced at the end of compilation increased from 2284k to 3722k, overall 62.93%
    Overall memory needed: 7551k -> 8384k
    Peak memory use before GGC: 2354k -> 3323k
    Peak memory use after GGC: 2019k -> 2988k
    Maximum of released memory in single GGC run: 335k
    Garbage: 495k
    Leak: 2284k -> 3722k
    Overhead: 405k -> 874k
    GGC runs: 3

comparing empty function compilation at -O2 level:
  Overall memory allocated via mmap and sbrk increased from 7567k to 8396k, overall 10.96%
  Peak amount of GGC memory allocated before garbage collecting increased from 2355k to 3323k, overall 41.10%
  Peak amount of GGC memory still allocated after garbage collecting increased from 2019k to 2988k, overall 47.99%
  Amount of memory still referenced at the end of compilation increased from 2284k to 3722k, overall 62.92%
    Overall memory needed: 7567k -> 8396k
    Peak memory use before GGC: 2355k -> 3323k
    Peak memory use after GGC: 2019k -> 2988k
    Maximum of released memory in single GGC run: 336k -> 335k
    Garbage: 498k
    Leak: 2284k -> 3722k
    Overhead: 406k -> 875k
    GGC runs: 4 -> 3

comparing empty function compilation at -O3 level:
  Overall memory allocated via mmap and sbrk increased from 7567k to 8396k, overall 10.96%
  Peak amount of GGC memory allocated before garbage collecting increased from 2355k to 3323k, overall 41.10%
  Peak amount of GGC memory still allocated after garbage collecting increased from 2019k to 2988k, overall 47.99%
  Amount of memory still referenced at the end of compilation increased from 2284k to 3722k, overall 62.92%
    Overall memory needed: 7567k -> 8396k
    Peak memory use before GGC: 2355k -> 3323k
    Peak memory use after GGC: 2019k -> 2988k
    Maximum of released memory in single GGC run: 336k -> 335k
    Garbage: 498k
    Leak: 2284k -> 3722k
    Overhead: 406k -> 875k
    GGC runs: 4 -> 3

comparing combine.c compilation at -O0 level:
  Overall memory allocated via mmap and sbrk increased from 22423k to 23396k, overall 4.34%
  Peak amount of GGC memory allocated before garbage collecting increased from 9019k to 9967k, overall 10.51%
  Peak amount of GGC memory still allocated after garbage collecting increased from 8221k to 9189k, overall 11.77%
  Amount of memory still referenced at the end of compilation increased from 6892k to 8337k, overall 20.97%
    Overall memory needed: 22423k -> 23396k
    Peak memory use before GGC: 9019k -> 9967k
    Peak memory use after GGC: 8221k -> 9189k
    Maximum of released memory in single GGC run: 1584k -> 1907k
    Garbage: 38291k -> 38260k
    Leak: 6892k -> 8337k
    Overhead: 4869k -> 5338k
    GGC runs: 280 -> 243

comparing combine.c compilation at -O0 -g level:
  Overall memory allocated via mmap and sbrk increased from 24351k to 25300k, overall 3.90%
  Peak amount of GGC memory allocated before garbage collecting increased from 10755k to 11631k, overall 8.15%
  Peak amount of GGC memory still allocated after garbage collecting increased from 10006k to 10975k, overall 9.68%
  Amount of memory still referenced at the end of compilation increased from 9785k to 11230k, overall 14.78%
    Overall memory needed: 24351k -> 25300k
    Peak memory use before GGC: 10755k -> 11631k
    Peak memory use after GGC: 10006k -> 10975k
    Maximum of released memory in single GGC run: 1879k -> 1880k
    Garbage: 38623k -> 38640k
    Leak: 9785k -> 11230k
    Overhead: 5574k -> 6044k
    GGC runs: 272 -> 241

comparing combine.c compilation at -O1 level:
  Overall memory allocated via mmap and sbrk increased from 35099k to 37100k, overall 5.70%
  Peak amount of GGC memory allocated before garbage collecting increased from 17670k to 18608k, overall 5.31%
  Peak amount of GGC memory still allocated after garbage collecting increased from 17481k to 18407k, overall 5.30%
  Amount of memory still referenced at the end of compilation increased from 6955k to 8393k, overall 20.67%
    Overall memory needed: 35099k -> 37100k
    Peak memory use before GGC: 17670k -> 18608k
    Peak memory use after GGC: 17481k -> 18407k
    Maximum of released memory in single GGC run: 1346k -> 1379k
    Garbage: 51740k -> 51724k
    Leak: 6955k -> 8393k
    Overhead: 5979k -> 6447k
    GGC runs: 358 -> 317

comparing combine.c compilation at -O2 level:
  Overall memory allocated via mmap and sbrk increased from 37023k to 39024k, overall 5.40%
  Peak amount of GGC memory allocated before garbage collecting increased from 17690k to 18649k, overall 5.42%
  Peak amount of GGC memory still allocated after garbage collecting increased from 17495k to 18464k, overall 5.54%
  Amount of memory still referenced at the end of compilation increased from 7069k to 8519k, overall 20.50%
    Overall memory needed: 37023k -> 39024k
    Peak memory use before GGC: 17690k -> 18649k
    Peak memory use after GGC: 17495k -> 18464k
    Maximum of released memory in single GGC run: 1348k -> 1356k
    Garbage: 68454k -> 68457k
    Leak: 7069k -> 8519k
    Overhead: 8105k -> 8577k
    GGC runs: 423 -> 373

comparing combine.c compilation at -O3 level:
  Overall memory allocated via mmap and sbrk increased from 43475k to 44480k, overall 2.31%
  Peak amount of GGC memory allocated before garbage collecting increased from 17891k to 18802k, overall 5.09%
  Peak amount of GGC memory still allocated after garbage collecting increased from 17634k to 18599k, overall 5.47%
  Amount of memory still referenced at the end of compilation increased from 7185k to 8631k, overall 20.12%
    Overall memory needed: 43475k -> 44480k
    Peak memory use before GGC: 17891k -> 18802k
    Peak memory use after GGC: 17634k -> 18599k
    Maximum of released memory in single GGC run: 3637k -> 3718k
    Garbage: 93745k -> 93707k
    Leak: 7185k -> 8631k
    Overhead: 11397k -> 11866k
    GGC runs: 449 -> 404

comparing insn-attrtab.c compilation at -O0 level:
  Peak amount of GGC memory allocated before garbage collecting increased from 59808k to 60776k, overall 1.62%
  Peak amount of GGC memory still allocated after garbage collecting increased from 32814k to 33782k, overall 2.95%
  Amount of memory still referenced at the end of compilation increased from 9744k to 11182k, overall 14.75%
    Overall memory needed: 142087k -> 143064k
    Peak memory use before GGC: 59808k -> 60776k
    Peak memory use after GGC: 32814k -> 33782k
    Maximum of released memory in single GGC run: 34625k -> 34624k
    Garbage: 131974k -> 131970k
    Leak: 9744k -> 11182k
    Overhead: 14266k -> 14735k
    GGC runs: 219 -> 186

comparing insn-attrtab.c compilation at -O0 -g level:
  Peak amount of GGC memory allocated before garbage collecting increased from 60970k to 61938k, overall 1.59%
  Peak amount of GGC memory still allocated after garbage collecting increased from 33975k to 34943k, overall 2.85%
  Amount of memory still referenced at the end of compilation increased from 11212k to 12890k, overall 14.96%
    Overall memory needed: 143331k -> 144332k
    Peak memory use before GGC: 60970k -> 61938k
    Peak memory use after GGC: 33975k -> 34943k
    Maximum of released memory in single GGC run: 34626k -> 34625k
    Garbage: 132465k -> 132202k
    Leak: 11212k -> 12890k
    Overhead: 14664k -> 15133k
    GGC runs: 218 -> 191

comparing insn-attrtab.c compilation at -O1 level:
  Peak amount of GGC memory allocated before garbage collecting increased from 58038k to 59006k, overall 1.67%
  Peak amount of GGC memory still allocated after garbage collecting increased from 53813k to 54779k, overall 1.80%
  Amount of memory still referenced at the end of compilation increased from 9630k to 11068k, overall 14.93%
    Overall memory needed: 153355k -> 153508k
    Peak memory use before GGC: 58038k -> 59006k
    Peak memory use after GGC: 53813k -> 54779k
    Maximum of released memory in single GGC run: 23623k
    Garbage: 215877k -> 215879k
    Leak: 9630k -> 11068k
    Overhead: 24621k -> 25091k
    GGC runs: 247 -> 217

