Page MenuHomePhabricator

Naghasan (Victor Lomuller)
User

Projects

User does not belong to any projects.

User Details

User Since
Feb 22 2018, 2:32 AM (266 w, 17 h)

Recent Activity

Apr 25 2022

Naghasan added a comment to D107668: [OpenMP]Fix PR50336: Remove temporary files in the offload bundler tool.

Thanks for your answer, make sense.

Apr 25 2022, 7:59 AM · Restricted Project, Restricted Project

Apr 21 2022

Herald added a project to D107668: [OpenMP]Fix PR50336: Remove temporary files in the offload bundler tool: Restricted Project.

Any rational for not calling addTempFile during the binding rather than the call setup ? I had a brief look there and it seems to me this is the root of the problem.

Apr 21 2022, 2:09 AM · Restricted Project, Restricted Project

Jan 25 2022

Naghasan accepted D114483: [SYCL] Add support for sycl_special_class attribute.

LGTM, thanks for the work

Jan 25 2022, 6:43 AM · Restricted Project

Jan 11 2022

Naghasan added inline comments to D112718: Add intrinsics and builtins for PTX atomics with semantic orders.
Jan 11 2022, 2:50 AM · Restricted Project, Restricted Project, Restricted Project

Dec 7 2021

Naghasan added a comment to D114483: [SYCL] Add support for sycl_special_class attribute.

Just a few comments

Dec 7 2021, 6:29 AM · Restricted Project

Nov 17 2021

Naghasan added a comment to D114080: [SYCL] Diagnose uses of zero length arrays.

right, TIL sizeof (ContainsArr) is 0 according to clang. I can see trouble arising.

@aaron.ballman maybe you know better, is that intended for the extension ?

I don't *know*, but I have two educated guesses that it is intended. GCC compatibility (GCC also has this behavior) and the relationship between zero-sized array members and flexible array members. Clang allows a flexible array member to be used in a structure with no members and it behaves the same as a zero-sized array in a structure with no members: https://godbolt.org/z/Gx7ozxYKz

Nov 17 2021, 6:17 AM · Restricted Project
Naghasan added a comment to D114080: [SYCL] Diagnose uses of zero length arrays.

right, TIL sizeof (ContainsArr) is 0 according to clang. I can see trouble arising.

Nov 17 2021, 5:42 AM · Restricted Project
Naghasan added a comment to D114080: [SYCL] Diagnose uses of zero length arrays.

Why the need for this restriction ?

Nov 17 2021, 5:27 AM · Restricted Project

Apr 7 2021

Naghasan accepted D99190: [SYCL] Add design document for SYCL mode.

LGTM

Apr 7 2021, 2:33 AM · Restricted Project

Apr 1 2021

Naghasan added inline comments to D99190: [SYCL] Add design document for SYCL mode.
Apr 1 2021, 7:54 AM · Restricted Project
Naghasan added inline comments to D99488: [SYCL][Doc] Add design document for SYCL mode.
Apr 1 2021, 4:22 AM · Restricted Project
Naghasan added inline comments to D99190: [SYCL] Add design document for SYCL mode.
Apr 1 2021, 2:48 AM · Restricted Project

Mar 29 2021

Naghasan added inline comments to D99488: [SYCL][Doc] Add design document for SYCL mode.
Mar 29 2021, 9:41 AM · Restricted Project
Naghasan requested changes to D99190: [SYCL] Add design document for SYCL mode.

Added comment for the front end support part.

Mar 29 2021, 6:14 AM · Restricted Project

Sep 29 2020

Naghasan accepted D87282: [SYCL] Assume SYCL device functions are convergent.

Change make sense to me (same rational as CUDA or OpenCL).

Sep 29 2020, 1:40 AM · Restricted Project

Jun 4 2020

Naghasan added a comment to D80932: [SYCL] Make default address space a superset of OpenCL address spaces..

Why? Can you explain what you are trying to achieve with this?

I think @asavonic can provide more detailed answer, but IIRC we spent a lot time trying to marry template meta-programming with OpenCL address space deductions, but even simplest template functions from C++ standard library breaks compilation. It's important to note one important difference in design goals for SYCL and C++ for OpenCL. SYCL aims to enable compilation of regular C++ as much as possible where as one of the C++ for OpenCL goals is to keep compatibility with OpenCL C. These two goals might lead to different design decisions.

I don't see a contradiction in those goals. We plan to support C++ libraries with C++ for OpenCL of course with functionality that is accepted by conformant OpenCL implementations. For example, libraries that use virtual functions leading to function pointers are not going to work portably in OpenCL devices. If SYCL aims to compile to conformant OpenCL devices then it should not be very different I imagine?

(Edited a bit)

Jun 4 2020, 7:04 AM · Restricted Project

Feb 18 2020

Naghasan added inline comments to D72857: [SYCL] Driver option to enable SYCL mode and select SYCL version.
Feb 18 2020, 1:46 AM · Restricted Project

Feb 6 2019

Naghasan accepted D57768: [SYCL] Add clang front-end option to enable SYCL device compilation flow..

Side note: might be good to also involve @Anastasia, as some of the future patches will overlap with OpenCL.

Feb 6 2019, 1:22 AM · Restricted Project