<feed xmlns='http://www.w3.org/2005/Atom'>
<title>linux-toradex.git/drivers/nvme/common, branch master</title>
<subtitle>Linux kernel for Apalis and Colibri modules</subtitle>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/'/>
<entry>
<title>nvme-auth: use the parsed offset for the DH-CHAP key</title>
<updated>2026-11-16T15:15:02+00:00</updated>
<author>
<name>Shao Mingyin</name>
<email>shao.mingyin@zte.com.cn</email>
</author>
<published>2026-11-09T14:59:19+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=f7141618d5d0b51bc2a64961bab738a0fe57e83f'/>
<id>f7141618d5d0b51bc2a64961bab738a0fe57e83f</id>
<content type='text'>
nvme_auth_parse_key() accepts the DH-HMAC-CHAP key with

	if (sscanf(secret, "DHHC-1:%hhd:%*s:", &amp;key_hash) != 1)
		return -EINVAL;

and then skips a hardcoded ten characters:

	key = nvme_auth_extract_key(secret + 10, key_hash);

The ten characters assume that the hash identifier is exactly two
characters long.  A string such as "DHHC-1:1:" still makes sscanf()
return 1, and secret + 10 then points one byte past the NUL terminator
of the buffer.  nvme_auth_extract_key() starts with
strlen(secret), so it reads past the end of the allocation.  The sysfs
dhchap_secret and dhchap_ctrl_secret stores only require seven
characters and a "DHHC-1:" prefix, so a short value reaches this code
(even though the nvme-fabrics option parser rejects secrets shorter than
eleven characters).

Use the %n conversion to obtain the real offset after the
"DHHC-1:&lt;hash&gt;:" prefix and reject a string where the prefix is not
complete.  This accepts both one and two character hash identifiers and
never starts the key past the terminator.

Fixes: f50fff73d620 ("nvme: implement In-Band authentication")
Cc: stable@vger.kernel.org
Signed-off-by: Shao Mingyin &lt;shao.mingyin@zte.com.cn&gt;
Reviewed-by: Hannes Reinecke &lt;hare@kernel.org&gt;
Reviewed-by: Christoph Hellwig &lt;hch@lst.de&gt;
Signed-off-by: Keith Busch &lt;kbusch@kernel.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
nvme_auth_parse_key() accepts the DH-HMAC-CHAP key with

	if (sscanf(secret, "DHHC-1:%hhd:%*s:", &amp;key_hash) != 1)
		return -EINVAL;

and then skips a hardcoded ten characters:

	key = nvme_auth_extract_key(secret + 10, key_hash);

The ten characters assume that the hash identifier is exactly two
characters long.  A string such as "DHHC-1:1:" still makes sscanf()
return 1, and secret + 10 then points one byte past the NUL terminator
of the buffer.  nvme_auth_extract_key() starts with
strlen(secret), so it reads past the end of the allocation.  The sysfs
dhchap_secret and dhchap_ctrl_secret stores only require seven
characters and a "DHHC-1:" prefix, so a short value reaches this code
(even though the nvme-fabrics option parser rejects secrets shorter than
eleven characters).

Use the %n conversion to obtain the real offset after the
"DHHC-1:&lt;hash&gt;:" prefix and reject a string where the prefix is not
complete.  This accepts both one and two character hash identifiers and
never starts the key past the terminator.

Fixes: f50fff73d620 ("nvme: implement In-Band authentication")
Cc: stable@vger.kernel.org
Signed-off-by: Shao Mingyin &lt;shao.mingyin@zte.com.cn&gt;
Reviewed-by: Hannes Reinecke &lt;hare@kernel.org&gt;
Reviewed-by: Christoph Hellwig &lt;hch@lst.de&gt;
Signed-off-by: Keith Busch &lt;kbusch@kernel.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>nvme-keyring: Add a missing key_put() call to nvme_tls_key_lookup()</title>
<updated>2026-11-04T15:51:10+00:00</updated>
<author>
<name>Jarkko Sakkinen</name>
<email>jarkko@kernel.org</email>
</author>
<published>2026-11-03T03:02:07+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=404a8f6caa63822c1d2a45adbf07fd9c0567a7ef'/>
<id>404a8f6caa63822c1d2a45adbf07fd9c0567a7ef</id>
<content type='text'>
Call key_put() before returning -EKEYREVOKED, as the reference count gets
incremented by key_lookup().

Cc: stable@vger.kernel.org # v6.12+
Fixes: 5bc46b49c828 ("nvme-tcp: check for invalidated or revoked key")
Reviewed-by: Hannes Reinecke &lt;hare@kernel.org&gt;
Signed-off-by: Jarkko Sakkinen &lt;jarkko@kernel.org&gt;
Signed-off-by: Keith Busch &lt;kbusch@kernel.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Call key_put() before returning -EKEYREVOKED, as the reference count gets
incremented by key_lookup().

Cc: stable@vger.kernel.org # v6.12+
Fixes: 5bc46b49c828 ("nvme-tcp: check for invalidated or revoked key")
Reviewed-by: Hannes Reinecke &lt;hare@kernel.org&gt;
Signed-off-by: Jarkko Sakkinen &lt;jarkko@kernel.org&gt;
Signed-off-by: Keith Busch &lt;kbusch@kernel.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>nvme-auth: Avoid C=1 warning in nvme_auth_derive_tls_psk()</title>
<updated>2026-07-06T18:28:31+00:00</updated>
<author>
<name>Eric Biggers</name>
<email>ebiggers@kernel.org</email>
</author>
<published>2026-06-25T17:59:11+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=ba6e9472b453ccea313c7dad5bf2ad98d6ad13f5'/>
<id>ba6e9472b453ccea313c7dad5bf2ad98d6ad13f5</id>
<content type='text'>
The following works fine with gcc and clang, but sparse warns about
label_len not being an actual constant expression:

    const size_t label_len = sizeof(label) - 1;
    ...
    static_assert(label_len &lt;= 255);

Avoid this by giving label an explicit length and using sizeof(label)
instead of label_len.

Reported-by: John Garry &lt;john.g.garry@oracle.com&gt;
Closes: https://lore.kernel.org/linux-nvme/965a37dd-f698-46b6-9623-1099a13f7e60@oracle.com
Fixes: d126cbaa7d9a ("nvme-auth: common: use crypto library in nvme_auth_derive_tls_psk()")
Signed-off-by: Eric Biggers &lt;ebiggers@kernel.org&gt;
Reviewed-by: Hannes Reinecke &lt;hare@kernel.org&gt;
Reviewed-by: John Garry &lt;john.g.garry@oracle.com&gt;
Signed-off-by: Keith Busch &lt;kbusch@kernel.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
The following works fine with gcc and clang, but sparse warns about
label_len not being an actual constant expression:

    const size_t label_len = sizeof(label) - 1;
    ...
    static_assert(label_len &lt;= 255);

Avoid this by giving label an explicit length and using sizeof(label)
instead of label_len.

Reported-by: John Garry &lt;john.g.garry@oracle.com&gt;
Closes: https://lore.kernel.org/linux-nvme/965a37dd-f698-46b6-9623-1099a13f7e60@oracle.com
Fixes: d126cbaa7d9a ("nvme-auth: common: use crypto library in nvme_auth_derive_tls_psk()")
Signed-off-by: Eric Biggers &lt;ebiggers@kernel.org&gt;
Reviewed-by: Hannes Reinecke &lt;hare@kernel.org&gt;
Reviewed-by: John Garry &lt;john.g.garry@oracle.com&gt;
Signed-off-by: Keith Busch &lt;kbusch@kernel.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>nvme-auth: Hash DH shared secret to create session key</title>
<updated>2026-04-22T20:02:16+00:00</updated>
<author>
<name>Chris Leech</name>
<email>cleech@redhat.com</email>
</author>
<published>2026-04-22T19:06:36+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=bd7b7ce96db4487bb77692a85ee4489fd2c395df'/>
<id>bd7b7ce96db4487bb77692a85ee4489fd2c395df</id>
<content type='text'>
The NVMe Base Specification 8.3.5.5.9 states that the session key Ks
shall be computed from the ephemeral DH key by applying the hash
function selected by the HashID parameter.

The current implementation stores the raw DH shared secret as the
session key without hashing it. This causes redundant hash operations:

1. Augmented challenge computation (section 8.3.5.5.4) requires
   Ca = HMAC(H(g^xy mod p), C). The code compensates by hashing the
   unhashed session key in nvme_auth_augmented_challenge() to produce
   the correct result.

2. PSK generation (section 8.3.5.5.9) requires PSK = HMAC(Ks, C1 || C2)
   where Ks should already be H(g^xy mod p). As the DH shared secret
   is always larger than the HMAC block size, HMAC internally hashes
   it before use, accidentally producing the correct result.

