Chris Lattner [Mon, 19 Sep 2005 06:56:21 +0000 (06:56 +0000)]
Teach the local spiller to turn stack slot loads into register-register copies
when possible, avoiding the load (and avoiding the copy if the value is already
in the right register).
This patch came about when I noticed code like the following being generated:
store R17 -> [SS1]
...blah...
R4 = load [SS1]
This was causing an LSU reject on the G5. This problem was due to the register
allocator folding spill code into a reg-reg copy (producing the load), which
prevented the spiller from being able to rewrite the load into a copy, despite
the fact that the value was already available in a register. In the case
above, we now rip out the R4 load and replace it with a R4 = R17 copy.
This speeds up several programs on X86 (which spills a lot :) ), e.g.
smg2k from 22.39->20.60s, povray from 12.93->12.66s, 168.wupwise from
68.54->53.83s (!), 197.parser from 7.33->6.62s (!), etc. This may have a larger
impact in some cases on the G5 (by avoiding LSU rejects), though it probably
won't trigger as often (less spilling in general).
Targets that implement folding of loads/stores into copies should implement
the isLoadFromStackSlot hook to get this.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23388
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Mon, 19 Sep 2005 05:23:44 +0000 (05:23 +0000)]
Implement the isLoadFromStackSlot interface
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23387
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sun, 18 Sep 2005 07:22:02 +0000 (07:22 +0000)]
Refactor this code a bit and make it more general. This now compiles:
struct S { unsigned int i : 6, j : 11, k : 15; } b;
void plus2 (unsigned int x) { b.j += x; }
To:
_plus2:
lis r2, ha16(L_b$non_lazy_ptr)
lwz r2, lo16(L_b$non_lazy_ptr)(r2)
lwz r4, 0(r2)
slwi r3, r3, 6
add r3, r4, r3
rlwimi r3, r4, 0, 26, 14
stw r3, 0(r2)
blr
instead of:
_plus2:
lis r2, ha16(L_b$non_lazy_ptr)
lwz r2, lo16(L_b$non_lazy_ptr)(r2)
lwz r4, 0(r2)
rlwinm r5, r4, 26, 21, 31
add r3, r5, r3
rlwimi r4, r3, 6, 15, 25
stw r4, 0(r2)
blr
by eliminating an 'and'.
I'm pretty sure this is as small as we can go :)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23386
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sun, 18 Sep 2005 06:30:59 +0000 (06:30 +0000)]
Compile
struct S { unsigned int i : 6, j : 11, k : 15; } b;
void plus2 (unsigned int x) {
b.j += x;
}
to:
plus2:
mov %EAX, DWORD PTR [b]
mov %ECX, %EAX
and %ECX, 131008
mov %EDX, DWORD PTR [%ESP + 4]
shl %EDX, 6
add %EDX, %ECX
and %EDX, 131008
and %EAX, -131009
or %EDX, %EAX
mov DWORD PTR [b], %EDX
ret
instead of:
plus2:
mov %EAX, DWORD PTR [b]
mov %ECX, %EAX
shr %ECX, 6
and %ECX, 2047
add %ECX, DWORD PTR [%ESP + 4]
shl %ECX, 6
and %ECX, 131008
and %EAX, -131009
or %ECX, %EAX
mov DWORD PTR [b], %ECX
ret
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23385
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sun, 18 Sep 2005 06:02:59 +0000 (06:02 +0000)]
Generalize this transform, using MaskedValueIsZero, allowing us to compile:
struct S { unsigned int i : 6, j : 11, k : 15; } b;
void plus3 (unsigned int x) { b.k += x; }
To:
plus3:
mov %EAX, DWORD PTR [%ESP + 4]
shl %EAX, 17
add DWORD PTR [b], %EAX
ret
instead of:
plus3:
mov %EAX, DWORD PTR [%ESP + 4]
shl %EAX, 17
mov %ECX, DWORD PTR [b]
add %EAX, %ECX
and %EAX, -131072
and %ECX, 131071
or %ECX, %EAX
mov DWORD PTR [b], %ECX
ret
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23384
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sun, 18 Sep 2005 05:25:20 +0000 (05:25 +0000)]
fix typeo
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23383
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sun, 18 Sep 2005 05:12:51 +0000 (05:12 +0000)]
Remove unintentionally committed code
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23382
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sun, 18 Sep 2005 05:12:10 +0000 (05:12 +0000)]
implement shift.ll:test25. This compiles:
struct S { unsigned int i : 6, j : 11, k : 15; } b;
void plus3 (unsigned int x) {
b.k += x;
}
to:
_plus3:
lis r2, ha16(L_b$non_lazy_ptr)
lwz r2, lo16(L_b$non_lazy_ptr)(r2)
lwz r3, 0(r2)
rlwinm r4, r3, 0, 0, 14
add r4, r4, r3
rlwimi r4, r3, 0, 15, 31
stw r4, 0(r2)
blr
instead of:
_plus3:
lis r2, ha16(L_b$non_lazy_ptr)
lwz r2, lo16(L_b$non_lazy_ptr)(r2)
lwz r4, 0(r2)
srwi r5, r4, 17
add r3, r5, r3
slwi r3, r3, 17
rlwimi r3, r4, 0, 15, 31
stw r3, 0(r2)
blr
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23381
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sun, 18 Sep 2005 05:10:39 +0000 (05:10 +0000)]
new testcase
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23380
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sun, 18 Sep 2005 04:24:45 +0000 (04:24 +0000)]
Implement add.ll:test29. Codegening:
struct S { unsigned int i : 6, j : 11, k : 15; } b;
void plus1 (unsigned int x) {
b.i += x;
}
as:
_plus1:
lis r2, ha16(L_b$non_lazy_ptr)
lwz r2, lo16(L_b$non_lazy_ptr)(r2)
lwz r4, 0(r2)
add r3, r4, r3
rlwimi r3, r4, 0, 0, 25
stw r3, 0(r2)
blr
instead of:
_plus1:
lis r2, ha16(L_b$non_lazy_ptr)
lwz r2, lo16(L_b$non_lazy_ptr)(r2)
lwz r4, 0(r2)
rlwinm r5, r4, 0, 26, 31
add r3, r5, r3
rlwimi r3, r4, 0, 0, 25
stw r3, 0(r2)
blr
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23379
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sun, 18 Sep 2005 04:22:59 +0000 (04:22 +0000)]
new testcase
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23378
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sun, 18 Sep 2005 03:50:25 +0000 (03:50 +0000)]
remove debug output
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23377
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sun, 18 Sep 2005 03:42:07 +0000 (03:42 +0000)]
Implement or.ll:test21. This teaches instcombine to be able to turn this:
struct {
unsigned int bit0:1;
unsigned int ubyte:31;
} sdata;
void foo() {
sdata.ubyte++;
}
into this:
foo:
add DWORD PTR [sdata], 2
ret
instead of this:
foo:
mov %EAX, DWORD PTR [sdata]
mov %ECX, %EAX
add %ECX, 2
and %ECX, -2
and %EAX, 1
or %EAX, %ECX
mov DWORD PTR [sdata], %EAX
ret
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23376
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sun, 18 Sep 2005 03:39:02 +0000 (03:39 +0000)]
new testcase
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23375
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sat, 17 Sep 2005 01:03:26 +0000 (01:03 +0000)]
Implement hook for ppc
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23374
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sat, 17 Sep 2005 01:02:45 +0000 (01:02 +0000)]
add a new callback
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23373
91177308-0d34-0410-b5e6-
96231b3b80d8
Nate Begeman [Fri, 16 Sep 2005 00:54:12 +0000 (00:54 +0000)]
More DAG combining. Still need the branch instructions, and select_cc
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23371
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Fri, 16 Sep 2005 00:29:46 +0000 (00:29 +0000)]
Fix a minor bug, add comments
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23370
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Thu, 15 Sep 2005 22:23:50 +0000 (22:23 +0000)]
teach the type inference code how to infer types for instructions and node
xforms. Run type inference on result patterns, so we always have fully typed
results (and to catch errors in .td files).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23369
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Thu, 15 Sep 2005 21:57:35 +0000 (21:57 +0000)]
put instructions into a map instead of a vector for quick lookup
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23368
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Thu, 15 Sep 2005 21:51:12 +0000 (21:51 +0000)]
when parsing instructions remember information about the types taken and
returned.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23367
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Thu, 15 Sep 2005 21:44:00 +0000 (21:44 +0000)]
disable this for now
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23366
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Thu, 15 Sep 2005 21:42:00 +0000 (21:42 +0000)]
Start parsing "Pattern" nodes
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23365
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Thu, 15 Sep 2005 02:38:02 +0000 (02:38 +0000)]
rename a couple of methods, add structure for pattern parsing
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23364
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Wed, 14 Sep 2005 23:05:13 +0000 (23:05 +0000)]
Verify that xform functions only occur in logical places
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23363
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Wed, 14 Sep 2005 23:01:59 +0000 (23:01 +0000)]
Promote xform fns to be explicit nodes in result patterns, and clean off
predicates since they will have already matched at this point.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23362
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Wed, 14 Sep 2005 22:55:26 +0000 (22:55 +0000)]
start building the instruction dest pattern correctly. Change the xform
functions to preserve the Record for the xform instead of making it into a
function name.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23361
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Wed, 14 Sep 2005 22:06:36 +0000 (22:06 +0000)]
catch unnamed inputs
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23360
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Wed, 14 Sep 2005 21:59:34 +0000 (21:59 +0000)]
check that there are no unexpected operands
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23359
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Wed, 14 Sep 2005 21:13:50 +0000 (21:13 +0000)]
force all instruction operands to be named.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23358
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Wed, 14 Sep 2005 21:11:13 +0000 (21:11 +0000)]
Give all operands names
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23357
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Wed, 14 Sep 2005 21:10:24 +0000 (21:10 +0000)]
give all operands names
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23356
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Wed, 14 Sep 2005 21:05:02 +0000 (21:05 +0000)]
Check that operands have unique names. REJECT instructions with broken operand
lists: only don't parse them if they are entirely missing (sparcv9).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23355
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Wed, 14 Sep 2005 21:04:12 +0000 (21:04 +0000)]
fix a broke range check
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23354
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Wed, 14 Sep 2005 20:53:42 +0000 (20:53 +0000)]
Parse significantly more of the instruction pattern, now collecting and
verifying information about the operands.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23353
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Wed, 14 Sep 2005 20:53:05 +0000 (20:53 +0000)]
Fix some issues exposed by more testing. XORIS had the wrong operands
specified. The various *imm operands defined by PPC are really all i32,
even though the actual immediate is restricted to a smaller value in it.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23352
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Wed, 14 Sep 2005 18:19:25 +0000 (18:19 +0000)]
Verify that set destinations occur first in the instruction operand list.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23351
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Wed, 14 Sep 2005 18:18:39 +0000 (18:18 +0000)]
Fix some bugs noticed by new checking code
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23350
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Wed, 14 Sep 2005 18:02:53 +0000 (18:02 +0000)]
add an accessor
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23349
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Wed, 14 Sep 2005 17:32:56 +0000 (17:32 +0000)]
Fix the regression last night compiling povray
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23348
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Wed, 14 Sep 2005 06:06:45 +0000 (06:06 +0000)]
fix a major regression from my patch this afternoon
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23347
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Wed, 14 Sep 2005 06:03:10 +0000 (06:03 +0000)]
remove some code that isn't ready for prime time
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23346
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Wed, 14 Sep 2005 04:03:16 +0000 (04:03 +0000)]
Switch to a slightly more structured representation for instructions
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23345
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Wed, 14 Sep 2005 02:11:12 +0000 (02:11 +0000)]
Add some more checking/verification code
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23344
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Wed, 14 Sep 2005 00:09:24 +0000 (00:09 +0000)]
start parsing instructions into patterns, start doing many more checks of
'set's.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23343
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Tue, 13 Sep 2005 22:05:21 +0000 (22:05 +0000)]
we don't need this proto any longer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23342
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Tue, 13 Sep 2005 22:05:02 +0000 (22:05 +0000)]
don't emit the namespace inside the class!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23341
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Tue, 13 Sep 2005 22:03:37 +0000 (22:03 +0000)]
Emit code suitable for emission into the ISel class, allowing us to use/define
methods.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23340
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Tue, 13 Sep 2005 22:03:06 +0000 (22:03 +0000)]
move the #include for the generated code into the isel class body so we
can use/define class methods
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23339
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Tue, 13 Sep 2005 21:59:15 +0000 (21:59 +0000)]
continue xform function parsing
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23338
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Tue, 13 Sep 2005 21:51:00 +0000 (21:51 +0000)]
Start parsing node transformation information
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23337
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Tue, 13 Sep 2005 21:44:28 +0000 (21:44 +0000)]
Add a new Record::getValueAsCode method to mirror the other getValueAs*
