Re: [RFC patch 14/19] bpf: Use migrate_disable() in hashtab code

Mathieu Desnoyers [off-list ref] writes:

On 14-Feb-2020 02:39:31 PM, Thomas Gleixner wrote:

quoted

Replace the preempt_disable/enable() pairs with migrate_disable/enable()
pairs to prepare BPF to work on PREEMPT_RT enabled kernels. On a non-RT
kernel this maps to preempt_disable/enable(), i.e. no functional change.

...

Having all those events randomly and silently discarded might be quite
unexpected from a user standpoint. This turns the deadlock prevention
mechanism into a random tracepoint-dropping facility, which is
unsettling.

Well, it randomly drops events which might be unrelated to the syscall
target today already, this will just drop some more. Shrug.

One alternative approach we could consider to solve this is to make
this deadlock prevention nesting counter per-thread rather than
per-cpu.

That should work both on !RT and RT.

Also, I don't think using __this_cpu_inc() without preempt-disable or
irq off is safe. You'll probably want to move to this_cpu_inc/dec
instead, which can be heavier on some architectures.

Good catch.

Thanks,

        tglx