similar to: [LLVMdev] PTX Backend in 2.9

Hi Justin, I am upstreaming the PTX backend. My plan is to have a working prototype (that mean you may compile non-trivial code with some workarounds) by the end of this year or by January 2011. I hope I could catch up next release of LLVM (version 2.9), so I will adjust my plan according to the release schedule once it is announced. Regards, Che-Liang On Wed, Oct 6, 2010 at 11:16 PM, Justin

Hi, The PTX backend we developed (CBackend approach, does not use the target independent code generator) is already more advanced. An older version is published here: http://sourceforge.net/projects/llvmptxbackend/ We recently eliminated a bug which increased the number of required registers per thread. Surprisingly, without that bug the generated code is already comparable to code generated

I read on the archives that a PTX back-end was discussed around August and I see that some initial code has been checked into LLVM trunk. What is the status of this project? Is it being actively worked on? I am ultimately interested in using LLVM for GPU code optimization/generation and am very interested in this particular project. -- Thanks, Justin Holewinski -------------- next part

On Mon, Nov 21, 2011 at 3:36 PM, Justin Holewinski <justin.holewinski at gmail.com> wrote: > On Mon, Nov 21, 2011 at 7:01 AM, Alberto Magni <alberto.magni86 at gmail.com> > wrote: >> >> Hi Justin, >> >> attached you find the patch for the integer max instruction. >> The multiclass PTX_INTRINSIC_INT3 in file PTXIntrinsicInstrInfo.td >> is almost

On Mon, Nov 21, 2011 at 5:31 PM, Justin Holewinski <justin.holewinski at gmail.com> wrote: > On Mon, Nov 21, 2011 at 11:45 AM, Alberto Magni <alberto.magni86 at gmail.com> > wrote: >> >> On Mon, Nov 21, 2011 at 3:36 PM, Justin Holewinski >> <justin.holewinski at gmail.com> wrote: >> > On Mon, Nov 21, 2011 at 7:01 AM, Alberto Magni >> >

On Thu, Dec 8, 2011 at 11:36 AM, Villmow, Micah <Micah.Villmow at amd.com>wrote: > It is my understanding that all you need to do is specify let isTarget = > 1 in your .td file and it will generate target specific intrinsics. This > should allow you to keep the IntrinsicsPTX.td file in the same location. > So we keep the intrinsics defined in include/llvm/IntrinsicsPTX.td?

On Sun, Dec 4, 2011 at 1:10 PM, Alberto Magni <alberto.magni86 at gmail.com>wrote: > Hi Justin, > > sorry for the delay, I have been busy. > > Micah's proposal requires to move the definitions of the intrinsics > from include/llvm/IntrinsicsPTX.td to lib/Target/PTX/PTXIntrinsics.td > thus allowing the generation of the file PTXGenIntrinsics.inc which > will be

On Nov 23, 2011 8:33 AM, "Justin Holewinski" <justin.holewinski at gmail.com> wrote: > > > On Nov 23, 2011 6:57 AM, "Alberto Magni" <alberto.magni86 at gmail.com> wrote: > > > > On Tue, Nov 22, 2011 at 5:01 PM, Villmow, Micah <Micah.Villmow at amd.com> wrote: > > > Alberto, > > > The AMDIL backend solves your problem

On Mon, Nov 21, 2011 at 7:01 AM, Alberto Magni <alberto.magni86 at gmail.com>wrote: > Hi Justin, > > attached you find the patch for the integer max instruction. > The multiclass PTX_INTRINSIC_INT3 in file PTXIntrinsicInstrInfo.td > is almost an exact copy of PTX_INT3 in PTXInstrInfo.td, maybe > a modification of this class can be defined in a separate file. > I'm

On Mon, Nov 21, 2011 at 11:45 AM, Alberto Magni <alberto.magni86 at gmail.com>wrote: > On Mon, Nov 21, 2011 at 3:36 PM, Justin Holewinski > <justin.holewinski at gmail.com> wrote: > > On Mon, Nov 21, 2011 at 7:01 AM, Alberto Magni < > alberto.magni86 at gmail.com> > > wrote: > >> > >> Hi Justin, > >> > >> attached you find

