linux-kernel - Re: [PATCH v3 3/3] nvme: handle connectivity loss in nvme_set_queue

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <bee71ac4-37bf-4182-ad7e-78efd9baff46@flourine.local>
Date: Tue, 17 Dec 2024 15:01:55 +0100
From: Daniel Wagner <dwagner@...e.de>
To: Hannes Reinecke <hare@...e.de>
Cc: Daniel Wagner <wagi@...nel.org>, 
	James Smart <james.smart@...adcom.com>, Keith Busch <kbusch@...nel.org>, Christoph Hellwig <hch@....de>, 
	Sagi Grimberg <sagi@...mberg.me>, Paul Ely <paul.ely@...adcom.com>, linux-nvme@...ts.infradead.org, 
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH v3 3/3] nvme: handle connectivity loss in
 nvme_set_queue_count

On Tue, Dec 17, 2024 at 10:45:45AM +0100, Hannes Reinecke wrote:
> On 12/17/24 09:35, Daniel Wagner wrote:
> > On Fri, Nov 29, 2024 at 12:10:33PM +0100, Hannes Reinecke wrote:
> > > > +	/*
> > > > +	 * It's either a kernel error or the host observed a connection
> > > > +	 * lost. In either case it's not possible communicate with the
> > > > +	 * controller and thus enter the error code path.
> > > > +	 */
> > > > +	if (status < 0 || status == NVME_SC_HOST_PATH_ERROR)
> > > >    		return status;
> > > >    	/*
> > > > 
> > > Hmm. Maybe checking for NVME_SC_DNR, too?
> > 
> > if no one complains I'll update the check to:
> > 
> > 	if (status < 0 || (status > 0 && (status & NVME_STATUS_DNR)) ||
> > 	    status == NVME_SC_HOST_PATH_ERROR)
> > 		return status;
> > 
> > okay?
> 
> Which really cries out for a helper. But otherwise okay.

As far I remember, Christoph is not a big fan of single users of tiny
helper functions. Anyway, what about this:

+static bool nvme_num_queue_updated(status)
+{
+       /*
+        * It's either a kernel error or the host observed a connection
+        * lost. In either case it's not possible communicate with the
+        * controller and thus enter the error code path.
+        */
+
+       if (status < 0 || status & NVME_STATUS_DNR ||
+           status == NVME_SC_HOST_PATH_ERROR)
+               return false;
+
+       return true;
+}
+
 int nvme_set_queue_count(struct nvme_ctrl *ctrl, int *count)
 {
        u32 q_count = (*count - 1) | ((*count - 1) << 16);
@@ -1701,13 +1716,8 @@ int nvme_set_queue_count(struct nvme_ctrl *ctrl, int *count)

        status = nvme_set_features(ctrl, NVME_FEAT_NUM_QUEUES, q_count, NULL, 0,
                        &result);
-       /*
-        * It's either a kernel error or the host observed a connection
-        * lost. In either case it's not possible communicate with the
-        * controller and thus enter the error code path.
-        */
-       if (status < 0 || (status > 0 && (status & NVME_STATUS_DNR)) ||
-           status == NVME_SC_HOST_PATH_ERROR)
+
+       if (!nvme_num_queue_updated(status))
                return status;

        /*