[PATCH v2 0/4] s390/crypto: Use -ENODEV instead of -EOPNOTSUPP

STALE2678d SUPERSEDED

Revision v2 of 2 in this series.

15 messages, 3 authors, 2019-06-12 · open the first message on its own page

[PATCH v2 0/4] s390/crypto: Use -ENODEV instead of -EOPNOTSUPP

From: David Hildenbrand <hidden>
Date: 2019-06-12 10:22:53

s390x crypto is one of the rare modules that returns -EOPNOTSUPP instead of
-ENODEV in case HW support is not available.

Convert to -ENODEV, so e.g., systemd's systemd-modules-load.service
ignores this error properly.

v1 -> v2:
- Include
-- "s390/crypto: ghash: Use -ENODEV instead of -EOPNOTSUPP"
-- "s390/crypto: prng: Use -ENODEV instead of -EOPNOTSUPP"
-- "s390/crypto: sha: Use -ENODEV instead of -EOPNOTSUPP"

Cc: Herbert Xu <herbert@gondor.apana.org.au>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Heiko Carstens <redacted>
Cc: Vasily Gorbik <gor@linux.ibm.com>
Cc: Christian Borntraeger <redacted>
Cc: Harald Freudenberger <freude@linux.ibm.com>
Cc: Cornelia Huck <cohuck@redhat.com>

David Hildenbrand (4):
  s390/pkey: Use -ENODEV instead of -EOPNOTSUPP
  s390/crypto: ghash: Use -ENODEV instead of -EOPNOTSUPP
  s390/crypto: prng: Use -ENODEV instead of -EOPNOTSUPP
  s390/crypto: sha: Use -ENODEV instead of -EOPNOTSUPP

 arch/s390/crypto/ghash_s390.c  | 2 +-
 arch/s390/crypto/prng.c        | 4 ++--
 arch/s390/crypto/sha1_s390.c   | 2 +-
 arch/s390/crypto/sha256_s390.c | 2 +-
 arch/s390/crypto/sha512_s390.c | 2 +-
 drivers/s390/crypto/pkey_api.c | 6 +++---
 6 files changed, 9 insertions(+), 9 deletions(-)

-- 
2.21.0

[PATCH v2 1/4] s390/pkey: Use -ENODEV instead of -EOPNOTSUPP

From: David Hildenbrand <hidden>
Date: 2019-06-12 10:22:57

systemd-modules-load.service automatically tries to load the pkey module
on systems that have MSA.

Pkey also requires the MSA3 facility and a bunch of subfunctions.
Failing with -EOPNOTSUPP makes "systemd-modules-load.service" fail on
any system that does not have all needed subfunctions. For example,
when running under QEMU TCG (but also on systems where protected keys
are disabled via the HMC).

Let's use -ENODEV, so systemd-modules-load.service properly ignores
failing to load the pkey module because of missing HW functionality.

Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: David Hildenbrand <redacted>
---
 drivers/s390/crypto/pkey_api.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/s390/crypto/pkey_api.c b/drivers/s390/crypto/pkey_api.c
index 45eb0c14b880..ddfcefb47284 100644
--- a/drivers/s390/crypto/pkey_api.c
+++ b/drivers/s390/crypto/pkey_api.c

@@ -1695,15 +1695,15 @@ static int __init pkey_init(void)
 	 * are able to work with protected keys.
 	 */
 	if (!cpacf_query(CPACF_PCKMO, &pckmo_functions))
-		return -EOPNOTSUPP;
+		return -ENODEV;
 
 	/* check for kmc instructions available */
 	if (!cpacf_query(CPACF_KMC, &kmc_functions))
-		return -EOPNOTSUPP;
+		return -ENODEV;
 	if (!cpacf_test_func(&kmc_functions, CPACF_KMC_PAES_128) ||
 	    !cpacf_test_func(&kmc_functions, CPACF_KMC_PAES_192) ||
 	    !cpacf_test_func(&kmc_functions, CPACF_KMC_PAES_256))
-		return -EOPNOTSUPP;
+		return -ENODEV;
 
 	pkey_debug_init();

-- 
2.21.0

[PATCH v2 3/4] s390/crypto: prng: Use -ENODEV instead of -EOPNOTSUPP

From: David Hildenbrand <hidden>
Date: 2019-06-12 10:23:04

Let's use the error value that is typically used if HW support is not
available when trying to load a module - this is also what systemd's
systemd-modules-load.service expects.

Signed-off-by: David Hildenbrand <redacted>
---
 arch/s390/crypto/prng.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/s390/crypto/prng.c b/arch/s390/crypto/prng.c
index 12cca467af7d..d977643fa627 100644
--- a/arch/s390/crypto/prng.c
+++ b/arch/s390/crypto/prng.c

@@ -824,7 +824,7 @@ static int __init prng_init(void)
 
 	/* check if the CPU has a PRNG */
 	if (!cpacf_query_func(CPACF_KMC, CPACF_KMC_PRNG))
-		return -EOPNOTSUPP;
+		return -ENODEV;
 
 	/* check if TRNG subfunction is available */
 	if (cpacf_query_func(CPACF_PRNO, CPACF_PRNO_TRNG))

@@ -837,7 +837,7 @@ static int __init prng_init(void)
 			if (prng_mode == PRNG_MODE_SHA512) {
 				pr_err("The prng module cannot "
 				       "start in SHA-512 mode\n");
-				return -EOPNOTSUPP;
+				return -ENODEV;
 			}
 			prng_mode = PRNG_MODE_TDES;
 		} else

-- 
2.21.0

[PATCH v2 2/4] s390/crypto: ghash: Use -ENODEV instead of -EOPNOTSUPP

From: David Hildenbrand <hidden>
Date: 2019-06-12 10:23:06

Let's use the error value that is typically used if HW support is not
available when trying to load a module - this is also what systemd's
systemd-modules-load.service expects.

Signed-off-by: David Hildenbrand <redacted>
---
 arch/s390/crypto/ghash_s390.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/s390/crypto/ghash_s390.c b/arch/s390/crypto/ghash_s390.c
index 86aed30fad3a..eeeb6a7737a4 100644
--- a/arch/s390/crypto/ghash_s390.c
+++ b/arch/s390/crypto/ghash_s390.c

@@ -137,7 +137,7 @@ static struct shash_alg ghash_alg = {
 static int __init ghash_mod_init(void)
 {
 	if (!cpacf_query_func(CPACF_KIMD, CPACF_KIMD_GHASH))
-		return -EOPNOTSUPP;
+		return -ENODEV;
 
 	return crypto_register_shash(&ghash_alg);
 }

-- 
2.21.0

[PATCH v2 4/4] s390/crypto: sha: Use -ENODEV instead of -EOPNOTSUPP

From: David Hildenbrand <hidden>
Date: 2019-06-12 10:23:10

Let's use the error value that is typically used if HW support is not
available when trying to load a module - this is also what systemd's
systemd-modules-load.service expects.

Signed-off-by: David Hildenbrand <redacted>
---
 arch/s390/crypto/sha1_s390.c   | 2 +-
 arch/s390/crypto/sha256_s390.c | 2 +-
 arch/s390/crypto/sha512_s390.c | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/arch/s390/crypto/sha1_s390.c b/arch/s390/crypto/sha1_s390.c
index 009572e8276d..7c15542d3685 100644
--- a/arch/s390/crypto/sha1_s390.c
+++ b/arch/s390/crypto/sha1_s390.c

@@ -86,7 +86,7 @@ static struct shash_alg alg = {
 static int __init sha1_s390_init(void)
 {
 	if (!cpacf_query_func(CPACF_KIMD, CPACF_KIMD_SHA_1))
-		return -EOPNOTSUPP;
+		return -ENODEV;
 	return crypto_register_shash(&alg);
 }

diff --git a/arch/s390/crypto/sha256_s390.c b/arch/s390/crypto/sha256_s390.c
index 62833a1d8724..af7505148f80 100644
--- a/arch/s390/crypto/sha256_s390.c
+++ b/arch/s390/crypto/sha256_s390.c

@@ -117,7 +117,7 @@ static int __init sha256_s390_init(void)
 	int ret;
 
 	if (!cpacf_query_func(CPACF_KIMD, CPACF_KIMD_SHA_256))
-		return -EOPNOTSUPP;
+		return -ENODEV;
 	ret = crypto_register_shash(&sha256_alg);
 	if (ret < 0)
 		goto out;

diff --git a/arch/s390/crypto/sha512_s390.c b/arch/s390/crypto/sha512_s390.c
index be589c340d15..ad29db085a18 100644
--- a/arch/s390/crypto/sha512_s390.c
+++ b/arch/s390/crypto/sha512_s390.c

