This is the mail archive of the libc-hacker@sources.redhat.com mailing list for the glibc project.
Note that libc-hacker is a closed list. You may look at the archives of this list, but subscription and posting are not open.
| Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
|---|---|---|
| Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |
| Other format: | [Raw text] | |
Hi!
ldconfig assumed loadaddr can be only 32-bit, while e.g. with prelinked
libs on Alpha it is usually 0x000002001XXXXXXX.
Furthermore, I don't see the point of masking off lower 12 bits of
p_vaddr/p_offset for computation. If anything it could be p_align - 1,
but then anyway, if a library is incorrectly linked (ie. to receive
ELF load command address/offset not properly aligned
error from dynamic linker), the dynamic strings will be way more likely
at the broken location then some other one we arbitrarily guess.
2002-08-29 Jakub Jelinek <jakub@redhat.com>
* sysdeps/generic/readelflib.c (process_elf_file): Make loadaddr
ElfW(Addr). Don't mask upper 32-bits and lower 12 bits off from
p_vaddr/p_offset when computing loadaddr.
--- libc/sysdeps/generic/readelflib.c.jj 2002-08-29 12:21:26.000000000 +0200
+++ libc/sysdeps/generic/readelflib.c 2002-08-29 12:21:26.000000000 +0200
@@ -46,7 +46,7 @@ process_elf_file (const char *file_name,
{
int i;
unsigned int j;
- int loadaddr;
+ ElfW(Addr) loadaddr;
unsigned int dynamic_addr;
size_t dynamic_size;
char *program_interpreter;
@@ -101,8 +101,7 @@ process_elf_file (const char *file_name,
{
case PT_LOAD:
if (loadaddr == -1)
- loadaddr = (segment->p_vaddr & 0xfffff000)
- - (segment->p_offset & 0xfffff000);
+ loadaddr = segment->p_vaddr - segment_p_offset;
break;
case PT_DYNAMIC:
Jakub
| Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
|---|---|---|
| Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |