thr3ads.net - llvm dev - [LLVMdev] Help!!!!Help!!!! " LLVM ERROR: Cannot select: 0x9fc9680: i32 = fp32_to_fp16 0x9fc0750 [ID=16] " problem!!!!!!!!!!!!!!!!!! [Jul 2014]

If this information is useful, please help other people find it:
Share via:

Renato Golin

2014-Jul-09 08:39 UTC

[LLVMdev] Help!!!!Help!!!! " LLVM ERROR: Cannot select: 0x9fc9680: i32 = fp32_to_fp16 0x9fc0750 [ID=16] " problem!!!!!!!!!!!!!!!!!!

On 9 July 2014 23:30, yalong at multicorewareinc.com
<yalong at multicorewareinc.com> wrote:>     If I use fptrunc and bitcast realise NEON vcvtt ( I can sure,
"fptrunc
> double %tmp to float" is right, but "fptrunc float %tmp to
half" is wrong).
> My target platform is MIPS.
AFAIK, MIPS doesn't support NEON. But that doesn't explain your
problem, which is probably due to hand-crafted IR.

Can you provide the full IR, with header and everything? Also the
command line that you're using with the LLVM tool you're using to
compile (llc?lli?).

It'd also be good to know how you generated that IR in the frist
place. Was it a tool? A front-end? Hand-crafted?

cheers,
--renato

Daniel Sanders

2014-Jul-09 09:05 UTC

head link

[LLVMdev] Help!!!!Help!!!! " LLVM ERROR: Cannot select: 0x9fc9680: i32 = fp32_to_fp16 0x9fc0750 [ID=16] " problem!!!!!!!!!!!!!!!!!!

Hi,

NEON is an ARM feature and is therefore not supported by MIPS so I assume you
are trying to achieve the same effect. As far as I know, the MIPS backend
doesn't support half-precision floating point at the moment.

There is limited support for the <8 x f16> type when MSA (MIPS SIMD
Architecture) is enabled but even then scalar half-precision is not currently
supported.

From: llvmdev-bounces at cs.uiuc.edu [mailto:llvmdev-bounces at cs.uiuc.edu] On
Behalf Of yalong at multicorewareinc.com
Sent: 09 July 2014 23:31
To: Kevin Qin
Cc: llvmdev
Subject: Re: [LLVMdev] Help!!!!Help!!!! " LLVM ERROR: Cannot select:
0x9fc9680: i32 = fp32_to_fp16 0x9fc0750 [ID=16] " problem!!!!!!!!!!!!!!!!!!

    Thank you Kevin!!!
    If I use fptrunc and bitcast realise NEON vcvtt ( I can sure, "fptrunc 
double %tmp to float" is right, but "fptrunc float %tmp to half"
is wrong). My target platform is MIPS.  The command as following:

NEON:
        vcvtt.f16.f32 s2, s0


llvm Code:

%Vt_2 = load float* %VFP_s0, align 4
%Vt3_1 = fptrunc float %Vt_2 to half
%Vt4_1 = bitcast half %Vt3_1 to i16
%Vt2_2 = bitcast float* %VFP_s2 to <2 x i16>*
%Vrti_1 = load <2 x i16>* %Vt2_2, align 4
%Vrti_2 = insertelement <2 x i16> %Vrti_1, i16 %Vt4_1, i32 1
%Vt2_3 = bitcast float* %VFP_s2 to <2 x i16>*
store <2 x i16> %Vrti_2, <2 x i16>* %Vt2_3, align 4


Error Log:
LLVM ERROR: Cannot select: 0x9f554b0: ch = store 0x9d0f28c, 0x9f5d900,
0x9f54ba8, 0x9f54b20<ST2[FixedStack0](align=4), trunc to f16> [ID=52]
0x9f5d900: f32,ch = load 0x9f5e290, 0x9f5dd40, 0x9f54b20<LD4[%sunkaddr69]>
[ORD=1810] [ID=51]
0x9f5dd40: i32 = add 0x9f55318, 0x9f5e0f8 [ORD=1808] [ID=31]
0x9f55318: i32,ch = CopyFromReg 0x9d0f28c, 0x9f6a3a0 [ORD=1796] [ID=26]
0x9f6a3a0: i32 = Register %vreg32 [ORD=1796] [ID=1]
0x9f5e0f8: i32 = Constant<64> [ORD=1808] [ID=17]
0x9f54b20: i32 = undef [ORD=1797] [ID=6]
0x9f54ba8: i32 = FrameIndex<0> [ID=24]
0x9f54b20: i32 = undef [ORD=1797] [ID=6]
In function: testVCVTT32TO16Function


________________________________
yalong at multicorewareinc.com<mailto:yalong at multicorewareinc.com>

From: Kevin Qin<mailto:kevinqindev at gmail.com>
Date: 2014-07-09 00:03
To: yalong at multicorewareinc.com<mailto:yalong at multicorewareinc.com>
CC: llvmdev<mailto:llvmdev at cs.uiuc.edu>
Subject: Re: [LLVMdev] Help!!!!Help!!!! " LLVM ERROR: Cannot select:
0x9fc9680: i32 = fp32_to_fp16 0x9fc0750 [ID=16] " problem!!!!!!!!!!!!!!!!!!
Hi,

Can you show us the command line you are using? At least can you tell us the
backend you tried on? If you can upload the test case as well, it will be very
useful to find out the problem.


Regards,
Kevin

2014-07-10 1:53 GMT+08:00 yalong at multicorewareinc.com<mailto:yalong at
multicorewareinc.com> <yalong at multicorewareinc.com<mailto:yalong at
multicorewareinc.com>>:
Hi all,
    I am new to llvm. I need help. Thank you every!

    I want to realize vcvtt.f16.f32  NEON instruction by llvm. This instruction
covert top-16bits of a single type to f16. I use the intrinsics function
llvm.convert.to.fp16, but cannot llc , I meet is following problem :

LLVM ERROR: Cannot select: 0x9fc9680: i32 = fp32_to_fp16 0x9fc0750 [ID=16]
0x9fc0750: f32,ch = load 0x3aafd68, 0x9fc2a20,
0x9feaab0<LD4[%sunkaddr85033]> [ORD=125117] [ID=15]
0x9fc2a20: i32 = add 0x9fed880, 0x9fd9ea0 [ORD=125115] [ID=13]
0x9fed880: i32,ch = CopyFromReg 0x3aafd68, 0x9fbea90 [ORD=125114] [ID=9]
0x9fbea90: i32 = Register %vreg13999 [ORD=125114] [ID=1]
0x9fd9ea0: i32 = Constant<80> [ORD=125115] [ID=2]
0x9feaab0: i32 = undef [ORD=125117] [ID=4]
In function: internal_function_69
Command exited with non-zero status 1

    If I change the method, I use " %1 = fptrunc float %0  to half ",
then " %2 = bitcast half %1 to i16", I meet samiliar problem, the log
is following:

LLVM ERROR: Cannot select: 0x9f554b0: ch = store 0x9d0f28c, 0x9f5d900,
0x9f54ba8, 0x9f54b20<ST2[FixedStack0](align=4), trunc to f16> [ID=52]
0x9f5d900: f32,ch = load 0x9f5e290, 0x9f5dd40, 0x9f54b20<LD4[%sunkaddr69]>
[ORD=1810] [ID=51]
0x9f5dd40: i32 = add 0x9f55318, 0x9f5e0f8 [ORD=1808] [ID=31]
0x9f55318: i32,ch = CopyFromReg 0x9d0f28c, 0x9f6a3a0 [ORD=1796] [ID=26]
0x9f6a3a0: i32 = Register %vreg32 [ORD=1796] [ID=1]
0x9f5e0f8: i32 = Constant<64> [ORD=1808] [ID=17]
0x9f54b20: i32 = undef [ORD=1797] [ID=6]
0x9f54ba8: i32 = FrameIndex<0> [ID=24]
0x9f54b20: i32 = undef [ORD=1797] [ID=6]
In function: testVCVTT32TO16Function


    Anyone can help me?? Thank you again.