methods. Use it to simplify some code.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23336
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Tue, 13 Sep 2005 21:20:49 +0000 (21:20 +0000)]
completely eliminate TreePattern::PatternType
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23335
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Tue, 13 Sep 2005 19:33:40 +0000 (19:33 +0000)]
Change the arg lowering code to use copyfromreg from vregs associated
with incoming arguments instead of the pregs themselves. This fixes
the scheduler from causing problems by moving a copyfromreg for an argument
to after a select_cc node (now it can, and bad things won't happen).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23334
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Tue, 13 Sep 2005 19:32:18 +0000 (19:32 +0000)]
This has been moved to the target-indep code
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23333
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Tue, 13 Sep 2005 19:31:44 +0000 (19:31 +0000)]
This code is no longer needed, it is moved to the target-indep code
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23332
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Tue, 13 Sep 2005 19:30:54 +0000 (19:30 +0000)]
If a function has liveins, and if the target requested that they be plopped
into particular vregs, emit copies into the entry MBB.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23331
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Tue, 13 Sep 2005 19:03:13 +0000 (19:03 +0000)]
Majik numbers are bad
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23330
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Tue, 13 Sep 2005 18:47:49 +0000 (18:47 +0000)]
Remove some dead vectors
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23329
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Tue, 13 Sep 2005 18:36:04 +0000 (18:36 +0000)]
Add a simple xform to simplify array accesses with casts in the way.
This is useful for 178.galgel where resolution of dope vectors (by the
optimizer) causes the scales to become apparent.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23328
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Tue, 13 Sep 2005 02:09:55 +0000 (02:09 +0000)]
Fix an issue where LSR would miss rewriting a use of an IV expression by a PHI node that is not the original PHI.
This fixes up a dot-product loop in galgel, speeding it up from 18.47s to
16.13s.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23327
91177308-0d34-0410-b5e6-
96231b3b80d8
Duraid Madina [Tue, 13 Sep 2005 01:03:53 +0000 (01:03 +0000)]
fails since linux-itanium headers are Different
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23326
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Tue, 13 Sep 2005 00:40:14 +0000 (00:40 +0000)]
Add a helper function, allowing us to simplify some code a bit, changing
indentation, no functionality change
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23325
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Mon, 12 Sep 2005 23:23:25 +0000 (23:23 +0000)]
Implement a simple xform to turn code like this:
if () { store A -> P; } else { store B -> P; }
into a PHI node with one store, in the most trival case. This implements
load.ll:test10.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23324
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Mon, 12 Sep 2005 23:22:17 +0000 (23:22 +0000)]
new testcase
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23323
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Mon, 12 Sep 2005 22:21:03 +0000 (22:21 +0000)]
Another load-peephole optimization: do gcse when two loads are next to
each other. This implements InstCombine/load.ll:test9
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23322
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Mon, 12 Sep 2005 22:19:46 +0000 (22:19 +0000)]
new testcase
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23321
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Mon, 12 Sep 2005 22:00:15 +0000 (22:00 +0000)]
Implement a trivial form of store->load forwarding where the store and the
load are exactly consequtive. This is picked up by other passes, but this
triggers thousands of times in fortran programs that use static locals
(and is thus a compile-time speedup).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23320
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Mon, 12 Sep 2005 21:59:22 +0000 (21:59 +0000)]
new testcase
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23319
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Mon, 12 Sep 2005 17:11:27 +0000 (17:11 +0000)]
Fix a regression from last night, which caused this pass to create invalid
code for IV uses outside of loops that are not dominated by the latch block.
We should only convert these uses to use the post-inc value if they ARE
dominated by the latch block.
Also use a new LoopInfo method to simplify some code.
