Displaying 16 results from an estimated 16 matches for "tmp45".
Did you mean:
tmp4
2012 Oct 17
4
[LLVMdev] Redundant Add Operation in Code Generation?
I'm curious why I am seeing this:
*%uglygep18.sum = add i32 %lsr_iv8, %tmp45*
%scevgep19 = getelementptr i8* %parBits_017, i32 %uglygep18_sum
%scevgep1920 = bitcast i8* %scevgep19 to i16*
%tmp78 = load i16* %scevgep1920, align 2
* %uglygep14.sum = add i32 %lsr_iv8, %tmp45*
%scevgep15 = getelementptr i8* %extIn_013, i32 %uglygep14_sum
%scevgep1516 = bitcast i8* %s...
2012 Oct 17
0
[LLVMdev] Redundant Add Operation in Code Generation?
On Wed, Oct 17, 2012 at 1:22 PM, Ryan Taylor <ryta1203 at gmail.com> wrote:
> I'm curious why I am seeing this:
>
> %uglygep18.sum = add i32 %lsr_iv8, %tmp45
> %scevgep19 = getelementptr i8* %parBits_017, i32 %uglygep18_sum
> %scevgep1920 = bitcast i8* %scevgep19 to i16*
> %tmp78 = load i16* %scevgep1920, align 2
> %uglygep14.sum = add i32 %lsr_iv8, %tmp45
> %scevgep15 = getelementptr i8* %extIn_013, i32 %uglygep14_sum
> %s...
2012 Oct 19
3
[LLVMdev] Redundant Add Operation in Code Generation?
...an itself up? Or am I mistaken?
On Fri, Oct 19, 2012 at 1:29 PM, Andrew Trick <atrick at apple.com> wrote:
>
> On Oct 17, 2012, at 1:22 PM, Ryan Taylor <ryta1203 at gmail.com> wrote:
>
> I'm curious why I am seeing this:
>
> *%uglygep18.sum = add i32 %lsr_iv8, %tmp45*
> %scevgep19 = getelementptr i8* %parBits_017, i32 %uglygep18_sum
> %scevgep1920 = bitcast i8* %scevgep19 to i16*
> %tmp78 = load i16* %scevgep1920, align 2
> * %uglygep14.sum = add i32 %lsr_iv8, %tmp45*
> %scevgep15 = getelementptr i8* %extIn_013, i32 %uglygep14_sum
>...
2012 Oct 19
0
[LLVMdev] Redundant Add Operation in Code Generation?
On Oct 17, 2012, at 1:22 PM, Ryan Taylor <ryta1203 at gmail.com> wrote:
> I'm curious why I am seeing this:
>
> %uglygep18.sum = add i32 %lsr_iv8, %tmp45
> %scevgep19 = getelementptr i8* %parBits_017, i32 %uglygep18_sum
> %scevgep1920 = bitcast i8* %scevgep19 to i16*
> %tmp78 = load i16* %scevgep1920, align 2
> %uglygep14.sum = add i32 %lsr_iv8, %tmp45
> %scevgep15 = getelementptr i8* %extIn_013, i32 %uglygep14_sum
> %s...
2012 Oct 19
2
[LLVMdev] Redundant Add Operation in Code Generation?
...> On Fri, Oct 19, 2012 at 1:29 PM, Andrew Trick <atrick at apple.com> wrote:
>
>>
>> On Oct 17, 2012, at 1:22 PM, Ryan Taylor <ryta1203 at gmail.com> wrote:
>>
>> I'm curious why I am seeing this:
>>
>> *%uglygep18.sum = add i32 %lsr_iv8, %tmp45*
>> %scevgep19 = getelementptr i8* %parBits_017, i32 %uglygep18_sum
>> %scevgep1920 = bitcast i8* %scevgep19 to i16*
>> %tmp78 = load i16* %scevgep1920, align 2
>> * %uglygep14.sum = add i32 %lsr_iv8, %tmp45*
>> %scevgep15 = getelementptr i8* %extIn_013, i32 %...
2012 Oct 17
2
[LLVMdev] Redundant Add Operation in Code Generation?
...to this)?
Thanks.