________________________________
yalong at multicorewareinc.com<mailto:yalong at multicorewareinc.com>

_______________________________________________
LLVM Developers mailing list
LLVMdev at cs.uiuc.edu<mailto:LLVMdev at cs.uiuc.edu>        
http://llvm.cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev



--
Best Regards,

Kevin Qin
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20140709/1b53e912/attachment.html>

Matt Arsenault

2014-Jul-09 19:41 UTC

head link

[LLVMdev] Help!!!!Help!!!! " LLVM ERROR: Cannot select: 0x9fc9680: i32 = fp32_to_fp16 0x9fc0750 [ID=16] " problem!!!!!!!!!!!!!!!!!!

On 07/09/2014 03:30 PM, yalong at multicorewareinc.com
wrote:>     Thank you Kevin!!!
>     If I use fptrunc and bitcast realise NEON vcvtt ( I can sure, 
> "fptrunc  double %tmp to float" is right, but "fptrunc float
%tmp to
> half" is wrong). My target platform is MIPS.  The command as
following:
>
> NEON:
> vcvtt.f16.f32 s2, s0
>
> llvm Code:
>
> %Vt_2 = load float* %VFP_s0, align 4
> %Vt3_1 = fptrunc float %Vt_2 to half
> %Vt4_1 = bitcast half %Vt3_1 to i16
> %Vt2_2 = bitcast float* %VFP_s2 to <2 x i16>*
> %Vrti_1 = load <2 x i16>* %Vt2_2, align 4
> %Vrti_2 = insertelement <2 x i16> %Vrti_1, i16 %Vt4_1, i32 1
> %Vt2_3 = bitcast float* %VFP_s2 to <2 x i16>*
> store <2 x i16> %Vrti_2, <2 x i16>* %Vt2_3, align 4
>
> Error Log:
> LLVM ERROR: Cannot select: 0x9f554b0: ch = store 0x9d0f28c, 0x9f5d900, 
> 0x9f54ba8, 0x9f54b20<ST2[FixedStack0](align=4), trunc to f16> [ID=52]
> 0x9f5d900: f32,ch = load 0x9f5e290, 0x9f5dd40, 
> 0x9f54b20<LD4[%sunkaddr69]> [ORD=1810] [ID=51]
> 0x9f5dd40: i32 = add 0x9f55318, 0x9f5e0f8 [ORD=1808] [ID=31]
> 0x9f55318: i32,ch = CopyFromReg 0x9d0f28c, 0x9f6a3a0 [ORD=1796] [ID=26]
> 0x9f6a3a0: i32 = Register %vreg32 [ORD=1796] [ID=1]
> 0x9f5e0f8: i32 = Constant<64> [ORD=1808] [ID=17]
> 0x9f54b20: i32 = undef [ORD=1797] [ID=6]
> 0x9f54ba8: i32 = FrameIndex<0> [ID=24]
> 0x9f54b20: i32 = undef [ORD=1797] [ID=6]
> In function: testVCVTT32TO16Function
>
>
I think that support for the half type is only implemented for ARM. Last 
I tried to use it, I found that none of it works even on x86, and the 
current handling of the half conversion SDNodes seem to rely on ARM 
specific assumptions
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20140709/e205111c/attachment.html>

Matt Arsenault

2014-Jul-09 19:51 UTC

head link

[LLVMdev] Help!!!!Help!!!! " LLVM ERROR: Cannot select: 0x9fc9680: i32 = fp32_to_fp16 0x9fc0750 [ID=16] " problem!!!!!!!!!!!!!!!!!!

