OpenCL Generic Address Spaces - CodeGen (was Re: [llvm-dev] AliasAnalysis does not look though a memcpy)

classic Classic list List threaded Threaded
9 messages Options
Reply | Threaded
Open this post in threaded view
|

OpenCL Generic Address Spaces - CodeGen (was Re: [llvm-dev] AliasAnalysis does not look though a memcpy)

Oleg Smolsky via cfe-dev
Hi, Andrew,

I'd like to fork this part of the thread and move it to cfe-dev. My best
advice is to handle this in Clang, not LLVM, and I've cc'd Richard and
John for their advice. More inline...

On 12/6/18 10:46 AM, Andrew Savonichev wrote:

> ...
>>>>>> Can you say more about the use case?
>>>>> OpenCL C has a notion of Generic Address Space (GAS), allowing you to
>>>>> cast a pointer from any (named) address space to a GAS pointer. Then you
>>>>> can use this GAS pointer instead of a named AS pointer. Compiler is
>>>>> responsible to infer the original address space of a GAS pointer when it
>>>>> is actually used (for load/store), otherwise this is a compilation
>>>>> error.
>>>> That seems scary :-) -- Can this inference not be done syntactically in
>>>> Clang?
>>>>
>>> From a frontend perspective, a GAS pointer is just a pointer:
>>>
>>>   void foo(int *p, int *q) { *p = 42; *q = 43; };
>>>
>>> Until Clang reaches a call site, it has no idea about real address
>>> spaces of `p' and `q'. When we do reach a call site, `foo()' can already
>>> be CodeGen'ed, so we can't really change anything.
>> Is this supposed to work like template instantiation? Are you guaranteed
>> to only get one (unique) set of address spaces for the function
>> arguments?
>>
> Yes, just like in C++ template, if `foo' is called with different sets
> of address spaces, a compiler have to create different function
> instantiations for each set.

I think that you should handle this in Clang using TreeTransform, in a
sense, just like C++ template instantiation. See
lib/Sema/TreeTransform.h, and there are a number of examples in lib/Sema
of transforms using this infrastructure. Using TreeTransform you would
create variants of each function with the right address spaces, based on
usage, and then emit them all during CodeGen. because you'd do this
prior to code generation, you don't need to worry about the emission
ordering.

 -Hal

>
>> We can change the order that functions are emitted in Clang if necessary.
>>
> I haven't thought this is actually configurable. I'd really appreciate
> if you can give me a pointer on how to do this.
>
>
>
--
Hal Finkel
Lead, Compiler Technology and Programming Languages
Leadership Computing Facility
Argonne National Laboratory

_______________________________________________
cfe-dev mailing list
[hidden email]
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev
Reply | Threaded
Open this post in threaded view
|

Re: OpenCL Generic Address Spaces - CodeGen (was Re: [llvm-dev] AliasAnalysis does not look though a memcpy)

Oleg Smolsky via cfe-dev
Hi Andrew,

Can you please provide a reference to the relevant part of the OpenCL specification describing this feature? This sounds like an extremely surprising and problematic language design choice, and I'd like to make sure we're not misinterpreting the specification.

(Some specific things that are unclear here: Where can GAS pointers be used? Can I put them in a struct? Can I make an array of them? Are all array elements required to point to the same address space? Are they mutable? Can I assign pointers from multiple different address spaces to the same GAS pointer? Must functions taking GAS pointers be defined in the same translation unit as the call? Can different GAS parameters resolve to different address spaces? Can you take the address of a function taking GAS pointers?)

On Thu, 6 Dec 2018 at 11:07, Finkel, Hal J. via cfe-dev <[hidden email]> wrote:
Hi, Andrew,

I'd like to fork this part of the thread and move it to cfe-dev. My best
advice is to handle this in Clang, not LLVM, and I've cc'd Richard and
John for their advice. More inline...

