All of lore.kernel.org
 help / color / mirror / Atom feed
From: Steven Whitehouse <swhiteho@redhat.com>
To: cluster-devel.redhat.com
Subject: [Cluster-devel] [PATCH] dlm: fix cross-platform error values
Date: Tue, 15 Jan 2008 08:48:51 +0000	[thread overview]
Message-ID: <1200386931.22038.137.camel@quoit> (raw)
In-Reply-To: <Pine.LNX.4.64.0801150712240.6826@trider-g7>

Hi,

On Tue, 2008-01-15 at 07:15 +0100, Fabio M. Di Nitto wrote:
> Hi guys,
> 
> On Mon, 14 Jan 2008, Patrick Caulfeld wrote:
> 
> > Some errno values differ across platforms. So if we return things like
> > -EINPROGRESS from one node it can get misinterpreted or rejected on
> > another one.
> >
> > This patch fixes up the errno values passed on the wire so that they
> > match the x86 ones (so as not to break the protocol), and re-instates
> > the platform-specific ones at the other end.
> >
> > Many thanks to Fabio for testing this patch.
> >
> > Signed-Off-By: Patrick Caulfield <pcaulfie@redhat.com
> > Signed-off-by: Fabio M. Di Nitto <fabbione@ubuntu.com>
> >
> 
> while doing some more testing in strange situations, i noticed that we had 
> more values going down the pipe.
> 
> This patch (based on Patrick's one) makes absolutely sure that all the -E 
> we use around fs/dlm/* are converted before hitting the wire. Mostlikely 
> not even half of them will go that far, but it's easier to catch them all 
> than finding them one at a time.
> 
> values and switch/case are also sorted in ascending order for pure 
> consistency paranoia.
> 
> Signed-off-by: Fabio M. Di Nitto <fabbione@ubuntu.com>
> 
> Thanks
> Fabio
> 
Perhaps it would be worth creating a translation table rather than
listing each value in a switch statement? Also I'd suggest to add a
"default" so that you can catch any values that you didn't expect and
BUG(); or something. That way if someone adds a new error value in the
future, you'll catch it before it hits the wire,

Steve.

> diff --git a/fs/dlm/dlm_internal.h b/fs/dlm/dlm_internal.h
> index ec61bba..d0de852 100644
> --- a/fs/dlm/dlm_internal.h
> +++ b/fs/dlm/dlm_internal.h
> @@ -220,6 +220,27 @@ struct dlm_args {
>   #define DLM_IFL_USER		0x00000001
>   #define DLM_IFL_ORPHAN		0x00000002
> 
> +
> +/* arch-independant errno values */
> +
> +#define DLM_ERRNO_EDEADLK		 35
> +#define DLM_ERRNO_EBADE			 52
> +#define DLM_ERRNO_EBADR			 53
> +#define DLM_ERRNO_EBADSLT		 57
> +#define DLM_ERRNO_EPROTO		 71
> +#define DLM_ERRNO_EBADMSG		 74
> +#define DLM_ERRNO_EPROTONOSUPPORT	 93
> +#define DLM_ERRNO_EOPNOTSUPP		 95
> +#define DLM_ERRNO_EADDRINUSE		 98
> +#define DLM_ERRNO_ENETDOWN		100
> +#define DLM_ERRNO_ENETUNREACH		101
> +#define DLM_ERRNO_ECONNABORTED		103
> +#define DLM_ERRNO_ENOBUFS		105
> +#define DLM_ERRNO_ENOTCONN		107
> +#define DLM_ERRNO_ETIMEDOUT		110
> +#define DLM_ERRNO_EHOSTUNREACH		113
> +#define DLM_ERRNO_EINPROGRESS		115
> +
>   struct dlm_lkb {
>   	struct dlm_rsb		*lkb_resource;	/* the rsb */
>   	struct kref		lkb_ref;
> diff --git a/fs/dlm/util.c b/fs/dlm/util.c
> index 963889c..e0f0ad4 100644
> --- a/fs/dlm/util.c
> +++ b/fs/dlm/util.c
> @@ -36,6 +36,61 @@ void dlm_message_out(struct dlm_message *ms)
> 
>   	header_out(hd);
> 
> +	/* Use arch-independant errno values on the wire */
> +	switch (ms->m_result) {
> +	case -EDEADLK:
> +		ms->m_result = -DLM_ERRNO_EDEADLK;
> +		break;
> +	case -EBADE:
> +		ms->m_result = -DLM_ERRNO_EBADE;
> +		break;
> +	case -EBADR:
> +		ms->m_result = -DLM_ERRNO_EBADR;
> +		break;
> +	case -EBADSLT:
> +		ms->m_result = -DLM_ERRNO_EBADSLT;
> +		break;
> +	case -EPROTO:
> +		ms->m_result = -DLM_ERRNO_EPROTO;
> +		break;
> +	case -EBADMSG:
> +		ms->m_result = -DLM_ERRNO_EBADMSG;
> +		break;
> +	case -EPROTONOSUPPORT:
> +		ms->m_result = -DLM_ERRNO_EPROTONOSUPPORT;
> +		break;
> +	case -EOPNOTSUPP:
> +		ms->m_result = -DLM_ERRNO_EOPNOTSUPP;
> +		break;
> +	case -EADDRINUSE:
> +		ms->m_result = -DLM_ERRNO_EADDRINUSE;
> +		break;
> +	case -ENETDOWN:
> +		ms->m_result = -DLM_ERRNO_ENETDOWN;
> +		break;
> +	case -ENETUNREACH:
> +		ms->m_result = -DLM_ERRNO_ENETUNREACH;
> +		break;
> +	case -ECONNABORTED:
> +		ms->m_result = -DLM_ERRNO_ECONNABORTED;
> +		break;
> +	case -ENOBUFS:
> +		ms->m_result = -DLM_ERRNO_ENOBUFS;
> +		break;
> +	case -ENOTCONN:
> +		ms->m_result = -DLM_ERRNO_ENOTCONN;
> +		break;
> +	case -ETIMEDOUT:
> +		ms->m_result = -DLM_ERRNO_ETIMEDOUT;
> +		break;
> +	case -EHOSTUNREACH:
> +		ms->m_result = -DLM_ERRNO_EHOSTUNREACH;
> +		break;
> +	case -EINPROGRESS:
> +		ms->m_result = -DLM_ERRNO_EINPROGRESS;
> +		break;
> +	}
> +
>   	ms->m_type		= cpu_to_le32(ms->m_type);
>   	ms->m_nodeid		= cpu_to_le32(ms->m_nodeid);
>   	ms->m_pid		= cpu_to_le32(ms->m_pid);
> @@ -80,6 +135,60 @@ void dlm_message_in(struct dlm_message *ms)
>   	ms->m_bastmode		= le32_to_cpu(ms->m_bastmode);
>   	ms->m_asts		= le32_to_cpu(ms->m_asts);
>   	ms->m_result		= le32_to_cpu(ms->m_result);
> +
> +	switch (ms->m_result) {
> +	case -DLM_ERRNO_EDEADLK:
> +		ms->m_result = -EDEADLK;
> +		break;
> +	case -DLM_ERRNO_EBADE:
> +		ms->m_result = -EBADE;
> +		break;
> +	case -DLM_ERRNO_EBADR:
> +		ms->m_result = -EBADR;
> +		break;
> +	case -DLM_ERRNO_EBADSLT:
> +		ms->m_result = -EBADSLT;
> +		break;
> +	case -DLM_ERRNO_EPROTO:
> +		ms->m_result = -EPROTO;
> +		break;
> +	case -DLM_ERRNO_EBADMSG:
> +		ms->m_result = -EBADMSG;
> +		break;
> +	case -DLM_ERRNO_EPROTONOSUPPORT:
> +		ms->m_result = -EPROTONOSUPPORT;
> +		break;
> +	case -DLM_ERRNO_EOPNOTSUPP:
> +		ms->m_result = -EOPNOTSUPP;
> +		break;
> +	case -DLM_ERRNO_EADDRINUSE:
> +		ms->m_result = -EADDRINUSE;
> +		break;
> +	case -DLM_ERRNO_ENETDOWN:
> +		ms->m_result = -ENETDOWN;
> +		break;
> +	case -DLM_ERRNO_ENETUNREACH:
> +		ms->m_result = -ENETUNREACH;
> +		break;
> +	case -DLM_ERRNO_ECONNABORTED:
> +		ms->m_result = -ECONNABORTED;
> +		break;
> +	case -DLM_ERRNO_ENOBUFS:
> +		ms->m_result = -ENOBUFS;
> +		break;
> +	case -DLM_ERRNO_ENOTCONN:
> +		ms->m_result = -ENOTCONN;
> +		break;
> +	case -DLM_ERRNO_ETIMEDOUT:
> +		ms->m_result = -ETIMEDOUT;
> +		break;
> +	case -DLM_ERRNO_EHOSTUNREACH:
> +		ms->m_result = -EHOSTUNREACH;
> +		break;
> +	case -DLM_ERRNO_EINPROGRESS:
> +		ms->m_result = -EINPROGRESS;
> +		break;
> +	}
>   }
> 
>   static void rcom_lock_out(struct rcom_lock *rl)
> 
> --
> I'm going to make him an offer he can't refuse.



      reply	other threads:[~2008-01-15  8:48 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-01-14 13:14 [Cluster-devel] [PATCH] dlm: fix cross-platform error values Patrick Caulfeld
2008-01-15  5:02 ` Fabio M. Di Nitto
2008-01-15  6:15 ` Fabio M. Di Nitto
2008-01-15  8:48   ` Steven Whitehouse [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=1200386931.22038.137.camel@quoit \
    --to=swhiteho@redhat.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.