On 07/09/2014 12:41 PM, Matt Arsenault wrote:> On 07/09/2014 03:30 PM, yalong at multicorewareinc.com wrote:
>>     Thank you Kevin!!!
>>     If I use fptrunc and bitcast realise NEON vcvtt ( I can sure, 
>> "fptrunc  double %tmp to float" is right, but "fptrunc
float %tmp to
>> half" is wrong). My target platform is MIPS.  The command as
following:
>>
>> NEON:
>> vcvtt.f16.f32 s2, s0
>>
>> llvm Code:
>>
>> %Vt_2 = load float* %VFP_s0, align 4
>> %Vt3_1 = fptrunc float %Vt_2 to half
>> %Vt4_1 = bitcast half %Vt3_1 to i16
>> %Vt2_2 = bitcast float* %VFP_s2 to <2 x i16>*
>> %Vrti_1 = load <2 x i16>* %Vt2_2, align 4
>> %Vrti_2 = insertelement <2 x i16> %Vrti_1, i16 %Vt4_1, i32 1
>> %Vt2_3 = bitcast float* %VFP_s2 to <2 x i16>*
>> store <2 x i16> %Vrti_2, <2 x i16>* %Vt2_3, align 4
>>
>> Error Log:
>> LLVM ERROR: Cannot select: 0x9f554b0: ch = store 0x9d0f28c, 
>> 0x9f5d900, 0x9f54ba8, 0x9f54b20<ST2[FixedStack0](align=4), trunc to 
>> f16> [ID=52]
>> 0x9f5d900: f32,ch = load 0x9f5e290, 0x9f5dd40, 
>> 0x9f54b20<LD4[%sunkaddr69]> [ORD=1810] [ID=51]
>> 0x9f5dd40: i32 = add 0x9f55318, 0x9f5e0f8 [ORD=1808] [ID=31]
>> 0x9f55318: i32,ch = CopyFromReg 0x9d0f28c, 0x9f6a3a0 [ORD=1796] [ID=26]
>> 0x9f6a3a0: i32 = Register %vreg32 [ORD=1796] [ID=1]
>> 0x9f5e0f8: i32 = Constant<64> [ORD=1808] [ID=17]
>> 0x9f54b20: i32 = undef [ORD=1797] [ID=6]
>> 0x9f54ba8: i32 = FrameIndex<0> [ID=24]
>> 0x9f54b20: i32 = undef [ORD=1797] [ID=6]
>> In function: testVCVTT32TO16Function
>>
>>
>
> I think that support for the half type is only implemented for ARM. 
> Last I tried to use it, I found that none of it works even on x86, and 
> the current handling of the half conversion SDNodes seem to rely on 
> ARM specific assumptions
Have you tried using the @llvm.convert.to/from.fp16 intrinsics instead?
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20140709/c963d972/attachment.html>

yalong at multicorewareinc.com

2014-Jul-09 22:30 UTC

head link

[LLVMdev] Help!!!!Help!!!! " LLVM ERROR: Cannot select: 0x9fc9680: i32 = fp32_to_fp16 0x9fc0750 [ID=16] " problem!!!!!!!!!!!!!!!!!!

    Thank you Kevin!!!    If I use fptrunc and bitcast realise NEON vcvtt ( I
can sure, "fptrunc  double %tmp to float" is right, but "fptrunc
float %tmp to half" is wrong). My target platform is MIPS.  The command as
following:
NEON:            vcvtt.f16.f32  s2, s0
llvm Code:
 %Vt_2 = load float* %VFP_s0, align 4

