Messages in this thread |  | | Date | Mon, 8 Oct 2018 12:31:40 -0400 | From | Steven Rostedt <> | Subject | Re: [POC][RFC][PATCH 1/2] jump_function: Addition of new feature "jump_function" |
| |
On Mon, 8 Oct 2018 17:57:57 +0200 Peter Zijlstra <peterz@infradead.org> wrote:
> On Mon, Oct 08, 2018 at 01:33:14AM -0700, Andy Lutomirski wrote: > > > Can't we hijack the relocation records for these functions before they > > > get thrown out in the (final) link pass or something? > > > > I could be talking out my arse here, but I thought we could do this, > > too, then changed my mind. The relocation records give us the > > location of the call or jump operand, but they don’t give the address > > of the beginning of the instruction. > > But that's like 1 byte before the operand, right? We could even double check > this by reading back that byte and ensuring it is in fact 0xE8 (CALL). > > AFAICT there is only the _1_ CALL encoding, and that is the 5 byte: E8 <PLT32>, > so if we have the PLT32 location, we also have the instruction location. Or am > I missing something?
Yes, this is exactly what I was thinking of doing. All we need to do is have objtool (or a modification of whatever we come up with), to find the call sites of a specific function (we can have a table to look up for), that creates a section listing all these call sites, and on boot up, we can confirm that they are indeed calls (e8 operations).
-- Steve
|  |