Re: [PATCH v3 05/14] virt: sev-guest: Add vmpck_id to snp_guest_dev struct

From: Tom Lendacky
Date: Wed Aug 02 2023 - 12:56:00 EST


On 8/1/23 23:12, Nikunj A. Dadhania wrote:
On 8/1/2023 10:00 PM, Tom Lendacky wrote:
On 7/22/23 06:19, Nikunj A Dadhania wrote:
Drop vmpck and os_area_msg_seqno pointers so that secret page layout
does not need to be exposed to the sev-guest driver after the rework.
Instead, add helper APIs to access vmpck and os_area_msg_seqno when
needed.

Signed-off-by: Nikunj A Dadhania <nikunj@xxxxxxx>
---
  drivers/virt/coco/sev-guest/sev-guest.c | 84 +++++++++++++------------
  1 file changed, 43 insertions(+), 41 deletions(-)

diff --git a/drivers/virt/coco/sev-guest/sev-guest.c b/drivers/virt/coco/sev-guest/sev-guest.c
index d4241048b397..8ad43e007d3b 100644
--- a/drivers/virt/coco/sev-guest/sev-guest.c
+++ b/drivers/virt/coco/sev-guest/sev-guest.c
@@ -50,8 +50,7 @@ struct snp_guest_dev {
        struct snp_secrets_page_layout *layout;
      struct snp_req_data input;
-    u32 *os_area_msg_seqno;
-    u8 *vmpck;
+    unsigned int vmpck_id;
  };
    static u32 vmpck_id;
@@ -67,12 +66,23 @@ static inline unsigned int get_ctx_authsize(struct snp_guest_dev *snp_dev)
      return 0;
  }
  -static bool is_vmpck_empty(struct snp_guest_dev *snp_dev)
+static inline u8 *snp_get_vmpck(struct snp_guest_dev *snp_dev)
+{
+    return snp_dev->layout->vmpck0 + snp_dev->vmpck_id * VMPCK_KEY_LEN;
+}
+
+static inline u32 *snp_get_os_area_msg_seqno(struct snp_guest_dev *snp_dev)
+{
+    return &snp_dev->layout->os_area.msg_seqno_0 + snp_dev->vmpck_id;
+}
+
+static bool snp_is_vmpck_empty(struct snp_guest_dev *snp_dev)

I noticed this name change from is_vmpck_empty() to snp_is_vmpck_empty(). Is that in prep for moving, too? Is so, maybe call that out in the commit message.

Yes, will add to the commit.


  {
      char zero_key[VMPCK_KEY_LEN] = {0};
+    u8 *key = snp_get_vmpck(snp_dev);
  -    if (snp_dev->vmpck)
-        return !memcmp(snp_dev->vmpck, zero_key, VMPCK_KEY_LEN);
+    if (key)
+        return !memcmp(key, zero_key, VMPCK_KEY_LEN);

I believe key can't be NULL, so this check isn't required.

Sure, will update.

Ah, although I noticed that when the various functions are moved over to the other file, the key return value can be NULL, so probably not worth changing here.

Thanks,
Tom


Regards
Nikunj