Re: [Opensg-users] OpenSG2: Accumulated changes... part 2: link flags

Hello Johannes, On 2017-02-01 09:03, Johannes wrote: > finally I have finished the link flag undertaking :-) > Attached you can find the final diff file. thank you for the patch. Everybody involved seemed happy with this, so I've applied it now. Cheers, Carsten

Re: [Opensg-users] OpenSG2: Accumulated changes... part 2: link flags

Hi Johannes! Thanks for your excellent work! Sorry for not responding earlier but I've been traveling and it was lost in the inbox. :( The space savings are great. Hopefully it will enable some performance benefit too. Cheers, /Marcus P.S. Good catch on incremental linking too! 2017-02-01

Re: [Opensg-users] OpenSG2: Accumulated changes... part 2: link flags

Hi, finally I have finished the link flag undertaking :-) Attached you can find the final diff file. What I have done: 1. Added cmake variable OSG_WINDOWS_LINK_OPTIMIZE which defaults to OFF. If the variable is ON the /OPT:REF /OPT:ICF are added to the CMAKE_SHARED_LINKER_FLAGS_DEBUGOPT,

Re: [Opensg-users] OpenSG2: Accumulated changes... part 2: link flags

Hi, attached you can find the size numbers for the release dlls and pdb files that I have extracted from the opensg master build with and without the patch of OSGSetupCompiler.cmake. So we would be gaining around 22.5% of size reduction for the dlls. Best, Johannes Microsoft Visual Studio

Re: [Opensg-users] OpenSG2: Accumulated changes... part 2: link flags

Hello Marcus, I will implement your suggestion and provide a new diff for that. But before I will take the time to build both versions so that I have numbers that I can provide. Please stay tuned, it will take one day or another. Carsten, please wait before check in. I will give it another

Re: [Opensg-users] OpenSG2: Accumulated changes... part 2: link flags

Hi Johannes, Thanks for the explanation! It sounds like /OPT:ICF is safe, even though the debugging experience might be a bit degraded (but not totally). Maybe check and see how much the binary sizes change. If it's only 5-10%, it's not really worth it, but if we gain more than that by COMDAT

Re: [Opensg-users] OpenSG2: Accumulated changes... part 2: link flags

Hello Marcus and Carsten, First of, we also do need debugging symbols in release builds. Microsoft writes on https://msdn.microsoft.com/en-us/library/windows/desktop/ee416349(v=vs.85).aspx "You also need to take steps to maintain full compiler optimizations while generating symbols." - set

Re: [Opensg-users] OpenSG2: Accumulated changes... part 2: link flags

Hi, We need debugging symbols in release builds. If OPT:REF / OPT:ICF doesn't conflict with /Zi and pdb-generation, I have no objections. Cheers, /Marcus 2017-01-25 0:51 GMT+01:00 Carsten Neumann : > Hello Johannes, > > On 2017-01-18 04:58, Johannes Brunen

Re: [Opensg-users] OpenSG2: Accumulated changes... part 2: link flags

Hello Johannes, On 2017-01-18 04:58, Johannes Brunen wrote: > This change does only affect the MSVC windows build. OpenSG is > traditionally build with the compiler flag /Zi on windows even in the > optimized cases. That results in generation of the .pdb database files > that can be used