ef28abefd0
Simplify a README.txt entry significantly to expose the core issue.
...
llvm-svn: 123556
2011-01-16 01:40:23 +00:00
b6c3aff1cb
typo
...
llvm-svn: 123406
2011-01-13 22:11:56 +00:00
b9cdf393a4
memcpy + metadata = bliss :)
...
llvm-svn: 123405
2011-01-13 22:08:15 +00:00
b1e7f557b7
Teach constant folding to perform conversions from constant floating
...
point values to their integer representation through the SSE intrinsic
calls. This is the last part of a README.txt entry for which I have real
world examples.
llvm-svn: 123206
2011-01-11 01:07:24 +00:00
d490c2d2ae
Fix a random missed optimization by making InstCombine more aggressive when determining which bits are demanded by
...
a comparison against a constant.
llvm-svn: 123203
2011-01-11 00:36:45 +00:00
78cdd2a6c6
+0.0 vs -0.0 differences can be handled by looking at the user of the
...
operation in some cases.
llvm-svn: 123190
2011-01-10 21:01:17 +00:00
eef1455020
expand on a note
...
llvm-svn: 123145
2011-01-10 00:33:01 +00:00
5b358c6825
typo
...
llvm-svn: 123142
2011-01-09 23:48:41 +00:00
320370e3ca
xref a PR #
...
llvm-svn: 123141
2011-01-09 23:42:22 +00:00
d011d5317c
Add a note about the inability to model FP -> int conversions which
...
perform rounding other than truncation in the IR. Common C code for this
turns into really an LLVM intrinsic call that blocks a lot of further
optimizations.
llvm-svn: 123135
2011-01-09 22:36:18 +00:00
0c68a668fa
Add a note about a missed FP optimization.
...
llvm-svn: 123126
2011-01-09 21:00:19 +00:00
82e6f6a325
Another missed memset in std::vector initialization.
...
llvm-svn: 123116
2011-01-09 11:29:57 +00:00
43f6d1b67e
Fix a cut-paste-o so that the sample code is correct for my last note.
...
Also, switch to a more clear 'sink' function with its declaration to
avoid any confusion about 'g'. Thanks for the suggestion Frits.
llvm-svn: 123113
2011-01-09 10:10:59 +00:00
ad6e1f0501
Another missed optimization of trivial vector code.
...
llvm-svn: 123112
2011-01-09 09:58:36 +00:00
f32619300a
Add a note about vector's size-constructor producing dead stores.
...
llvm-svn: 123111
2011-01-09 09:58:33 +00:00
5d684c17a7
Add a note about a missed memset optimization from std::fill.
...
llvm-svn: 123103
2011-01-09 01:32:55 +00:00
134cde912a
Revert 122959, it needs more thought. Add it back to README.txt with additional notes.
...
llvm-svn: 123030
2011-01-07 20:42:20 +00:00
84184b7207
With Benjamin's recent amazing patches, we should be able to do even better things :)
...
llvm-svn: 122978
2011-01-06 22:25:00 +00:00
1e01ade2e8
Add a note from llvmdev, this time with more info.
...
llvm-svn: 122966
2011-01-06 17:35:50 +00:00
605f21a6c8
EarlyCSE does this now (and GVN always did it).
...
llvm-svn: 122960
2011-01-06 13:19:46 +00:00
799b011276
InstCombine: If we call llvm.objectsize on a malloc call we can replace it with the size passed to malloc.
...
llvm-svn: 122959
2011-01-06 13:11:05 +00:00
245de78e06
add a note about object size from drystone, add a poorly optimized loop from 179.art.
...
llvm-svn: 122954
2011-01-06 07:41:22 +00:00
73552c2cce
add a trivial instcombine missed in Dhrystone
...
llvm-svn: 122953
2011-01-06 07:09:23 +00:00
51415d26f1
update a bunch of entries.
...
llvm-svn: 122700
2011-01-02 18:31:38 +00:00
ddf58010bd
Allow loop-idiom to run on multiple BB loops, but still only scan the loop
...
header for now for memset/memcpy opportunities. It turns out that loop-rotate
is successfully rotating loops, but *DOESN'T MERGE THE BLOCKS*, turning "for
loops" into 2 basic block loops that loop-idiom was ignoring.