On 12/6/18 10:46 AM, Andrew Savonichev wrote:
> ...
>>>>>> Can you say more about the use case?
>>>>> OpenCL C has a notion of Generic Address Space (GAS), allowing you to
>>>>> cast a pointer from any (named) address space to a GAS pointer. Then you
>>>>> can use this GAS pointer instead of a named AS pointer. Compiler is
>>>>> responsible to infer the original address space of a GAS pointer when it
>>>>> is actually used (for load/store), otherwise this is a compilation
>>>>> error.
>>>> That seems scary :-) -- Can this inference not be done syntactically in
>>>> Clang?
>>>>
>>> From a frontend perspective, a GAS pointer is just a pointer:
>>>
>>>   void foo(int *p, int *q) { *p = 42; *q = 43; };
>>>
>>> Until Clang reaches a call site, it has no idea about real address
>>> spaces of `p' and `q'. When we do reach a call site, `foo()' can already
>>> be CodeGen'ed, so we can't really change anything.
>> Is this supposed to work like template instantiation? Are you guaranteed
>> to only get one (unique) set of address spaces for the function
>> arguments?
>>
> Yes, just like in C++ template, if `foo' is called with different sets
> of address spaces, a compiler have to create different function
> instantiations for each set.

I think that you should handle this in Clang using TreeTransform, in a
sense, just like C++ template instantiation. See
lib/Sema/TreeTransform.h, and there are a number of examples in lib/Sema
of transforms using this infrastructure. Using TreeTransform you would
create variants of each function with the right address spaces, based on
usage, and then emit them all during CodeGen. because you'd do this
prior to code generation, you don't need to worry about the emission
ordering.

 -Hal

>
>> We can change the order that functions are emitted in Clang if necessary.
>>
> I haven't thought this is actually configurable. I'd really appreciate
> if you can give me a pointer on how to do this.
>
>
>
--
Hal Finkel
Lead, Compiler Technology and Programming Languages
Leadership Computing Facility
Argonne National Laboratory

_______________________________________________
cfe-dev mailing list
[hidden email]
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev

_______________________________________________
cfe-dev mailing list
[hidden email]
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev
Reply | Threaded
Open this post in threaded view
|

Re: OpenCL Generic Address Spaces - CodeGen (was Re: [llvm-dev] AliasAnalysis does not look though a memcpy)

Oleg Smolsky via cfe-dev
On 12/6/18 1:36 PM, Richard Smith wrote:
Hi Andrew,

Can you please provide a reference to the relevant part of the OpenCL specification describing this feature? This sounds like an extremely surprising and problematic language design choice, and I'd like to make sure we're not misinterpreting the specification.

(Some specific things that are unclear here: Where can GAS pointers be used? Can I put them in a struct? Can I make an array of them? Are all array elements required to point to the same address space? Are they mutable? Can I assign pointers from multiple different address spaces to the same GAS pointer? Must functions taking GAS pointers be defined in the same translation unit as the call? Can different GAS parameters resolve to different address spaces? Can you take the address of a function taking GAS pointers?)

Andrew, please correct me if I'm wrong... It looks like the answer to all of Richard's questions is: yes . This doesn't look like template instantiation. I retract my recommendation in that regard.

  https://www.khronos.org/registry/OpenCL/specs/2.2/html/OpenCL_C.html#the-generic-address-space

which explicitly says that you get to do this:

kernel void bar(global int *g, local int *l)
{
    int *var;

    if (...)
        var = g;
    else
        var = l;
    *var = 42;
    ...
}

where the address space associated with a particular variable can be control-dependent. Also, it can change over time:

global int *gp;
local int *lp;
private int *pp;

int *p;
p = gp; // legal
p = lp; // legal
p = pp; // legal

If you can't represent these directly (e.g., your global address space is also your generic address space) then you might need a fat-pointer representation which you optimize, where possible, by propagating AS info where possible.

 -Hal


On Thu, 6 Dec 2018 at 11:07, Finkel, Hal J. via cfe-dev <[hidden email]> wrote:
Hi, Andrew,

I'd like to fork this part of the thread and move it to cfe-dev. My best
advice is to handle this in Clang, not LLVM, and I've cc'd Richard and
John for their advice. More inline...

On 12/6/18 10:46 AM, Andrew Savonichev wrote:
> ...
>>>>>> Can you say more about the use case?
>>>>> OpenCL C has a notion of Generic Address Space (GAS), allowing you to
>>>>> cast a pointer from any (named) address space to a GAS pointer. Then you
>>>>> can use this GAS pointer instead of a named AS pointer. Compiler is
>>>>> responsible to infer the original address space of a GAS pointer when it
>>>>> is actually used (for load/store), otherwise this is a compilation
>>>>> error.
>>>> That seems scary :-) -- Can this inference not be done syntactically in
>>>> Clang?
>>>>
>>> From a frontend perspective, a GAS pointer is just a pointer:
>>>
>>>   void foo(int *p, int *q) { *p = 42; *q = 43; };
>>>
>>> Until Clang reaches a call site, it has no idea about real address
>>> spaces of `p' and `q'. When we do reach a call site, `foo()' can already
>>> be CodeGen'ed, so we can't really change anything.
>> Is this supposed to work like template instantiation? Are you guaranteed
>> to only get one (unique) set of address spaces for the function
>> arguments?
>>
> Yes, just like in C++ template, if `foo' is called with different sets
> of address spaces, a compiler have to create different function
> instantiations for each set.

I think that you should handle this in Clang using TreeTransform, in a
sense, just like C++ template instantiation. See
lib/Sema/TreeTransform.h, and there are a number of examples in lib/Sema
of transforms using this infrastructure. Using TreeTransform you would
create variants of each function with the right address spaces, based on
usage, and then emit them all during CodeGen. because you'd do this
prior to code generation, you don't need to worry about the emission
ordering.

 -Hal

>
>> We can change the order that functions are emitted in Clang if necessary.
>>
> I haven't thought this is actually configurable. I'd really appreciate
> if you can give me a pointer on how to do this.
>
>
>
--
Hal Finkel
Lead, Compiler Technology and Programming Languages
Leadership Computing Facility
Argonne National Laboratory

_______________________________________________
cfe-dev mailing list
[hidden email]
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev
-- 
Hal Finkel
Lead, Compiler Technology and Programming Languages
Leadership Computing Facility
Argonne National Laboratory

_______________________________________________
cfe-dev mailing list
[hidden email]
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev
Reply | Threaded
Open this post in threaded view
|

Re: OpenCL Generic Address Spaces - CodeGen (was Re: [llvm-dev] AliasAnalysis does not look though a memcpy)

Oleg Smolsky via cfe-dev
In reply to this post by Oleg Smolsky via cfe-dev
>>>>> On Thu, 6 Dec 2018 19:07:30 +0000, "Finkel, Hal J. via cfe-dev" <[hidden email]> said:

    Hal> Hi, Andrew, I'd like to fork this part of the thread and move
    Hal> it to cfe-dev. My best advice is to handle this in Clang, not
    Hal> LLVM, and I've cc'd Richard and John for their advice. More
    Hal> inline...

    On 12/6/18 10:46 AM, Andrew Savonichev wrote:
    >> ...
    >>>>>>> Can you say more about the use case?
>>>>> OpenCL C has a notion of Generic Address Space (GAS), allowing you
    >>>>>>> to
>>>>> cast a pointer from any (named) address space to a GAS
    >>>>>>> pointer. Then you
>>>>> can use this GAS pointer instead of a named AS pointer. Compiler
    >>>>>>> is
>>>>> responsible to infer the original address space of a GAS pointer
    >>>>>>> when it
>>>>> is actually used (for load/store), otherwise this is a compilation
>>>>> error.

    Hal:
    >>>>> That seems scary :-) -- Can this inference not be done
    >>>>> syntactically in Clang?

Actually the generic address space is an OpenCL 2.x feature and does not
require address space inference since it is expected that the hardware
will be able to address any kind of address spaces (except the constant
one, but it is a borderline case). Or it could even be done with run-time
resolution, probably in a less efficient way.

