Is there something like forceinline in PTX? I have a PTX-only kernel that would really benefit from funciton inlining. Looking at SASS source I see no inlining at all, forced or not.
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| asm inlining in CUDA code? | 5 | 6559 | July 19, 2010 | |
| Inline PTX assembly example | 1 | 14821 | August 3, 2010 | |
| What is __noinline__ equivalent on PTX ISA? | 1 | 533 | May 8, 2021 | |
| Call inline ptx function? | 5 | 2317 | June 19, 2012 | |
| ptxas -v output : kernel calling device functions | 3 | 1299 | January 23, 2015 | |
| Inline functions not inlined in CUDA 6.5? | 7 | 6084 | November 29, 2014 | |
| __forceinline__ and __forceinline | 1 | 9618 | July 16, 2013 | |
| __forceinline__, __noinline__ and "Too many resources" | 1 | 888 | May 20, 2023 | |
| How functions are compiled? Are function calls expanded inline or are actually CALLED? | 3 | 1306 | April 8, 2011 | |
| why CUDA 2.0 does not expose all PTX ISA 1.3 ? | 20 | 27985 | November 5, 2008 |