forked from luck/tmp_suning_uos_patched
c65916fe35
I can't see the reason ". = VDSO_PRELINK + 0x900;" was ever there in the linker script for the x86_64 vDSO. I can't find anything that depends on this magic offset, or that should care at all about the particular location of of the .data section (all from vvar.c) in the vDSO image. If it is really desireable to place .data at 0x900, then it should be after all the other sections so they fill in the space up to 0x900. This removes the 0x900 magic and cleans up the output sections generally in the vDSO linker script. This saves a few hundred bytes in the size of the vDSO file, bringing it back well under 4kb total so that its vma only needs one page. Signed-off-by: Roland McGrath <roland@redhat.com> Cc: Andi Kleen <ak@suse.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> |
||
|---|---|---|
| .. | ||
| .gitignore | ||
| Makefile | ||
| vclock_gettime.c | ||
| vdso-note.S | ||
| vdso-start.S | ||
| vdso.lds.S | ||
| vdso.S | ||
| vextern.h | ||
| vgetcpu.c | ||
| vma.c | ||
| voffset.h | ||
| vvar.c | ||