Yorick Hardy a écrit :
Not to worry, I am now more confused than ever :-).
By the way, what is STN_UNDEF for ?
it's used as an index in the Elf32_Sym array to identify the undefined
symbol, whereas SHN_UNDEF is a potential value of Elf32_Sym.st_shndx
(which is the section in which the symbol is d
> seems like i misread some of the ELF specs. Your fix is right (sorry for
> the noise). BTW, you could also remove the use of STN_UNDEF in elf.c
> while you're at it.
Not to worry, I am now more confused than ever :-).
By the way, what is STN_UNDEF for ?
Changelog: Use SHN_UNDEF instead of STN
Yorick Hardy a écrit :
Thanks for your input. It is very likely that I misunderstood, but I found
the following in /usr/include/elf.h on NetBSD 1.6
[snip]
Is this consistent with what you said?
seems like i misread some of the ELF specs. Your fix is right (sorry for
the noise). BTW, you could als
Thanks for your input. It is very likely that I misunderstood, but I found
the following in /usr/include/elf.h on NetBSD 1.6
/*
* Symbol Table
*/
typedef struct {
...
Elf32_Half st_shndx; /* section index of symbol */
} Elf32_Sym;
...
/*
* Special sectio
Yorick Hardy a écrit :
Changelog: Use SHN_UNDEF instead of STN_UNDEF when it is defined
and STN_UNDEF is not. Fixes compiling on NetBSD 1.6.
Kind regards,
both STN_UNDEF and SHN_UNDEF exist in ELF definition and are meant to be
used in different cases, which your patch mixes. Hence the
