diff options
author | Hal Finkel <hfinkel@anl.gov> | 2014-02-28 00:27:01 +0000 |
---|---|---|
committer | Hal Finkel <hfinkel@anl.gov> | 2014-02-28 00:27:01 +0000 |
commit | 36e1825e686c8c9bf47bcf4dfc9a634c4a91dcb2 (patch) | |
tree | 593ff59cbbcedbf8a1db750f43358a53cdf06beb /test/CodeGen | |
parent | c72cf87285d46a60c3f672f5a99f1d85f6074f30 (diff) | |
download | llvm-36e1825e686c8c9bf47bcf4dfc9a634c4a91dcb2.tar.gz llvm-36e1825e686c8c9bf47bcf4dfc9a634c4a91dcb2.tar.bz2 llvm-36e1825e686c8c9bf47bcf4dfc9a634c4a91dcb2.tar.xz |
Add CR-bit tracking to the PowerPC backend for i1 values
This change enables tracking i1 values in the PowerPC backend using the
condition register bits. These bits can be treated on PowerPC as separate
registers; individual bit operations (and, or, xor, etc.) are supported.
Tracking booleans in CR bits has several advantages:
- Reduction in register pressure (because we no longer need GPRs to store
boolean values).
- Logical operations on booleans can be handled more efficiently; we used to
have to move all results from comparisons into GPRs, perform promoted
logical operations in GPRs, and then move the result back into condition
register bits to be used by conditional branches. This can be very
inefficient, because the throughput of these CR <-> GPR moves have high
latency and low throughput (especially when other associated instructions
are accounted for).
- On the POWER7 and similar cores, we can increase total throughput by using
the CR bits. CR bit operations have a dedicated functional unit.
Most of this is more-or-less mechanical: Adjustments were needed in the
calling-convention code, support was added for spilling/restoring individual
condition-register bits, and conditional branch instruction definitions taking
specific CR bits were added (plus patterns and code for generating bit-level
operations).
This is enabled by default when running at -O2 and higher. For -O0 and -O1,
where the ability to debug is more important, this feature is disabled by
default. Individual CR bits do not have assigned DWARF register numbers,
and storing values in CR bits makes them invisible to the debugger.
It is critical, however, that we don't move i1 values that have been promoted
to larger values (such as those passed as function arguments) into bit
registers only to quickly turn around and move the values back into GPRs (such
as happens when values are returned by functions). A pair of target-specific
DAG combines are added to remove the trunc/extends in:
trunc(binary-ops(binary-ops(zext(x), zext(y)), ...)
and:
zext(binary-ops(binary-ops(trunc(x), trunc(y)), ...)
In short, we only want to use CR bits where some of the i1 values come from
comparisons or are used by conditional branches or selects. To put it another
way, if we can do the entire i1 computation in GPRs, then we probably should
(on the POWER7, the GPR-operation throughput is higher, and for all cores, the
CR <-> GPR moves are expensive).
POWER7 test-suite performance results (from 10 runs in each configuration):
SingleSource/Benchmarks/Misc/mandel-2: 35% speedup
MultiSource/Benchmarks/Prolangs-C++/city/city: 21% speedup
MultiSource/Benchmarks/MiBench/automotive-susan: 23% speedup
SingleSource/Benchmarks/CoyoteBench/huffbench: 13% speedup
SingleSource/Benchmarks/Misc-C++/Large/sphereflake: 13% speedup
SingleSource/Benchmarks/Misc-C++/mandel-text: 10% speedup
SingleSource/Benchmarks/Misc-C++-EH/spirit: 10% slowdown
MultiSource/Applications/lemon/lemon: 8% slowdown
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@202451 91177308-0d34-0410-b5e6-96231b3b80d8
Diffstat (limited to 'test/CodeGen')
-rw-r--r-- | test/CodeGen/PowerPC/bdzlr.ll | 9 | ||||
-rw-r--r-- | test/CodeGen/PowerPC/crbits.ll | 174 | ||||
-rw-r--r-- | test/CodeGen/PowerPC/early-ret2.ll | 6 | ||||
-rw-r--r-- | test/CodeGen/PowerPC/fold-zero.ll | 13 | ||||
-rw-r--r-- | test/CodeGen/PowerPC/optcmp.ll | 2 | ||||
-rw-r--r-- | test/CodeGen/PowerPC/rlwimi-and.ll | 2 | ||||
-rw-r--r-- | test/CodeGen/PowerPC/sdag-ppcf128.ll | 2 | ||||
-rw-r--r-- | test/CodeGen/PowerPC/setcc_no_zext.ll | 4 | ||||
-rw-r--r-- | test/CodeGen/PowerPC/seteq-0.ll | 7 | ||||
-rw-r--r-- | test/CodeGen/PowerPC/subsumes-pred-regs.ll | 2 |
10 files changed, 212 insertions, 9 deletions
diff --git a/test/CodeGen/PowerPC/bdzlr.ll b/test/CodeGen/PowerPC/bdzlr.ll index e487558e94..29b74c6c8c 100644 --- a/test/CodeGen/PowerPC/bdzlr.ll +++ b/test/CodeGen/PowerPC/bdzlr.ll @@ -1,4 +1,5 @@ -; RUN: llc < %s -mtriple=powerpc64-unknown-linux-gnu -mcpu=pwr7 | FileCheck %s +; RUN: llc < %s -mtriple=powerpc64-unknown-linux-gnu -mcpu=pwr7 -mattr=-crbits | FileCheck %s +; RUN: llc < %s -mtriple=powerpc64-unknown-linux-gnu -mcpu=pwr7 | FileCheck %s -check-prefix=CHECK-CRB target datalayout = "E-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-f128:128:128-v128:128:128-n32:64" target triple = "powerpc64-unknown-linux-gnu" @@ -54,6 +55,12 @@ for.end: ; preds = %for.body, %if.end, ; CHECK: bnelr ; CHECK: bdzlr ; CHECK-NOT: blr + +; CHECK-CRB: @lua_xmove +; CHECK-CRB: bclr 12, +; CHECK-CRB: bclr 12, +; CHECK-CRB: bdzlr +; CHECK-CRB-NOT: blr } attributes #0 = { nounwind } diff --git a/test/CodeGen/PowerPC/crbits.ll b/test/CodeGen/PowerPC/crbits.ll new file mode 100644 index 0000000000..998e940e8a --- /dev/null +++ b/test/CodeGen/PowerPC/crbits.ll @@ -0,0 +1,174 @@ +; RUN: llc -mcpu=pwr7 < %s | FileCheck %s +target datalayout = "E-m:e-i64:64-n32:64" +target triple = "powerpc64-unknown-linux-gnu" + +; FIXME: For a number of these we load (1, 0) for the isel into two registers, +; whereas if we reverse the condition, we could use only one register (using ZERO +; for 0 in the isel). + +; Function Attrs: nounwind readnone +define zeroext i1 @test1(float %v1, float %v2) #0 { +entry: + %cmp = fcmp oge float %v1, %v2 + %cmp2 = fcmp ole float %v2, 0.000000e+00 + %and5 = and i1 %cmp, %cmp2 + ret i1 %and5 + +; CHECK-LABEL: @test1 +; CHECK-DAG: fcmpu {{[0-9]+}}, 1, 2 +; CHECK-DAG: li [[REG1:[0-9]+]], 1 +; CHECK-DAG: lfs [[REG2:[0-9]+]], +; CHECK-DAG: li [[REG3:[0-9]+]], 0 +; CHECK-DAG: fcmpu {{[0-9]+}}, 2, [[REG2]] +; CHECK: crnor +; CHECK: crnor +; CHECK: crand [[REG4:[0-9]+]], +; CHECK: isel 3, [[REG1]], [[REG3]], [[REG4]] +; CHECK: blr +} + +; Function Attrs: nounwind readnone +define zeroext i1 @test2(float %v1, float %v2) #0 { +entry: + %cmp = fcmp oge float %v1, %v2 + %cmp2 = fcmp ole float %v2, 0.000000e+00 + %xor5 = xor i1 %cmp, %cmp2 + ret i1 %xor5 + +; CHECK-LABEL: @test2 +; CHECK-DAG: fcmpu {{[0-9]+}}, 1, 2 +; CHECK-DAG: li [[REG1:[0-9]+]], 1 +; CHECK-DAG: lfs [[REG2:[0-9]+]], +; CHECK-DAG: li [[REG3:[0-9]+]], 0 +; CHECK-DAG: fcmpu {{[0-9]+}}, 2, [[REG2]] +; CHECK: crnor +; CHECK: crnor +; CHECK: crxor [[REG4:[0-9]+]], +; CHECK: isel 3, [[REG1]], [[REG3]], [[REG4]] +; CHECK: blr +} + +; Function Attrs: nounwind readnone +define zeroext i1 @test3(float %v1, float %v2, i32 signext %x) #0 { +entry: + %cmp = fcmp oge float %v1, %v2 + %cmp2 = fcmp ole float %v2, 0.000000e+00 + %cmp4 = icmp ne i32 %x, -2 + %and7 = and i1 %cmp2, %cmp4 + %xor8 = xor i1 %cmp, %and7 + ret i1 %xor8 + +; CHECK-LABEL: @test3 +; CHECK-DAG: fcmpu {{[0-9]+}}, 1, 2 +; CHECK-DAG: li [[REG1:[0-9]+]], 1 +; CHECK-DAG: lfs [[REG2:[0-9]+]], +; CHECK-DAG: li [[REG3:[0-9]+]], 0 +; CHECK-DAG: fcmpu {{[0-9]+}}, 2, [[REG2]] +; CHECK: crnor +; CHECK: crnor +; CHECK: crandc +; CHECK: crxor [[REG4:[0-9]+]], +; CHECK: isel 3, [[REG1]], [[REG3]], [[REG4]] +; CHECK: blr +} + +; Function Attrs: nounwind readnone +define zeroext i1 @test4(i1 zeroext %v1, i1 zeroext %v2, i1 zeroext %v3) #0 { +entry: + %and8 = and i1 %v1, %v2 + %or9 = or i1 %and8, %v3 + ret i1 %or9 + +; CHECK-DAG: @test4 +; CHECK: and [[REG1:[0-9]+]], 3, 4 +; CHECK: or 3, [[REG1]], 5 +; CHECK: blr +} + +; Function Attrs: nounwind readnone +define zeroext i1 @test5(i1 zeroext %v1, i1 zeroext %v2, i32 signext %v3) #0 { +entry: + %and6 = and i1 %v1, %v2 + %cmp = icmp ne i32 %v3, -2 + %or7 = or i1 %and6, %cmp + ret i1 %or7 + +; CHECK-LABEL: @test5 +; CHECK-DAG: and [[REG1:[0-9]+]], 3, 4 +; CHECK-DAG: cmpwi {{[0-9]+}}, 5, -2 +; CHECK: li [[REG3:[0-9]+]], 1 +; CHECK: andi. {{[0-9]+}}, [[REG1]], 1 +; CHECK: li [[REG4:[0-9]+]], 0 +; CHECK: crorc [[REG5:[0-9]+]], +; CHECK: isel 3, [[REG3]], [[REG4]], [[REG5]] +; CHECK: blr +} + +; Function Attrs: nounwind readnone +define zeroext i1 @test6(i1 zeroext %v1, i1 zeroext %v2, i32 signext %v3) #0 { +entry: + %cmp = icmp ne i32 %v3, -2 + %or6 = or i1 %cmp, %v2 + %and7 = and i1 %or6, %v1 + ret i1 %and7 + +; CHECK-LABEL: @test6 +; CHECK-DAG: andi. {{[0-9]+}}, 3, 1 +; CHECK-DAG: cmpwi {{[0-9]+}}, 5, -2 +; CHECK-DAG: cror [[REG1:[0-9]+]], 1, 1 +; CHECK: andi. {{[0-9]+}}, 4, 1 +; CHECK: li [[REG2:[0-9]+]], 1 +; CHECK: li [[REG3:[0-9]+]], 0 +; CHECK: crorc [[REG4:[0-9]+]], 1, +; CHECK: crand [[REG5:[0-9]+]], [[REG4]], [[REG1]] +; CHECK: isel 3, [[REG2]], [[REG3]], [[REG5]] +; CHECK: blr +} + +; Function Attrs: nounwind readnone +define signext i32 @test7(i1 zeroext %v2, i32 signext %i1, i32 signext %i2) #0 { +entry: + %cond = select i1 %v2, i32 %i1, i32 %i2 + ret i32 %cond + +; CHECK-LABEL: @test7 +; CHECK: andi. {{[0-9]+}}, 3, 1 +; CHECK: isel [[REG1:[0-9]+]], 4, 5, 1 +; CHECK: extsw 3, [[REG1]] +; CHECK: blr +} + +; Function Attrs: nounwind readnone +define float @test8(i1 zeroext %v2, float %v1, float %v3) #0 { +entry: + %cond = select i1 %v2, float %v1, float %v3 + ret float %cond + +; CHECK-LABEL: @test8 +; CHECK: andi. {{[0-9]+}}, 3, 1 +; CHECK: bclr 12, 1, 0 +; CHECK: fmr 1, 2 +; CHECK: blr +} + +; Function Attrs: nounwind readnone +define signext i32 @test10(i32 signext %v1, i32 signext %v2) #0 { +entry: + %tobool = icmp ne i32 %v1, 0 + %lnot = icmp eq i32 %v2, 0 + %and3 = and i1 %tobool, %lnot + %and = zext i1 %and3 to i32 + ret i32 %and + +; CHECK-LABEL: @test10 +; CHECK-DAG: cmpwi {{[0-9]+}}, 3, 0 +; CHECK-DAG: cmpwi {{[0-9]+}}, 4, 0 +; CHECK-DAG: li [[REG1:[0-9]+]], 0 +; CHECK-DAG: li [[REG2:[0-9]+]], 1 +; CHECK: crandc [[REG3:[0-9]+]], +; CHECK: isel 3, [[REG2]], [[REG1]], [[REG3]] +; CHECK: blr +} + +attributes #0 = { nounwind readnone } + diff --git a/test/CodeGen/PowerPC/early-ret2.ll b/test/CodeGen/PowerPC/early-ret2.ll index a274e2c265..a8e456fea6 100644 --- a/test/CodeGen/PowerPC/early-ret2.ll +++ b/test/CodeGen/PowerPC/early-ret2.ll @@ -1,4 +1,5 @@ -; RUN: llc < %s -mtriple=powerpc64-unknown-linux-gnu -mcpu=pwr7 | FileCheck %s +; RUN: llc < %s -mtriple=powerpc64-unknown-linux-gnu -mcpu=pwr7 -mattr=-crbits | FileCheck %s +; RUN: llc < %s -mtriple=powerpc64-unknown-linux-gnu -mcpu=pwr7 | FileCheck %s -check-prefix=CHECK-CRB target datalayout = "E-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-f128:128:128-v128:128:128-n32:64" target triple = "powerpc64-unknown-linux-gnu" @@ -17,6 +18,9 @@ while.end: ; preds = %while.body, %while. ; CHECK: @_Z8example3iPiS_ ; CHECK: bnelr + +; CHECK-CRB: @_Z8example3iPiS_ +; CHECK-CRB: bclr 12, } attributes #0 = { noinline nounwind } diff --git a/test/CodeGen/PowerPC/fold-zero.ll b/test/CodeGen/PowerPC/fold-zero.ll index c7ec6fade5..c1eea43017 100644 --- a/test/CodeGen/PowerPC/fold-zero.ll +++ b/test/CodeGen/PowerPC/fold-zero.ll @@ -1,4 +1,5 @@ -; RUN: llc < %s -mtriple=powerpc64-unknown-linux-gnu -mcpu=pwr7 | FileCheck %s +; RUN: llc < %s -mtriple=powerpc64-unknown-linux-gnu -mcpu=pwr7 -mattr=-crbits | FileCheck %s +; RUN: llc < %s -mtriple=powerpc64-unknown-linux-gnu -mcpu=pwr7 | FileCheck -check-prefix=CHECK-CRB %s target datalayout = "E-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-f128:128:128-v128:128:128-n32:64" target triple = "powerpc64-unknown-linux-gnu" @@ -12,3 +13,13 @@ define i32 @test1(i1 %a, i32 %c) nounwind { ; CHECK: blr } +define i32 @test2(i1 %a, i32 %c) nounwind { + %x = select i1 %a, i32 0, i32 %c + ret i32 %x + +; CHECK-CRB: @test2 +; CHECK-CRB-NOT: li {{[0-9]+}}, 0 +; CHECK-CRB: isel 3, 0, +; CHECK-CRB: blr +} + diff --git a/test/CodeGen/PowerPC/optcmp.ll b/test/CodeGen/PowerPC/optcmp.ll index 35aabfa52c..d929eae206 100644 --- a/test/CodeGen/PowerPC/optcmp.ll +++ b/test/CodeGen/PowerPC/optcmp.ll @@ -1,4 +1,4 @@ -; RUN: llc < %s -mtriple=powerpc64-unknown-linux-gnu -mcpu=a2 -disable-ppc-cmp-opt=0 | FileCheck %s +; RUN: llc < %s -mtriple=powerpc64-unknown-linux-gnu -mcpu=a2 -mattr=-crbits -disable-ppc-cmp-opt=0 | FileCheck %s target datalayout = "E-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-f128:128:128-v128:128:128-n32:64" target triple = "powerpc64-unknown-linux-gnu" diff --git a/test/CodeGen/PowerPC/rlwimi-and.ll b/test/CodeGen/PowerPC/rlwimi-and.ll index 7963249ddf..213363ee81 100644 --- a/test/CodeGen/PowerPC/rlwimi-and.ll +++ b/test/CodeGen/PowerPC/rlwimi-and.ll @@ -1,4 +1,4 @@ -; RUN: llc -mcpu=pwr7 < %s | FileCheck %s +; RUN: llc -mcpu=pwr7 -mattr=-crbits < %s | FileCheck %s target datalayout = "E-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-f128:128:128-v128:128:128-n32:64" target triple = "powerpc64-bgq-linux" diff --git a/test/CodeGen/PowerPC/sdag-ppcf128.ll b/test/CodeGen/PowerPC/sdag-ppcf128.ll index 535ece6d3d..c46bc6b22d 100644 --- a/test/CodeGen/PowerPC/sdag-ppcf128.ll +++ b/test/CodeGen/PowerPC/sdag-ppcf128.ll @@ -1,4 +1,4 @@ -; RUN: llc -mtriple=powerpc64-unknown-linux-gnu < %s | FileCheck %s +; RUN: llc -mtriple=powerpc64-unknown-linux-gnu -mattr=-crbits < %s | FileCheck %s ; ; PR14751: Unsupported type in SelectionDAG::getConstantFP() diff --git a/test/CodeGen/PowerPC/setcc_no_zext.ll b/test/CodeGen/PowerPC/setcc_no_zext.ll index 9b2036e1dc..467e921f74 100644 --- a/test/CodeGen/PowerPC/setcc_no_zext.ll +++ b/test/CodeGen/PowerPC/setcc_no_zext.ll @@ -1,5 +1,9 @@ ; RUN: llc < %s -march=ppc32 | not grep rlwinm +; FIXME: This optimization has temporarily regressed with crbits enabled by +; default at the default CodeOpt level. +; XFAIL: * + define i32 @setcc_one_or_zero(i32* %a) { entry: %tmp.1 = icmp ne i32* %a, null ; <i1> [#uses=1] diff --git a/test/CodeGen/PowerPC/seteq-0.ll b/test/CodeGen/PowerPC/seteq-0.ll index 731958374e..b7dd78085e 100644 --- a/test/CodeGen/PowerPC/seteq-0.ll +++ b/test/CodeGen/PowerPC/seteq-0.ll @@ -1,9 +1,12 @@ -; RUN: llc < %s -march=ppc32 -mtriple=powerpc-apple-darwin8 | \ -; RUN: grep "srwi r., r., 5" +; RUN: llc < %s -march=ppc32 -mtriple=powerpc-apple-darwin8 | FileCheck %s define i32 @eq0(i32 %a) { %tmp.1 = icmp eq i32 %a, 0 ; <i1> [#uses=1] %tmp.2 = zext i1 %tmp.1 to i32 ; <i32> [#uses=1] ret i32 %tmp.2 + +; CHECK: cntlzw [[REG:r[0-9]+]], r3 +; CHECK: rlwinm r3, [[REG]], 27, 31, 31 +; CHECK: blr } diff --git a/test/CodeGen/PowerPC/subsumes-pred-regs.ll b/test/CodeGen/PowerPC/subsumes-pred-regs.ll index 97ac788164..da637cd254 100644 --- a/test/CodeGen/PowerPC/subsumes-pred-regs.ll +++ b/test/CodeGen/PowerPC/subsumes-pred-regs.ll @@ -1,4 +1,4 @@ -; RUN: llc < %s -mcpu=ppc64 | FileCheck %s +; RUN: llc < %s -mcpu=ppc64 -mattr=-crbits | FileCheck %s target datalayout = "E-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-f128:128:128-v128:128:128-n32:64" target triple = "powerpc64-unknown-linux-gnu" |