With this fix, we form many *many* more memcpy and memsets than before, including
on the "history" loops in the viterbi benchmark, which look like this:
for (j=0; j<MAX_history; ++j) {
history_new[i][j+1] = history[2*i][j];
}
Transforming these loops into memcpy's speeds up the viterbi benchmark from
11.98s to 3.55s on my machine. Woo.
llvm-svn: 122685
2011-01-02 07:58:36 +00:00
6c3fc0a52d
a missed __builtin_object_size case.
...
llvm-svn: 122676
2011-01-01 22:57:31 +00:00
e5d5a41a58
various updates.
...
llvm-svn: 122675
2011-01-01 22:52:11 +00:00
772749aea1
Revert commit 122654 at the request of Chris, who reckons that instsimplify
...
is the wrong hammer for this nail, and is probably right.
llvm-svn: 122661
2011-01-01 20:08:02 +00:00
e3c539581c
Fix a README item by having InstructionSimplify do a mild form of value
...
numbering, in which it considers (for example) "%a = add i32 %x, %y" and
"%b = add i32 %x, %y" to be equal because the operands are equal and the
result of the instructions only depends on the values of the operands.
This has almost no effect (it removes 4 instructions from gcc-as-one-file),
and perhaps slows down compilation: I measured a 0.4% slowdown on the large
gcc-as-one-file testcase, but it wasn't statistically significant.
llvm-svn: 122654
2011-01-01 16:12:09 +00:00
102bc01900
add a note from llvmdev
...
llvm-svn: 122603
2010-12-28 18:45:02 +00:00
dfa40f8f19
Remove/fix invalid README entries. The well thought out strcpy function doesn't return a pointer to the end of the string.
...
llvm-svn: 122496
2010-12-23 15:32:07 +00:00
5e0c0c72e9
recognize an unsigned add with overflow idiom into uadd.
...
This resolves a README entry and technically resolves PR4916,
but we still get poor code for the testcase in that PR because
GVN isn't CSE'ing uadd with add, filed as PR8817.
Previously we got:
_test7: ## @test7
addq %rsi, %rdi
cmpq %rdi, %rsi
movl $42, %eax
cmovaq %rsi, %rax
ret
Now we get:
_test7: ## @test7
addq %rsi, %rdi
movl $42, %eax
cmovbq %rsi, %rax
ret
llvm-svn: 122182
2010-12-19 19:37:52 +00:00
5174921b5b
add another overflow idiom
...
llvm-svn: 121854
2010-12-15 07:28:58 +00:00
2e33985300
add a note about overflow idiom recognition.
...
llvm-svn: 121853
2010-12-15 07:25:55 +00:00
27ecda1efd
add a shift/imul missed optimization
...
llvm-svn: 121850
2010-12-15 07:10:43 +00:00
aded09f27f
add a note about a SPEC hack that gcc mainline does.
...
llvm-svn: 121849
2010-12-15 06:38:24 +00:00
14cb11ddb2
add a note
...
llvm-svn: 121656
2010-12-13 00:15:25 +00:00
c4169cebe3
Generalize the and-icmp-select instcombine further by allowing selects of the form
...
(x & 2^n) ? 2^m+C : C
we can offset both arms by C to get the "(x & 2^n) ? 2^m : 0" form, optimize the
select to a shift and apply the offset afterwards.
llvm-svn: 121609
2010-12-11 10:49:22 +00:00
94a622af4c
The srem -> urem transform is not safe for any divisor that's not a power of two.
...
E.g. -5 % 5 is 0 with srem and 1 with urem.
Also addresses Frits van Bommel's comments.
llvm-svn: 120049
2010-11-23 20:33:57 +00:00
b5afa65b0a
InstCombine: Reduce "X shift (A srem B)" to "X shift (A urem B)" iff B is positive.
...
This allows to transform the rem in "1 << ((int)x % 8);" to an and.
llvm-svn: 120028
2010-11-23 18:52:42 +00:00
f1ebb63161
InstCombine: Implement X - A*-B -> X + A*B.
...
llvm-svn: 119984
2010-11-22 20:31:27 +00:00
24656c9583
Implement the "if (X == 6 || X == 4)" -> "if ((X|2) == 6)" optimization.
...
This currently only catches the most basic case, a two-case switch, but can be
extended later.
llvm-svn: 119964
2010-11-22 09:45:38 +00:00
9165d9d2ac
add some random notes.
...
llvm-svn: 119925
2010-11-21 07:05:31 +00:00
f7e896138e
optimize:
...
void a(int x) { if (((1<<x)&8)==0) b(); }
into "x != 3", which occurs over 100 times in 403.gcc but in no
other program in llvm-test.
llvm-svn: 119922
2010-11-21 06:44:42 +00:00
9de0176ef8
tail calls on x86 are implemented.
...
llvm-svn: 119920
2010-11-21 06:10:27 +00:00
932aab3cbf
add a note
...
llvm-svn: 118806
2010-11-11 18:23:57 +00:00
1d6aa32b87
add pr#
...
llvm-svn: 118797
2010-11-11 17:17:56 +00:00
4d94e47368
add a case we fail to devirt.
...
llvm-svn: 118608
2010-11-09 19:37:28 +00:00
f532d31198
Fix a README item: when doing a comparison with the result
...
of a select instruction, see if doing the compare with the
true and false values of the select gives the same result.
If so, that can be used as the value of the comparison.
llvm-svn: 118378
2010-11-07 16:12:23 +00:00
8628e2a19c
Add a note.
...
llvm-svn: 118337
2010-11-06 10:37:16 +00:00
2b76c66fd6
Add constant folding for strspn and strcspn to SimplifyLibCalls.
...
llvm-svn: 115116
2010-09-30 00:58:35 +00:00
011f146419
idiom recognition should catch this.
...
llvm-svn: 114304
2010-09-19 00:37:34 +00:00
bb10e90487
Add optimization to Target/README.txt.
...
llvm-svn: 110543
2010-08-08 07:04:25 +00:00
2321e6a4d4
Teach instcombine to transform
...
(X >s -1) ? C1 : C2 and (X <s 0) ? C2 : C1
into ((X >>s 31) & (C2 - C1)) + C1, avoiding the conditional.
This optimization could be extended to take non-const C1 and C2 but we better
stay conservative to avoid code size bloat for now.
for
int sel(int n) {
return n >= 0 ? 60 : 100;
}
we now generate
sarl $31, %edi
andl $40, %edi
leal 60(%rdi), %eax
instead of
testl %edi, %edi
movl $60, %ecx
movl $100, %eax
cmovnsl %ecx, %eax
llvm-svn: 107866
2010-07-08 11:39:10 +00:00
c8f595212f
Minor amendment to switch-lowering improvement.
...
llvm-svn: 107569
2010-07-03 08:43:32 +00:00
836fdbc85b
Note switch-lowering inefficiency.
...
llvm-svn: 107565
2010-07-03 07:38:12 +00:00
e34471bb31
Add another bswap idiom that isn't matched.
...
llvm-svn: 107213
2010-06-29 22:22:22 +00:00
41476410c9
TODO--
...
llvm-svn: 106102
2010-06-16 15:47:00 +00:00
e17e4aea2a
Add README entry; based on testcase from Bill Hart.
...
llvm-svn: 105878
2010-06-12 05:54:27 +00:00
4dc833c607
add a note
...
llvm-svn: 104404
2010-05-21 23:16:21 +00:00
73c8145505
Add a README entry.
...
llvm-svn: 102906
2010-05-03 14:31:00 +00:00
cfc921cd2a
add a note
...
llvm-svn: 101581
2010-04-16 23:52:30 +00:00
4041ab6e00
Implement rdar://7860110 (also in target/readme.txt) narrowing
...
a load/or/and/store sequence into a narrower store when it is
safe. Daniel tells me that clang will start producing this sort
of thing with bitfields, and this does trigger a few dozen times
on 176.gcc produced by llvm-gcc even now.
This compiles code like CodeGen/X86/2009-05-28-DAGCombineCrash.ll
into:
movl %eax, 36(%rdi)
instead of:
movl $4294967295, %eax ## imm = 0xFFFFFFFF
andq 32(%rdi), %rax
shlq $32, %rcx
addq %rax, %rcx
movq %rcx, 32(%rdi)
and each of the testcases into a single store. Each of them used
to compile into craziness like this:
_test4:
movl $65535, %eax ## imm = 0xFFFF
andl (%rdi), %eax
shll $16, %esi
addl %eax, %esi
movl %esi, (%rdi)
ret
llvm-svn: 101343
2010-04-15 04:48:01 +00:00
1f6689a8ba
move PR6576 here.
...
llvm-svn: 98194
2010-03-10 21:42:42 +00:00
187242b3ab
move PR6212 to this file.
...
llvm-svn: 95624
2010-02-09 00:11:10 +00:00
0de0b3677a
Remove a completed item, add a couple new ones.
...
llvm-svn: 94945
2010-01-31 04:55:32 +00:00
7c42b9d51e
Improve isSafeToLoadUnconditionally to recognize that GEPs with constant
...
indices are safe if the result is known to be within the bounds of the
underlying object.
llvm-svn: 94829
2010-01-29 19:19:08 +00:00
e3a68d1063
reassociate should do this.
...
llvm-svn: 94374
2010-01-24 20:17:09 +00:00
7e3f8b60d6
add a note.
...
llvm-svn: 94373
2010-01-24 20:01:41 +00:00
249da5cb73
implement a simple instcombine xform that has been in the
...
readme forever.
llvm-svn: 94318
2010-01-23 18:49:30 +00:00
082da53f9a
add some notes, making posix-memalign be nocapture would be an easy improvement.
...
llvm-svn: 94312
2010-01-23 17:59:23 +00:00
9ed49c5c8f
Add some potentially interesting transformations to README.
...
llvm-svn: 93797
2010-01-18 22:36:59 +00:00
c8493da5b1
Fix a README item: have functionattrs look through selects and
...
phi nodes when deciding which pointers point to local memory.
I actually checked long ago how useful this is, and it isn't
very: it hardly ever fires in the testsuite, but since Chris
wants it here it is!
llvm-svn: 92836
2010-01-06 15:37:47 +00:00
78376ad7e1
Partially address a README by having functionattrs consider calls to
...
memcpy, memset and other intrinsics that only access their arguments
to be readnone if the intrinsic's arguments all point to local memory.
This improves the testcase in the README to readonly, but it could in
theory be made readnone, however this would involve more sophisticated
analysis that looks through the memcpy.
llvm-svn: 92829
2010-01-06 08:45:52 +00:00
2d91231d82
implement an instcombine xform needed by clang's codegen
...
on the example in PR4216. This doesn't trigger in the testsuite,
so I'd really appreciate someone scrutinizing the logic for
correctness.
llvm-svn: 92458
2010-01-04 06:03:59 +00:00
39f18e545e
Teach codegen to lower llvm.powi to an efficient (but not optimal)
...
multiply sequence when the power is a constant integer. Before, our
codegen for std::pow(.., int) always turned into a libcall, which was
really inefficient.
This should also make many gfortran programs happier I'd imagine.
llvm-svn: 92388
2010-01-01 03:32:16 +00:00
71cf7c256f
update this. To take the next step, llvm.powi should be generalized to work
...
on integers as well and codegen should lower them to branch trees.
llvm-svn: 92382
2010-01-01 01:29:26 +00:00
96cf7f42b0
More info on this transformation.
...
llvm-svn: 91230
2009-12-12 23:23:43 +00:00
8eada9f580
Remove some stuff that's already implemented. Also, remove the note about
...
merging x >u 5 and x <s 20 because it's impossible to implement.
llvm-svn: 91228
2009-12-12 21:41:48 +00:00
f05330a5c8
expand note.
...
llvm-svn: 90429
2009-12-03 07:43:46 +00:00
d1e4ee3c2b
add a note
...
llvm-svn: 90428
2009-12-03 07:41:54 +00:00
58ccf88c36
update and consolidate the load pre notes.
...
llvm-svn: 90050
2009-11-29 02:19:52 +00:00
83a4a9868f
add a deadargelim note.
...
llvm-svn: 90009
2009-11-27 17:12:30 +00:00
ca9e0e83b3
This testcase is actually only partially redundant, and requires
...
the FIXME I added yesterday to be implemented.
llvm-svn: 90008
2009-11-27 16:53:57 +00:00
cc6d29286c
this (and probably several others) are now done.
...
llvm-svn: 89982
2009-11-27 00:35:04 +00:00
9bd2136ca3
Teach memdep to phi translate bitcasts. This allows us to compile
...
the example in GCC PR16799 to:
LBB1_2: ## %bb1
movl %eax, %eax
subq %rax, %rdi
movq %rdi, (%rcx)
movl (%rdi), %eax
testl %eax, %eax
je LBB1_2
instead of:
LBB1_2: ## %bb1
movl (%rdi), %ecx
subq %rcx, %rdi
movq %rdi, (%rax)
cmpl $0, (%rdi)
je LBB1_2
llvm-svn: 89978
2009-11-26 23:41:07 +00:00
29bc8a91d3
Teach basicaa that x|c == x+c when the c bits of x are clear. This
...
allows us to compile the example in readme.txt into:
LBB1_1: ## %bb
movl 4(%rdx,%rax), %ecx
movl %ecx, %esi
imull (%rdx,%rax), %esi
imull %esi, %ecx
movl %esi, 8(%rdx,%rax)
imull %ecx, %esi
movl %ecx, 12(%rdx,%rax)
movl %esi, 16(%rdx,%rax)
imull %ecx, %esi
movl %esi, 20(%rdx,%rax)
addq $16, %rax
cmpq $4000, %rax
jne LBB1_1
instead of:
LBB1_1:
movl (%rdx,%rax), %ecx
imull 4(%rdx,%rax), %ecx
movl %ecx, 8(%rdx,%rax)
imull 4(%rdx,%rax), %ecx
movl %ecx, 12(%rdx,%rax)
imull 8(%rdx,%rax), %ecx
movl %ecx, 16(%rdx,%rax)
imull 12(%rdx,%rax), %ecx
movl %ecx, 20(%rdx,%rax)
addq $16, %rax
cmpq $4000, %rax
jne LBB1_1
GCC (4.2) doesn't seem to be able to eliminate the loads in this
testcase either, it generates:
L2:
movl (%rdx), %eax
imull 4(%rdx), %eax
movl %eax, 8(%rdx)
imull 4(%rdx), %eax
movl %eax, 12(%rdx)
imull 8(%rdx), %eax
movl %eax, 16(%rdx)
imull 12(%rdx), %eax
movl %eax, 20(%rdx)
addl $4, %ecx
addq $16, %rdx
cmpl $1002, %ecx
jne L2
llvm-svn: 89952
2009-11-26 16:26:43 +00:00
12dacdd359
teach basicaa that A[i] != A[i+1].
...
llvm-svn: 89951
2009-11-26 16:18:10 +00:00
8e09ad6f3c
update some notes slightly
...
llvm-svn: 89913
2009-11-26 01:51:18 +00:00
ef4ea9a2a9
Add a complex missed optimization opportunity I came across while investigating
...
bug 5438.
llvm-svn: 88855
2009-11-15 17:51:23 +00:00
7a09964e81
another const prop failure.
...
llvm-svn: 86848
2009-11-11 17:54:02 +00:00
539bdf0487
add a note
...
llvm-svn: 86847
2009-11-11 17:51:27 +00:00
0169fd7c62
add a note
...
llvm-svn: 86756
2009-11-10 23:47:45 +00:00
8ff26038ef
I did this a week or two ago
...
llvm-svn: 86754
2009-11-10 23:40:49 +00:00
b9397262b7
Improve tail call elimination to handle the switch statement.
...
llvm-svn: 86403
2009-11-07 21:10:15 +00:00
06c26d982e
add a note from PR5313
...
llvm-svn: 86146
2009-11-05 18:19:19 +00:00
2e5198ff09
Add new note.
...
llvm-svn: 85341
2009-10-27 23:30:07 +00:00
fd2730ee8c
Move and clarify note.
...
llvm-svn: 85334
2009-10-27 22:48:31 +00:00
13b8b56dd4
this is done.
...
llvm-svn: 85041
2009-10-25 06:17:51 +00:00
851193b873
some stuff is done, we still have constantexpr simplification to do.
...
llvm-svn: 84943
2009-10-23 07:00:55 +00:00
Chandler Carruth