On Tue, Nov 22, 2011 at 5:01 PM, Villmow, Micah <Micah.Villmow at amd.com> wrote: > Alberto, >  The AMDIL backend solves your problem with intrinsic overloading this way: > def int_AMDIL_mad     : GCCBuiltin<"__amdil_mad">, TernaryIntFloat; > > Where TernaryIntFloat is defined as: > class TernaryIntFloat : >          Intrinsic<[llvm_anyfloat_ty],

Ah, that's helpful. I knew that I'd need to end up with PTX as text, not a true binary, but I would have figured that it would come out of MCJIT. Thanks for helping to steer me away from the wrong trail. OK, one more question: Can anybody clarify the pros and cons of generating the PTX through the standard LLVM distro, versus using the "libnvvm" that comes with the Cuda SDK?

Justin, Add this to your TargetLowering constructor, this fixes the mem* issue. maxStoresPerMemcpy = 4096; maxStoresPerMemmove = 4096; maxStoresPerMemset = 4096; From: llvmdev-bounces at cs.uiuc.edu [mailto:llvmdev-bounces at cs.uiuc.edu] On Behalf Of Justin Holewinski Sent: Monday, November 14, 2011 7:12 AM To: Alberto Magni Cc: llvmdev at cs.uiuc.edu Subject: Re: [LLVMdev] PTX backend

Alberto, The AMDIL backend solves your problem with intrinsic overloading this way: def int_AMDIL_mad : GCCBuiltin<"__amdil_mad">, TernaryIntFloat; Where TernaryIntFloat is defined as: class TernaryIntFloat : Intrinsic<[llvm_anyfloat_ty], [LLVMMatchType<0>, LLVMMatchType<0>, LLVMMatchType<0>], []>; This allows us to write a

Hi Justin, sorry for the delay, I have been busy. Micah's proposal requires to move the definitions of the intrinsics from include/llvm/IntrinsicsPTX.td to lib/Target/PTX/PTXIntrinsics.td thus allowing the generation of the file PTXGenIntrinsics.inc which will be included by PTXIntrinsicInfo.cpp. This is a quite big modification, do you agree with this ? Or do you have a better solution.

On Nov 23, 2011 6:57 AM, "Alberto Magni" <alberto.magni86 at gmail.com> wrote: > > On Tue, Nov 22, 2011 at 5:01 PM, Villmow, Micah <Micah.Villmow at amd.com> wrote: > > Alberto, > > The AMDIL backend solves your problem with intrinsic overloading this way: > > def int_AMDIL_mad : GCCBuiltin<"__amdil_mad">, TernaryIntFloat; >

Alright, couple of points here: 1. Address space 0 is invalid for global variables. This is causing a crash in llc where we use llvm_unreachable() on this case. This is most likely why you're seeing llc run forever. The fix for this is to use address space 1 for globals, which puts them into PTX global memory. On our side, we should provide a meaningful error message in this case. 2. The

It is my understanding that all you need to do is specify let isTarget = 1 in your .td file and it will generate target specific intrinsics. This should allow you to keep the IntrinsicsPTX.td file in the same location. Micah From: Justin Holewinski [mailto:justin.holewinski at gmail.com] Sent: Monday, December 05, 2011 6:13 AM To: Alberto Magni Cc: Villmow, Micah; LLVM Developers Mailing List

On Tue, Dec 13, 2011 at 12:54 PM, Villmow, Micah <Micah.Villmow at amd.com>wrote: > ** ** > > ** ** > > *From:* Justin Holewinski [mailto:justin.holewinski at gmail.com] > *Sent:* Tuesday, December 13, 2011 9:48 AM > *To:* Villmow, Micah > *Cc:* LLVM Developers Mailing List > *Subject:* Re: [LLVMdev] Changes to the PTX calling conventions**** > > ** ** >

Hi Nikolaos, Following commands work great for me. $ clang -S -emit-llvm -target nvptx -x cl -include clc/clctypes.h ../data-types/scalar.cl $ llc -mcpu=sm_30 scalar.s You can follow Justin's blog [1]. It helped me a lot to understand where to start. [1] http://jholewinski.org/blog/llvm-3-0-ptx-backend/ Best, Ankur On Mon, Feb 4, 2013 at 11:40 PM, Justin Holewinski < justin.holewinski