From: Jack Bond-Preston <jack.bond-preston@foss.arm.com>
To: Kai Ji <kai.ji@intel.com>, Fan Zhang <fanzhang.oss@gmail.com>,
Akhil Goyal <gakhil@marvell.com>
Cc: dev@dpdk.org, stable@dpdk.org,
Wathsala Vithanage <wathsala.vithanage@arm.com>
Subject: [PATCH 1/5] crypto/openssl: fix GCM and CCM thread unsafe ctxs
Date: Mon, 3 Jun 2024 16:01:15 +0000 [thread overview]
Message-ID: <20240603160119.1279476-2-jack.bond-preston@foss.arm.com> (raw)
In-Reply-To: <20240603160119.1279476-1-jack.bond-preston@foss.arm.com>
Commit 67ab783b5d70 ("crypto/openssl: use local copy for session
contexts") introduced a fix for concurrency bugs which could occur when
using one OpenSSL PMD session across multiple cores simultaneously. The
solution was to clone the EVP contexts per-buffer to avoid them being
used concurrently.
However, part of commit 75adf1eae44f ("crypto/openssl: update HMAC
routine with 3.0 EVP API") reverted this fix, only for combined ops
(AES-GCM and AES-CCM), with no explanation. This commit fixes the issue
again, essentially reverting this part of the commit.
Throughput performance uplift measurements for AES-GCM-128 encrypt on
Ampere Altra Max platform:
1 worker lcore
| buffer sz (B) | prev (Gbps) | optimised (Gbps) | uplift |
|-----------------+---------------+--------------------+----------|
| 64 | 2.60 | 1.31 | -49.5% |
| 256 | 7.69 | 4.45 | -42.1% |
| 1024 | 15.33 | 11.30 | -26.3% |
| 2048 | 18.74 | 15.37 | -18.0% |
| 4096 | 21.11 | 18.80 | -10.9% |
8 worker lcores
| buffer sz (B) | prev (Gbps) | optimised (Gbps) | uplift |
|-----------------+---------------+--------------------+----------|
| 64 | 19.94 | 2.83 | -85.8% |
| 256 | 58.84 | 11.00 | -81.3% |
| 1024 | 119.71 | 42.46 | -64.5% |
| 2048 | 147.69 | 80.91 | -45.2% |
| 4096 | 167.39 | 121.25 | -27.6% |
Fixes: 75adf1eae44f ("crypto/openssl: update HMAC routine with 3.0 EVP API")
Cc: stable@dpdk.org
Signed-off-by: Jack Bond-Preston <jack.bond-preston@foss.arm.com>
Reviewed-by: Wathsala Vithanage <wathsala.vithanage@arm.com>
---
drivers/crypto/openssl/rte_openssl_pmd.c | 13 +++++++++----
1 file changed, 9 insertions(+), 4 deletions(-)
diff --git a/drivers/crypto/openssl/rte_openssl_pmd.c b/drivers/crypto/openssl/rte_openssl_pmd.c
index e8cb09defc..ca7ed30ec4 100644
--- a/drivers/crypto/openssl/rte_openssl_pmd.c
+++ b/drivers/crypto/openssl/rte_openssl_pmd.c
@@ -1590,6 +1590,9 @@ process_openssl_combined_op
return;
}
+ EVP_CIPHER_CTX *ctx = EVP_CIPHER_CTX_new();
+ EVP_CIPHER_CTX_copy(ctx, sess->cipher.ctx);
+
iv = rte_crypto_op_ctod_offset(op, uint8_t *,
sess->iv.offset);
if (sess->auth.algo == RTE_CRYPTO_AUTH_AES_GMAC) {
@@ -1623,12 +1626,12 @@ process_openssl_combined_op
status = process_openssl_auth_encryption_gcm(
mbuf_src, offset, srclen,
aad, aadlen, iv,
- dst, tag, sess->cipher.ctx);
+ dst, tag, ctx);
else
status = process_openssl_auth_encryption_ccm(
mbuf_src, offset, srclen,
aad, aadlen, iv,
- dst, tag, taglen, sess->cipher.ctx);
+ dst, tag, taglen, ctx);
} else {
if (sess->auth.algo == RTE_CRYPTO_AUTH_AES_GMAC ||
@@ -1636,14 +1639,16 @@ process_openssl_combined_op
status = process_openssl_auth_decryption_gcm(
mbuf_src, offset, srclen,
aad, aadlen, iv,
- dst, tag, sess->cipher.ctx);
+ dst, tag, ctx);
else
status = process_openssl_auth_decryption_ccm(
mbuf_src, offset, srclen,
aad, aadlen, iv,
- dst, tag, taglen, sess->cipher.ctx);
+ dst, tag, taglen, ctx);
}
+ EVP_CIPHER_CTX_free(ctx);
+
if (status != 0) {
if (status == (-EFAULT) &&
sess->auth.operation ==
--
2.34.1
next prev parent reply other threads:[~2024-06-03 16:01 UTC|newest]
Thread overview: 49+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-06-03 16:01 [PATCH 0/5] OpenSSL PMD Optimisations Jack Bond-Preston
2024-06-03 16:01 ` Jack Bond-Preston [this message]
2024-06-03 16:12 ` [PATCH 1/5] crypto/openssl: fix GCM and CCM thread unsafe ctxs Jack Bond-Preston
2024-06-03 16:01 ` [PATCH 2/5] crypto/openssl: only init 3DES-CTR key + impl once Jack Bond-Preston
2024-06-03 16:01 ` [PATCH 3/5] crypto/openssl: per-qp cipher context clones Jack Bond-Preston
2024-06-03 16:01 ` [PATCH 4/5] crypto/openssl: per-qp auth " Jack Bond-Preston
2024-06-03 16:30 ` Jack Bond-Preston
2024-06-03 16:01 ` [PATCH 5/5] crypto/openssl: only set cipher padding once Jack Bond-Preston
2024-06-03 18:43 ` [PATCH v2 0/5] OpenSSL PMD Optimisations Jack Bond-Preston
2024-06-03 18:43 ` [PATCH v2 1/5] crypto/openssl: fix GCM and CCM thread unsafe ctxs Jack Bond-Preston
2024-06-03 18:43 ` [PATCH v2 2/5] crypto/openssl: only init 3DES-CTR key + impl once Jack Bond-Preston
2024-06-03 18:43 ` [PATCH v2 3/5] crypto/openssl: per-qp cipher context clones Jack Bond-Preston
2024-06-03 18:43 ` [PATCH v2 4/5] crypto/openssl: per-qp auth " Jack Bond-Preston
2024-06-03 18:43 ` [PATCH v2 5/5] crypto/openssl: only set cipher padding once Jack Bond-Preston
2024-06-03 18:59 ` [PATCH v2 0/5] OpenSSL PMD Optimisations Jack Bond-Preston
2024-06-03 18:59 ` [PATCH v2 1/5] crypto/openssl: fix GCM and CCM thread unsafe ctxs Jack Bond-Preston
2024-06-03 18:59 ` [PATCH v2 2/5] crypto/openssl: only init 3DES-CTR key + impl once Jack Bond-Preston
2024-06-03 18:59 ` [PATCH v2 3/5] crypto/openssl: per-qp cipher context clones Jack Bond-Preston
2024-06-03 18:59 ` [PATCH v2 4/5] crypto/openssl: per-qp auth " Jack Bond-Preston
2024-06-03 18:59 ` [PATCH v2 5/5] crypto/openssl: only set cipher padding once Jack Bond-Preston
2024-06-06 10:20 ` [PATCH v3 0/5] OpenSSL PMD Optimisations Jack Bond-Preston
2024-06-06 10:20 ` [PATCH v3 1/5] crypto/openssl: fix GCM and CCM thread unsafe ctxs Jack Bond-Preston
2024-06-06 10:44 ` [EXTERNAL] " Akhil Goyal
2024-06-06 10:20 ` [PATCH v3 2/5] crypto/openssl: only init 3DES-CTR key + impl once Jack Bond-Preston
2024-06-06 10:20 ` [PATCH v3 3/5] crypto/openssl: per-qp cipher context clones Jack Bond-Preston
2024-06-06 10:20 ` [PATCH v3 4/5] crypto/openssl: per-qp auth " Jack Bond-Preston
2024-06-06 10:20 ` [PATCH v3 5/5] crypto/openssl: only set cipher padding once Jack Bond-Preston
2024-06-07 12:47 ` [PATCH v4 0/5] OpenSSL PMD Optimisations Jack Bond-Preston
2024-06-07 12:47 ` [PATCH v4 1/5] crypto/openssl: fix GCM and CCM thread unsafe ctxs Jack Bond-Preston
2024-07-01 12:55 ` Ji, Kai
2024-07-02 15:39 ` [EXTERNAL] " Akhil Goyal
2024-07-03 10:49 ` Jack Bond-Preston
2024-07-03 13:46 ` Jack Bond-Preston
2024-06-07 12:47 ` [PATCH v4 2/5] crypto/openssl: only init 3DES-CTR key + impl once Jack Bond-Preston
2024-07-01 12:55 ` Ji, Kai
2024-06-07 12:47 ` [PATCH v4 3/5] crypto/openssl: per-qp cipher context clones Jack Bond-Preston
2024-07-01 12:56 ` Ji, Kai
2024-06-07 12:47 ` [PATCH v4 4/5] crypto/openssl: per-qp auth " Jack Bond-Preston
2024-07-01 12:55 ` Ji, Kai
2024-06-07 12:47 ` [PATCH v4 5/5] crypto/openssl: only set cipher padding once Jack Bond-Preston
2024-07-01 12:56 ` Ji, Kai
2024-06-24 16:14 ` [PATCH 0/5] OpenSSL PMD Optimisations Ji, Kai
2024-07-03 13:45 ` [PATCH v5 " Jack Bond-Preston
2024-07-03 13:45 ` [PATCH v5 1/5] crypto/openssl: fix GCM and CCM thread unsafe ctxs Jack Bond-Preston
2024-07-03 15:20 ` [EXTERNAL] " Akhil Goyal
2024-07-03 13:45 ` [PATCH v5 2/5] crypto/openssl: only init 3DES-CTR key + impl once Jack Bond-Preston
2024-07-03 13:45 ` [PATCH v5 3/5] crypto/openssl: per-qp cipher context clones Jack Bond-Preston
2024-07-03 13:45 ` [PATCH v5 4/5] crypto/openssl: per-qp auth " Jack Bond-Preston
2024-07-03 13:45 ` [PATCH v5 5/5] crypto/openssl: only set cipher padding once Jack Bond-Preston
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=20240603160119.1279476-2-jack.bond-preston@foss.arm.com \
--to=jack.bond-preston@foss.arm.com \
--cc=dev@dpdk.org \
--cc=fanzhang.oss@gmail.com \
--cc=gakhil@marvell.com \
--cc=kai.ji@intel.com \
--cc=stable@dpdk.org \
--cc=wathsala.vithanage@arm.com \
/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).