 %Vt3_1 = fptrunc float %Vt_2 to half

 %Vt4_1 = bitcast half %Vt3_1 to i16

 %Vt2_2 = bitcast float* %VFP_s2 to  <2 x i16>*

 %Vrti_1 = load <2 x i16>* %Vt2_2, align 4

 %Vrti_2 = insertelement <2 x i16> %Vrti_1, i16 %Vt4_1, i32 1

 %Vt2_3 = bitcast float* %VFP_s2 to <2 x i16>*

 store <2 x i16> %Vrti_2, <2 x i16>* %Vt2_3, align 4 

Error Log:LLVM ERROR: Cannot select: 0x9f554b0: ch = store 0x9d0f28c, 0x9f5d900,
0x9f54ba8, 0x9f54b20<ST2[FixedStack0](align=4), trunc to f16> [ID=52] 
0x9f5d900: f32,ch = load 0x9f5e290, 0x9f5dd40, 0x9f54b20<LD4[%sunkaddr69]>
[ORD=1810] [ID=51] 
0x9f5dd40: i32 = add 0x9f55318, 0x9f5e0f8 [ORD=1808] [ID=31] 
0x9f55318: i32,ch = CopyFromReg 0x9d0f28c, 0x9f6a3a0 [ORD=1796] [ID=26] 
0x9f6a3a0: i32 = Register %vreg32 [ORD=1796] [ID=1] 
0x9f5e0f8: i32 = Constant<64> [ORD=1808] [ID=17] 
0x9f54b20: i32 = undef [ORD=1797] [ID=6] 
0x9f54ba8: i32 = FrameIndex<0> [ID=24] 
0x9f54b20: i32 = undef [ORD=1797] [ID=6] 
In function: testVCVTT32TO16Function 




yalong at multicorewareinc.com
 From: Kevin QinDate: 2014-07-09 00:03To: yalong at
multicorewareinc.comCC: llvmdevSubject: Re: [LLVMdev] Help!!!!Help!!!! "
LLVM ERROR: Cannot select: 0x9fc9680: i32 = fp32_to_fp16 0x9fc0750 [ID=16]
" problem!!!!!!!!!!!!!!!!!!Hi,
Can you show us the command line you are using? At least can you tell us the
backend you tried on? If you can upload the test case as well, it will be very
useful to find out the problem.


Regards,Kevin

2014-07-10 1:53 GMT+08:00 yalong at multicorewareinc.com <yalong at
multicorewareinc.com>:


Hi all,     I am new to llvm. I need help. Thank you every!
    I want to realize vcvtt.f16.f32  NEON instruction by llvm. This instruction
covert top-16bits of a single type to f16. I use the intrinsics function
llvm.convert.to.fp16, but cannot llc , I meet is following problem :

LLVM ERROR: Cannot select: 0x9fc9680: i32 = fp32_to_fp16 0x9fc0750 [ID=16]

  0x9fc0750: f32,ch = load 0x3aafd68, 0x9fc2a20,
0x9feaab0<LD4[%sunkaddr85033]> [ORD=125117] [ID=15]

    0x9fc2a20: i32 = add 0x9fed880, 0x9fd9ea0 [ORD=125115] [ID=13]

      0x9fed880: i32,ch = CopyFromReg 0x3aafd68, 0x9fbea90 [ORD=125114] [ID=9]

        0x9fbea90: i32 = Register %vreg13999 [ORD=125114] [ID=1]

      0x9fd9ea0: i32 = Constant<80> [ORD=125115] [ID=2]

    0x9feaab0: i32 = undef [ORD=125117] [ID=4]

In function: internal_function_69

Command exited with non-zero status 1
    If I change the method, I use " %1 = fptrunc float %0  to half ",
then " %2 = bitcast half %1 to i16", I meet samiliar problem, the log
is following:

LLVM ERROR: Cannot select: 0x9f554b0: ch = store 0x9d0f28c, 0x9f5d900,
0x9f54ba8, 0x9f54b20<ST2[FixedStack0](align=4), trunc to f16> [ID=52]

  0x9f5d900: f32,ch = load 0x9f5e290, 0x9f5dd40,
0x9f54b20<LD4[%sunkaddr69]> [ORD=1810] [ID=51]

    0x9f5dd40: i32 = add 0x9f55318, 0x9f5e0f8 [ORD=1808] [ID=31]

      0x9f55318: i32,ch = CopyFromReg 0x9d0f28c, 0x9f6a3a0 [ORD=1796] [ID=26]

        0x9f6a3a0: i32 = Register %vreg32 [ORD=1796] [ID=1]

      0x9f5e0f8: i32 = Constant<64> [ORD=1808] [ID=17]

    0x9f54b20: i32 = undef [ORD=1797] [ID=6]

  0x9f54ba8: i32 = FrameIndex<0> [ID=24]

  0x9f54b20: i32 = undef [ORD=1797] [ID=6]

