[llvm-dev] [LLVM][RFC] Representing the target device information in the LLVM IR

Fri Apr 27 06:41:49 PDT 2018

I think because it is prohibited by the standard. According to OpenMP
standard the new copies of the variables must be of the same type and
the same size. If the type has the different size on the device, we
become not compatible with the standard.

I agree with Eric Christofer. From my point of view, it just breaks the
existing ABI. Instead of breaking the existing ABI, I think, it would
better to introduce a new, portable, ABI. The classic ABI could be used
to get a little bit more performance, while the new one could be used to
get the compatibility between pointer-size incompatible targets. In this
ABI all variables must be passed by reference. I think it will solve all
the problems.

With the proposed change there would be a difference between the code
compiled for (1) 64 bit host+64 bit device and (2) 64 bit host +64|32
bit device. You won't be able to make it work properly the program
linked from the (2)nd host object + the (1)st device object because of
the different ABIs. I think it is better to explicitly specify that
we're going to use some special ABI rather than doing such tricky and
dangerous things as in this proposal.

-------------
Best regards,
Alexey Bataev

26.04.2018 20:14, Hal Finkel via llvm-dev пишет:
>
>
> On 04/26/2018 07:03 PM, Narayanaswamy, Ravi wrote:
>>
>> Hi Hal,
>>
>>    We are not trying to address issues where the object mapped are of
>> different sizes between host and target with different ABI.
>>
>
> Why are you not trying to address that issue?
>
>  -Hal
>
>> The issue is when the objects are of same size like double which is
>> 8bytes on both 32bit and 64bit platform.  If a double is used in a
>> first_private on a target clause,  the 64 bit side will pass it as
>> value whereas on the 32bit side since the value does not fit in the
>> argument it will be passed as pointer to a double. There will be a
>> mis-match at the call site and entry site on this value.
>>
>>   The main reason for this change is that when we do backend
>> outlining for target pragmas the targets information needs to be
>> communicated to the backend to generate the tables with the right
>> names.  Generate LLVM IR for passing this information is one
>> mechanism and other is passing the command option to the backend. 
>> For the later each pass which needs this info will have to change.
>>
>> Thanks
>>
>> Ravi
>>
>>  
>>
>> *From:*Hal Finkel [mailto:hfinkel at anl.gov]
>> *Sent:* Wednesday, April 25, 2018 5:50 PM
>> *To:* Lin, Jin <jin.lin at intel.com <mailto:jin.lin at intel.com>>;
>> Friedman, Eli <efriedma at codeaurora.org
>> <mailto:efriedma at codeaurora.org>>; 'llvm-dev at lists.llvm.org'
>> <llvm-dev at lists.llvm.org <mailto:llvm-dev at lists.llvm.org>>
>> *Subject:* Re: [llvm-dev] [LLVM][RFC] Representing the target device
>> information in the LLVM IR
>>
>>  
>>
>> Hi, Jin,
>>
>> Can you please back up a bit and talk about the programming
>> environment in which this problem manifests?
>>
>> If I have a host and a target with different ABIs, then it seems we
>> have lots of problems. For one thing, the layouts of structures are
>> different, the sizes of some integer types are different, the sizes
>> of pointers are different, and so on. It seems like a solution in
>> this space should address, somehow, this general translation problem.
>> Fixing this particular problem with the dispatch function's
>> parameters feels like only the tip of the iceberg. What if I'm
>> passing a pointer to some structure, or a pointer to other pointers,
>> etc.?
>>
>> I understand that OpenMP v5 is expected to have some custom "mappers"
>> to handle deep copying and translation. Is this related to the design
>> space here?
>>
>> Thanks again,
>>
>> Hal
>>
>>  
>>
>> On 04/25/2018 07:22 PM, Lin, Jin via llvm-dev wrote:
>>
>>     For the firstprivate clause, the compiler generates code to pass
>>     it  by value or by reference to the outlined function. The reason
>>     the first private scalars is generally passed by value is for the
>>     performance reason.
>>
>>     For this particular case, the compiler cannot generate code to
>>     pass the double @gg by value under i386-pc-linux-gnu since the
>>     value is 64 bit while the architecture is 32bit.
>>
>>     For the host compilation, the compiler generates the code to pass
>>     the data as well as the outlined function name to the OMP runtime.
>>
>>     For the target compilation, the compiler generates the outlined
>>     function so that it can be called by the OMP runtime. 
>>
>>     So, the compiler is required to generate a single call on the
>>     host to support all the targets. All the target versions must
>>     have the same interface. So the common interface of the outline
>>     function should be used. For this particular example, the
>>     variable @gcc should be passed by reference under x86_64-mic.
>>
>>     Please let me know if you have more questions.
>>
>>     Jin
>>
>>      
>>
>>     *From:*Friedman, Eli [mailto:efriedma at codeaurora.org]
>>     *Sent:* Wednesday, April 25, 2018 4:14 PM
>>     *To:* Lin, Jin <jin.lin at intel.com> <mailto:jin.lin at intel.com>;
>>     'llvm-dev at lists.llvm.org <mailto:llvm-dev at lists.llvm.org>'
>>     <llvm-dev at lists.llvm.org> <mailto:llvm-dev at lists.llvm.org>
>>     *Subject:* Re: [llvm-dev] [LLVM][RFC] Representing the target
>>     device information in the LLVM IR
>>
>>      
>>
>>     On 4/25/2018 3:48 PM, Lin, Jin wrote:
>>
>>         Given a global variable @gg, the compiler has to generate
>>         code on the host to specify whether it is passed by value or
>>         passed by reference. In the following example, if the
>>         compiler generates the code for passing by value, the
>>         outlined function on the target i386-pc-linux-gnucannot get
>>         the correct value since it assumes the variable @gg is passed
>>         by reference.  
>>
>>          
>>
>>         Here is the corresponding IR on the host side.
>>
>>           %0 = load double, double* @gg, align 8, !tbaa !3
>>
>>           %1 = bitcast double %0 to i64
>>
>>            …
>>
>>           %12 = getelementptr inbounds [4 x i8*], [4 x i8*]*
>>         %.offload_baseptrs, i32 0, i32 2
>>
>>           %13 = bitcast i8** %12 to i64*
>>
>>           store i64 %1, i64* %13, align 8
>>
>>
>>     Could you describe the overall process of calling an offloaded
>>     function in a bit more detail?  How do you describe the ABI of
>>     the called function to the OpenMP runtime?
>>
>>     I suspect you shouldn't be trying to store things which aren't
>>     pointers into offload_baseptrs.
>>
>>     -Eli
>>
>>
>>     -- 
>>
>>     Employee of Qualcomm Innovation Center, Inc.
>>
>>     Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, a Linux Foundation Collaborative Project
>>
>>
>>
>>     _______________________________________________
>>
>>     LLVM Developers mailing list
>>
>>     llvm-dev at lists.llvm.org <mailto:llvm-dev at lists.llvm.org>
>>
>>     http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev
>>
>>  
>>
>> -- 
>> Hal Finkel
>> Lead, Compiler Technology and Programming Languages
>> Leadership Computing Facility
>> Argonne National Laboratory
>
> -- 
> Hal Finkel
> Lead, Compiler Technology and Programming Languages
> Leadership Computing Facility
> Argonne National Laboratory

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20180427/3cd5c385/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: OpenPGP digital signature
URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20180427/3cd5c385/attachment.sig>