Fresh clone of AlloSystem *segfaults clang*

[kybr]

here's a gist

https://gist.github.com/52acbe5004f7233500195f4f4df8a30f

here's a "verbose" gist:

https://gist.github.com/82fc04f6b1c7d756cb3ac0f9ba9f98ce

here's the machine info:

$ c++ --version
Apple LLVM version 6.0 (clang-600.0.54) (based on LLVM 3.5svn)
Target: x86_64-apple-darwin13.4.0
Thread model: posix

$ uname -a
Darwin audio.10g 13.4.0 Darwin Kernel Version 13.4.0: Sun Aug 17 19:50:11 PDT 2014; root:xnu-2422.115.4~1/RELEASE_X86_64 x86_64

$ sw_vers -productVersion
10.9.5

[kybr]

84055fd is the first bad commit (according to git bisect)

[konhyong]

oh wow i was actually wondering why lighting was suddenly broken in omniapp and started looking into this. good find. just as a PSA any non-zero lighting amount in omniapp/graphicsrenderer will not work currently.

…

On Fri, Mar 16, 2018 at 9:04 AM, karl yerkes ***@***.***> wrote: 84055fd <84055fd> is the first bad commit (according to git bisect) — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#76 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABifzXYn7s-YKU6JSlo32xdx0f4IcdXBks5te-KVgaJpZM4Ss7PR> .

[kybr]

Can you explain more what you at talking about? How is this crash in the compiler related to omniapp and lighting?

[LancePutnam]

From what I can tell, this machine is rather old: OS X Mavericks (released 2014) and Apple LLVM 6.0. Everything works for me on El Capitan w/ Apple LLVM 8.0.0. Maybe try on a more current version of OS X? Another thought, try moving ShaderData out of Graphics. It's somewhat unusual to have a nested template class, so could be exposing a bug.

[LancePutnam]

Re. lighting not working: From a quick look at OmniApp, it looks like the shader assumes the light ID is 0. If you are using the Light class, this needs to match its ID. I will add a public accessor for this so it can be passed into shaders.

[LancePutnam]

Commit 4b7e4f9 adds the light ID accessor.

[konhyong]

Thanks Lance, works fine now it seems. and Karl: sorry for not explaining enough. Lighting had broken down in some of the programs, so I was backtracking commits one by one. ultimately the point code broke down was that commit you found. I assumed that the compilation issues were tied, since I was getting some shader compilation errors and the commit was tackling some fixed pipeline elements. sounds like the original cause might be different though.

…

On Mon, Mar 19, 2018 at 11:17 AM, Lance Putnam ***@***.***> wrote: Commit 4b7e4f9 <4b7e4f9> adds the light ID accessor. — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#76 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABifzShbWQBWC3_gbiAkyyVJIifNYEszks5tf_Y0gaJpZM4Ss7PR> .