Thread (3 messages) 3 messages, 2 authors, 2018-01-31

Re: [PATCH] tls: Add support for encryption using async offload accelerator

From: Gilad Ben-Yossef <gilad@benyossef.com>
Date: 2018-01-31 06:10:17
Also in: linux-crypto

Hi Vakul,

On Wed, Jan 31, 2018 at 12:36 PM, Vakul Garg [off-list ref] wrote:
Async crypto accelerators (e.g. drivers/crypto/caam) support offloading
GCM operation. If they are enabled, crypto_aead_encrypt() return error
code -EINPROGRESS. In this case tls_do_encryption() needs to wait on a
completion till the time the response for crypto offload request is
received.
Thank you for this patch. I think it is actually a bug fix and should
probably go into stable
since I believe current code is broken on machines where any async GCM
transformation
is registered with a high enough priority as it will try to call a
NULL callback.

But, please use the new crypto_wait_req() and friends to implement the waiting.
Take a look at the code example in
Documentation/crypto/api-samples.rst to see how.

Also, maybe add an explanation why  CRYPTO_TFM_REQ_MAY_BACKLOG
flag should not be used here, if this was on purpose?

I mean I'm not sure using backlog here is a good idea but probably
should explain why
not.

Thanks!
Gilad

quoted hunk ↗ jump to hunk
Signed-off-by: Vakul Garg <redacted>
---
 net/tls/tls_sw.c | 22 ++++++++++++++++++++++
 1 file changed, 22 insertions(+)
diff --git a/net/tls/tls_sw.c b/net/tls/tls_sw.c
index 73d19210dd49..390e6dc7b135 100644
--- a/net/tls/tls_sw.c
+++ b/net/tls/tls_sw.c
@@ -39,6 +39,11 @@

 #include <net/tls.h>

+struct crypto_completion {
+       struct completion comp;
+       int result;
+};
+
 static void trim_sg(struct sock *sk, struct scatterlist *sg,
                    int *sg_num_elem, unsigned int *sg_size, int target_size)
 {
@@ -194,6 +199,14 @@ static void tls_free_both_sg(struct sock *sk)
                &ctx->sg_plaintext_size);
 }

+static void crypto_complete(struct crypto_async_request *req, int err)
+{
+       struct crypto_completion *x = req->data;
+
+       x->result = err;
+       complete(&x->comp);
+}
+
 static int tls_do_encryption(struct tls_context *tls_ctx,
                             struct tls_sw_context *ctx, size_t data_len,
                             gfp_t flags)
@@ -202,6 +215,7 @@ static int tls_do_encryption(struct tls_context *tls_ctx,
                crypto_aead_reqsize(ctx->aead_send);
        struct aead_request *aead_req;
        int rc;
+       struct crypto_completion crypto_done;

        aead_req = kzalloc(req_size, flags);
        if (!aead_req)
@@ -214,7 +228,15 @@ static int tls_do_encryption(struct tls_context *tls_ctx,
        aead_request_set_ad(aead_req, TLS_AAD_SPACE_SIZE);
        aead_request_set_crypt(aead_req, ctx->sg_aead_in, ctx->sg_aead_out,
                               data_len, tls_ctx->iv);
+       aead_request_set_callback(aead_req, 0, crypto_complete, &crypto_done);
+
+       init_completion(&crypto_done.comp);
+
        rc = crypto_aead_encrypt(aead_req);
+       if (rc == -EINPROGRESS) {
+               wait_for_completion(&crypto_done.comp);
+               rc = crypto_done.result;
+       }

        ctx->sg_encrypted_data[0].offset -= tls_ctx->prepend_size;
        ctx->sg_encrypted_data[0].length += tls_ctx->prepend_size;
--
2.13.6


-- 
Gilad Ben-Yossef
Chief Coffee Drinker

"If you take a class in large-scale robotics, can you end up in a
situation where the homework eats your dog?"
 -- Jean-Baptiste Queru
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help