This fixes Transforms/LoopStrengthReduce/2005-09-12-UsesOutOutsideOfLoop.ll
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23318
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Mon, 12 Sep 2005 17:09:40 +0000 (17:09 +0000)]
relax pattern match on name
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23317
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Mon, 12 Sep 2005 17:08:15 +0000 (17:08 +0000)]
new testcase
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23316
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Mon, 12 Sep 2005 17:03:55 +0000 (17:03 +0000)]
Add a new getLoopLatch() method.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23315
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Mon, 12 Sep 2005 17:03:16 +0000 (17:03 +0000)]
new method
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23314
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Mon, 12 Sep 2005 06:04:47 +0000 (06:04 +0000)]
_test:
li r2, 0
LBB_test_1: ; no_exit.2
li r5, 0
stw r5, 0(r3)
addi r2, r2, 1
addi r3, r3, 4
cmpwi cr0, r2, 701
blt cr0, LBB_test_1 ; no_exit.2
LBB_test_2: ; loopexit.2.loopexit
addi r2, r2, 1
stw r2, 0(r4)
blr
[zion ~/llvm]$ cat > ~/xx
Uses of IV's outside of the loop should use hte post-incremented version
of the IV, not the preincremented version. This helps many loops (e.g. in sixtrack)
which used to generate code like this (this is the code from the
dont-hoist-simple-loop-constants.ll testcase):
_test:
li r2, 0 **** IV starts at 0
LBB_test_1: ; no_exit.2
or r5, r2, r2 **** Copy for loop exit
li r2, 0
stw r2, 0(r3)
addi r3, r3, 4
addi r2, r5, 1
addi r6, r5, 2 **** IV+2
cmpwi cr0, r6, 701
blt cr0, LBB_test_1 ; no_exit.2
LBB_test_2: ; loopexit.2.loopexit
addi r2, r5, 2 **** IV+2
stw r2, 0(r4)
blr
And now generated code like this:
_test:
li r2, 1 *** IV starts at 1
LBB_test_1: ; no_exit.2
li r5, 0
stw r5, 0(r3)
addi r2, r2, 1
addi r3, r3, 4
cmpwi cr0, r2, 701 *** IV.postinc + 0
blt cr0, LBB_test_1
LBB_test_2: ; loopexit.2.loopexit
stw r2, 0(r4) *** IV.postinc + 0
blr
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23313
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Mon, 12 Sep 2005 05:50:15 +0000 (05:50 +0000)]
new testcase
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23312
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Mon, 12 Sep 2005 05:30:06 +0000 (05:30 +0000)]
Regenerate
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23311
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Mon, 12 Sep 2005 05:29:43 +0000 (05:29 +0000)]
Rearrange two rules, which apparently makes some versions of bison happier.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23310
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sun, 11 Sep 2005 03:50:38 +0000 (03:50 +0000)]
Make sure to disable 64-bit extensions for this test
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23309
91177308-0d34-0410-b5e6-
96231b3b80d8
Jeff Cohen [Sat, 10 Sep 2005 02:33:17 +0000 (02:33 +0000)]
Fix more Visual Studio build problems.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23308
91177308-0d34-0410-b5e6-
96231b3b80d8
Jeff Cohen [Sat, 10 Sep 2005 02:00:02 +0000 (02:00 +0000)]
Fix miscellaneous Visual Studio build problems.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23307
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sat, 10 Sep 2005 01:18:45 +0000 (01:18 +0000)]
implement Transforms/LoopStrengthReduce/dont-hoist-simple-loop-constants.ll.
We used to emit this code for it:
_test:
li r2, 1 ;; Value tying up a register for the whole loop
li r5, 0
LBB_test_1: ; no_exit.2
or r6, r5, r5
li r5, 0
stw r5, 0(r3)
addi r5, r6, 1
addi r3, r3, 4
add r7, r2, r5 ;; should be addi r7, r5, 1
cmpwi cr0, r7, 701
blt cr0, LBB_test_1 ; no_exit.2
LBB_test_2: ; loopexit.2.loopexit
addi r2, r6, 2
stw r2, 0(r4)
blr
now we emit this:
_test:
li r2, 0
LBB_test_1: ; no_exit.2
or r5, r2, r2
li r2, 0
stw r2, 0(r3)
addi r3, r3, 4
addi r2, r5, 1
addi r6, r5, 2 ;; whoa, fold those adds!
