[En-Nut-Discussion] Anyone ever compiled contrib/arm-crypto-library?

Ole Reinhardt ole.reinhardt at embedded-it.de
Wed Oct 30 14:13:10 CET 2013

Hi Harald,

>>> Really? 4 seconds only? That didn't coincide with my experience. I
>>> didn't test the Cortex, but the ARM7TDMI took about 1 minute @ 50MHz.
>> Does the --insecure parameter speeds up things here?
> It may be possible that this speeds up the client side. But the server
> still needs to do the same calculations.

In this case, the 4 seconds are real.

Perhaps the thumb instructions speed things up and perhaps the M3 has a
better pipelining and a better memory bandwith? Otherwise I can't
explain this havy difference...

>> With either google-chrome and firefox I did not manage to establish a
>> connection at all.
> Note, that you need at least Firefox version 24 and you need to enable
> TLS 1.2. Google Chrome should work.

Thanks for this hint!

> However, using the trunk right now with Ethernut 5 seems to crash the
> TLS client. I'm investigating this.

Have you enabled debugging? There is at least one error in one of the
debugging outputs.

This is my patch:

--- a/nut/app/tls/tls_handshake.c	2012-11-27 11:33:38.791667229 +0100
+++ b/nut/app/tls/tls_handshake.c	2013-10-26 02:04:36.281511440 +0200
@@ -57,6 +57,7 @@

 #include "hexdump.h"

+#include "tls_error_report.h"
 #include "tls_record_layer.h"
 #include "tls_driver.h"
 #include <crypto/scal-basic.h>
@@ -756,9 +757,9 @@
 #if DEBUG
     fprint_ciphersuite(TLS_ERROR_REPORT_STREAM, selected_cipher);
-    TLS_ERROR_REPORT("selected hash-signature: %s - %s\n",
-            hash_names[selected_hashsignature->id.names.hash_algorithm],
+    TLS_ERROR_REPORT("selected hash-signature: %d - %d\n",
+            selected_hashsignature->id.names.hash_algorithm,
+            selected_hashsignature->id.names.signature_algorithm);
     set_pending_state(tls_con, selected_cipher, selected_hashsignature);

kernel concepts GmbH            Tel: +49-271-771091-14
Sieghuetter Hauptweg 48         Mob: +49-177-7420433
D-57072 Siegen

More information about the En-Nut-Discussion mailing list