On Wed, Feb 28, 2024 at 06:51:49PM +0100, Alexandre Ghiti wrote:

For now, we use stop_machine() to patch the text and when we use IPIs for
remote icache flushes (which is emitted in patch_text_nosync()), the system
hangs.

So instead, make sure every cpu executes the stop_machine() patching
function and emit a local icache flush there.

Co-developed-by: Björn Töpel <redacted>
Signed-off-by: Björn Töpel <redacted>
Signed-off-by: Alexandre Ghiti <redacted>

Modulo the removal of the hunk discussed with Samuel,

Reviewed-by: Andrea Parri <parri.andrea@gmail.com>

Some nits / amendments to the inline comments below:

+		/*
+		 * Make sure the patching store is effective *before* we
+		 * increment the counter which releases all waiting cpus
+		 * by using the release version of atomic increment.
+		 */

s/cpus/CPUs
s/release version/release variant

The comment could be amended with a description of the matching barrier(s), say,
"The release pairs with the call to local_flush_icache_all() on the waiting CPU".

(Same for the comment in patch_text_cb().)

  Andrea