2015-06-05 09:39:29

by Corentin Labbe

[permalink] [raw]
Subject: [PATCH RFC] crypto: testmgr - Document struct cipher_testvec

Hello

I was adding a test to testmgr for cbc(aes) but I hesitate on the cipher_testvec parameter also_non_np.

Since its functions is not trivial, I propose a patch for documenting the struture.

So does I have well understand the purpose of also_non_np ?
And if yes what it is the interest of doing a test twice (one with and one without SG splitting) ?

Regards

>From 0156e32be7dd7ffd0a4fdad8d0a3eaf3c0bf92c7 Mon Sep 17 00:00:00 2001
From: LABBE Corentin <[email protected]>
Date: Wed, 3 Jun 2015 09:09:13 +0200
Subject: [PATCH 02/11] crypto: testmgr - Document struct cipher_testvec

Signed-off-by: LABBE Corentin <[email protected]>
---
crypto/testmgr.h | 18 ++++++++++++++++++
1 file changed, 18 insertions(+)

diff --git a/crypto/testmgr.h b/crypto/testmgr.h
index 8a31e9a..ab2019e 100644
--- a/crypto/testmgr.h
+++ b/crypto/testmgr.h
@@ -46,6 +46,24 @@ struct hash_testvec {
unsigned char ksize;
};

+/*
+ * cipher_testvec: structure to describe a cipher test
+ * @key: A pointer to a key used by the test
+ * @klen: The length of @key
+ * @iv: A pointer to the IV used by the test
+ * @input: A pointer to data used as input
+ * @ilen The length of data in @input
+ * @result: A pointer to what the test need to produce
+ * @rlen: The length of data in @result
+ * @fail: If set to one, the test need to fail
+ * @wk: Does the test need CRYPTO_TFM_REQ_WEAK_KEY
+ * ( e.g. test needs to fail due to a weak key )
+ * @np: numbers of SG to distribute data in (from 1 to MAX_TAP)
+ * @tap: How to distribute data in @np SGs
+ * @also_non_np: if set to 1, the test will be also done without
+ * splitting data in @np SGs
+ */
+
struct cipher_testvec {
char *key;
char *iv;
--
2.3.6


2015-06-09 14:34:18

by Herbert Xu

[permalink] [raw]
Subject: Re: [PATCH RFC] crypto: testmgr - Document struct cipher_testvec

LABBE Corentin <[email protected]> wrote:
> Hello
>
> I was adding a test to testmgr for cbc(aes) but I hesitate on the cipher_testvec parameter also_non_np.
>
> Since its functions is not trivial, I propose a patch for documenting the struture.
>
> So does I have well understand the purpose of also_non_np ?
> And if yes what it is the interest of doing a test twice (one with and one without SG splitting) ?
>
> Regards
>
> From 0156e32be7dd7ffd0a4fdad8d0a3eaf3c0bf92c7 Mon Sep 17 00:00:00 2001
> From: LABBE Corentin <[email protected]>
> Date: Wed, 3 Jun 2015 09:09:13 +0200
> Subject: [PATCH 02/11] crypto: testmgr - Document struct cipher_testvec
>
> Signed-off-by: LABBE Corentin <[email protected]>

Applied. Please don't include nested headers like this without
using scissors (see git help am).

Thanks,
--
Email: Herbert Xu <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt