From: Thomas Gleixner <tglx@linutronix.de>
To: LKML <linux-kernel@vger.kernel.org>
Cc: David Miller <davem@davemloft.net>,
bpf@vger.kernel.org, netdev@vger.kernel.org,
Alexei Starovoitov <ast@kernel.org>,
Daniel Borkmann <daniel@iogearbox.net>,
Sebastian Sewior <bigeasy@linutronix.de>,
Peter Zijlstra <peterz@infradead.org>,
Clark Williams <williams@redhat.com>,
Steven Rostedt <rostedt@goodmis.org>,
Juri Lelli <juri.lelli@redhat.com>,
Ingo Molnar <mingo@kernel.org>,
Mathieu Desnoyers <mathieu.desnoyers@efficios.com>,
Vinicius Costa Gomes <vinicius.gomes@intel.com>,
Jakub Kicinski <kuba@kernel.org>
Subject: [patch V2 14/20] bpf: Use migrate_disable() in hashtab code
Date: Thu, 20 Feb 2020 21:45:31 +0100 [thread overview]
Message-ID: <20200220204618.703606294@linutronix.de> (raw)
In-Reply-To: 20200220204517.863202864@linutronix.de
The required protection is that the caller cannot be migrated to a
different CPU as these places take either a hash bucket lock or might
trigger a kprobe inside the memory allocator. Both scenarios can lead to
deadlocks. The deadlock prevention is per CPU by incrementing a per CPU
variable which temporarily blocks the invocation of BPF programs from perf
and kprobes.
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.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
kernel/bpf/hashtab.c | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
--- a/kernel/bpf/hashtab.c
+++ b/kernel/bpf/hashtab.c
@@ -1319,7 +1319,7 @@ static int
}
again:
- preempt_disable();
+ migrate_disable();
this_cpu_inc(bpf_prog_active);
rcu_read_lock();
again_nocopy:
@@ -1339,7 +1339,7 @@ static int
raw_spin_unlock_irqrestore(&b->lock, flags);
rcu_read_unlock();
this_cpu_dec(bpf_prog_active);
- preempt_enable();
+ migrate_enable();
goto after_loop;
}
@@ -1348,7 +1348,7 @@ static int
raw_spin_unlock_irqrestore(&b->lock, flags);
rcu_read_unlock();
this_cpu_dec(bpf_prog_active);
- preempt_enable();
+ migrate_enable();
kvfree(keys);
kvfree(values);
goto alloc;
@@ -1398,7 +1398,7 @@ static int
rcu_read_unlock();
this_cpu_dec(bpf_prog_active);
- preempt_enable();
+ migrate_enable();
if (bucket_cnt && (copy_to_user(ukeys + total * key_size, keys,
key_size * bucket_cnt) ||
copy_to_user(uvalues + total * value_size, values,
next prev parent reply other threads:[~2020-02-20 20:57 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-02-20 20:45 [patch V2 00/20] bpf: Make BPF and PREEMPT_RT co-exist Thomas Gleixner
2020-02-20 20:45 ` [patch V2 01/20] bpf: Enforce preallocation for all instrumentation programs Thomas Gleixner
2020-02-22 4:29 ` Alexei Starovoitov
2020-02-22 8:40 ` Thomas Gleixner
2020-02-23 22:40 ` Alexei Starovoitov
2020-02-20 20:45 ` [patch V2 02/20] bpf: Update locking comment in hashtab code Thomas Gleixner
2020-02-20 20:45 ` [patch V2 03/20] bpf/tracing: Remove redundant preempt_disable() in __bpf_trace_run() Thomas Gleixner
2020-02-20 20:45 ` [patch V2 04/20] perf/bpf: Remove preempt disable around BPF invocation Thomas Gleixner
2020-02-20 20:45 ` [patch V2 05/20] bpf: Remove recursion prevention from rcu free callback Thomas Gleixner
2020-02-20 20:45 ` [patch V2 06/20] bpf: Dont iterate over possible CPUs with interrupts disabled Thomas Gleixner
2020-02-20 20:45 ` [patch V2 07/20] bpf: Provide bpf_prog_run_pin_on_cpu() helper Thomas Gleixner
2020-02-20 20:45 ` [patch V2 08/20] bpf: Replace cant_sleep() with cant_migrate() Thomas Gleixner
2020-02-20 20:45 ` [patch V2 09/20] bpf: Use bpf_prog_run_pin_on_cpu() at simple call sites Thomas Gleixner
2020-02-20 20:45 ` [patch V2 10/20] trace/bpf: Use migrate disable in trace_call_bpf() Thomas Gleixner
2020-02-20 20:45 ` [patch V2 11/20] bpf/tests: Use migrate disable instead of preempt disable Thomas Gleixner
2020-02-20 20:45 ` [patch V2 12/20] bpf: Use migrate_disable/enabe() in trampoline code Thomas Gleixner
2020-02-20 20:45 ` [patch V2 13/20] bpf: Use migrate_disable/enable in array macros and cgroup/lirc code Thomas Gleixner
2020-02-20 20:45 ` Thomas Gleixner [this message]
2020-02-20 20:45 ` [patch V2 15/20] bpf: Provide recursion prevention helpers Thomas Gleixner
2020-02-20 20:45 ` [patch V2 16/20] bpf: Replace open coded recursion prevention Thomas Gleixner
2020-02-20 20:45 ` [patch V2 17/20] bpf: Factor out hashtab bucket lock operations Thomas Gleixner
2020-02-20 20:45 ` [patch V2 18/20] bpf: Prepare hashtab locking for PREEMPT_RT Thomas Gleixner
2020-02-20 20:45 ` [patch V2 19/20] bpf, lpm: Make locking RT friendly Thomas Gleixner
2020-02-20 20:45 ` [patch V2 20/20] bpf/stackmap: Dont trylock mmap_sem with PREEMPT_RT and interrupts disabled Thomas Gleixner
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20200220204618.703606294@linutronix.de \
--to=tglx@linutronix.de \
--cc=ast@kernel.org \
--cc=bigeasy@linutronix.de \
--cc=bpf@vger.kernel.org \
--cc=daniel@iogearbox.net \
--cc=davem@davemloft.net \
--cc=juri.lelli@redhat.com \
--cc=kuba@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mathieu.desnoyers@efficios.com \
--cc=mingo@kernel.org \
--cc=netdev@vger.kernel.org \
--cc=peterz@infradead.org \
--cc=rostedt@goodmis.org \
--cc=vinicius.gomes@intel.com \
--cc=williams@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).