@@ -127,7 +127,7 @@ static int __init init(void)
 	int ret;
 
 	if (!cpacf_query_func(CPACF_KIMD, CPACF_KIMD_SHA_512))
-		return -EOPNOTSUPP;
+		return -ENODEV;
 	if ((ret = crypto_register_shash(&sha512_alg)) < 0)
 		goto out;
 	if ((ret = crypto_register_shash(&sha384_alg)) < 0)

-- 
2.21.0

Re: [PATCH v2 4/4] s390/crypto: sha: Use -ENODEV instead of -EOPNOTSUPP

From: Harald Freudenberger <freude@linux.ibm.com>
Date: 2019-06-12 10:30:33

On 12.06.19 12:22, David Hildenbrand wrote:

quoted hunk

Let's use the error value that is typically used if HW support is not
available when trying to load a module - this is also what systemd's
systemd-modules-load.service expects.

Signed-off-by: David Hildenbrand <redacted>
---
 arch/s390/crypto/sha1_s390.c   | 2 +-
 arch/s390/crypto/sha256_s390.c | 2 +-
 arch/s390/crypto/sha512_s390.c | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/arch/s390/crypto/sha1_s390.c b/arch/s390/crypto/sha1_s390.c
index 009572e8276d..7c15542d3685 100644
--- a/arch/s390/crypto/sha1_s390.c
+++ b/arch/s390/crypto/sha1_s390.c

@@ -86,7 +86,7 @@ static struct shash_alg alg = {
 static int __init sha1_s390_init(void)
 {
 	if (!cpacf_query_func(CPACF_KIMD, CPACF_KIMD_SHA_1))
-		return -EOPNOTSUPP;
+		return -ENODEV;
 	return crypto_register_shash(&alg);
 }

diff --git a/arch/s390/crypto/sha256_s390.c b/arch/s390/crypto/sha256_s390.c
index 62833a1d8724..af7505148f80 100644
--- a/arch/s390/crypto/sha256_s390.c
+++ b/arch/s390/crypto/sha256_s390.c

@@ -117,7 +117,7 @@ static int __init sha256_s390_init(void)
 	int ret;
 
 	if (!cpacf_query_func(CPACF_KIMD, CPACF_KIMD_SHA_256))
-		return -EOPNOTSUPP;
+		return -ENODEV;
 	ret = crypto_register_shash(&sha256_alg);
 	if (ret < 0)
 		goto out;

diff --git a/arch/s390/crypto/sha512_s390.c b/arch/s390/crypto/sha512_s390.c
index be589c340d15..ad29db085a18 100644
--- a/arch/s390/crypto/sha512_s390.c
+++ b/arch/s390/crypto/sha512_s390.c

@@ -127,7 +127,7 @@ static int __init init(void)
 	int ret;
 
 	if (!cpacf_query_func(CPACF_KIMD, CPACF_KIMD_SHA_512))
-		return -EOPNOTSUPP;
+		return -ENODEV;
 	if ((ret = crypto_register_shash(&sha512_alg)) < 0)
 		goto out;
 	if ((ret = crypto_register_shash(&sha384_alg)) < 0)

fine with me
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>

Re: [PATCH v2 3/4] s390/crypto: prng: Use -ENODEV instead of -EOPNOTSUPP

From: Harald Freudenberger <freude@linux.ibm.com>
Date: 2019-06-12 10:31:22

On 12.06.19 12:22, David Hildenbrand wrote:

quoted hunk

Let's use the error value that is typically used if HW support is not
available when trying to load a module - this is also what systemd's
systemd-modules-load.service expects.

Signed-off-by: David Hildenbrand <redacted>
---
 arch/s390/crypto/prng.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/s390/crypto/prng.c b/arch/s390/crypto/prng.c
index 12cca467af7d..d977643fa627 100644
--- a/arch/s390/crypto/prng.c
+++ b/arch/s390/crypto/prng.c

@@ -824,7 +824,7 @@ static int __init prng_init(void)
 
 	/* check if the CPU has a PRNG */
 	if (!cpacf_query_func(CPACF_KMC, CPACF_KMC_PRNG))
-		return -EOPNOTSUPP;
+		return -ENODEV;
 
 	/* check if TRNG subfunction is available */
 	if (cpacf_query_func(CPACF_PRNO, CPACF_PRNO_TRNG))

@@ -837,7 +837,7 @@ static int __init prng_init(void)
 			if (prng_mode == PRNG_MODE_SHA512) {
 				pr_err("The prng module cannot "
 				       "start in SHA-512 mode\n");
-				return -EOPNOTSUPP;
+				return -ENODEV;
 			}
 			prng_mode = PRNG_MODE_TDES;
 		} else

fine with me
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>

Re: [PATCH v2 2/4] s390/crypto: ghash: Use -ENODEV instead of -EOPNOTSUPP

From: Harald Freudenberger <freude@linux.ibm.com>
Date: 2019-06-12 10:31:52

On 12.06.19 12:22, David Hildenbrand wrote:

quoted hunk

Let's use the error value that is typically used if HW support is not
available when trying to load a module - this is also what systemd's
systemd-modules-load.service expects.

Signed-off-by: David Hildenbrand <redacted>
---
 arch/s390/crypto/ghash_s390.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/s390/crypto/ghash_s390.c b/arch/s390/crypto/ghash_s390.c
index 86aed30fad3a..eeeb6a7737a4 100644
--- a/arch/s390/crypto/ghash_s390.c
+++ b/arch/s390/crypto/ghash_s390.c

@@ -137,7 +137,7 @@ static struct shash_alg ghash_alg = {
 static int __init ghash_mod_init(void)
 {
 	if (!cpacf_query_func(CPACF_KIMD, CPACF_KIMD_GHASH))
-		return -EOPNOTSUPP;
+		return -ENODEV;
 
 	return crypto_register_shash(&ghash_alg);
 }

fine with me
Reviewed-by: Harald Freudenberger <freude@linux.ibm.com>

Re: [PATCH v2 2/4] s390/crypto: ghash: Use -ENODEV instead of -EOPNOTSUPP

From: Cornelia Huck <cohuck@redhat.com>
Date: 2019-06-12 10:32:14

On Wed, 12 Jun 2019 12:22:46 +0200
David Hildenbrand [off-list ref] wrote:

Let's use the error value that is typically used if HW support is not
available when trying to load a module - this is also what systemd's
systemd-modules-load.service expects.

Signed-off-by: David Hildenbrand <redacted>
---
 arch/s390/crypto/ghash_s390.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

Reviewed-by: Cornelia Huck <cohuck@redhat.com>

Re: [PATCH v2 3/4] s390/crypto: prng: Use -ENODEV instead of -EOPNOTSUPP

From: Cornelia Huck <cohuck@redhat.com>
Date: 2019-06-12 10:35:24

On Wed, 12 Jun 2019 12:22:47 +0200
David Hildenbrand [off-list ref] wrote:

Let's use the error value that is typically used if HW support is not
available when trying to load a module - this is also what systemd's
systemd-modules-load.service expects.

Signed-off-by: David Hildenbrand <redacted>
---
 arch/s390/crypto/prng.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

Reviewed-by: Cornelia Huck <cohuck@redhat.com>

Re: [PATCH v2 4/4] s390/crypto: sha: Use -ENODEV instead of -EOPNOTSUPP

From: Cornelia Huck <cohuck@redhat.com>
Date: 2019-06-12 10:37:25

On Wed, 12 Jun 2019 12:22:48 +0200
David Hildenbrand [off-list ref] wrote:

Let's use the error value that is typically used if HW support is not
available when trying to load a module - this is also what systemd's
systemd-modules-load.service expects.

Signed-off-by: David Hildenbrand <redacted>
---
 arch/s390/crypto/sha1_s390.c   | 2 +-
 arch/s390/crypto/sha256_s390.c | 2 +-
 arch/s390/crypto/sha512_s390.c | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

Reviewed-by: Cornelia Huck <cohuck@redhat.com>

Re: [PATCH v2 1/4] s390/pkey: Use -ENODEV instead of -EOPNOTSUPP

From: Harald Freudenberger <freude@linux.ibm.com>
Date: 2019-06-12 10:39:12

On 12.06.19 12:22, David Hildenbrand wrote:

quoted hunk

systemd-modules-load.service automatically tries to load the pkey module
on systems that have MSA.

Pkey also requires the MSA3 facility and a bunch of subfunctions.
Failing with -EOPNOTSUPP makes "systemd-modules-load.service" fail on
any system that does not have all needed subfunctions. For example,
when running under QEMU TCG (but also on systems where protected keys
are disabled via the HMC).

Let's use -ENODEV, so systemd-modules-load.service properly ignores
failing to load the pkey module because of missing HW functionality.

Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: David Hildenbrand <redacted>
---
 drivers/s390/crypto/pkey_api.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/s390/crypto/pkey_api.c b/drivers/s390/crypto/pkey_api.c
index 45eb0c14b880..ddfcefb47284 100644
--- a/drivers/s390/crypto/pkey_api.c
+++ b/drivers/s390/crypto/pkey_api.c

@@ -1695,15 +1695,15 @@ static int __init pkey_init(void)
 	 * are able to work with protected keys.
 	 */
 	if (!cpacf_query(CPACF_PCKMO, &pckmo_functions))
-		return -EOPNOTSUPP;
+		return -ENODEV;
 
 	/* check for kmc instructions available */
 	if (!cpacf_query(CPACF_KMC, &kmc_functions))
-		return -EOPNOTSUPP;
+		return -ENODEV;
 	if (!cpacf_test_func(&kmc_functions, CPACF_KMC_PAES_128) ||
 	    !cpacf_test_func(&kmc_functions, CPACF_KMC_PAES_192) ||
 	    !cpacf_test_func(&kmc_functions, CPACF_KMC_PAES_256))
-		return -EOPNOTSUPP;
+		return -ENODEV;
 
 	pkey_debug_init();

You missed one match in this file. Function pkey_clr2protkey()
also does a cpacf_test_func() and may return -EOPNOTSUPP.
I checked the call chain, it's save to change the returncode there also.
If done, Thanks and add my
reviewed-by: Harald Freudenberger <freude@linux.ibm.com>

Re: [PATCH v2 1/4] s390/pkey: Use -ENODEV instead of -EOPNOTSUPP

From: David Hildenbrand <hidden>
Date: 2019-06-12 10:41:25

On 12.06.19 12:39, Harald Freudenberger wrote:

On 12.06.19 12:22, David Hildenbrand wrote:

quoted

systemd-modules-load.service automatically tries to load the pkey module
on systems that have MSA.

Pkey also requires the MSA3 facility and a bunch of subfunctions.
Failing with -EOPNOTSUPP makes "systemd-modules-load.service" fail on
any system that does not have all needed subfunctions. For example,
when running under QEMU TCG (but also on systems where protected keys
are disabled via the HMC).

Let's use -ENODEV, so systemd-modules-load.service properly ignores
failing to load the pkey module because of missing HW functionality.

Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: David Hildenbrand <redacted>
---
 drivers/s390/crypto/pkey_api.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/s390/crypto/pkey_api.c b/drivers/s390/crypto/pkey_api.c
index 45eb0c14b880..ddfcefb47284 100644
--- a/drivers/s390/crypto/pkey_api.c
+++ b/drivers/s390/crypto/pkey_api.c

@@ -1695,15 +1695,15 @@ static int __init pkey_init(void)
 	 * are able to work with protected keys.
 	 */
 	if (!cpacf_query(CPACF_PCKMO, &pckmo_functions))
-		return -EOPNOTSUPP;
+		return -ENODEV;
 
 	/* check for kmc instructions available */
 	if (!cpacf_query(CPACF_KMC, &kmc_functions))
-		return -EOPNOTSUPP;
+		return -ENODEV;
 	if (!cpacf_test_func(&kmc_functions, CPACF_KMC_PAES_128) ||
 	    !cpacf_test_func(&kmc_functions, CPACF_KMC_PAES_192) ||
 	    !cpacf_test_func(&kmc_functions, CPACF_KMC_PAES_256))
-		return -EOPNOTSUPP;
+		return -ENODEV;
 
 	pkey_debug_init();

You missed one match in this file. Function pkey_clr2protkey()
also does a cpacf_test_func() and may return -EOPNOTSUPP.
I checked the call chain, it's save to change the returncode there also.

That's unrelated to module loading (if I am not wrong), shall we still
include this change here?

Thanks!

If done, Thanks and add my
reviewed-by: Harald Freudenberger <freude@linux.ibm.com>


-- 

Thanks,

David / dhildenb

Re: [PATCH v2 1/4] s390/pkey: Use -ENODEV instead of -EOPNOTSUPP

From: Harald Freudenberger <freude@linux.ibm.com>
Date: 2019-06-12 11:08:00

On 12.06.19 12:41, David Hildenbrand wrote:

On 12.06.19 12:39, Harald Freudenberger wrote:

quoted

On 12.06.19 12:22, David Hildenbrand wrote:

quoted

systemd-modules-load.service automatically tries to load the pkey module
on systems that have MSA.

Pkey also requires the MSA3 facility and a bunch of subfunctions.
Failing with -EOPNOTSUPP makes "systemd-modules-load.service" fail on
any system that does not have all needed subfunctions. For example,
when running under QEMU TCG (but also on systems where protected keys
are disabled via the HMC).

