28/10/2019 11:47, Andrzej Ostruszka: > On 10/27/19 12:47 PM, Thomas Monjalon wrote: > >> +turned on. This depends obviously on the capabilities of the compiler > > > > Not sure what you are talking about. > > If it is about fat objects, it is not obvious. > > > >> +to do "whole program" optimization at link time and is available only > > That was not a reference to some formally defined "capabilities" just an > "obvious" statement that in order to benefit from LTO one needs a > compiler "capable of" performing it. I will rephrase this paragraph as: > > -8<------------- > The DPDK supports compilation with link time optimization turned on. > This depends obviously on the ability of the compiler to do "whole > program" optimization at link time and is available only for compilers > that support that feature (gcc and icc). > To be more specific, compiler (in addition to performing LTO) have to > support creation of ELF objects containing both normal code and internal > representation (fat-lto-objects). This is required since during build > some code is generated by parsing produced ELF objects (pmdinfogen). > -8<------------- > > Will it suffice? If not, then please suggest wording to be used.
There is a confusion here: "compilers that support that feature (gcc and icc)" In my understanding, clang supports LTO but is not mentioned because of the lack of fat object support. I think you should mention gcc and icc only after explaining the second constraint (fat objects). > >> +for compilers that support that feature (gcc and icc). To be more > > > > Please start a new sentence at the beginning of a line in the RsT file. > > It is easier to review and update. > > OK Please consider such comment to apply to the whole file :) [...] > OK. I'll send the next version today. > > Thank you again for the comments. Thank you
