All of lore.kernel.org
 help / color / mirror / Atom feed
From: Michael Roth <michael.roth@amd.com>
To: <linux-crypto@vger.kernel.org>
Cc: <herbert@gondor.apana.org.au>, <davem@davemloft.net>,
	<ackerleytng@google.com>, <thomas.lendacky@amd.com>,
	<ashish.kalra@amd.com>, <aik@amd.com>, <seanjc@google.com>
Subject: [PATCH v2 9/9] crypto: ccp - Allocate status page outside sev_cmd_mutex in SNP_PLATFORM_STATUS
Date: Thu, 15 Oct 2026 23:09:37 -0500	[thread overview]
Message-ID: <20261016041109.1778673-10-michael.roth@amd.com> (raw)
In-Reply-To: <20261016041109.1778673-1-michael.roth@amd.com>

Allocating memory while holding the global SEV command mutex
(sev_cmd_mutex) unnecessarily blocks concurrent SEV and SNP operations
across the system if page allocation stalls. Handle allocation/freeing
outside of the mutex instead.

Co-developed-by: Ackerley Tng <ackerleytng@google.com>
Signed-off-by: Ackerley Tng <ackerleytng@google.com>
Signed-off-by: Michael Roth <michael.roth@amd.com>
---
 drivers/crypto/ccp/sev-dev.c | 62 ++++++++++++++++++------------------
 1 file changed, 31 insertions(+), 31 deletions(-)

diff --git a/drivers/crypto/ccp/sev-dev.c b/drivers/crypto/ccp/sev-dev.c
index 1e017783dd03..aa4f0fd731c0 100644
--- a/drivers/crypto/ccp/sev-dev.c
+++ b/drivers/crypto/ccp/sev-dev.c
@@ -2532,8 +2532,8 @@ static int sev_ioctl_do_pdh_export(struct sev_issue_cmd *argp, bool writable)
 	return ret;
 }
 
-static int __sev_do_snp_platform_status(struct sev_user_data_snp_status *status,
-					int *error)
+static int sev_do_snp_platform_status(struct sev_user_data_snp_status *status,
+				      int *error)
 {
 	struct sev_device *sev = psp_master->sev_data;
 	void *data __free(free_page) = NULL;
@@ -2544,31 +2544,36 @@ static int __sev_do_snp_platform_status(struct sev_user_data_snp_status *status,
 	if (!data)
 		return -ENOMEM;
 
-	/*
-	 * SNP_PLATFORM_STATUS can be executed in any SNP state. But if executed
-	 * when SNP has been initialized, the status page must be firmware-owned.
-	 */
-	if (sev->snp_initialized) {
+	scoped_guard(mutex, &sev_cmd_mutex) {
 		/*
-		 * Firmware expects the status page to be in Firmware state,
-		 * otherwise it will report an error INVALID_PAGE_STATE.
+		 * SNP_PLATFORM_STATUS can be executed in any SNP state. But if executed
+		 * when SNP has been initialized, the status page must be firmware-owned.
 		 */
-		if (rmp_mark_pages_firmware(__pa(data), 1, true))
-			return -EFAULT;
-	}
+		if (sev->snp_initialized) {
+			/*
+			 * Firmware expects the status page to be in Firmware state,
+			 * otherwise it will report an error INVALID_PAGE_STATE.
+			 */
+			if (rmp_mark_pages_firmware(__pa(data), 1, true))
+				return -EFAULT;
+		}
 
-	buf.address = __psp_pa(data);
-	ret = __sev_do_cmd_locked(SEV_CMD_SNP_PLATFORM_STATUS, &buf, error);
+		buf.address = __psp_pa(data);
+		ret = __sev_do_cmd_locked(SEV_CMD_SNP_PLATFORM_STATUS, &buf, error);
 
-	if (sev->snp_initialized) {
-		/*
-		 * The status page will be in Reclaim state on success, or left
-		 * in Firmware state on failure. Use snp_reclaim_pages() to
-		 * transition either case back to Hypervisor-owned state.
-		 */
-		if (snp_reclaim_pages(__pa(data), 1, true)) {
-			data = NULL;
-			return -EFAULT;
+		if (sev->snp_initialized) {
+			/*
+			 * The status page will be in Reclaim state on success, or left
+			 * in Firmware state on failure. Use snp_reclaim_pages() to
+			 * transition either case back to Hypervisor-owned state. If
+			 * for some reason that fails, it will have been added to the
+			 * leaked pages list, so indicate to the caller to not free
+			 * the page.
+			 */
+			if (snp_reclaim_pages(__pa(data), 1, true)) {
+				data = NULL;
+				return -EFAULT;
+			}
 		}
 	}
 
