ggml: thread syncronization on Arm

For x86 slaren was genereous enough to add _mm_pause() to the busy
spin wait loop in ggml_barrier(), but everything else just busy
spins, loading an atomic int on every iteration, thus forcing cache
sync between the cores. This results in a massive drop in performance
on my M2-Max laptop when using 8 threads. The closest approximation
to _mm_pause() on Arm seems to be
     __asm__ __volatile__("isb\n");
After adding this to the busy spin loop, performance for 8 threads
recovers back to expected levels.

1 files changed, 2 insertions, 0 deletions

diff --git a/ggml.c b/ggml.c
index 9a83059d..0f68cea3 100644
--- a/ggml.c
+++ b/ggml.c
@@ -19142,6 +19142,8 @@ static void ggml_barrier(struct ggml_compute_state * state) {
                 }
             #if defined(__SSE3__)
                 _mm_pause();
+            #elif defined __ARM_NEON
+                __asm__ __volatile__("isb\n");
             #endif
             }
             sched_yield();