diff options
author | Matthias Braun <matze@braunis.de> | 2015-07-14 02:08:26 +0000 |
---|---|---|
committer | Matthias Braun <matze@braunis.de> | 2015-07-14 02:08:26 +0000 |
commit | 4ac4ecdadf2ed0e8110085affe6485d231d0e14b (patch) | |
tree | 059b57e040de92feca3c18821f60b1d1acac06ca /llvm/test | |
parent | b457ed33123bc77077b3794a5c72c83b40eec618 (diff) | |
download | bcm5719-llvm-4ac4ecdadf2ed0e8110085affe6485d231d0e14b.tar.gz bcm5719-llvm-4ac4ecdadf2ed0e8110085affe6485d231d0e14b.zip |
LegalizeDAG: Fix and improve FCOPYSIGN/FABS legalization
- Factor out code to query and modify the sign bit of a floatingpoint
value as an integer. This also works if none of the targets integer
types is big enough to hold all bits of the floatingpoint value.
- Legalize FABS(x) as FCOPYSIGN(x, 0.0) if FCOPYSIGN is available,
otherwise perform bit manipulation on the sign bit. The previous code
used "x >u 0 ? x : -x" which is incorrect for x being -0.0! It also
takes 34 instructions on ARM Cortex-M4. With this patch we only
require 5:
vldr d0, LCPI0_0
vmov r2, r3, d0
lsrs r2, r3, #31
bfi r1, r2, #31, #1
bx lr
(This could be further improved if the compiler would recognize that
r2, r3 is zero).
- Only lower FCOPYSIGN(x, y) = sign(x) ? -FABS(x) : FABS(x) if FABS is
available otherwise perform bit manipulation on the sign bit.
- Perform the sign(x) test by masking out the sign bit and comparing
with 0 rather than shifting the sign bit to the highest position and
testing for "<s 0". For x86 copysignl (on 80bit values) this gets us:
testl $32768, %eax
rather than:
shlq $48, %rax
sets %al
testb %al, %al
llvm-svn: 242107
Diffstat (limited to 'llvm/test')
-rw-r--r-- | llvm/test/CodeGen/Thumb2/float-intrinsics-double.ll | 9 | ||||
-rw-r--r-- | llvm/test/CodeGen/X86/pr13577.ll | 4 |
2 files changed, 7 insertions, 6 deletions
diff --git a/llvm/test/CodeGen/Thumb2/float-intrinsics-double.ll b/llvm/test/CodeGen/Thumb2/float-intrinsics-double.ll index 01a23bd0fe6..b34176d41a4 100644 --- a/llvm/test/CodeGen/Thumb2/float-intrinsics-double.ll +++ b/llvm/test/CodeGen/Thumb2/float-intrinsics-double.ll @@ -109,9 +109,12 @@ declare double @llvm.fabs.f64(double %Val) define double @abs_d(double %a) { ; CHECK-LABEL: abs_d: ; NONE: bic r1, r1, #-2147483648 -; SP: bl __aeabi_dcmpgt -; SP: bl __aeabi_dcmpun -; SP: bl __aeabi_dsub +; SP: vldr d1, .LCPI{{.*}} +; SP: vmov r0, r1, d0 +; SP: vmov r2, r3, d1 +; SP: lsrs r2, r3, #31 +; SP: bfi r1, r2, #31, #1 +; SP: vmov d0, r0, r1 ; DP: vabs.f64 d0, d0 %1 = call double @llvm.fabs.f64(double %a) ret double %1 diff --git a/llvm/test/CodeGen/X86/pr13577.ll b/llvm/test/CodeGen/X86/pr13577.ll index 090c7262d6e..13b3665d1a7 100644 --- a/llvm/test/CodeGen/X86/pr13577.ll +++ b/llvm/test/CodeGen/X86/pr13577.ll @@ -7,9 +7,7 @@ ; CHECK-LABEL: foo: ; CHECK: movq {{.*}}, %rax -; CHECK: shlq $48, %rax -; CHECK: sets %al -; CHECK: testb %al, %al +; CHECK: testl $32768, %eax ; CHECK: flds LCPI0_0(%rip) ; CHECK: flds LCPI0_1(%rip) ; CHECK: fcmovne %st(1), %st(0) |