RE: [PATCH bpf-next 08/13] uprobes/x86: Add support to optimize uprobes

[David Laight <David.Laight@ACULAB.COM>]

From: Jiri Olsa
> Sent: 11 December 2024 13:34
> 
> Putting together all the previously added pieces to support optimized
> uprobes on top of 5-byte nop instruction.
> 
> The current uprobe execution goes through following:
>   - installs breakpoint instruction over original instruction
>   - exception handler hit and calls related uprobe consumers
>   - and either simulates original instruction or does out of line single step
>     execution of it
>   - returns to user space
> 
> The optimized uprobe path
> 
>   - checks the original instruction is 5-byte nop (plus other checks)
>   - adds (or uses existing) user space trampoline and overwrites original
>     instruction (5-byte nop) with call to user space trampoline
>   - the user space trampoline executes uprobe syscall that calls related uprobe
>     consumers
>   - trampoline returns back to next instruction
...

How on earth can you safely overwrite a randomly aligned 5 byte instruction
that might be being prefetched and executed by another thread of the
same process.

If the instruction doesn't cross a cache line boundary then you might
manage to convince people that an 8-byte write will always be atomic
wrt other cpu reading instructions.
But you can't guarantee the alignment.

You might manage with the 7 byte sequence:
	br .+7; call addr
and then update 'addr' before changing the branch offset from 05 to 00.
But even that may not be safe if 'addr' crosses a cache line boundary.

You could replace a one byte nop (0x90) with a breakpoint (0xcc) and
then return to the instruction after the breakpoint.
That would save having to emulate or single stap the overwritten
instruction.

	David

-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)