]> git.baikalelectronics.ru Git - kernel.git/commit
tracing: Simplify & fix saved_tgids logic
authorPaul Burton <paulburton@google.com>
Wed, 30 Jun 2021 00:34:05 +0000 (17:34 -0700)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Mon, 19 Jul 2021 06:53:17 +0000 (08:53 +0200)
commitc2461118ff6c15bbd38b8d39c7b8b9a6fc3b3ec6
tree29c90b2a7fbcc3b80294e2cc6bc1ebeac2c9d72b
parenteb2bbe9cd4e4f415da4f62e1fbd603987a0192de
tracing: Simplify & fix saved_tgids logic

commit b81b3e959adb107cd5b36c7dc5ba1364bbd31eb2 upstream.

The tgid_map array records a mapping from pid to tgid, where the index
of an entry within the array is the pid & the value stored at that index
is the tgid.

The saved_tgids_next() function iterates over pointers into the tgid_map
array & dereferences the pointers which results in the tgid, but then it
passes that dereferenced value to trace_find_tgid() which treats it as a
pid & does a further lookup within the tgid_map array. It seems likely
that the intent here was to skip over entries in tgid_map for which the
recorded tgid is zero, but instead we end up skipping over entries for
which the thread group leader hasn't yet had its own tgid recorded in
tgid_map.

A minimal fix would be to remove the call to trace_find_tgid, turning:

  if (trace_find_tgid(*ptr))

into:

  if (*ptr)

..but it seems like this logic can be much simpler if we simply let
seq_read() iterate over the whole tgid_map array & filter out empty
entries by returning SEQ_SKIP from saved_tgids_show(). Here we take that
approach, removing the incorrect logic here entirely.

Link: https://lkml.kernel.org/r/20210630003406.4013668-1-paulburton@google.com
Fixes: 86e3f32bac53 ("tracing: Add support for recording tgid of tasks")
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Joel Fernandes <joelaf@google.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Paul Burton <paulburton@google.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
kernel/trace/trace.c