@@ -2586,8 +2591,7 @@ static int sev_ioctl_do_snp_platform_status(struct sev_issue_cmd *argp)
 	if (!argp->data)
 		return -EINVAL;
 
-	scoped_guard(mutex, &sev_cmd_mutex)
-		ret = __sev_do_snp_platform_status(&status, &argp->error);
+	ret = sev_do_snp_platform_status(&status, &argp->error);
 	if (ret < 0)
 		return ret;
 
@@ -3086,9 +3090,7 @@ static int get_v1_svn(struct sev_device *sev)
 	struct sev_user_data_snp_status status;
 	int ret, error = 0;
 
-	mutex_lock(&sev_cmd_mutex);
-	ret = __sev_do_snp_platform_status(&status, &error);
-	mutex_unlock(&sev_cmd_mutex);
+	ret = sev_do_snp_platform_status(&status, &error);
 	if (ret < 0)
 		return ret;
 
@@ -3103,9 +3105,7 @@ static int get_v2_svn(struct sev_device *sev)
 	struct sev_snp_tcb_version_turin *tcb;
 	int ret, error = 0;
 
-	mutex_lock(&sev_cmd_mutex);
-	ret = __sev_do_snp_platform_status(&status, &error);
-	mutex_unlock(&sev_cmd_mutex);
+	ret = sev_do_snp_platform_status(&status, &error);
 	if (ret < 0)
 		return ret;
 
-- 
2.43.0


  parent reply	other threads:[~2026-10-16  4:13 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-10-16  4:09 [PATCH v2 0/9] crypto: ccp - Make sev_cmd_mutex more granular Michael Roth
2026-10-16  4:09 ` [PATCH v2 1/9] crypto: ccp - Shut down SEV firmware before copying to user Michael Roth
2026-10-16  4:09 ` [PATCH v2 2/9] crypto: ccp - Check SNP initialization state after copying user inputs Michael Roth
2026-10-16  4:09 ` [PATCH v2 3/9] crypto: ccp - Use __free() handlers for sev_ioctl_do_pek_csr() allocations Michael Roth
2026-10-19 13:33   ` Tom Lendacky
2026-10-16  4:09 ` [PATCH v2 4/9] crypto: ccp - Use __free() handlers for sev_ioctl_do_pek_import() allocations Michael Roth
2026-10-19 13:38   ` Tom Lendacky
2026-10-16  4:09 ` [PATCH v2 5/9] crypto: ccp - Use __free() handlers for sev_ioctl_do_pdh_export() allocations Michael Roth
2026-10-19 13:40   ` Tom Lendacky
2026-10-16  4:09 ` [PATCH v2 6/9] crypto: ccp: - Rework sev_ioctl_do_pdh_export() to avoid goto usage Michael Roth
2026-10-19 13:49   ` Tom Lendacky
2026-10-16  4:09 ` [PATCH v2 7/9] crypto: ccp - Make usage of sev_cmd_mutex more granular Michael Roth
2026-10-16  4:09 ` [PATCH v2 8/9] crypto: ccp - Use __free() handlers for __sev_do_snp_platform_status() allocations Michael Roth
2026-10-19 14:03   ` Tom Lendacky
2026-10-16  4:09 ` Michael Roth [this message]
2026-10-19 14:05   ` [PATCH v2 9/9] crypto: ccp - Allocate status page outside sev_cmd_mutex in SNP_PLATFORM_STATUS Tom Lendacky
2026-10-20 21:42 ` [PATCH v2 0/9] crypto: ccp - Make sev_cmd_mutex more granular Michael Roth
2026-10-21  5:26   ` Herbert Xu
2026-10-21 14:39     ` Sean Christopherson
2026-10-21 15:00       ` Michael Roth

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=20261016041109.1778673-10-michael.roth@amd.com \
    --to=michael.roth@amd.com \
    --cc=ackerleytng@google.com \
    --cc=aik@amd.com \
    --cc=ashish.kalra@amd.com \
    --cc=davem@davemloft.net \
    --cc=herbert@gondor.apana.org.au \
    --cc=linux-crypto@vger.kernel.org \
    --cc=seanjc@google.com \
    --cc=thomas.lendacky@amd.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.