cmpwi cr0, r6, 701
blt cr0, LBB_test_1 ; no_exit.2
LBB_test_2: ; loopexit.2.loopexit
addi r2, r5, 2
stw r2, 0(r4)
blr
more improvement coming.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23306
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sat, 10 Sep 2005 01:14:37 +0000 (01:14 +0000)]
new testcase
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23305
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sat, 10 Sep 2005 00:21:06 +0000 (00:21 +0000)]
PowerPC cannot truncstore i1 natively
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23304
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sat, 10 Sep 2005 00:20:18 +0000 (00:20 +0000)]
Allow targets to say they don't support truncstore i1 (which includes a mask
when storing to an 8-bit memory location), as most don't.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23303
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Fri, 9 Sep 2005 23:53:39 +0000 (23:53 +0000)]
Add a missing #include, patch courtesy of Baptiste Lepilleur.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23302
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Fri, 9 Sep 2005 23:00:07 +0000 (23:00 +0000)]
Fix a problem duraid encountered on itanium where this folding:
select (x < y), 1, 0 -> (x < y) incorrectly: the setcc returns i1 but the
select returned i32. Add the zero extend as needed.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23301
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Fri, 9 Sep 2005 22:35:03 +0000 (22:35 +0000)]
Fix a crash viewing dags that have target nodes in them
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23300
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Fri, 9 Sep 2005 21:59:44 +0000 (21:59 +0000)]
I forgot that we always spill fp values as 64-bits. Implement spill folding
for FP as well. This triggers a couple dozen times on 177.mesa (for example).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23299
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Fri, 9 Sep 2005 21:46:49 +0000 (21:46 +0000)]
Fix a problem that Nate noticed, where spill code was not getting coallesced
with copies, leading to code like this:
lwz r4, 380(r1)
or r10, r4, r4 ;; Last use of r4
By teaching the PPC backend how to fold spills into copies, we now get this
code:
lwz r10, 380(r1)
wow. :)
This reduces a testcase nate sent me from 1505 instructions to 1484.
Note that this could handle FP values but doesn't currently, for reasons
mentioned in the patch
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23298
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Fri, 9 Sep 2005 20:51:08 +0000 (20:51 +0000)]
code cleanup
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23297
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Fri, 9 Sep 2005 20:29:51 +0000 (20:29 +0000)]
Use continue in the use-processing loop to make it clear what the early exits
are, simplify logic, and cause things to not be nested as deeply. This also
uses MRI->areAliases instead of an explicit loop.
No functionality change, just code cleanup.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23296
91177308-0d34-0410-b5e6-
96231b3b80d8
Nate Begeman [Fri, 9 Sep 2005 19:49:52 +0000 (19:49 +0000)]
Last round of 2-node folds from SD.cpp. Will move on to 3 node ops such
as setcc and select next.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23295
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Fri, 9 Sep 2005 19:19:20 +0000 (19:19 +0000)]
remove debugging code *slaps head*
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23294
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Fri, 9 Sep 2005 19:17:47 +0000 (19:17 +0000)]
When spilling a live range that is used multiple times by one instruction,
only add a reload live range once for the instruction. This is one step
towards fixing a regalloc pessimization that Nate notice, but is later undone
by the spiller (so no code is changed).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23293
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Fri, 9 Sep 2005 18:17:41 +0000 (18:17 +0000)]
Teach the code generator that rlwimi is commutable if the rotate amount
is zero. This lets the register allocator elide some copies in some cases.
This implements CodeGen/PowerPC/rlwimi-commute.ll
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23292
91177308-0d34-0410-b5e6-
96231b3b80d8
Jim Laskey [Fri, 9 Sep 2005 17:50:20 +0000 (17:50 +0000)]
Added targets to speed up build of llc.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23291
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Fri, 9 Sep 2005 17:48:57 +0000 (17:48 +0000)]
New testcase, neither should require a register-register copy
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23290
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Fri, 9 Sep 2005 01:15:01 +0000 (01:15 +0000)]
add an accessor to provide more checking
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23289
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Fri, 9 Sep 2005 01:11:44 +0000 (01:11 +0000)]
use new accessors to simplify code. Add checking to make sure top-level instr
definitions are void
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23288
91177308-0d34-0410-b5e6-
96231b3b80d8