When using secure channel concatenation with bidirectional
authentication, this results in hashing the DH value three times: twice
for augmented challenge calculations and once during PSK generation.

Fix this by:
- Modifying nvme_auth_gen_shared_secret() to hash the DH shared secret
  once after computation: Ks = H(g^xy mod p)
- Removing the hash operation from nvme_auth_augmented_challenge()
  as the session key is now already hashed
- Updating session key buffer size from DH key size to hash output size
- Adding specification references in comments

This avoid storing the raw DH shared secret and reduces the number of
hash operations from three to one when using secure channel
concatenation.

Reviewed-by: Hannes Reinecke &lt;hare@suse.de&gt;
Reviewed-by: Eric Biggers &lt;ebiggers@kernel.org&gt;
Signed-off-by: Chris Leech &lt;cleech@redhat.com&gt;
Signed-off-by: Keith Busch &lt;kbusch@kernel.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
The NVMe Base Specification 8.3.5.5.9 states that the session key Ks
shall be computed from the ephemeral DH key by applying the hash
function selected by the HashID parameter.

The current implementation stores the raw DH shared secret as the
session key without hashing it. This causes redundant hash operations:

1. Augmented challenge computation (section 8.3.5.5.4) requires
   Ca = HMAC(H(g^xy mod p), C). The code compensates by hashing the
   unhashed session key in nvme_auth_augmented_challenge() to produce
   the correct result.

2. PSK generation (section 8.3.5.5.9) requires PSK = HMAC(Ks, C1 || C2)
   where Ks should already be H(g^xy mod p). As the DH shared secret
   is always larger than the HMAC block size, HMAC internally hashes
   it before use, accidentally producing the correct result.

When using secure channel concatenation with bidirectional
authentication, this results in hashing the DH value three times: twice
for augmented challenge calculations and once during PSK generation.

Fix this by:
- Modifying nvme_auth_gen_shared_secret() to hash the DH shared secret
  once after computation: Ks = H(g^xy mod p)
- Removing the hash operation from nvme_auth_augmented_challenge()
  as the session key is now already hashed
- Updating session key buffer size from DH key size to hash output size
- Adding specification references in comments

This avoid storing the raw DH shared secret and reduces the number of
hash operations from three to one when using secure channel
concatenation.

Reviewed-by: Hannes Reinecke &lt;hare@suse.de&gt;
Reviewed-by: Eric Biggers &lt;ebiggers@kernel.org&gt;
Signed-off-by: Chris Leech &lt;cleech@redhat.com&gt;
Signed-off-by: Keith Busch &lt;kbusch@kernel.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>nvme-auth: common: remove selections of no-longer used crypto modules</title>
<updated>2026-03-27T14:35:03+00:00</updated>
<author>
<name>Eric Biggers</name>
<email>ebiggers@kernel.org</email>
</author>
<published>2026-03-02T07:59:58+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=26c8c2dddecb016a6115b30cf2cee633b311222b'/>
<id>26c8c2dddecb016a6115b30cf2cee633b311222b</id>
<content type='text'>
Now that nvme-auth uses the crypto library instead of crypto_shash,
remove obsolete selections from the NVME_AUTH kconfig option.

Acked-by: Ard Biesheuvel &lt;ardb@kernel.org&gt;
Acked-by: Christoph Hellwig &lt;hch@lst.de&gt;
Reviewed-by: Hannes Reinecke &lt;hare@suse.de&gt;
Signed-off-by: Eric Biggers &lt;ebiggers@kernel.org&gt;
Signed-off-by: Keith Busch &lt;kbusch@kernel.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Now that nvme-auth uses the crypto library instead of crypto_shash,
remove obsolete selections from the NVME_AUTH kconfig option.

Acked-by: Ard Biesheuvel &lt;ardb@kernel.org&gt;
Acked-by: Christoph Hellwig &lt;hch@lst.de&gt;
Reviewed-by: Hannes Reinecke &lt;hare@suse.de&gt;
Signed-off-by: Eric Biggers &lt;ebiggers@kernel.org&gt;
Signed-off-by: Keith Busch &lt;kbusch@kernel.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>nvme-auth: common: remove nvme_auth_digest_name()</title>
<updated>2026-03-27T14:35:02+00:00</updated>
<author>
<name>Eric Biggers</name>
<email>ebiggers@kernel.org</email>
</author>
<published>2026-03-02T07:59:57+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=844d950bb2cb1fc5b8973369de59cbfb7eecd94d'/>
<id>844d950bb2cb1fc5b8973369de59cbfb7eecd94d</id>
<content type='text'>
Since nvme_auth_digest_name() is no longer used, remove it and the
associated data from the hash_map array.

Acked-by: Ard Biesheuvel &lt;ardb@kernel.org&gt;
Acked-by: Christoph Hellwig &lt;hch@lst.de&gt;
Reviewed-by: Hannes Reinecke &lt;hare@suse.de&gt;
Signed-off-by: Eric Biggers &lt;ebiggers@kernel.org&gt;
Signed-off-by: Keith Busch &lt;kbusch@kernel.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Since nvme_auth_digest_name() is no longer used, remove it and the
associated data from the hash_map array.

Acked-by: Ard Biesheuvel &lt;ardb@kernel.org&gt;
Acked-by: Christoph Hellwig &lt;hch@lst.de&gt;
Reviewed-by: Hannes Reinecke &lt;hare@suse.de&gt;
Signed-off-by: Eric Biggers &lt;ebiggers@kernel.org&gt;
Signed-off-by: Keith Busch &lt;kbusch@kernel.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>nvme-auth: common: use crypto library in nvme_auth_derive_tls_psk()</title>
<updated>2026-03-27T14:35:01+00:00</updated>
<author>
<name>Eric Biggers</name>
<email>ebiggers@kernel.org</email>
</author>
<published>2026-03-02T07:59:50+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=d126cbaa7d9a971dedc8535d4f2529c799de8f85'/>
<id>d126cbaa7d9a971dedc8535d4f2529c799de8f85</id>
<content type='text'>
For the HKDF-Expand-Label computation in nvme_auth_derive_tls_psk(), use
the crypto library instead of crypto_shash and crypto/hkdf.c.

While this means the HKDF "helper" functions are no longer utilized,
they clearly weren't buying us much: it's simpler to just inline the
HMAC computations directly, and this code needs to be tested anyway.  (A
similar result was seen in fs/crypto/.  As a result, this eliminates the
last user of crypto/hkdf.c, which we'll be able to remove as well.)

As usual this is also a lot more efficient, eliminating the allocation
of a transformation object and multiple other dynamic allocations.

Acked-by: Ard Biesheuvel &lt;ardb@kernel.org&gt;
Acked-by: Christoph Hellwig &lt;hch@lst.de&gt;
Reviewed-by: Hannes Reinecke &lt;hare@suse.de&gt;
Signed-off-by: Eric Biggers &lt;ebiggers@kernel.org&gt;
Signed-off-by: Keith Busch &lt;kbusch@kernel.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
For the HKDF-Expand-Label computation in nvme_auth_derive_tls_psk(), use
the crypto library instead of crypto_shash and crypto/hkdf.c.

While this means the HKDF "helper" functions are no longer utilized,
they clearly weren't buying us much: it's simpler to just inline the
HMAC computations directly, and this code needs to be tested anyway.  (A
similar result was seen in fs/crypto/.  As a result, this eliminates the
last user of crypto/hkdf.c, which we'll be able to remove as well.)

As usual this is also a lot more efficient, eliminating the allocation
of a transformation object and multiple other dynamic allocations.

Acked-by: Ard Biesheuvel &lt;ardb@kernel.org&gt;
Acked-by: Christoph Hellwig &lt;hch@lst.de&gt;
Reviewed-by: Hannes Reinecke &lt;hare@suse.de&gt;
Signed-off-by: Eric Biggers &lt;ebiggers@kernel.org&gt;
Signed-off-by: Keith Busch &lt;kbusch@kernel.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>nvme-auth: common: use crypto library in nvme_auth_generate_digest()</title>
<updated>2026-03-27T14:35:01+00:00</updated>
<author>
<name>Eric Biggers</name>
<email>ebiggers@kernel.org</email>
</author>
<published>2026-03-02T07:59:49+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=0002764c2faa769cd41b45e95af3dd5e1777df9f'/>
<id>0002764c2faa769cd41b45e95af3dd5e1777df9f</id>
<content type='text'>
For the HMAC computation in nvme_auth_generate_digest(), use the crypto
library instead of crypto_shash.  This is simpler, faster, and more
reliable.  Notably, this eliminates the crypto transformation object
allocation for every call, which was very slow.

Acked-by: Ard Biesheuvel &lt;ardb@kernel.org&gt;
Acked-by: Christoph Hellwig &lt;hch@lst.de&gt;
Reviewed-by: Hannes Reinecke &lt;hare@suse.de&gt;
Signed-off-by: Eric Biggers &lt;ebiggers@kernel.org&gt;
Signed-off-by: Keith Busch &lt;kbusch@kernel.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
For the HMAC computation in nvme_auth_generate_digest(), use the crypto
library instead of crypto_shash.  This is simpler, faster, and more
reliable.  Notably, this eliminates the crypto transformation object
allocation for every call, which was very slow.

Acked-by: Ard Biesheuvel &lt;ardb@kernel.org&gt;
Acked-by: Christoph Hellwig &lt;hch@lst.de&gt;
Reviewed-by: Hannes Reinecke &lt;hare@suse.de&gt;
Signed-off-by: Eric Biggers &lt;ebiggers@kernel.org&gt;
Signed-off-by: Keith Busch &lt;kbusch@kernel.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>nvme-auth: common: use crypto library in nvme_auth_generate_psk()</title>
<updated>2026-03-27T14:35:01+00:00</updated>
<author>
<name>Eric Biggers</name>
<email>ebiggers@kernel.org</email>
</author>
<published>2026-03-02T07:59:48+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=be01b841d3dd667d873cbcd984d9839b7e98ef4f'/>
<id>be01b841d3dd667d873cbcd984d9839b7e98ef4f</id>
<content type='text'>
For the HMAC computation in nvme_auth_generate_psk(), use the crypto
library instead of crypto_shash.  This is simpler, faster, and more
reliable.  Notably, this eliminates the crypto transformation object
allocation for every call, which was very slow.

Acked-by: Ard Biesheuvel &lt;ardb@kernel.org&gt;
Acked-by: Christoph Hellwig &lt;hch@lst.de&gt;
Reviewed-by: Hannes Reinecke &lt;hare@suse.de&gt;
Signed-off-by: Eric Biggers &lt;ebiggers@kernel.org&gt;
Signed-off-by: Keith Busch &lt;kbusch@kernel.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
For the HMAC computation in nvme_auth_generate_psk(), use the crypto
library instead of crypto_shash.  This is simpler, faster, and more
reliable.  Notably, this eliminates the crypto transformation object
allocation for every call, which was very slow.

Acked-by: Ard Biesheuvel &lt;ardb@kernel.org&gt;
Acked-by: Christoph Hellwig &lt;hch@lst.de&gt;
Reviewed-by: Hannes Reinecke &lt;hare@suse.de&gt;
Signed-off-by: Eric Biggers &lt;ebiggers@kernel.org&gt;
Signed-off-by: Keith Busch &lt;kbusch@kernel.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>nvme-auth: common: use crypto library in nvme_auth_augmented_challenge()</title>
<updated>2026-03-27T14:35:01+00:00</updated>
<author>
<name>Eric Biggers</name>
<email>ebiggers@kernel.org</email>
</author>
<published>2026-03-02T07:59:47+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=a67d096fe9761e3e503f40643228bca6d69c7c4e'/>
<id>a67d096fe9761e3e503f40643228bca6d69c7c4e</id>
<content type='text'>
For the hash and HMAC computations in nvme_auth_augmented_challenge(),
use the crypto library instead of crypto_shash.  This is simpler,
faster, and more reliable.  Notably, this eliminates two crypto
transformation object allocations for every call, which was very slow.

Acked-by: Ard Biesheuvel &lt;ardb@kernel.org&gt;
Acked-by: Christoph Hellwig &lt;hch@lst.de&gt;
Reviewed-by: Hannes Reinecke &lt;hare@suse.de&gt;
Signed-off-by: Eric Biggers &lt;ebiggers@kernel.org&gt;
Signed-off-by: Keith Busch &lt;kbusch@kernel.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
For the hash and HMAC computations in nvme_auth_augmented_challenge(),
use the crypto library instead of crypto_shash.  This is simpler,
faster, and more reliable.  Notably, this eliminates two crypto
transformation object allocations for every call, which was very slow.

Acked-by: Ard Biesheuvel &lt;ardb@kernel.org&gt;
Acked-by: Christoph Hellwig &lt;hch@lst.de&gt;
Reviewed-by: Hannes Reinecke &lt;hare@suse.de&gt;
Signed-off-by: Eric Biggers &lt;ebiggers@kernel.org&gt;
Signed-off-by: Keith Busch &lt;kbusch@kernel.org&gt;
</pre>
</div>
</content>
</entry>
</feed>