But of course in some case it might be useful to remove some generic
address space using some address space inference at compile time and
replace it by some more concrete address spaces.

An important use case could be to compile OpenCL 2.x programs to run on
OpenCL 1.x-level hardware, without generic address space support, or
perhaps because instructions using generic address space are less
efficient on some OpenCL 2.x hardware, or...

Is it this the kind of usage Andrew you are working on?
--
  Ronan KERYELL
_______________________________________________
cfe-dev mailing list
[hidden email]
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev
Reply | Threaded
Open this post in threaded view
|

Re: OpenCL Generic Address Spaces - CodeGen (was Re: [llvm-dev] AliasAnalysis does not look though a memcpy)

Oleg Smolsky via cfe-dev
On Thu, 6 Dec 2018 at 14:36, Ronan KERYELL via cfe-dev <[hidden email]> wrote:
>>>>> On Thu, 6 Dec 2018 19:07:30 +0000, "Finkel, Hal J. via cfe-dev" <[hidden email]> said:

    Hal> Hi, Andrew, I'd like to fork this part of the thread and move
    Hal> it to cfe-dev. My best advice is to handle this in Clang, not
    Hal> LLVM, and I've cc'd Richard and John for their advice. More
    Hal> inline...

    On 12/6/18 10:46 AM, Andrew Savonichev wrote:
    >> ...
    >>>>>>> Can you say more about the use case?
>>>>> OpenCL C has a notion of Generic Address Space (GAS), allowing you
    >>>>>>> to
>>>>> cast a pointer from any (named) address space to a GAS
    >>>>>>> pointer. Then you
>>>>> can use this GAS pointer instead of a named AS pointer. Compiler
    >>>>>>> is
>>>>> responsible to infer the original address space of a GAS pointer
    >>>>>>> when it
>>>>> is actually used (for load/store), otherwise this is a compilation
>>>>> error.

    Hal:
    >>>>> That seems scary :-) -- Can this inference not be done
    >>>>> syntactically in Clang?

Actually the generic address space is an OpenCL 2.x feature and does not
require address space inference since it is expected that the hardware
will be able to address any kind of address spaces (except the constant
one, but it is a borderline case). Or it could even be done with run-time
resolution, probably in a less efficient way.

But of course in some case it might be useful to remove some generic
address space using some address space inference at compile time and
replace it by some more concrete address spaces.

An important use case could be to compile OpenCL 2.x programs to run on
OpenCL 1.x-level hardware, without generic address space support, or
perhaps because instructions using generic address space are less
efficient on some OpenCL 2.x hardware, or...

Sounds to me like this doesn't require frontend changes, then. (Well, mostly: if we need to use fat pointers to encode the address space, then the frontend might need changes to encode that in the data layout and to increase the size of pointers and such, but that sounds like the extent of the frontend's responsibilities.) A pass to pick a concrete address space for a pointer based on usage would belong in the middle-end.
 
Is it this the kind of usage Andrew you are working on?
--
  Ronan KERYELL
_______________________________________________
cfe-dev mailing list
[hidden email]
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev

_______________________________________________
cfe-dev mailing list
[hidden email]
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev
Reply | Threaded
Open this post in threaded view
|

Re: OpenCL Generic Address Spaces - CodeGen (was Re: [llvm-dev] AliasAnalysis does not look though a memcpy)

Oleg Smolsky via cfe-dev
In reply to this post by Oleg Smolsky via cfe-dev
On 6 Dec 2018, at 14:07, Finkel, Hal J. wrote:

> Hi, Andrew,
>
> I'd like to fork this part of the thread and move it to cfe-dev. My
> best
> advice is to handle this in Clang, not LLVM, and I've cc'd Richard and
> John for their advice. More inline...
>
> On 12/6/18 10:46 AM, Andrew Savonichev wrote:
>> ...
>>>>>>> Can you say more about the use case?
>>>>>> OpenCL C has a notion of Generic Address Space (GAS), allowing
>>>>>> you to
>>>>>> cast a pointer from any (named) address space to a GAS pointer.
>>>>>> Then you
>>>>>> can use this GAS pointer instead of a named AS pointer. Compiler
>>>>>> is
>>>>>> responsible to infer the original address space of a GAS pointer
>>>>>> when it
>>>>>> is actually used (for load/store), otherwise this is a
>>>>>> compilation
>>>>>> error.
>>>>> That seems scary :-) -- Can this inference not be done
>>>>> syntactically in
>>>>> Clang?
>>>>>
>>>> From a frontend perspective, a GAS pointer is just a pointer:
>>>>
>>>>   void foo(int *p, int *q) { *p = 42; *q = 43; };
>>>>
>>>> Until Clang reaches a call site, it has no idea about real address
>>>> spaces of `p' and `q'. When we do reach a call site, `foo()' can
>>>> already
>>>> be CodeGen'ed, so we can't really change anything.
>>> Is this supposed to work like template instantiation? Are you
>>> guaranteed
>>> to only get one (unique) set of address spaces for the function
>>> arguments?
>>>
>> Yes, just like in C++ template, if `foo' is called with different
>> sets
>> of address spaces, a compiler have to create different function
>> instantiations for each set.
>
> I think that you should handle this in Clang using TreeTransform, in a
> sense, just like C++ template instantiation. See
> lib/Sema/TreeTransform.h, and there are a number of examples in
> lib/Sema
> of transforms using this infrastructure. Using TreeTransform you would
> create variants of each function with the right address spaces, based
> on
> usage, and then emit them all during CodeGen. because you'd do this
> prior to code generation, you don't need to worry about the emission
> ordering.

I think it depends a lot on the language rules.  If the language rules
are set up so that we can easily propagate qualifiers from the arguments
without a complex analysis, well, okay, TreeTransform makes sense.  But
I
think it's much more likely that this would have to be a
data-flow-sensitive,
best-effort analysis that simply fails in arbitrary ways if the
optimizer
isn't able to fully eliminate a use of the generic address space.

To enable a TreeTransform-based implementation, we'd have to be able to
infer
a concrete address space immediately for every place where the GAS would
otherwise be used in the function, and when those places are e.g. the
types
of local variables, that inference must prove to be consistent with
other uses
of the variable/whatever as we propagate type information forward.

I don't really know how that inference would work; it sounds incredibly
complicated.  But then, admittedly, so does a data-flow-sensitive
rewrite.

John.
_______________________________________________
cfe-dev mailing list
[hidden email]
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev
Reply | Threaded
Open this post in threaded view
|

Re: OpenCL Generic Address Spaces - CodeGen (was Re: [llvm-dev] AliasAnalysis does not look though a memcpy)

Oleg Smolsky via cfe-dev
On 6 Dec 2018, at 23:47, John McCall wrote:

> On 6 Dec 2018, at 14:07, Finkel, Hal J. wrote:
>> Hi, Andrew,
>>
>> I'd like to fork this part of the thread and move it to cfe-dev. My
>> best
>> advice is to handle this in Clang, not LLVM, and I've cc'd Richard
>> and
>> John for their advice. More inline...
>>
>> On 12/6/18 10:46 AM, Andrew Savonichev wrote:
>>> ...
>>>>>>>> Can you say more about the use case?
>>>>>>> OpenCL C has a notion of Generic Address Space (GAS), allowing
>>>>>>> you to
>>>>>>> cast a pointer from any (named) address space to a GAS pointer.
>>>>>>> Then you
>>>>>>> can use this GAS pointer instead of a named AS pointer. Compiler
>>>>>>> is
>>>>>>> responsible to infer the original address space of a GAS pointer
>>>>>>> when it
>>>>>>> is actually used (for load/store), otherwise this is a
>>>>>>> compilation
>>>>>>> error.
>>>>>> That seems scary :-) -- Can this inference not be done
>>>>>> syntactically in
>>>>>> Clang?
>>>>>>
>>>>> From a frontend perspective, a GAS pointer is just a pointer:
>>>>>
>>>>>   void foo(int *p, int *q) { *p = 42; *q = 43; };
>>>>>
>>>>> Until Clang reaches a call site, it has no idea about real address
>>>>> spaces of `p' and `q'. When we do reach a call site, `foo()' can
>>>>> already
>>>>> be CodeGen'ed, so we can't really change anything.
>>>> Is this supposed to work like template instantiation? Are you
>>>> guaranteed
>>>> to only get one (unique) set of address spaces for the function
>>>> arguments?
>>>>
>>> Yes, just like in C++ template, if `foo' is called with different
>>> sets
>>> of address spaces, a compiler have to create different function
>>> instantiations for each set.
>>
>> I think that you should handle this in Clang using TreeTransform, in
>> a
>> sense, just like C++ template instantiation. See
>> lib/Sema/TreeTransform.h, and there are a number of examples in
>> lib/Sema
>> of transforms using this infrastructure. Using TreeTransform you
>> would
>> create variants of each function with the right address spaces, based
>> on
>> usage, and then emit them all during CodeGen. because you'd do this
>> prior to code generation, you don't need to worry about the emission
>> ordering.
>
> I think it depends a lot on the language rules.  If the language rules
> are set up so that we can easily propagate qualifiers from the
> arguments
> without a complex analysis, well, okay, TreeTransform makes sense.  
> But I
> think it's much more likely that this would have to be a
> data-flow-sensitive,
> best-effort analysis that simply fails in arbitrary ways if the
> optimizer
> isn't able to fully eliminate a use of the generic address space.
>
> To enable a TreeTransform-based implementation, we'd have to be able
> to infer
> a concrete address space immediately for every place where the GAS
> would
> otherwise be used in the function, and when those places are e.g. the
> types
> of local variables, that inference must prove to be consistent with
> other uses
> of the variable/whatever as we propagate type information forward.
>
> I don't really know how that inference would work; it sounds
> incredibly
> complicated.  But then, admittedly, so does a data-flow-sensitive
> rewrite.

Oh, and now I've found the existing discussion where we've come to the
conclusion that a first-class GAS is acceptable.  Great!

John.
_______________________________________________
cfe-dev mailing list
[hidden email]
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev
Reply | Threaded
Open this post in threaded view
|

Re: OpenCL Generic Address Spaces - CodeGen (was Re: [llvm-dev] AliasAnalysis does not look though a memcpy)

Oleg Smolsky via cfe-dev
In reply to this post by Oleg Smolsky via cfe-dev
Ronan KERYELL writes:

>>>>>> On Thu, 6 Dec 2018 19:07:30 +0000, "Finkel, Hal J. via cfe-dev" <[hidden email]> said:
>
>     Hal> Hi, Andrew, I'd like to fork this part of the thread and move
>     Hal> it to cfe-dev. My best advice is to handle this in Clang, not
>     Hal> LLVM, and I've cc'd Richard and John for their advice. More
>     Hal> inline...
>
>     On 12/6/18 10:46 AM, Andrew Savonichev wrote:
>     >> ...
>     >>>>>>> Can you say more about the use case?
>>>>>> OpenCL C has a notion of Generic Address Space (GAS), allowing you
>     >>>>>>> to
>>>>>> cast a pointer from any (named) address space to a GAS
>     >>>>>>> pointer. Then you
>>>>>> can use this GAS pointer instead of a named AS pointer. Compiler
>     >>>>>>> is
>>>>>> responsible to infer the original address space of a GAS pointer
>     >>>>>>> when it
>>>>>> is actually used (for load/store), otherwise this is a compilation
>>>>>> error.
>
>     Hal:
>     >>>>> That seems scary :-) -- Can this inference not be done
>     >>>>> syntactically in Clang?
>
> Actually the generic address space is an OpenCL 2.x feature and does not
> require address space inference since it is expected that the hardware
> will be able to address any kind of address spaces (except the constant
> one, but it is a borderline case). Or it could even be done with run-time
> resolution, probably in a less efficient way.
>
> But of course in some case it might be useful to remove some generic
> address space using some address space inference at compile time and
> replace it by some more concrete address spaces.
>
> An important use case could be to compile OpenCL 2.x programs to run on
> OpenCL 1.x-level hardware, without generic address space support, or
> perhaps because instructions using generic address space are less
> efficient on some OpenCL 2.x hardware, or...
>
> Is it this the kind of usage Andrew you are working on?

You're right, my definition of GAS was too strict - OpenCL specification
does allow dynamic GAS resolution, but it requires backend (or hardware)
support to do that.

For static GAS resolution my use cases are:
1) backends (e.g. based on OpenCL 1.x) which do not support GAS natively.
2) performance and code size improvements that can be archived by GAS
   static resolution.

For (2) I'm probably fine with running this analysis after
optimizations, but support for debug (no-opt) can be important for (1).

--
Andrew

--------------------------------------------------------------------
Joint Stock Company Intel A/O
Registered legal address: Krylatsky Hills Business Park,
17 Krylatskaya Str., Bldg 4, Moscow 121614,
Russian Federation

This e-mail and any attachments may contain confidential material for
the sole use of the intended recipient(s). Any review or distribution
by others is strictly prohibited. If you are not the intended
recipient, please contact the sender and delete all copies.

_______________________________________________
cfe-dev mailing list
[hidden email]
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev
Reply | Threaded
Open this post in threaded view
|

Re: OpenCL Generic Address Spaces - CodeGen (was Re: [llvm-dev] AliasAnalysis does not look though a memcpy)

Oleg Smolsky via cfe-dev
In reply to this post by Oleg Smolsky via cfe-dev
Finkel, Hal J. writes:
> On 12/6/18 1:36 PM, Richard Smith wrote:
> Hi Andrew,
>
> Can you please provide a reference to the relevant part of the OpenCL
> specification describing this feature? This sounds like an extremely
> surprising and problematic language design choice, and I'd like to
> make sure we're not misinterpreting the specification.
>
Sorry, GAS compile-time resolution (inference) is not a mandatory as I
believed, and OpenCL 2.0 compiler is allowed to fallback to dynamic
resolution (generate instructions which can work with any AS).

Though it is still beneficial to infer as much as possible at compile
time.

As Hal already mentioned, GAS is described here:
  https://www.khronos.org/registry/OpenCL/specs/2.2/html/OpenCL_C.html#the-generic-address-space

> (Some specific things that are unclear here:
> Where can GAS pointers be used?
>
Everywhere where a named pointer can be used. Only casts from GAS to a
named AS are disallowed.

> Can I put them in a struct? Can I make an array of them?
>
Yes.

> Are all array elements required to point to the same address space?
>
Specification does not explicitly forbid this, so the answer should be:
no, elements of an array can have different address space.

> Are they mutable? Can I assign pointers from multiple different
> address spaces to the same GAS pointer?
>
Yes.

> Must functions taking GAS pointers be defined in the same translation
> unit as the call?
>
No, they can be defined in different translation units.

> Can different GAS parameters resolve to different address spaces?
>
Yes.

> Can you take the address of a function taking GAS pointers?)
>
Luckily, function pointers are not allowed in OpenCL at all.

> Andrew, please correct me if I'm wrong... It looks like the answer to
> all of Richard's questions is: yes.
>
> This doesn't look like template instantiation. I retract my
> recommendation in that regard.
>
Sorry, I was not clear about this. Specification is not clear about this
either, but from my understanding, this should work like a templates
when a GAS pointer is a function argument: if a function is called twice
with parameters of different AS, we have to duplicate the function.

>   https://www.khronos.org/registry/OpenCL/specs/2.2/html/OpenCL_C.html#the-generic-address-space
>
> which explicitly says that you get to do this:
>
> kernel void bar(global int *g, local int *l)
> {
>     int *var;
>
>     if (...)
>         var = g;
>     else
>         var = l;
>     *var = 42;
>     ...
> }
>
> where the address space associated with a particular variable can be control-dependent. Also, it can change over time:
>
> global int *gp;
> local int *lp;
> private int *pp;
>
> int *p;
> p = gp; // legal
> p = lp; // legal
> p = pp; // legal
>
> If you can't represent these directly (e.g., your global address space
> is also your generic address space) then you might need a fat-pointer
> representation which you optimize, where possible, by propagating AS
> info where possible.
>
So you suggest to optimize address spaces in LLVM IR, and if the
optimization fails somewhere, then generate a generic code in a device
backend. Is this accurate?

This is probably the best approach, and from my understanding, this is
how it works for many OpenCL 2.0 implementations.

What I'm trying to figure out, is how to write an analysis that can
statically infer all address spaces in an "average code".

So it looks like I cannot modify Clang to make IR more friendly for this
analysis, and I need to run some optimizations to infer an address space
in more complicated cases.

> On Thu, 6 Dec 2018 at 11:07, Finkel, Hal J. via cfe-dev <[hidden email]<mailto:[hidden email]>> wrote:
> Hi, Andrew,
>
> I'd like to fork this part of the thread and move it to cfe-dev. My best
> advice is to handle this in Clang, not LLVM, and I've cc'd Richard and
> John for their advice. More inline...
>
> On 12/6/18 10:46 AM, Andrew Savonichev wrote:
>> ...
>>>>>>> Can you say more about the use case?
>>>>>> OpenCL C has a notion of Generic Address Space (GAS), allowing you to
>>>>>> cast a pointer from any (named) address space to a GAS pointer. Then you
>>>>>> can use this GAS pointer instead of a named AS pointer. Compiler is
>>>>>> responsible to infer the original address space of a GAS pointer when it
>>>>>> is actually used (for load/store), otherwise this is a compilation
>>>>>> error.
>>>>> That seems scary :-) -- Can this inference not be done syntactically in
>>>>> Clang?
>>>>>
>>>> From a frontend perspective, a GAS pointer is just a pointer:
>>>>
>>>>   void foo(int *p, int *q) { *p = 42; *q = 43; };
>>>>
>>>> Until Clang reaches a call site, it has no idea about real address
>>>> spaces of `p' and `q'. When we do reach a call site, `foo()' can already
>>>> be CodeGen'ed, so we can't really change anything.
>>> Is this supposed to work like template instantiation? Are you guaranteed
>>> to only get one (unique) set of address spaces for the function
>>> arguments?
>>>
>> Yes, just like in C++ template, if `foo' is called with different sets
>> of address spaces, a compiler have to create different function
>> instantiations for each set.
>
> I think that you should handle this in Clang using TreeTransform, in a
> sense, just like C++ template instantiation. See
> lib/Sema/TreeTransform.h, and there are a number of examples in lib/Sema
> of transforms using this infrastructure. Using TreeTransform you would
> create variants of each function with the right address spaces, based on
> usage, and then emit them all during CodeGen. because you'd do this
> prior to code generation, you don't need to worry about the emission
> ordering.
>
>  -Hal
>
>>
>>> We can change the order that functions are emitted in Clang if necessary.
>>>
>> I haven't thought this is actually configurable. I'd really appreciate
>> if you can give me a pointer on how to do this.

--
Andrew

--------------------------------------------------------------------
Joint Stock Company Intel A/O
Registered legal address: Krylatsky Hills Business Park,
17 Krylatskaya Str., Bldg 4, Moscow 121614,
Russian Federation

This e-mail and any attachments may contain confidential material for
the sole use of the intended recipient(s). Any review or distribution
by others is strictly prohibited. If you are not the intended
recipient, please contact the sender and delete all copies.

_______________________________________________
cfe-dev mailing list
[hidden email]
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev