public inbox for qemu-devel@nongnu.org
 help / color / mirror / Atom feed
From: ihanzhzh@gmail.com
To: qemu-devel@nongnu.org
Cc: kbusch@kernel.org, its@irrelevant.dk, foss@defmacro.it,
	qemu-block@nongnu.org, Han Zhang <ihanzh@outlook.com>,
	Han Zhang <ihanzhzh@gmail.com>
Subject: [PATCH v2] hw/nvme: keep switch handling in nvme_directive_receive()
Date: Mon, 23 Mar 2026 23:57:37 +0800	[thread overview]
Message-ID: <20260323155736.1558483-1-ihanzhzh@gmail.com> (raw)
In-Reply-To: <20260317010721.232247-1-ihanzh@outlook.com>

From: Han Zhang <ihanzh@outlook.com>

Static analysis reported that the default branches in
nvme_directive_receive() were unreachable because dtype and doper were
validated before the switch statements.

Keep the switch statements as explicit documentation of supported
directive type/operation combinations, and remove the redundant early
dtype/doper checks instead.

Also move namespace lookup into the RETURN_PARAMS path so invalid
dtype/doper values are rejected via switch defaults without an
unnecessary nvme_ns() lookup.

Reported-by: Ekaterina Zilotina
Fixes: https://gitlab.com/qemu-project/qemu/-/issues/2472
Signed-off-by: Han Zhang <ihanzhzh@gmail.com>
---
 hw/nvme/ctrl.c | 15 +++++++--------
 1 file changed, 7 insertions(+), 8 deletions(-)

diff --git a/hw/nvme/ctrl.c b/hw/nvme/ctrl.c
index cc4593cd42..479a4b8725 100644
--- a/hw/nvme/ctrl.c
+++ b/hw/nvme/ctrl.c
@@ -7540,13 +7540,7 @@ static uint16_t nvme_directive_receive(NvmeCtrl *n, NvmeRequest *req)
 
     trans_len = MIN(sizeof(NvmeDirectiveIdentify), numd << 2);
 
-    if (nsid == NVME_NSID_BROADCAST || dtype != NVME_DIRECTIVE_IDENTIFY ||
-        doper != NVME_DIRECTIVE_RETURN_PARAMS) {
-        return NVME_INVALID_FIELD | NVME_DNR;
-    }
-
-    ns = nvme_ns(n, nsid);
-    if (!ns) {
+    if (nsid == NVME_NSID_BROADCAST) {
         return NVME_INVALID_FIELD | NVME_DNR;
     }
 
@@ -7554,6 +7548,11 @@ static uint16_t nvme_directive_receive(NvmeCtrl *n, NvmeRequest *req)
     case NVME_DIRECTIVE_IDENTIFY:
         switch (doper) {
         case NVME_DIRECTIVE_RETURN_PARAMS:
+            ns = nvme_ns(n, nsid);
+            if (!ns) {
+                return NVME_INVALID_FIELD | NVME_DNR;
+            }
+
             if (ns->endgrp && ns->endgrp->fdp.enabled) {
                 id.supported |= 1 << NVME_DIRECTIVE_DATA_PLACEMENT;
                 id.enabled |= 1 << NVME_DIRECTIVE_DATA_PLACEMENT;
@@ -7567,7 +7566,7 @@ static uint16_t nvme_directive_receive(NvmeCtrl *n, NvmeRequest *req)
         }
 
     default:
-        return NVME_INVALID_FIELD;
+        return NVME_INVALID_FIELD | NVME_DNR;
     }
 }
 
-- 
2.34.1



      parent reply	other threads:[~2026-03-23 16:02 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-03-17  1:07 [PATCH] hw/nvme: remove unreachable default cases in nvme_directive_receive() Han Zhang
2026-03-23 11:38 ` Klaus Jensen
2026-03-23 15:57 ` ihanzhzh [this message]

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=20260323155736.1558483-1-ihanzhzh@gmail.com \
    --to=ihanzhzh@gmail.com \
    --cc=foss@defmacro.it \
    --cc=ihanzh@outlook.com \
    --cc=its@irrelevant.dk \
    --cc=kbusch@kernel.org \
    --cc=qemu-block@nongnu.org \
    --cc=qemu-devel@nongnu.org \
    /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