Re: [PATCH bpf v2] bpf: fix nested bpf tracepoints with per-cpu data

2019-06-15 Thread Alexei Starovoitov
On Tue, Jun 11, 2019 at 2:54 PM Matt Mullins wrote: > > BPF_PROG_TYPE_RAW_TRACEPOINTs can be executed nested on the same CPU, as > they do not increment bpf_prog_active while executing. > > This enables three levels of nesting, to support > - a kprobe or raw tp or perf event, > - another one o

Re: [PATCH bpf v2] bpf: fix nested bpf tracepoints with per-cpu data

2019-06-14 Thread Matt Mullins
On Fri, 2019-06-14 at 16:50 +0200, Daniel Borkmann wrote: > On 06/14/2019 02:51 AM, Matt Mullins wrote: > > On Fri, 2019-06-14 at 00:47 +0200, Daniel Borkmann wrote: > > > On 06/12/2019 07:00 AM, Andrii Nakryiko wrote: > > > > On Tue, Jun 11, 2019 at 8:48 PM Matt Mullins wrote: > > > > > > > > >

Re: [PATCH bpf v2] bpf: fix nested bpf tracepoints with per-cpu data

2019-06-14 Thread Daniel Borkmann
On 06/14/2019 02:55 AM, Alexei Starovoitov wrote: > On Thu, Jun 13, 2019 at 5:52 PM Matt Mullins wrote: >> On Fri, 2019-06-14 at 00:47 +0200, Daniel Borkmann wrote: >>> On 06/12/2019 07:00 AM, Andrii Nakryiko wrote: On Tue, Jun 11, 2019 at 8:48 PM Matt Mullins wrote: > > BPF_PROG_TYP

Re: [PATCH bpf v2] bpf: fix nested bpf tracepoints with per-cpu data

2019-06-14 Thread Daniel Borkmann
On 06/14/2019 02:51 AM, Matt Mullins wrote: > On Fri, 2019-06-14 at 00:47 +0200, Daniel Borkmann wrote: >> On 06/12/2019 07:00 AM, Andrii Nakryiko wrote: >>> On Tue, Jun 11, 2019 at 8:48 PM Matt Mullins wrote: BPF_PROG_TYPE_RAW_TRACEPOINTs can be executed nested on the same CPU, as

Re: [PATCH bpf v2] bpf: fix nested bpf tracepoints with per-cpu data

2019-06-13 Thread Alexei Starovoitov
On Thu, Jun 13, 2019 at 5:52 PM Matt Mullins wrote: > > On Fri, 2019-06-14 at 00:47 +0200, Daniel Borkmann wrote: > > On 06/12/2019 07:00 AM, Andrii Nakryiko wrote: > > > On Tue, Jun 11, 2019 at 8:48 PM Matt Mullins wrote: > > > > > > > > BPF_PROG_TYPE_RAW_TRACEPOINTs can be executed nested on th

Re: [PATCH bpf v2] bpf: fix nested bpf tracepoints with per-cpu data

2019-06-13 Thread Matt Mullins
On Fri, 2019-06-14 at 00:47 +0200, Daniel Borkmann wrote: > On 06/12/2019 07:00 AM, Andrii Nakryiko wrote: > > On Tue, Jun 11, 2019 at 8:48 PM Matt Mullins wrote: > > > > > > BPF_PROG_TYPE_RAW_TRACEPOINTs can be executed nested on the same CPU, as > > > they do not increment bpf_prog_active while

Re: [PATCH bpf v2] bpf: fix nested bpf tracepoints with per-cpu data

2019-06-13 Thread Daniel Borkmann
On 06/12/2019 07:00 AM, Andrii Nakryiko wrote: > On Tue, Jun 11, 2019 at 8:48 PM Matt Mullins wrote: >> >> BPF_PROG_TYPE_RAW_TRACEPOINTs can be executed nested on the same CPU, as >> they do not increment bpf_prog_active while executing. >> >> This enables three levels of nesting, to support >>

Re: [PATCH bpf v2] bpf: fix nested bpf tracepoints with per-cpu data

2019-06-11 Thread Andrii Nakryiko
On Tue, Jun 11, 2019 at 8:48 PM Matt Mullins wrote: > > BPF_PROG_TYPE_RAW_TRACEPOINTs can be executed nested on the same CPU, as > they do not increment bpf_prog_active while executing. > > This enables three levels of nesting, to support > - a kprobe or raw tp or perf event, > - another one o

[PATCH bpf v2] bpf: fix nested bpf tracepoints with per-cpu data

2019-06-11 Thread Matt Mullins
BPF_PROG_TYPE_RAW_TRACEPOINTs can be executed nested on the same CPU, as they do not increment bpf_prog_active while executing. This enables three levels of nesting, to support - a kprobe or raw tp or perf event, - another one of the above that irq context happens to call, and - another one