|
ID
|
Sev
|
Pri
|
Host
|
Assignee
|
Status
|
Resolution
|
Summary
|
|
37273
|
nor
|
P2
|
|
|
UNCO
|
|
[4.4/4.5 Regression] IRA does not re-materializes address...
|
|
37490
|
nor
|
P2
|
|
|
UNCO
|
|
[4.4/4.5 Regression] IRA causes FP code to have more spills
|
|
34087
|
nor
|
P3
|
powerpc64-linux
|
|
UNCO
|
|
ICE in regrename.c for movdf_hardfloat64_mfpgpr
|
|
36539
|
enh
|
P3
|
|
|
UNCO
|
|
IRA doesn't account for earlyclobber asm conflicts
|
|
40842
|
nor
|
P3
|
powerpc64*-*-*
|
|
UNCO
|
|
Poor register class choice in IRA
|
|
42502
|
nor
|
P3
|
i686-linux
|
|
UNCO
|
|
Bad register allocation in a very simple code
|
|
42536
|
nor
|
P3
|
|
|
UNCO
|
|
[4.4/4.5 regression] ICE in spill_failure, at reload1.c:2141
|
|
38532
|
nor
|
P3
|
all
|
vmakarov@gcc.gnu.org
|
UNCO
|
|
dse broken for frame related stores
|
|
38603
|
nor
|
P4
|
x86_64-unknown-linux-gnu
|
|
UNCO
|
|
[4.4/4.5 Regression] IRA does not accommodate LOAD_EXTEND...
|
|
11488
|
nor
|
P2
|
|
|
NEW
|
|
Pre-regalloc scheduling severely worsens performance
|
|
11873
|
enh
|
P2
|
|
|
NEW
|
|
inefficient use of registers induces size and time overhead
|
|
12754
|
nor
|
P2
|
powerpc-apple-darwin
|
|
NEW
|
|
Faulty register allocation under certain circumstances
|
|
14060
|
enh
|
P2
|
i686-pc-linux-gnu
|
|
NEW
|
|
An unused register is saved to the stack.
|
|
16541
|
enh
|
P2
|
|
|
NEW
|
|
code quality issue for bit manipulations with 64bit
|
|
17838
|
enh
|
P2
|
|
|
NEW
|
|
spills are not re-used
|
|
19680
|
enh
|
P2
|
cygwin
|
|
NEW
|
|
sub-optimial register allocation with sse
|
|
20634
|
nor
|
P2
|
|
|
NEW
|
|
code fails to compile/code generation issue with -funit-a...
|
|
21150
|
nor
|
P2
|
|
|
NEW
|
|
Suboptimal byte extraction from 64bits
|
|
21182
|
nor
|
P2
|
|
|
NEW
|
|
gcc can use registers but uses stack instead
|
|
23813
|
nor
|
P2
|
|
|
NEW
|
|
redundant register assignments not eliminated
|
|
27855
|
nor
|
P2
|
|
|
NEW
|
|
[4.3/4.4/4.5 regression] reassociation causes the RA to b...
|
|
37272
|
nor
|
P2
|
|
|
NEW
|
|
[4.4/4.5 Regression] IRA has caused ppc64-double-1.c to fail
|
|
37534
|
nor
|
P2
|
powerpc64-linux
|
|
NEW
|
|
[4.4/4.5 Regression] IRA causes 17% degradation in 187.fa...
|
|
38666
|
nor
|
P2
|
x86_64-linux-gnu
|
|
NEW
|
|
[4.3 Regression] internal compiler error: Segmentation fa...
|
|
39871
|
nor
|
P2
|
x86_64-unknown-linux-gnu
|
|
NEW
|
|
[4.3/4.4/4.5 regression] Code size increase on ARM due to...
|
|
42258
|
nor
|
P2
|
i686-linux
|
|
NEW
|
|
[4.5 Regression] redundant register move around mul instr...
|
|
5739
|
nor
|
P3
|
|
|
NEW
|
|
"const" and "pure" function attributes pessimize code due...
