This is the mail archive of the gcc-bugs@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[Bug c++/58575] gcc should compile firefox code faster.


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58575

Manuel LÃpez-IbÃÃez <manu at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |manu at gcc dot gnu.org

--- Comment #2 from Manuel LÃpez-IbÃÃez <manu at gcc dot gnu.org> ---
That GCC could be much faster is well-known but how to make it so is less clear
and there is simply not enough people to do the work required. Currently there
is as many people working just on Clang than on the whole GCC (including all
languages, middle-end and backends).

Dimitrios Apostolou did some nice work on benchmarking, but his actual patches
did not bring impressive improvements:

http://gcc.gnu.org/wiki/OptimisingGCC

I don't know how much of this work ended up in GCC (if any).

There are many ideas on what could speed-up C++ parsing:

http://gcc.gnu.org/wiki/Speedup_areas#C.2B-.2B-_improvements

BTW, reducing memory allocation (or using smarter data structures) is likely to
bring significant improvements:

http://gcc.gnu.org/wiki/Speedup_areas#Some_general_thoughts_on_memory_usage

But at the end, someone needs to come up with very detailed benchmark data to
identify the real compile-time hogs and patches to address them.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]