comparing insn-attrtab.c compilation at -O2 level:
  Peak amount of GGC memory allocated before garbage collecting increased from 57917k to 58885k, overall 1.67%
  Peak amount of GGC memory still allocated after garbage collecting increased from 53902k to 54870k, overall 1.80%
  Amount of memory still referenced at the end of compilation increased from 9620k to 11057k, overall 14.94%
    Overall memory needed: 193179k -> 194232k
    Peak memory use before GGC: 57917k -> 58885k
    Peak memory use after GGC: 53902k -> 54870k
    Maximum of released memory in single GGC run: 21313k -> 21314k
    Garbage: 248440k -> 248433k
    Leak: 9620k -> 11057k
    Overhead: 29973k -> 30441k
    GGC runs: 271 -> 240

comparing insn-attrtab.c compilation at -O3 level:
  Peak amount of GGC memory allocated before garbage collecting increased from 69164k to 70104k, overall 1.36%
  Peak amount of GGC memory still allocated after garbage collecting increased from 64704k to 65673k, overall 1.50%
  Amount of memory still referenced at the end of compilation increased from 9630k to 11068k, overall 14.93%
    Overall memory needed: 196963k -> 198004k
    Peak memory use before GGC: 69164k -> 70104k
    Peak memory use after GGC: 64704k -> 65673k
    Maximum of released memory in single GGC run: 22922k -> 22921k
    Garbage: 277526k -> 277483k
    Leak: 9630k -> 11068k
    Overhead: 31698k -> 32160k
    GGC runs: 274 -> 242

comparing Gerald's testcase PR8361 compilation at -O0 level:
  Peak amount of GGC memory allocated before garbage collecting increased from 88872k to 89734k, overall 0.97%
  Peak amount of GGC memory still allocated after garbage collecting increased from 87992k to 88846k, overall 0.97%
  Amount of memory still referenced at the end of compilation increased from 51076k to 52513k, overall 2.81%
    Overall memory needed: 153453k -> 154244k
    Peak memory use before GGC: 88872k -> 89734k
    Peak memory use after GGC: 87992k -> 88846k
    Maximum of released memory in single GGC run: 17970k -> 17863k
    Garbage: 208365k -> 208375k
    Leak: 51076k -> 52513k
    Overhead: 24076k -> 24545k
    GGC runs: 410 -> 399

comparing Gerald's testcase PR8361 compilation at -O0 -g level:
  Peak amount of GGC memory allocated before garbage collecting increased from 101492k to 102488k, overall 0.98%
  Peak amount of GGC memory still allocated after garbage collecting increased from 100487k to 101473k, overall 0.98%
  Amount of memory still referenced at the end of compilation increased from 74385k to 75823k, overall 1.93%
    Overall memory needed: 173921k -> 174640k
    Peak memory use before GGC: 101492k -> 102488k
    Peak memory use after GGC: 100487k -> 101473k
    Maximum of released memory in single GGC run: 18337k -> 18365k
    Garbage: 214057k -> 214040k
    Leak: 74385k -> 75823k
    Overhead: 29972k -> 30441k
    GGC runs: 381 -> 374

comparing Gerald's testcase PR8361 compilation at -O1 level:
  Peak amount of GGC memory allocated before garbage collecting increased from 100334k to 101386k, overall 1.05%
  Peak amount of GGC memory still allocated after garbage collecting increased from 99339k to 100381k, overall 1.05%
  Amount of memory still referenced at the end of compilation increased from 52214k to 53628k, overall 2.71%
    Overall memory needed: 149743k -> 150850k
    Peak memory use before GGC: 100334k -> 101386k
    Peak memory use after GGC: 99339k -> 100381k
    Maximum of released memory in single GGC run: 17358k -> 17436k
    Garbage: 336358k -> 336350k
    Leak: 52214k -> 53628k
    Overhead: 30953k -> 31419k
    GGC runs: 526 -> 511

comparing Gerald's testcase PR8361 compilation at -O2 level:
  Peak amount of GGC memory allocated before garbage collecting increased from 100724k to 101759k, overall 1.03%
  Peak amount of GGC memory still allocated after garbage collecting increased from 99723k to 100750k, overall 1.03%
  Amount of memory still referenced at the end of compilation increased from 52933k to 54371k, overall 2.72%
    Overall memory needed: 158627k -> 159582k
    Peak memory use before GGC: 100724k -> 101759k
    Peak memory use after GGC: 99723k -> 100750k
    Maximum of released memory in single GGC run: 17358k -> 17436k
    Garbage: 384157k -> 384144k
    Leak: 52933k -> 54371k
    Overhead: 36579k -> 37046k
    GGC runs: 578 -> 562

comparing Gerald's testcase PR8361 compilation at -O3 level:
  Peak amount of GGC memory allocated before garbage collecting increased from 102751k to 103720k, overall 0.94%
  Peak amount of GGC memory still allocated after garbage collecting increased from 101724k to 102692k, overall 0.95%
  Amount of memory still referenced at the end of compilation increased from 53185k to 54623k, overall 2.70%
    Overall memory needed: 159991k -> 161070k
    Peak memory use before GGC: 102751k -> 103720k
    Peak memory use after GGC: 101724k -> 102692k
    Maximum of released memory in single GGC run: 17782k -> 17798k
    Garbage: 416336k -> 416230k
    Leak: 53185k -> 54623k
    Overhead: 39163k -> 39619k
    GGC runs: 598 -> 587

comparing PR rtl-optimization/28071 testcase compilation at -O0 level:
  Peak amount of GGC memory allocated before garbage collecting increased from 102418k to 103386k, overall 0.95%
  Peak amount of GGC memory still allocated after garbage collecting increased from 58071k to 59040k, overall 1.67%
  Amount of memory still referenced at the end of compilation increased from 7439k to 8877k, overall 19.33%
    Overall memory needed: 386599k -> 387574k
    Peak memory use before GGC: 102418k -> 103386k
    Peak memory use after GGC: 58071k -> 59040k
    Maximum of released memory in single GGC run: 50583k -> 50582k
    Garbage: 179454k -> 179453k
    Leak: 7439k -> 8877k
    Overhead: 30910k -> 31379k
    GGC runs: 76 -> 64

comparing PR rtl-optimization/28071 testcase compilation at -O0 -g level:
  Peak amount of GGC memory allocated before garbage collecting increased from 103064k to 104033k, overall 0.94%
  Peak amount of GGC memory still allocated after garbage collecting increased from 58718k to 59686k, overall 1.65%
  Amount of memory still referenced at the end of compilation increased from 9207k to 10645k, overall 15.62%
    Overall memory needed: 387399k -> 388378k
    Peak memory use before GGC: 103064k -> 104033k
    Peak memory use after GGC: 58718k -> 59686k
    Maximum of released memory in single GGC run: 50582k -> 50583k
    Garbage: 179559k -> 179509k
    Leak: 9207k -> 10645k
    Overhead: 31406k -> 31875k
    GGC runs: 84 -> 72

comparing PR rtl-optimization/28071 testcase compilation at -O1 level:
  Peak amount of GGC memory allocated before garbage collecting increased from 83384k to 84352k, overall 1.16%
  Peak amount of GGC memory still allocated after garbage collecting increased from 74918k to 75887k, overall 1.29%
  Amount of memory still referenced at the end of compilation increased from 20874k to 22312k, overall 6.89%
    Overall memory needed: 306531k -> 311684k
    Peak memory use before GGC: 83384k -> 84352k
    Peak memory use after GGC: 74918k -> 75887k
    Maximum of released memory in single GGC run: 39404k -> 39403k
    Garbage: 232481k -> 232481k
    Leak: 20874k -> 22312k
    Overhead: 31637k -> 32107k
    GGC runs: 81 -> 71