In function: testVCVTT32TO16Function 



    Anyone can help me?? Thank you again.

yalong at multicorewareinc.com

_______________________________________________

LLVM Developers mailing list

LLVMdev at cs.uiuc.edu         http://llvm.cs.uiuc.edu

http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev





-- 
Best Regards,
Kevin Qin


-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20140709/be30ec2a/attachment.html>

yalong at multicorewareinc.com

2014-Jul-10 00:40 UTC

head link

[LLVMdev] Help!!!!Help!!!! " LLVM ERROR: Cannot select: 0x9fc9680: i32 = fp32_to_fp16 0x9fc0750 [ID=16] " problem!!!!!!!!!!!!!!!!!!

Hi Renato,
    Thank you your replying.    Yes, we are making a tool in order to generate
MIPS IR by ARM asembler. The ir is  Hand-crafted. We want to realise NEON
instruction "vcvtt.f16.f32" by llvm.     Yes, MIPS is not supported
NEON, but the problem is not related with mips platform. I think this problem
only relate with llc of llvm, so this problem is about convertion between half
float and float.  The problem I don't understand is  if llvm support half
float, how to convert float to half float.        Robin Lau

yalong at multicorewareinc.com
 From: Renato GolinDate: 2014-07-09 01:39To: yalong at
multicorewareinc.comCC: Kevin Qin; llvmdevSubject: Re: [LLVMdev]
Help!!!!Help!!!! " LLVM ERROR: Cannot select: 0x9fc9680: i32 = fp32_to_fp16
0x9fc0750 [ID=16] " problem!!!!!!!!!!!!!!!!!!On 9 July 2014 23:30, yalong
at multicorewareinc.com
<yalong at multicorewareinc.com> wrote:>     If I use fptrunc and bitcast realise NEON vcvtt ( I can sure,
"fptrunc
> double %tmp to float" is right, but "fptrunc float %tmp to
half" is wrong).
> My target platform is MIPS. 
AFAIK, MIPS doesn't support NEON. But that doesn't explain your
problem, which is probably due to hand-crafted IR.

Can you provide the full IR, with header and everything? Also the
command line that you're using with the LLVM tool you're using to
compile (llc?lli?).

It'd also be good to know how you generated that IR in the frist
place. Was it a tool? A front-end? Hand-crafted?

cheers,
--renato

-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20140709/6c3168bf/attachment.html>

yalong at multicorewareinc.com

2014-Jul-10 00:49 UTC

head link

[LLVMdev] Help!!!!Help!!!! " LLVM ERROR: Cannot select: 0x9fc9680: i32 = fp32_to_fp16 0x9fc0750 [ID=16] " problem!!!!!!!!!!!!!!!!!!

Hi Daniel,    Thank you your replying.     Yes, the problem is about MIPS
backend. You give me this message "There is limited support for the <8 x
f16> type when MSA (MIPS SIMD Architecture) is enabled but even then scalar
half-precision is not currently supported."  Could you give me some
official link or some evidence? Thank you very much.
Robin


yalong at multicorewareinc.com
 From: Daniel SandersDate: 2014-07-09 02:05To: yalong at multicorewareinc.com;
Kevin QinCC: llvmdevSubject: RE: [LLVMdev] Help!!!!Help!!!! " LLVM ERROR:
Cannot select: 0x9fc9680: i32 = fp32_to_fp16 0x9fc0750 [ID=16] "
problem!!!!!!!!!!!!!!!!!!







Hi,
 
NEON is an ARM feature and is therefore not supported by MIPS so I assume you
are trying to achieve the same effect. As far as I know, the MIPS backend
doesn't support half-precision
 floating point at the moment.
 
There is limited support for the <8 x f16> type when MSA (MIPS SIMD
Architecture) is enabled but even then scalar half-precision is not currently
supported.
 



From: llvmdev-bounces at cs.uiuc.edu [mailto:llvmdev-bounces at cs.uiuc.edu]
On Behalf Of yalong at multicorewareinc.com

Sent: 09 July 2014 23:31

To: Kevin Qin

Cc: llvmdev

Subject: Re: [LLVMdev] Help!!!!Help!!!! " LLVM ERROR: Cannot select:
0x9fc9680: i32 = fp32_to_fp16 0x9fc0750 [ID=16] " problem!!!!!!!!!!!!!!!!!!


 

    Thank you Kevin!!!


    If I use fptrunc and bitcast realise NEON vcvtt ( I can sure, "fptrunc
 double %tmp to float" is right, but "fptrunc float %tmp to half"
is wrong). My target
 platform is MIPS.  The command as following:


 


NEON:    


        vcvtt.f16.f32 s2, s0









llvm Code:




%Vt_2 = load float* %VFP_s0, align 4 

%Vt3_1 = fptrunc float %Vt_2 to half 

%Vt4_1 = bitcast half %Vt3_1 to i16 

%Vt2_2 = bitcast float* %VFP_s2 to <2 x i16>* 

%Vrti_1 = load <2 x i16>* %Vt2_2, align 4 

%Vrti_2 = insertelement <2 x i16> %Vrti_1, i16 %Vt4_1, i32 1 

%Vt2_3 = bitcast float* %VFP_s2 to <2 x i16>* 

store <2 x i16> %Vrti_2, <2 x i16>* %Vt2_3, align 4 






 


Error Log:



LLVM ERROR: Cannot select: 0x9f554b0: ch = store 0x9d0f28c, 0x9f5d900,
0x9f54ba8, 0x9f54b20<ST2[FixedStack0](align=4), trunc to f16> [ID=52] 

0x9f5d900: f32,ch = load 0x9f5e290, 0x9f5dd40, 0x9f54b20<LD4[%sunkaddr69]>
[ORD=1810] [ID=51] 

0x9f5dd40: i32 = add 0x9f55318, 0x9f5e0f8 [ORD=1808] [ID=31] 

0x9f55318: i32,ch = CopyFromReg 0x9d0f28c, 0x9f6a3a0 [ORD=1796] [ID=26] 

0x9f6a3a0: i32 = Register %vreg32 [ORD=1796] [ID=1] 

0x9f5e0f8: i32 = Constant<64> [ORD=1808] [ID=17] 

0x9f54b20: i32 = undef [ORD=1797] [ID=6] 

0x9f54ba8: i32 = FrameIndex<0> [ID=24] 

0x9f54b20: i32 = undef [ORD=1797] [ID=6] 

In function: testVCVTT32TO16Function 



 


 







yalong at multicorewareinc.com





 




From: Kevin
 Qin


Date: 2014-07-09 00:03


To: yalong at multicorewareinc.com


CC: llvmdev


Subject: Re: [LLVMdev] Help!!!!Help!!!! " LLVM
 ERROR: Cannot select: 0x9fc9680: i32 = fp32_to_fp16 0x9fc0750 [ID=16] "
problem!!!!!!!!!!!!!!!!!!






Hi,

 


Can you show us the command line you are using? At least can you tell us the
backend you tried on? If you can upload the test case as well, it will be very
useful
 to find out the problem.


 


 


Regards,


Kevin



 

2014-07-10 1:53 GMT+08:00
yalong at multicorewareinc.com <yalong at multicorewareinc.com>:


Hi all, 


    I am new to llvm. I need help. Thank you every!


 


    I want to realize vcvtt.f16.f32  NEON instruction by llvm. This instruction
covert top-16bits of a single type to f16. I use the intrinsics function
llvm.convert.to.fp16,
 but cannot llc , I meet is following problem :


 


LLVM ERROR: Cannot select: 0x9fc9680: i32 = fp32_to_fp16 0x9fc0750 [ID=16]


0x9fc0750: f32,ch = load 0x3aafd68, 0x9fc2a20,
0x9feaab0<LD4[%sunkaddr85033]> [ORD=125117] [ID=15]


0x9fc2a20: i32 = add 0x9fed880, 0x9fd9ea0 [ORD=125115] [ID=13] 

