https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84487
Wilco <wilco at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |wilco at gcc dot gnu.org
--- Comment #7 from Wilco <wilco at gcc dot gnu.org> ---
(In reply to Martin Liška from comment #5)
> Sorry I pointed to a wrong revision. With the mentioned revision and #c4
> test-case I have:
>
> $ nm good.o
> $ nm bad.o
> U free
> U _gfortran_runtime_error_at
> U memcpy
> 0000000000000c00 T __types_module_MOD___copy_types_module_Archive_type
> 0000000000000be0 T __types_module_MOD___copy_types_module_Atom_type
> 0000000000000b40 T __types_module_MOD___copy_types_module_Basis_type
> 0000000000000b20 T __types_module_MOD___copy_types_module_Buffer_type
> ...
>
> Is it a regression or it's intentional?
I see a similar regression on cam4_r:
text data bss dec hex
5291209 384776 57000384 62676369 3bc5d91 before
53805556 389304 58141504 112336364 6b21dec after
All of the huge increase is due to rodata:
13 .rodata 0006ad88 00000000008a10d0 00000000008a10d0 004a10d0 2**4
CONTENTS, ALLOC, LOAD, READONLY, DATA
13 .rodata 02e81dc8 0000000000888f70 0000000000888f70 00488f70 2**4
CONTENTS, ALLOC, LOAD, READONLY, DATA
Almost all of the 48MBytes of extra rodata is zeroes.