|
|
26479
|
enh
|
P3
|
|
|
NEW
|
|
suboptimal register allocation for return register
|
|
26525
|
enh
|
P3
|
|
|
NEW
|
|
missed opportunity to use value in fp reg
|
|
32593
|
enh
|
P3
|
|
|
NEW
|
|
Missed optimization of 'y = constant - x' operation
|
|
32647
|
nor
|
P3
|
|
|
NEW
|
|
spill failures with hard-register variable
|
|
34011
|
nor
|
P3
|
|
|
NEW
|
|
Memory load is not eliminated from tight vectorized loop
|
|
35926
|
nor
|
P3
|
|
|
NEW
|
|
Pushing / Poping ebx without using it.
|
|
36680
|
nor
|
P3
|
x86_64-unknown-linux-gnu
|
|
NEW
|
|
ICE in spill_failure, reload1.c:1995
|
|
37488
|
nor
|
P3
|
i686-pc-linux-gnu
|
|
NEW
|
|
register allocation spills floats needlessly
|
|
41299
|
nor
|
P3
|
|
|
NEW
|
|
code size regression (from 4.3) with -Os and -mregparm=3
|
|
42500
|
nor
|
P3
|
i686-linux
|
|
NEW
|
|
Unnecessary mov of sp to a register
|
|
42575
|
nor
|
P3
|
i686-linux
|
|
NEW
|
|
arm-eabi-gcc 64-bit multiply weirdness
|
|
42592
|
enh
|
P3
|
x86_64-linux
|
|
NEW
|
|
really bad register allocation for x86
|
|
38711
|
nor
|
P3
|
all
|
vmakarov@gcc.gnu.org
|
NEW
|
|
ira should not be using df-lr except at -O1.
|
|
17863
|
nor
|
P4
|
|
|
NEW
|
|
[4.3/4.4/4.5 Regression] performance loss (TER register p...
|
|
24319
|
nor
|
P5
|
|
|
NEW
|
|
[4.3/4.4/4.5 regression] amd64 register spill error with ...
|
|
17381
|
enh
|
P2
|
|
pinskia@gcc.gnu.org
|
ASSI
|
|
Unnecessary register move for float extend
|
|
18316
|
enh
|
P2
|
|
rakdver@gcc.gnu.org
|
ASSI
|
|
Missed IV optimization
|
|
31827
|
nor
|
P3
|
hppa64-hp-hpux11.11
|
bonzini@gnu.org
|
ASSI
|
|
limits-exprparen.c: Pid 2297 received a SIGSEGV for stack...
|
|
4079
|
enh
|
P3
|
|
dje@gcc.gnu.org
|
SUSP
|
|
unnecessary register move on simple code
|
|
18617
|
nor
|
P2
|
|
|
WAIT
|
|
missed volatile variable optimizations
|
|
20972
|
nor
|
P3
|
|
|
WAIT
|
|
Register allocator/reload uses auto-inc register in non-a...
|
|
42499
|
nor
|
P3
|
i686-linux
|
|
WAIT
|
|
Bad register allocation in multiplication code by constant
|
|
38130
|
nor
|
P1
|
i686-pc-mingw32
|
jakub@gcc.gnu.org
|
RESO
|
FIXE
|
[4.4 regression]__builtin_alloca (vs IRA?) testsuite fai...
|
|
24160
|
nor
|
P1
|
|
rth@gcc.gnu.org
|
RESO
|
FIXE
|
[4.1 Regression] ICE with -O1 -ftree-vectorize -msse
|
|
37447
|
blo
|
P1
|
|
rth@gcc.gnu.org
|
RESO
|
FIXE
|
[4.4 Regression] test pr28982b.c fails execution on power...
|
|
27158
|
nor
|
P1
|
powerpc-linux-gnu
|
|
RESO
|
FIXE
|
[4.1/4.2 regression] ICE in extract_insn with -maltivec
|
|
32004
|
nor
|
P1
|
|
|
RESO
|
FIXE
|
[4.1/4.2/4.3 regression] : can't find a register in class...
|
|
34529
|
cri
|
P1
|
|
|
RESO
|
FIXE
|
[4.2 Regression] Wrong code with altivec stores and offsets
|
|
37422
|
maj
|
P1
|
i386-pc-solaris2.10
|
|
RESO
|
DUPL
|
[4.4 regression] IRA merge breaks Solaris/x86 bootstrap
|
|
37424
|
blo
|
P1
|
sparc-sun-solaris2.11
|
|
RESO
|
FIXE
|
[4.4 regression] IRA merge breaks Solaris/SPARC bootstrap
|
|
38495
|
maj
|
P1
|
|
|
RESO
|
FIXE
|
[4.4 Regression] ACATS tests cxa4004 cxa4005 cxa4026 fail
|
|
37884
|
cri
|
P1
|
s390x-ibm-linux
|
vmakarov@gcc.gnu.org
|
RESO
|
FIXE
|
[4.4 Regression] Bootstrap failure due to miscompilation ...
|
|
19653
|
min
|
P2
|
|
bonzini@gnu.org
|
RESO
|
FIXE
|
x87 reg allocated for constants for -mfpmath=sse
|
|
26778
|
nor
|
P2
|
|
bonzini@gnu.org
|
RESO
|
WONT
|
[4.0/4.1 regression] GCC4 moves the result of a condition...
|
|
27827
|
nor
|
P2
|
|
bonzini@gnu.org
|
RESO
|
FIXE
|
[4.0 Regression] gcc 4 produces worse x87 code on all pla...
|
|
21529
|
nor
|
P2
|
i486-pc-linux-gnu
|
rth@gcc.gnu.org
|
RESO
|
FIXE
|
[4.0/4.1 Regression] code size regression (+40%) with -Os...
|
|
19464
|
nor
|
P2
|
|
steven@gcc.gnu.org
|
RESO
|
FIXE
|
[3.3/3.4/4.0 Regression] gcse causes poor register alloca...
|
|
14121
|
enh
|
P2
|
|
|
RESO
|
FIXE
|
use inline assembler register constraints as home registe...
|
|
15431
|
enh
|
P2
|
|
|
RESO
|
FIXE
|
inefficient register allocation
|
|
15464
|
nor
|
P2
|
i686-pc-linux
|
|
RESO
|
FIXE
|
ICE: unable to find a register to spill in class `DREG'
|
|
15792
|
enh
|
P2
|
|
|
RESO
|
FIXE
|
missed subreg optimization
|
|
16185
|
min
|
P2
|
|
|
RESO
|
WORK
|
ICE: in spill_failure, at reload1.c:1892, global register...
|
|
18427
|
enh
|
P2
|
|
|
RESO
|
FIXE
|
[meta-bug] We need a better register allocator
|
|
18560
|
enh
|
P2
|
|
|
RESO
|
FIXE
|
better optimalization of EOR/MOV block.
|
|
18766
|
min
|
P2
|
|
|
RESO
|
FIXE
|
Inefficient code with -mfpmath=387,sse
|
|
19780
|
enh
|
P2
|
|
|
RESO
|
WONT
|
Floating point computation far slower for -mfpmath=sse
|
|
20945
|
nor
|
P2
|
|
|
RESO
|
DUPL
|
[4.0/4.1 Regresson] about 2x perfomance regression in com...
|
|
21202
|
nor
|
P2
|
|
|
RESO
|
FIXE
|
Extra register moves generated with long long
|
|
21596
|
min
|
P2
|
|
|
RESO
|
FIXE
|
[4.2 Regression] extra temporaries when using global regi...
|
|
21715
|
min
|
P2
|
|
|
RESO
|
FIXE
|
[4.0/4.1 regression] code-generation performance regression
|
|
22072
|
enh
|
P2
|
|
|
RESO
|
FIXE
|
bizarre code for int*int/2 for -Os
|
|
22453
|
nor
|
P2
|
|
|
RESO
|
DUPL
|
reload uses FLOAT_REGS class eagerly even if -mfpmath=sse
|
|
23100
|
enh
|
P2
|
|
|
RESO
|
FIXE
|
poor code generation for i686
|
|
23181
|
nor
|
P2
|
|
|
RESO
|
FIXE
|
[4.1 Regression] Slowdown of the bresenham line drawing b...
|
|
23322
|
nor
|
P2
|
|
|
RESO
|
FIXE
|
[4.3 regression] performance regression
|
|
23451
|
enh
|
P2
|
|
|
RESO
|
FIXE
|
Redundant reloading from stack frame on i386
|
|
23812
|
enh
|
P2
|
|
|
RESO
|
FIXE
|
swapping DImode halves produces poor x86 register allocation
|
|
24007
|
nor
|
P2
|
|
|
RESO
|
FIXE
|
very weird register allocation, putting a fp in the ctr r...
|
|
24082
|
enh
|
P2
|
|
|
RESO
|
FIXE
|
Not that good register allocating on powerpc for vectors
|
|
28481
|
nor
|
P2
|
i386-pc-linux-gnu
|
|
RESO
|
DUPL
|
[4.3/4.4/4.5 Regression] uses memory where it can use reg...
|
|
37364
|
nor
|
P2
|
|
|
RESO
|
FIXE
|
[4.4 Regression] IRA generates inefficient code due to mi...
|
|
37435
|
nor
|
P2
|
|
|
RESO
|
FIXE
|
[4.4 regression] ICE with volatiles
|
|
37437
|
nor
|
P2
|
|
|
RESO
|
FIXE
|
[4.4 regression] speed regression
|
|
37948
|
nor
|
P2
|
|
|
RESO
|
FIXE
|
[4.4 Regression] IRA generates slower code
|
|
38426
|
nor
|
P2
|
i686-pc-linux-gnu
|
|
RESO
|
FIXE
|
[4.4 Regression] Incorrect code produced with -momit-leaf...
|
|
38583
|
nor
|
P2
|
|
|
RESO
|
FIXE
|
huge test case makes register allocator run out of memory...
|
|
37669
|
nor
|
P3
|
|
nemet@gcc.gnu.org
|
RESO
|
FIXE
|
[4.4 Regression] ice for legal code with -O2
|
|
27946
|
enh
|
P3
|
|
pinskia@gcc.gnu.org
|
RESO
|
FIXE
|
double to long long and back to double stores to the stack
|
|
6585
|
enh
|
P3
|
i686-pc-linux-gnu
|
rask@gcc.gnu.org
|
RESO
|
DUPL
|
Redundant store/load instruction pairs on ix86
|
|
34312
|
nor
|
P3
|
i686-pc-linux-gnu
|
ubizjak@gmail.com
|
RESO
|
FIXE
|
[4.3 regression] spill failure with -O2 -fPIC -march=pent...
|
|
9085
|
nor
|
P3
|
|
|
RESO
|
FIXE
|
Unable to find register to spill when optimizing
|
|
11825
|
enh
|
P3
|
|
|
RESO
|
FIXE
|
Redundant move in some cases (dead code before return) fo...
|
|
11826
|
enh
|
P3
|
|
|
RESO
|
FIXE
|
[ARM] Minor register allocation problem before function r...
|
|
ID
|
Sev
|
Pri
|
Host
|
Assignee
|
Status
|
Resolution
|
Summary
|
|
17272
|
enh
|
P3
|
athlon-gnu-linux
|
|
RESO
|
FIXE
|
Extra store emitted when concatenating inline assembly se...
|
|
25225
|
enh
|
P3
|
|
|
RESO
|
FIXE
|
code generation -- extra instructions generated
|
|
25381
|
nor
|
P3
|
|
|
RESO
|
FIXE
|
local alloc is not really that good
|
|
25848
|
enh
|
P3
|
|
|
RESO
|
DUPL
|
local alloc causing an extra move
|
|
27001
|
nor
|
P3
|
|
|
RESO
|
FIXE
|
ICE with -fschedule-insns -fstack-protector-all
|
|
27432
|
min
|
P3
|
i686-pc-linux-gnu
|
|
RESO
|
WORK
|
-fschedule-insns -O2 -march=athlon cause compilation error
|
|
27856
|
enh
|
P3
|
i686-pc-linux-gnu
|
|
RESO
|
DUPL
|
With -Os, loading a constant to a register can use anothe...
