On Wed, 2 Mar 2016 17:24:12 +0100 Peter Zijlstra <pet...@infradead.org> wrote:
> On Wed, Mar 02, 2016 at 04:53:36PM +0100, Sedat Dilek wrote: > > ffffffff8110f570 <del_timer_sync>: > > ffffffff8110f570: 55 push %rbp > > ffffffff8110f571: 48 89 e5 mov %rsp,%rbp > > ffffffff8110f574: 41 57 push %r15 > > ffffffff8110f576: 41 56 push %r14 > > ffffffff8110f578: 53 push %rbx > > ffffffff8110f579: 48 83 ec 28 sub $0x28,%rsp > > stack offset is 0x28 bytes [*] Actually, isn't it really 0x40 bytes? The stack pushed 3 words (8 bytes each) before doing the subtract. 0x28 == 40 bytes, 3 * 8 = 24, 40 + 24 = 64 == 0x40. > > > ffffffff8110f57d: 48 89 fb mov %rdi,%rbx > > ffffffff8110f580: e8 6b 6e 80 00 callq ffffffff819163f0 <mcount> > > ffffffff8110f585: e8 66 6e 80 00 callq ffffffff819163f0 <mcount> > > ffffffff8110f58a: e8 61 6e 80 00 callq ffffffff819163f0 <mcount> > > ffffffff8110f58f: e8 5c 6e 80 00 callq ffffffff819163f0 > > <mcount> > > Your compiler is on drugs! Totally agree! [..] > > > ffffffff8110f5fd: 4c 89 75 d0 mov %r14,-0x30(%rbp) > > ffffffff8110f601: ff 75 d0 pushq -0x30(%rbp) > > ffffffff8110f604: 9d popfq > > put r14 into -0x30(rbp) and pushes/pops that, see [*] this is 8 bytes > over stack ?! But from rbp, the stack is 0x40 bytes. This may be fine. -- Steve