Let's use -ENODEV, so systemd-modules-load.service properly ignores
failing to load the pkey module because of missing HW functionality.

Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: David Hildenbrand <redacted>
---
 drivers/s390/crypto/pkey_api.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/s390/crypto/pkey_api.c b/drivers/s390/crypto/pkey_api.c
index 45eb0c14b880..ddfcefb47284 100644
--- a/drivers/s390/crypto/pkey_api.c
+++ b/drivers/s390/crypto/pkey_api.c

@@ -1695,15 +1695,15 @@ static int __init pkey_init(void)
 	 * are able to work with protected keys.
 	 */
 	if (!cpacf_query(CPACF_PCKMO, &pckmo_functions))
-		return -EOPNOTSUPP;
+		return -ENODEV;
 
 	/* check for kmc instructions available */
 	if (!cpacf_query(CPACF_KMC, &kmc_functions))
-		return -EOPNOTSUPP;
+		return -ENODEV;
 	if (!cpacf_test_func(&kmc_functions, CPACF_KMC_PAES_128) ||
 	    !cpacf_test_func(&kmc_functions, CPACF_KMC_PAES_192) ||
 	    !cpacf_test_func(&kmc_functions, CPACF_KMC_PAES_256))
-		return -EOPNOTSUPP;
+		return -ENODEV;
 
 	pkey_debug_init();

You missed one match in this file. Function pkey_clr2protkey()
also does a cpacf_test_func() and may return -EOPNOTSUPP.
I checked the call chain, it's save to change the returncode there also.

That's unrelated to module loading (if I am not wrong), shall we still
include this change here?

Thanks!

That would be nice.
However, I agree it is not related to module loading.

quoted

If done, Thanks and add my
reviewed-by: Harald Freudenberger <freude@linux.ibm.com>

Re: [PATCH v2 1/4] s390/pkey: Use -ENODEV instead of -EOPNOTSUPP

From: David Hildenbrand <hidden>
Date: 2019-06-12 11:08:45

On 12.06.19 13:07, Harald Freudenberger wrote:

On 12.06.19 12:41, David Hildenbrand wrote:

quoted

On 12.06.19 12:39, Harald Freudenberger wrote:

quoted

On 12.06.19 12:22, David Hildenbrand wrote:

quoted

systemd-modules-load.service automatically tries to load the pkey module
on systems that have MSA.

Pkey also requires the MSA3 facility and a bunch of subfunctions.
Failing with -EOPNOTSUPP makes "systemd-modules-load.service" fail on
any system that does not have all needed subfunctions. For example,
when running under QEMU TCG (but also on systems where protected keys
are disabled via the HMC).

Let's use -ENODEV, so systemd-modules-load.service properly ignores
failing to load the pkey module because of missing HW functionality.

Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: David Hildenbrand <redacted>
---
 drivers/s390/crypto/pkey_api.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/s390/crypto/pkey_api.c b/drivers/s390/crypto/pkey_api.c
index 45eb0c14b880..ddfcefb47284 100644
--- a/drivers/s390/crypto/pkey_api.c
+++ b/drivers/s390/crypto/pkey_api.c

@@ -1695,15 +1695,15 @@ static int __init pkey_init(void)
 	 * are able to work with protected keys.
 	 */
 	if (!cpacf_query(CPACF_PCKMO, &pckmo_functions))
-		return -EOPNOTSUPP;
+		return -ENODEV;
 
 	/* check for kmc instructions available */
 	if (!cpacf_query(CPACF_KMC, &kmc_functions))
-		return -EOPNOTSUPP;
+		return -ENODEV;
 	if (!cpacf_test_func(&kmc_functions, CPACF_KMC_PAES_128) ||
 	    !cpacf_test_func(&kmc_functions, CPACF_KMC_PAES_192) ||
 	    !cpacf_test_func(&kmc_functions, CPACF_KMC_PAES_256))
-		return -EOPNOTSUPP;
+		return -ENODEV;
 
 	pkey_debug_init();

You missed one match in this file. Function pkey_clr2protkey()
also does a cpacf_test_func() and may return -EOPNOTSUPP.
I checked the call chain, it's save to change the returncode there also.

That's unrelated to module loading (if I am not wrong), shall we still
include this change here?

Thanks!

That would be nice.
However, I agree it is not related to module loading.

I can include that, thanks!


-- 

Thanks,

David / dhildenb

`h`	back out one level
`j`	next message in thread
`k`	previous message in thread
`l`	drill in
`Esc`	close help / fold thread tree
`?`	toggle this help