xen-devel.lists.xenproject.org archive mirror
 help / color / mirror / Atom feed
From: Ian Campbell <ian.campbell@citrix.com>
To: Joshua Otto <jtotto@uwaterloo.ca>, xen-devel@lists.xen.org
Cc: wei.liu2@citrix.com, stefano.stabellini@eu.citrix.com,
	george.dunlap@eu.citrix.com, dario.faggioli@citrix.com,
	ian.jackson@eu.citrix.com, czylin@uwaterloo.ca,
	hjarmstr@uwaterloo.ca
Subject: Re: [PATCH 2/5] libxl: make GC_FREE reachable in libxl_get_scheduler()
Date: Mon, 4 Jan 2016 16:29:33 +0000	[thread overview]
Message-ID: <1451924973.13361.183.camel@citrix.com> (raw)
In-Reply-To: <1451279808-25264-3-git-send-email-jtotto@uwaterloo.ca>

On Mon, 2015-12-28 at 00:16 -0500, Joshua Otto wrote:
> Coverity CID 1343309
> 
> This patch preserves the multiple error paths in order to avoid
> meaninglessly assigning the ERROR_FAIL libxl error code to the
> return variable, which is of type libxl_scheduler.

Which makes me think that the existing code is bogus to return an error
code as a libxl_scheduler too, since that is not very different to the
bogus assignment.

The function ought to return LIBXL_SCHEDLER_UNKNOWN. However that might be
considered an API breakage (since at least xl checks for errors with < 0).

A _really_ correct libxl function API would take an output libxl_scheduler
pointer and return an int error code, but that is definitely an API change.

Given that a caller really ought to be handling LIBXL_SCHEDULER_UNKNOWN as
a return value, even if it is also written to expect negative error values
as well, so I reckon we can get away with changing the return to
SCHEDULER_UNKNOWN in the error case. Either the caller already handles
that, or it was already buggy in not doing so.

That would also allow us to move to a single error path, since you'd no
longer worry about clobbering.

Wei, Ian, any thoughts?

> Signed-off-by: Joshua Otto <jtotto@uwaterloo.ca>
> ---
>  tools/libxl/libxl.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/tools/libxl/libxl.c b/tools/libxl/libxl.c
> index ca4679b..60a2509 100644
> --- a/tools/libxl/libxl.c
> +++ b/tools/libxl/libxl.c
> @@ -5590,8 +5590,8 @@ libxl_scheduler libxl_get_scheduler(libxl_ctx *ctx)
>      int r = xc_sched_id(ctx->xch, (int *)&sched);
>      if (r != 0) {
>          LOGE(ERROR, "getting current scheduler id");
> -        return ERROR_FAIL;
>          GC_FREE;
> +        return ERROR_FAIL;
>      }
>      GC_FREE;
>      return sched;

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel

  reply	other threads:[~2016-01-04 16:29 UTC|newest]

Thread overview: 51+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-12-04 20:26 Taking on a Xen development project jtotto
2015-12-10 10:59 ` Wei Liu
2015-12-10 17:23 ` Andrew Cooper
2015-12-12  2:19   ` Yang Hongyang
2015-12-12 22:30     ` Joshua Otto
2015-12-12 23:02       ` Andrew Cooper
2015-12-14 22:49         ` Joshua Otto
2015-12-11 13:52 ` Ian Campbell
2015-12-12 22:07   ` Joshua Otto
2015-12-14 11:08     ` Ian Campbell
2015-12-14 22:59       ` Joshua Otto
2015-12-15 15:48         ` Ian Campbell
2015-12-28  5:16       ` Coverity tidying Joshua Otto
2015-12-28  5:16         ` [PATCH 1/5] libxl: tidy libxl_get_scheduler() according to CODING_STYLE Joshua Otto
2016-01-04 16:23           ` Ian Campbell
2016-01-05  8:20             ` Dario Faggioli
2016-01-19  5:57               ` [PATCH v2 " Chester Lin
2016-01-19  9:14                 ` Dario Faggioli
2016-01-19 11:28                 ` Wei Liu
2016-01-19 11:35                   ` Ian Campbell
2015-12-28  5:16         ` [PATCH 2/5] libxl: make GC_FREE reachable in libxl_get_scheduler() Joshua Otto
2016-01-04 16:29           ` Ian Campbell [this message]
2016-01-05  8:49             ` Dario Faggioli
2016-01-05 11:16               ` Ian Campbell
2016-01-19  5:57                 ` [PATCH v2 " Chester Lin
2016-01-19  9:08                   ` Dario Faggioli
2016-01-19 14:15                   ` Ian Jackson
2015-12-28  5:16         ` [PATCH 3/5] ns16550: widen an integer constant for Coverity Joshua Otto
2016-01-04 16:36           ` Ian Campbell
2016-01-06  9:26             ` Jan Beulich
2016-01-19  5:57               ` [PATCH v2 3/5] n16550: add sanity check for reg_shift Chester Lin
2016-01-19 13:32                 ` Jan Beulich
2016-01-25  0:41                   ` czylin
2015-12-28  5:16         ` [PATCH 4/5] credit: remove pointless local variable initialization Joshua Otto
2015-12-28  5:16         ` [PATCH 5/5] libxl: Add explicit cast to libxl_psr_cat_set_cbm Joshua Otto
2016-01-04 16:40           ` Ian Campbell
2016-01-19  5:58             ` [PATCH v2 " Chester Lin
2016-01-19  8:34               ` Dario Faggioli
2016-01-19 14:06               ` Ian Jackson
2016-01-19 14:21                 ` Ian Campbell
2016-01-19 14:28                   ` Dario Faggioli
2016-01-19 14:33                     ` Ian Jackson
2016-01-19 14:31                   ` George Dunlap
2016-01-19 14:31                 ` Ian Campbell
2016-01-19 14:35                   ` Ian Jackson
2017-01-12 18:08                     ` George Dunlap
2017-01-13  9:05                       ` Dario Faggioli
2015-12-28  9:34         ` Coverity tidying Andrew Cooper
2016-01-01  3:14           ` [PATCH] svm: rephrase local variable use for Coverity Joshua Otto
2016-01-06 13:24             ` Jan Beulich
2016-01-06 14:33               ` Boris Ostrovsky

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=1451924973.13361.183.camel@citrix.com \
    --to=ian.campbell@citrix.com \
    --cc=czylin@uwaterloo.ca \
    --cc=dario.faggioli@citrix.com \
    --cc=george.dunlap@eu.citrix.com \
    --cc=hjarmstr@uwaterloo.ca \
    --cc=ian.jackson@eu.citrix.com \
    --cc=jtotto@uwaterloo.ca \
    --cc=stefano.stabellini@eu.citrix.com \
    --cc=wei.liu2@citrix.com \
    --cc=xen-devel@lists.xen.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;
as well as URLs for NNTP newsgroup(s).