Re: [PATCH] x86/alternatives: Fix optimize_nops() checking

[Borislav Petkov]

On Wed, Jan 10, 2018 at 12:20:40PM -0800, Linus Torvalds wrote:
> No, that's not valid. That could never work anyway. The ".long 2f"
> would be the absolute address in the alternative section, but opcode
> E9 takes a relative 32-bit offset.

Ah, right, doh. I remember now. We used to do those jmps by computing the
relative offset:

From: 090a3f615524c3f75d09fdb37f15ea1868d79f7e
-       .section .altinstr_replacement,"ax"
-1:     .byte 0xeb                                      /* jmp <disp8> */
-       .byte (copy_page_rep - copy_page) - (2f - 1b)   /* offset */

-- 
Regards/Gruss,
    Boris.

Good mailing practices for 400: avoid top-posting and trim the reply.