On Wed, Oct 17, 2012 at 1:44 PM, Eli Friedman <eli.friedman at gmail.com>wrote:
> On Wed, Oct 17, 2012 at 1:22 PM, Ryan Taylor <ryta1203 at gmail.com> wrote:
> > I'm curious why I am seeing this:
> >
> > %uglygep18.sum = add i32 %lsr_iv8, %tmp45
> > %scevgep19 = getelementptr i8* %parBits_017, i32 %uglygep18_sum
> > %scevgep1920 = bitcast i8* %scevgep19 to i16*
> > %tmp78 = load i16* %scevgep1920, align 2
> > %uglygep14.sum = add i32 %lsr_iv8, %tmp45
> > %scevgep15 = getelementptr i8* %extIn_013, i32...
2012 Oct 19
0
[LLVMdev] Redundant Add Operation in Code Generation?
...staken?
>
> On Fri, Oct 19, 2012 at 1:29 PM, Andrew Trick <atrick at apple.com> wrote:
>
> On Oct 17, 2012, at 1:22 PM, Ryan Taylor <ryta1203 at gmail.com> wrote:
>
>> I'm curious why I am seeing this:
>>
>> %uglygep18.sum = add i32 %lsr_iv8, %tmp45
>> %scevgep19 = getelementptr i8* %parBits_017, i32 %uglygep18_sum
>> %scevgep1920 = bitcast i8* %scevgep19 to i16*
>> %tmp78 = load i16* %scevgep1920, align 2
>> %uglygep14.sum = add i32 %lsr_iv8, %tmp45
>> %scevgep15 = getelementptr i8* %extIn_013, i32 %ugl...
2012 Oct 19
0
[LLVMdev] Redundant Add Operation in Code Generation?
...i, Oct 19, 2012 at 1:29 PM, Andrew Trick <atrick at apple.com> wrote:
>>
>> On Oct 17, 2012, at 1:22 PM, Ryan Taylor <ryta1203 at gmail.com> wrote:
>>
>>> I'm curious why I am seeing this:
>>>
>>> %uglygep18.sum = add i32 %lsr_iv8, %tmp45
>>> %scevgep19 = getelementptr i8* %parBits_017, i32 %uglygep18_sum
>>> %scevgep1920 = bitcast i8* %scevgep19 to i16*
>>> %tmp78 = load i16* %scevgep1920, align 2
>>> %uglygep14.sum = add i32 %lsr_iv8, %tmp45
>>> %scevgep15 = getelementptr i8*...
2012 Oct 17
0
[LLVMdev] Redundant Add Operation in Code Generation?
...On Wed, Oct 17, 2012 at 1:44 PM, Eli Friedman <eli.friedman at gmail.com>wrote:
>
>> On Wed, Oct 17, 2012 at 1:22 PM, Ryan Taylor <ryta1203 at gmail.com> wrote:
>> > I'm curious why I am seeing this:
>> >
>> > %uglygep18.sum = add i32 %lsr_iv8, %tmp45
>> > %scevgep19 = getelementptr i8* %parBits_017, i32 %uglygep18_sum
>> > %scevgep1920 = bitcast i8* %scevgep19 to i16*
>> > %tmp78 = load i16* %scevgep1920, align 2
>> > %uglygep14.sum = add i32 %lsr_iv8, %tmp45
>> > %scevgep15 = getelementptr...
2007 Nov 29
2
[LLVMdev] LLVM and OpenMP
...%.omp_data_i) {
entry:
%tmp1 = tail call i32 (...)*
@omp_get_thread_num( ) ; <i32> [#uses=1]
%tmp3 = bitcast i8* %.omp_data_i to i32** ;
<i32**> [#uses=1]
%tmp4 = load i32** %tmp3, align 4 ; <i32*>
[#uses=1]
%tmp45 = bitcast i32* %tmp4 to i8* ; <i8*>
[#uses=1]
%tmp6 = tail call i32 @__sync_fetch_and_add_4( i8* %tmp45, i32 %
tmp1 ) nounwind ; <i32> [#uses=0]
ret void
}
--
With best regards, Anton Korobeynikov.
Faculty of Mathematics & Mechanics, S...
2008 Mar 04
0
[LLVMdev] Deleting Instructions after Intrinsic Creation
...; <i32> [#uses=3]
%tmp12.i = sitofp i32 %tmp5.i to double ; <double> [#uses=1]
%tmp3.i = sub double %tmp12.i, 1.000000e+00 ; <double>
[#uses=3]
%tmp4.i1 = fdiv double %tmp3.i, 4.000000e+00 ; <double>
[#uses=1]
%tmp45.i = fptosi double %tmp4.i1 to i32 ; <i32>
[#uses=1]
%tmp9.i = fdiv double %tmp3.i, 1.000000e+02 ; <double>
[#uses=1]
%tmp910.i = fptosi double %tmp9.i to i32 ; <i32>
[#uses=1]
%tmp14.i = fdiv double %tmp3.i, 4.00...
