search for: nolto

...ns We can potentially implement the analysis/runtime check with and without LTO. It is better to implement runtime check without LTO as we are expecting developers to use runtime-check build locally. LTO will slow down developer’s working cycle. For analysis, it is not clear which option is better. noLTO Implementation We create a special section containing those information for each TU. This section for the link unit will have the combined information. We then post-process this information to emit error messages. For each direct method, we create a struct containing the method name and the name of...

> On Jan 16, 2017, at 10:00 AM, Sean Silva via llvm-dev <llvm-dev at lists.llvm.org> wrote: > > What is the end goal? If the goal is roughly "if a user passes -O0 when compiling a TU in LTO mode, their final binary should have functions that look like the result of -O0 noLTO compilation", then the frontend can just emit a normal -O0 object file I think. What is the advantage of passing bitcode all the way to the linker only to jump through hoops to prevent LTO code generation from modifying it? That’s a good point, that addresses the “I want LTO for optimization”...

Looking at the existing flow of passes for LTO, it appears that most all passes are run on a per file basis, before the call to the gold linker. I'm looking to get people's feedback on whether there would be an advantage to waiting to run a number of these passes until the linking stage. For example, I believe I saw a post a little while back about postponing vectorization until the

In D28404, Mehdi wanted to use the 'optnone' attribute as a way to record "I was compiled with -O0" in the IR, because it seems like a good idea to remember that fact in an LTO compilation and there is no way to remember that fact currently. A couple of people felt it might be better to have this idea discussed on the dev list, where it might get better exposure, so I'm