|
|
28019
|
nor
|
P3
|
|
|
RESO
|
WONT
|
register allocator does not reschedule for x86 imull
|
|
31396
|
nor
|
P3
|
|
|
RESO
|
FIXE
|
Inline code performance much worse than out-of-line
|
|
31455
|
nor
|
P3
|
|
|
RESO
|
FIXE
|
[4.3 Regression] lower subreg causes a performance regres...
|
|
31508
|
nor
|
P3
|
x86_64-pc-linux-gnu
|
|
RESO
|
DUPL
|
Error during compilation with flag -fschedule-insns
|
|
31575
|
nor
|
P3
|
|
|
RESO
|
WONT
|
Extra push+pop generated on x86
|
|
31704
|
enh
|
P3
|
|
|
RESO
|
FIXE
|
x86_64 poor floating point register allocation across fun...
|
|
32201
|
nor
|
P3
|
i686-pc-linux-gnu
|
|
RESO
|
FIXE
|
Can not allocate %xmm0 register for variable blend insn
|
|
32725
|
nor
|
P3
|
|
|
RESO
|
FIXE
|
Unnecessary reg-reg moves
|
|
33791
|
min
|
P3
|
|
|
RESO
|
DUPL
|
x86 out of registers ICE with -fschedule-insns -march=core2
|
|
36222
|
nor
|
P3
|
|
|
RESO
|
FIXE
|
x86 fails to optimize out __v4si -> __m128i move
|
|
36223
|
enh
|
P3
|
i686-pc-linux-gnu
|
|
RESO
|
WORK
|
IV-opt is not optimal for mips
|
|
37243
|
blo
|
P3
|
|
|
RESO
|
FIXE
|
[4.4 Regression] IRA causes wrong code generation
|
|
37270
|
nor
|
P3
|
|
|
RESO
|
FIXE
|
[4.4 Regression] SH: spill failure for R0_REGS
|
|
37344
|
nor
|
P3
|
sparc-unknown-linux-gnu
|
|
RESO
|
FIXE
|
[4.4 Regression] sparc bootstrap fails with Bus error in ...
|
|
37359
|
blo
|
P3
|
|
|
RESO
|
FIXE
|
[4.4 Regression] IRA miscompiled transfer.o in libgfortra...
|
|
37397
|
nor
|
P3
|
|
|
RESO
|
FIXE
|
[4.4 Regression] IRA performance impact on SPEC CPU 2K/2006
|
|
37426
|
maj
|
P3
|
alpha-dec-osf5.1b
|
|
RESO
|
FIXE
|
[4.4 regression] IRA merge breaks Tru64 UNIX bootstrap
|
|
37512
|
nor
|
P3
|
|
|
RESO
|
FIXE
|
Useless instructions in member call by pointer
|
|
38034
|
nor
|
P3
|
|
|
RESO
|
FIXE
|
Unnecessary register move
|
|
38587
|
nor
|
P3
|
x86_64-unknown-linux-gnu
|
|
RESO
|
FIXE
|
[4.4 Regression] IRA doesn't preserve local variables aft...
|
|
38601
|
nor
|
P3
|
|
|
RESO
|
DUPL
|
[4.4 Regression] FAIL: gcc.target/i386/pr34256.c
|
|
37514
|
nor
|
P4
|
|
|
RESO
|
FIXE
|
[4.4 Regression] Wrong code generated for 20021120-1.c wi...
|
|
39836
|
min
|
P4
|
x86_64-unknown-linux-gnu
|
|
RESO
|
WONT
|
[4.4/4.5 regression] unoptimal code generated
|
|
16796
|
enh
|
P5
|
|
|
RESO
|
FIXE
|
PowerPC - Unnecessary Floating Point Register Copy
|
|
37674
|
blo
|
P1
|
s390x-ibm-linux
|
vmakarov@gcc.gnu.org
|
VERI
|
FIXE
|
[4.4 Regression] Bootstrap failure due to miscompilation ...
|
|
19530
|
nor
|
P2
|
i686-pc-mingw32
|
|
VERI
|
FIXE
|
MMX load intrinsic produces SSE superfluous instructions ...
|
133 bugs found.