From: Iuliana Prodan <iuliana.prodan@nxp.com>
To: Herbert Xu <herbert@gondor.apana.org.au>,
Horia Geanta <horia.geanta@nxp.com>,
Aymen Sghaier <aymen.sghaier@nxp.com>
Cc: "David S. Miller" <davem@davemloft.net>,
linux-crypto@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-imx <linux-imx@nxp.com>
Subject: [PATCH v3 12/14] crypto: caam - execute module exit point only if necessary
Date: Thu, 25 Jul 2019 16:58:24 +0300 [thread overview]
Message-ID: <1564063106-9552-13-git-send-email-iuliana.prodan@nxp.com> (raw)
In-Reply-To: <1564063106-9552-1-git-send-email-iuliana.prodan@nxp.com>
Commit 1b46c90c8e00 ("crypto: caam - convert top level drivers to libraries")
changed entry and exit points behavior for caamalg,
caamalg_qi, caamalg_qi2, caamhash, caampkc, caamrng.
For example, previously caam_pkc_init() and caam_pkc_exit() were
module entry/exit points. This means that if an error would happen
in caam_pkc_init(), then caam_pkc_exit() wouldn't have been called.
After the mentioned commit, caam_pkc_init() and caam_pkc_exit()
are manually called - from jr.c. caam_pkc_exit() is called
unconditionally, even if caam_pkc_init() failed.
Added a global variable to keep the status of the algorithm
registration and free of resources.
The exit point of caampkc/caamrng module is executed only if the
registration was successful. Therefore we avoid double free of
resources in case the algorithm registration failed.
Fixes: 1b46c90c8e00 ("crypto: caam - convert top level drivers to libraries")
Signed-off-by: Iuliana Prodan <iuliana.prodan@nxp.com>
Reviewed-by: Horia Geanta <horia.geanta@nxp.com>
---
drivers/crypto/caam/caampkc.c | 11 +++++++++++
drivers/crypto/caam/caamrng.c | 14 +++++++++++++-
2 files changed, 24 insertions(+), 1 deletion(-)
diff --git a/drivers/crypto/caam/caampkc.c b/drivers/crypto/caam/caampkc.c
index 574428c7..cfdf7a2 100644
--- a/drivers/crypto/caam/caampkc.c
+++ b/drivers/crypto/caam/caampkc.c
@@ -29,6 +29,12 @@
/* buffer filled with zeros, used for padding */
static u8 *zero_buffer;
+/*
+ * variable used to avoid double free of resources in case
+ * algorithm registration was unsuccessful
+ */
+static bool init_done;
+
static void rsa_io_unmap(struct device *dev, struct rsa_edesc *edesc,
struct akcipher_request *req)
{
@@ -1081,6 +1087,7 @@ int caam_pkc_init(struct device *ctrldev)
struct caam_drv_private *priv = dev_get_drvdata(ctrldev);
u32 pk_inst;
int err;
+ init_done = false;
/* Determine public key hardware accelerator presence. */
if (priv->era < 10)
@@ -1105,6 +1112,7 @@ int caam_pkc_init(struct device *ctrldev)
dev_warn(ctrldev, "%s alg registration failed\n",
caam_rsa.base.cra_driver_name);
} else {
+ init_done = true;
dev_info(ctrldev, "caam pkc algorithms registered in /proc/crypto\n");
}
@@ -1113,6 +1121,9 @@ int caam_pkc_init(struct device *ctrldev)
void caam_pkc_exit(void)
{
+ if (!init_done)
+ return;
+
kfree(zero_buffer);
crypto_unregister_akcipher(&caam_rsa);
}
diff --git a/drivers/crypto/caam/caamrng.c b/drivers/crypto/caam/caamrng.c
index 54c32d5..7fbda1b 100644
--- a/drivers/crypto/caam/caamrng.c
+++ b/drivers/crypto/caam/caamrng.c
@@ -80,6 +80,12 @@ struct caam_rng_ctx {
static struct caam_rng_ctx *rng_ctx;
+/*
+ * Variable used to avoid double free of resources in case
+ * algorithm registration was unsuccessful
+ */
+static bool init_done;
+
static inline void rng_unmap_buf(struct device *jrdev, struct buf_data *bd)
{
if (bd->addr)
@@ -296,6 +302,9 @@ static struct hwrng caam_rng = {
void caam_rng_exit(void)
{
+ if (!init_done)
+ return;
+
caam_jr_free(rng_ctx->jrdev);
hwrng_unregister(&caam_rng);
kfree(rng_ctx);
@@ -307,6 +316,7 @@ int caam_rng_init(struct device *ctrldev)
u32 rng_inst;
struct caam_drv_private *priv = dev_get_drvdata(ctrldev);
int err;
+ init_done = false;
/* Check for an instantiated RNG before registration */
if (priv->era < 10)
@@ -335,8 +345,10 @@ int caam_rng_init(struct device *ctrldev)
dev_info(dev, "registering rng-caam\n");
err = hwrng_register(&caam_rng);
- if (!err)
+ if (!err) {
+ init_done = true;
return err;
+ }
free_rng_ctx:
kfree(rng_ctx);
--
2.1.0
next prev parent reply other threads:[~2019-07-25 13:59 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-07-25 13:58 [PATCH v3 00/14] crypto: caam - fixes for kernel v5.3 Iuliana Prodan
2019-07-25 13:58 ` [PATCH v3 01/14] crypto: caam/qi - fix error handling in ERN handler Iuliana Prodan
2019-07-25 13:58 ` [PATCH v3 02/14] crypto: caam - fix return code in completion callbacks Iuliana Prodan
2019-07-25 13:58 ` [PATCH v3 03/14] crypto: caam - update IV only when crypto operation succeeds Iuliana Prodan
2019-07-26 14:57 ` Horia Geanta
2019-07-25 13:58 ` [PATCH v3 04/14] crypto: caam - check key length Iuliana Prodan
2019-07-26 15:19 ` Horia Geanta
2019-07-25 13:58 ` [PATCH v3 05/14] crypto: caam - check authsize Iuliana Prodan
2019-07-26 15:25 ` Horia Geanta
2019-07-25 13:58 ` [PATCH v3 06/14] crypto: caam - check assoclen Iuliana Prodan
2019-07-26 15:28 ` Horia Geanta
2019-07-25 13:58 ` [PATCH v3 07/14] crypto: caam - check zero-length input Iuliana Prodan
2019-07-25 13:58 ` [PATCH v3 08/14] crypto: caam - update rfc4106 sh desc to support zero length input Iuliana Prodan
2019-07-26 15:44 ` Horia Geanta
2019-07-25 13:58 ` [PATCH v3 09/14] crypto: caam - keep both virtual and dma key addresses Iuliana Prodan
2019-07-25 13:58 ` [PATCH v3 10/14] crypto: caam - fix MDHA key derivation for certain user key lengths Iuliana Prodan
2019-07-26 15:04 ` Horia Geanta
2019-07-25 13:58 ` [PATCH v3 11/14] crypto: caam - free resources in case caam_rng registration failed Iuliana Prodan
2019-07-26 15:53 ` Horia Geanta
2019-07-25 13:58 ` Iuliana Prodan [this message]
2019-07-25 13:58 ` [PATCH v3 13/14] crypto: caam - unregister algorithm only if the registration succeeded Iuliana Prodan
2019-07-25 13:58 ` [PATCH v3 14/14] crypto: caam - change return value in case CAAM has no MDHA Iuliana Prodan
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1564063106-9552-13-git-send-email-iuliana.prodan@nxp.com \
--to=iuliana.prodan@nxp.com \
--cc=aymen.sghaier@nxp.com \
--cc=davem@davemloft.net \
--cc=herbert@gondor.apana.org.au \
--cc=horia.geanta@nxp.com \
--cc=linux-crypto@vger.kernel.org \
--cc=linux-imx@nxp.com \
--cc=linux-kernel@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).