0x9fed880: i32,ch = CopyFromReg 0x3aafd68, 0x9fbea90 [ORD=125114] [ID=9] 

0x9fbea90: i32 = Register %vreg13999 [ORD=125114] [ID=1] 

0x9fd9ea0: i32 = Constant<80> [ORD=125115] [ID=2] 

0x9feaab0: i32 = undef [ORD=125117] [ID=4] 

In function: internal_function_69 

Command exited with non-zero status 1


 


    If I change the method, I use " %1 = fptrunc float %0  to half ",
then " %2 = bitcast half %1 to i16", I meet samiliar problem,
the log is following:


 


LLVM ERROR: Cannot select: 0x9f554b0: ch = store 0x9d0f28c, 0x9f5d900,
0x9f54ba8, 0x9f54b20<ST2[FixedStack0](align=4), trunc to f16> [ID=52]


0x9f5d900: f32,ch = load 0x9f5e290, 0x9f5dd40, 0x9f54b20<LD4[%sunkaddr69]>
[ORD=1810] [ID=51]


0x9f5dd40: i32 = add 0x9f55318, 0x9f5e0f8 [ORD=1808] [ID=31] 

0x9f55318: i32,ch = CopyFromReg 0x9d0f28c, 0x9f6a3a0 [ORD=1796] [ID=26] 

0x9f6a3a0: i32 = Register %vreg32 [ORD=1796] [ID=1] 

0x9f5e0f8: i32 = Constant<64> [ORD=1808] [ID=17] 

0x9f54b20: i32 = undef [ORD=1797] [ID=6] 

0x9f54ba8: i32 = FrameIndex<0> [ID=24] 

0x9f54b20: i32 = undef [ORD=1797] [ID=6] 

In function: testVCVTT32TO16Function 


 


 


    Anyone can help me?? Thank you again.


 







yalong at multicorewareinc.com






_______________________________________________

LLVM Developers mailing list

LLVMdev at cs.uiuc.edu         
http://llvm.cs.uiuc.edu

http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev







 

--


Best Regards,

 


Kevin Qin











-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20140709/51fffa6a/attachment.html>

llvm dev - Jul 2014 - [LLVMdev] Help!!!!Help!!!! " LLVM ERROR: Cannot select: 0x9fc9680: i32 = fp32_to_fp16 0x9fc0750 [ID=16] " problem!!!!!!!!!!!!!!!!!!

[LLVMdev] Help!!!!Help!!!! " LLVM ERROR: Cannot select: 0x9fc9680: i32 = fp32_to_fp16 0x9fc0750 [ID=16] " problem!!!!!!!!!!!!!!!!!!

[LLVMdev] Help!!!!Help!!!! " LLVM ERROR: Cannot select: 0x9fc9680: i32 = fp32_to_fp16 0x9fc0750 [ID=16] " problem!!!!!!!!!!!!!!!!!!

[LLVMdev] Help!!!!Help!!!! " LLVM ERROR: Cannot select: 0x9fc9680: i32 = fp32_to_fp16 0x9fc0750 [ID=16] " problem!!!!!!!!!!!!!!!!!!

[LLVMdev] Help!!!!Help!!!! " LLVM ERROR: Cannot select: 0x9fc9680: i32 = fp32_to_fp16 0x9fc0750 [ID=16] " problem!!!!!!!!!!!!!!!!!!

[LLVMdev] Help!!!!Help!!!! " LLVM ERROR: Cannot select: 0x9fc9680: i32 = fp32_to_fp16 0x9fc0750 [ID=16] " problem!!!!!!!!!!!!!!!!!!

[LLVMdev] Help!!!!Help!!!! " LLVM ERROR: Cannot select: 0x9fc9680: i32 = fp32_to_fp16 0x9fc0750 [ID=16] " problem!!!!!!!!!!!!!!!!!!

[LLVMdev] Help!!!!Help!!!! " LLVM ERROR: Cannot select: 0x9fc9680: i32 = fp32_to_fp16 0x9fc0750 [ID=16] " problem!!!!!!!!!!!!!!!!!!