[PATCH bpf-next v3 2/2] docs: bpf: Clarify -mcpu=v3 requirement for atomic ops

From: Brendan Jackman
Date: Wed Jan 20 2021 - 16:40:15 EST

Alexei pointed out [1] that this wording is pretty confusing. Here's
an attempt to be more explicit and clear.

[1] https://lore.kernel.org/bpf/CAADnVQJVvwoZsE1K+6qRxzF7+6CvZNzygnoBW9tZNWJELk5c=Q@xxxxxxxxxxxxxx/T/#m07264fc18fdc43af02fc1320968afefcc73d96f4

Signed-off-by: Brendan Jackman <jackmanb@xxxxxxxxxx>
Documentation/networking/filter.rst | 7 ++++---
1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/Documentation/networking/filter.rst b/Documentation/networking/filter.rst
index 4c2bb4c6364d..b3f457802836 100644
--- a/Documentation/networking/filter.rst
+++ b/Documentation/networking/filter.rst
@@ -1081,9 +1081,10 @@ before is loaded back to ``R0``.

Note that 1 and 2 byte atomic operations are not supported.

-Except ``BPF_ADD`` *without* ``BPF_FETCH`` (for legacy reasons), all 4 byte
-atomic operations require alu32 mode. Clang enables this mode by default in
-architecture v3 (``-mcpu=v3``). For older versions it can be enabled with
+Clang can generate atomic instructions by default when ``-mcpu=v3`` is
+enabled. If a lower version for ``-mcpu`` is set, the only atomic instruction
+Clang can generate is ``BPF_ADD`` *without* ``BPF_FETCH``. If you need to enable
+the atomics features, while keeping a lower ``-mcpu`` version, you can use
``-Xclang -target-feature -Xclang +alu32``.

You may encounter ``BPF_XADD`` - this is a legacy name for ``BPF_ATOMIC``,