From 606cb1626f49b00d5c246b03bdda1c4ad086545e Mon Sep 17 00:00:00 2001
From: Hanno Becker <hanno.becker@arm.com>
Date: Sun, 17 Apr 2022 06:57:34 +0100
Subject: [PATCH] Add comment explaining structure of UMAAL assembly
Signed-off-by: Hanno Becker <hanno.becker@arm.com>
---
library/bn_mul.h | 10 ++++++++++
1 file changed, 10 insertions(+)
diff --git a/library/bn_mul.h b/library/bn_mul.h
index 275be6048..962d7a97b 100644
--- a/library/bn_mul.h
+++ b/library/bn_mul.h
@@ -739,6 +739,16 @@
mbedtls_mpi_uint tmp_a1, tmp_b1; \
asm volatile (
+ /* - Make sure loop is 4-byte aligned to avoid stalls
+ * upon repeated non-word aligned instructions in
+ * some microarchitectures.
+ * - Don't use ldm with post-increment or back-to-back
+ * loads with post-increment and same address register
+ * to avoid stalls on some microarchitectures.
+ * - Bunch loads and stores to reduce latency on some
+ * microarchitectures. E.g., on Cortex-M4, the first
+ * in a series of load/store operations has latency
+ * 2 cycles, while subsequent loads/stores are single-cycle. */
#define MULADDC_X2_CORE \
".p2align 2 \n\t" \
"ldr.w %[a0], [%[in]], #+8 \n\t" \