comparing PR rtl-optimization/28071 testcase compilation at -O2 level:
  Peak amount of GGC memory allocated before garbage collecting increased from 79904k to 80873k, overall 1.21%
  Peak amount of GGC memory still allocated after garbage collecting increased from 74918k to 75887k, overall 1.29%
  Amount of memory still referenced at the end of compilation increased from 20954k to 22392k, overall 6.86%
    Overall memory needed: 319363k -> 320064k
    Peak memory use before GGC: 79904k -> 80873k
    Peak memory use after GGC: 74918k -> 75887k
    Maximum of released memory in single GGC run: 33013k -> 33014k
    Garbage: 240335k -> 240335k
    Leak: 20954k -> 22392k
    Overhead: 33872k -> 34341k
    GGC runs: 92 -> 81

comparing PR rtl-optimization/28071 testcase compilation at -O3 -fno-tree-pre -fno-tree-fre level:
  Peak amount of GGC memory allocated before garbage collecting increased from 183658k to 184627k, overall 0.53%
  Peak amount of GGC memory still allocated after garbage collecting increased from 171174k to 172143k, overall 0.57%
  Amount of memory still referenced at the end of compilation increased from 46332k to 47770k, overall 3.10%
    Overall memory needed: 1031399k -> 1032492k
    Peak memory use before GGC: 183658k -> 184627k
    Peak memory use after GGC: 171174k -> 172143k
    Maximum of released memory in single GGC run: 80733k
    Garbage: 348761k -> 348761k
    Leak: 46332k -> 47770k
    Overhead: 44716k -> 45186k
    GGC runs: 75 -> 66

Head of the ChangeLog is:

--- /usr/src/SpecTests/sandbox-britten-memory/x86_64/mem-result/ChangeLog	2007-06-25 10:52:00.000000000 +0000
+++ /usr/src/SpecTests/sandbox-britten-memory/gcc/gcc/ChangeLog	2007-06-26 02:59:58.000000000 +0000
@@ -1,3 +1,78 @@
+2007-06-25  Andrew Pinski  <andrew_pinski@playstation.sony.com>
+
+	PR tree-opt/32421
+	* tree-vect-transform.c (vectorizable_operation): Convert
+	POINTER_PLUS_EXPR over to PLUS_EXPR.
+
+2007-06-25  Chao-ying Fu  <fu@mips.com>
+
+	* doc/rtl.texi (Machine Modes): Document QQ, HQ, SQ, DQ, TQ,
+	UQQ, UHQ, USQ, UDQ, UTQ, HA, SA, DA, TA, UHA, USA, UDA, and UTAmodes.
+	Document MODE_FRACT, MODE_UFRACT, MODE_ACCUM, MODE_UACCUM.
+	Document GET_MODE_IBIT, and GET_MODE_FBIT.
+
+	* machmode.h (VECTOR_MODE_P): Test MODE_VECTOR_FRACT,
+	MODE_VECTOR_UFRACT, MODE_VECTOR_ACCUM, MODE_VECTOR_UACCUM.
+	(SCALAR_FRACT_MODE_P, SCALAR_UFRACT_MODE_P, ALL_SCALAR_FRACT_MODE_P,
+	SCALAR_ACCUM_MODE_P, SCALAR_UACCUM_MODE_P, ALL_SCALAR_ACCUM_MODE_P,
+	SIGNED_SCALAR_FIXED_POINT_MODE_P, UNSIGNED_SCALAR_FIXED_POINT_MODE_P,
+	ALL_SCALAR_FIXED_POINT_MODE_P, FRACT_MODE_P, UFRACT_MODE_P,
+	ALL_FRACT_MODE_P, ACCUM_MODE_P, UACCUM_MODE_P, ALL_ACCUM_MODE_P,
+	SIGNED_FIXED_POINT_MODE_P, UNSIGNED_FIXED_POINT_MODE_P,
+	ALL_FIXED_POINT_MODE_P): New define.
+	(CLASS_HAS_WIDER_MODES_P): Test MODE_FRACT, MODE_UFRACT, MODE_ACCUM,
+	MODE_UACCUM.
+	(GET_MODE_IBIT, GET_MODE_FBIT): New define.
+
+	* mode-classes.def (MODE_FRACT, MODE_UFRACT, MODE_ACCUM, MODE_UACCUM,
+	MODE_VECTOR_FRACT, MODE_VECTOR_UFRACT, MODE_VECTOR_ACCUM,
+	MODE_VECTOR_UACCUM): New mode classes.
+
+	* machmode.def: Document FRACT_MODE, UFRACT_MODE, ACCUM_MODE,
+	UACCUM_MODE, ADJUST_IBIT, and ADJUST_FBIT.
+	Add QQ, HQ, SQ, DQ, TQ, UQQ, UHQ, USQ, UDQ, UTQ, HA, SA, DA, TA, UHA,
+	USA, UDA, and UTA.
+
+	* genmodes.c (struct mode_data): Add ibit and fbit fields.
+	(blank_mode): Initialize ibit and fbit.
+	(adj_ibit, adj_fbit): New to adjust ibit and fbit.
+	(vector_class): Handle MODE_FRACT, MODE_UFRACT, MODE_ACCUM, MODE_UACCUM.
+	(new_adjust): Change required_class to required_class_from and
+	required_class_to for testing within a range.
+	(complete_mode): Handle MODE_FRACT, MODE_UFRACT, MODE_ACCUM,
+	MODE_UACCUM, MODE_VECTOR_FRACT, MODE_VECTOR_UFRACT, MODE_VECTOR_ACCUM,
+	MODE_VECTOR_UACCUM.
+	(FRACT_MODE, UFRACT_MODE, ACCUM_MODE, UACCUM_MODE): New define.
+	(make_fixed_point_mode): New.
+	(_ADD_ADJUST): Change C to C1 and C2.
+	(ADJUST_BYTESIZE, ADJUST_ALIGNMENT, ADJUST_FLOAT_FORMAT): Change to
+	use a range for machine classes.
+	(ADJUST_IBIT, ADJUST_FBIT): New.
+	(emit_insn_modes_h): Output defines of CONST_MODE_IBIT and
+	CONST_MODE_FBIT.
+	(emit_mode_adjustments): Handle MODE_VECTOR_FRACT, MODE_VECTOR_UFRACT,
+	MODE_VECTOR_ACCUM, MODE_VECTOR_UACCUM.
+	Emit adjustment for ibit and fbit.
+	(emit_mode_ibit, emit_mode_fbit): New.
+	(emit_insn_modes_c): Add emit_mode_ibit and emit_mode_fbit.
+
+2007-06-25  Nathan Froyd  <froydnj@codesourcery.com>
+
+	* config/rs6000/spe.md (*frob_ti_tf_2): Specify an input_operand
+	as the source of the set.
+
+2007-06-25  Roman Zippel <zippel@linux-m68k.org>
+
+	* config/m68k/m68k.h (DATA_REGNO_P, ADDRESS_REGNO_P, INT_REGNO_P,
+	FP_REGNO_P): Use IN_RANGE.
+	(REGNO_OK_FOR_DATA_P, REGNO_OK_FOR_FP_P): Remove.
+	(REGNO_OK_FOR_INDEX_NONSTRICT_P, REGNO_OK_FOR_BASE_NONSTRICT_P): New.
+	(DATA_REG_P): Use DATA_REGNO_P.
+	(FP_REG_P): Use FP_REGNO_P.
+	(ADDRESS_REG_P): Use ADDRESS_REGNO_P.
+	* config/m68k/m68k.c (m68k_legitimate_base_reg_p): Use
+	REGNO_OK_FOR_INDEX_NONSTRICT_P, REGNO_OK_FOR_BASE_NONSTRICT_P.
+
 2007-06-24  Jan Hubicka  <jh@suse.cz>
 
 	PR middle-end/30563
--- /usr/src/SpecTests/sandbox-britten-memory/x86_64/mem-result/ChangeLog.cp	2007-06-24 02:40:52.000000000 +0000
+++ /usr/src/SpecTests/sandbox-britten-memory/gcc/gcc/cp/ChangeLog	2007-06-26 02:59:58.000000000 +0000
@@ -1,3 +1,9 @@
+2007-06-26  Simon Martin  <simartin@users.sourceforge.net>
+
+	PR c++/32111
+	* decl.c (grokdeclarator): Reset friendp for member functions declared
+	friend of their own class.
+
 2007-06-23  Mark Mitchell  <mark@codesourcery.com>
 
 	* decl2.c (determine_visibility): Don't look for dllexport here.


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]