2008 Mar 04
1
[LLVMdev] Deleting Instructions after Intrinsic Creation
...; <i32> [#uses=3]
%tmp12.i = sitofp i32 %tmp5.i to double
; <double> [#uses=1]
%tmp3.i = sub double %tmp12.i, 1.000000e+00
; <double> [#uses=3]
%tmp4.i1 = fdiv double %tmp3.i, 4.000000e+00
; <double> [#uses=1]
%tmp45.i = fptosi double %tmp4.i1 to i32
; <i32> [#uses=1]
%tmp9.i = fdiv double %tmp3.i, 1.000000e+02
; <double> [#uses=1]
%tmp910.i = fptosi double %tmp9.i to i32
; <i32> [#uses=1]
%tmp14.i = fdiv double %tmp3.i, 4.00000...
2008 Jan 12
1
[LLVMdev] Labels
...tmp1133 = icmp slt i32 %tmp931, %tmp1032
br i1 %tmp1133, label %bb13, label %bb
bb:
%tmp.rle = phi i32 [ %tmp1032, %entry ], [ %tmp10, %bb ]
%tmp1 = shl i32 %tmp.rle, 1
store i32 %tmp1, i32* @yythunkslen, align 4
%tmp3 = shl i32 %tmp.rle, 5
%tmp4 = load %struct.yythunk** @yythunks, align 4
%tmp45 = bitcast %struct.yythunk* %tmp4 to i8*
%tmp6 = tail call i8* @realloc( i8* %tmp45, i32 %tmp3 )
%tmp67 = bitcast i8* %tmp6 to %struct.yythunk*
store %struct.yythunk* %tmp67, %struct.yythunk** @yythunks, align 4
%tmp9 = load i32* @yythunkpos, align 4
%tmp10 = load i32* @yythunkslen, align 4
%t...
2010 Sep 29
0
[LLVMdev] spilling & xmm register usage
...00000
> %tmp33.i = fadd float %tmp23.i, %tmp32.i
> %call36.i = tail call float @llvm.sqrt.f32(float %tmp21.i) nounwind
> %tmp37.i = fmul float %tmp33.i, %call36.i
> %tmp40.i = fdiv float %tmp9.i, %tmp9.i
> %call41.i = tail call float @llvm.log.f32(float %tmp40.i) nounwind
> %tmp45.i = fmul float %tmp33.i, %tmp33.i
> %tmp47.i = fdiv float %tmp45.i, 2.000000e+00
> %tmp48.i = fadd float %tmp27.i, %tmp47.i
> %tmp50.i = fmul float %tmp48.i, %tmp21.i
> %tmp51.i = fadd float %call41.i, %tmp50.i
> %tmp53.i = fdiv float %tmp51.i, %tmp37.i
> %tmp56.i = fsub flo...
2010 Sep 29
3
[LLVMdev] spilling & xmm register usage
Hello everybody,
I have stumbled upon a test case (the attached module is a slightly
reduced version) that shows extremely reduced performance on linux
compared to windows when executed using LLVM's JIT.
We narrowed the problem down to the actual code being generated, the
source IR on both systems is the same.
Try compiling the attached module:
llc -O3 -filetype=asm -o BAD.s BAD.ll
Under
2008 Jun 10
3
[LLVMdev] DejaGNU test fixes
...s=0]
store i32 %a, i32* %a_addr
%tmp1 = load i32* %a_addr, align 4 ; <i32> [#uses=1]
store i32 %tmp1, i32* %a.0, align 4
%tmp2 = load i32* %a.0, align 4 ; <i32> [#uses=1]
%tmp3 = sub i32 %tmp2, 1 ; <i32> [#uses=0]
%tmp4 = load i32* %a.0, align 4 ; <i32> [#uses=1]
%tmp45 = zext i32 %tmp4 to i64 ; <i64> [#uses=1]
%mask = and i64 %tmp45, 68719476735 ; <i64> [#uses=2]
%mask6 = and i64 %mask, 68719476735 ; <i64> [#uses=0]
%mask7 = and i64 %mask, 68719476735 ; <i64> [#uses=1]
%tmp8 = mul i64 %mask7, 32 ; <i64> [#uses=1]
%mask9 = a...