driverdev-devel.linuxdriverproject.org archive mirror
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: Jerome Pouiller <Jerome.Pouiller@silabs.com>
Cc: "devel@driverdev.osuosl.org" <devel@driverdev.osuosl.org>,
	"kbuild@01.org" <kbuild@01.org>,
	Dan Carpenter <dan.carpenter@oracle.com>,
	"kbuild-all@01.org" <kbuild-all@01.org>
Subject: Re: [staging:staging-testing 57/111] drivers/staging/wfx/scan.c:207 wfx_scan_work() warn: inconsistent returns 'sem:&wvif->scan.lock'.
Date: Thu, 10 Oct 2019 10:42:47 +0200	[thread overview]
Message-ID: <20191010084247.GA365617@kroah.com> (raw)
In-Reply-To: <18946799.ioaMs8cUNb@pc-42>

On Wed, Oct 09, 2019 at 03:07:39PM +0000, Jerome Pouiller wrote:
> On Wednesday 9 October 2019 09:38:31 CEST kbuild test robot wrote:
> > tree:   https://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging.git staging-testing
> > head:   d49d1c76b96ebf39539e93d5ab7943a01ef70e4f
> > commit: 1a61af0f8cbecd1610c6fc380d0fb00f57fd43f2 [57/111] staging: wfx: allow to scan networks
> > 
> > If you fix the issue, kindly add following tag
> > Reported-by: kbuild test robot <lkp@intel.com>
> > Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
> > 
> > New smatch warnings:
> > drivers/staging/wfx/scan.c:207 wfx_scan_work() warn: inconsistent returns 'sem:&wvif->scan.lock'.
> >   Locked on:   line 201
> >   Unlocked on: line 145
> > 
> > # https://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging.git/commit/?id=1a61af0f8cbecd1610c6fc380d0fb00f57fd43f2
> > git remote add staging https://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging.git
> > git remote update staging
> > git checkout 1a61af0f8cbecd1610c6fc380d0fb00f57fd43f2
> > vim +207 drivers/staging/wfx/scan.c
> > 
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  116  void wfx_scan_work(struct work_struct *work)
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  117  {
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  118          struct wfx_vif *wvif = container_of(work, struct wfx_vif, scan.work);
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  119          struct ieee80211_channel **it;
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  120          struct wfx_scan_params scan = {
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  121                  .scan_req.scan_type.type = 0,    /* Foreground */
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  122          };
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  123          struct ieee80211_channel *first;
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  124          int i;
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  125
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  126          down(&wvif->scan.lock);
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  127          mutex_lock(&wvif->wdev->conf_mutex);
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  128
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  129          if (!wvif->scan.req || wvif->scan.curr == wvif->scan.end) {
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  130                  if (wvif->scan.output_power != wvif->wdev->output_power)
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  131                          hif_set_output_power(wvif, wvif->wdev->output_power * 10);
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  132
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  133                  if (wvif->scan.status < 0)
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  134                          dev_warn(wvif->wdev->dev, "scan failed\n");
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  135                  else if (wvif->scan.req)
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  136                          dev_dbg(wvif->wdev->dev, "scan completed\n");
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  137                  else
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  138                          dev_dbg(wvif->wdev->dev, "scan canceled\n");
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  139
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  140                  wvif->scan.req = NULL;
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  141                  wfx_tx_unlock(wvif->wdev);
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  142                  mutex_unlock(&wvif->wdev->conf_mutex);
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  143                  __ieee80211_scan_completed_compat(wvif->wdev->hw, wvif->scan.status ? 1 : 0);
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  144                  up(&wvif->scan.lock);
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  145                  return;
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  146          }
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  147          first = *wvif->scan.curr;
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  148
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  149          for (it = wvif->scan.curr + 1, i = 1;
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  150               it != wvif->scan.end && i < HIF_API_MAX_NB_CHANNELS;
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  151               ++it, ++i) {
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  152                  if ((*it)->band != first->band)
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  153                          break;
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  154                  if (((*it)->flags ^ first->flags) &
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  155                                  IEEE80211_CHAN_NO_IR)
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  156                          break;
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  157                  if (!(first->flags & IEEE80211_CHAN_NO_IR) &&
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  158                      (*it)->max_power != first->max_power)
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  159                          break;
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  160          }
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  161          scan.scan_req.band = first->band;
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  162
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  163          if (wvif->scan.req->no_cck)
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  164                  scan.scan_req.max_transmit_rate = API_RATE_INDEX_G_6MBPS;
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  165          else
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  166                  scan.scan_req.max_transmit_rate = API_RATE_INDEX_B_1MBPS;
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  167          scan.scan_req.num_of_probe_requests =
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  168                  (first->flags & IEEE80211_CHAN_NO_IR) ? 0 : 2;
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  169          scan.scan_req.num_of_ssi_ds = wvif->scan.n_ssids;
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  170          scan.ssids = &wvif->scan.ssids[0];
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  171          scan.scan_req.num_of_channels = it - wvif->scan.curr;
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  172          scan.scan_req.probe_delay = 100;
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  173
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  174          scan.ch = kcalloc(scan.scan_req.num_of_channels, sizeof(u8), GFP_KERNEL);
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  175
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  176          if (!scan.ch) {
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  177                  wvif->scan.status = -ENOMEM;
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  178                  goto fail;
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  179          }
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  180          for (i = 0; i < scan.scan_req.num_of_channels; ++i)
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  181                  scan.ch[i] = wvif->scan.curr[i]->hw_value;
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  182
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  183          if (wvif->scan.curr[0]->flags & IEEE80211_CHAN_NO_IR) {
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  184                  scan.scan_req.min_channel_time = 50;
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  185                  scan.scan_req.max_channel_time = 150;
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  186          } else {
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  187                  scan.scan_req.min_channel_time = 10;
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  188                  scan.scan_req.max_channel_time = 50;
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  189          }
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  190          if (!(first->flags & IEEE80211_CHAN_NO_IR) &&
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  191              wvif->scan.output_power != first->max_power) {
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  192                  wvif->scan.output_power = first->max_power;
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  193                  hif_set_output_power(wvif, wvif->scan.output_power * 10);
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  194          }
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  195          wvif->scan.status = wfx_scan_start(wvif, &scan);
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  196          kfree(scan.ch);
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  197          if (wvif->scan.status)
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  198                  goto fail;
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  199          wvif->scan.curr = it;
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  200          mutex_unlock(&wvif->wdev->conf_mutex);
> > 
> > Smatch says that we should have an:  up(&wvif->scan.lock);.  That seems
> > reasonable, but I'm not positive that it's correct.
> > 
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  201          return;
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  202
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  203  fail:
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  204          wvif->scan.curr = wvif->scan.end;
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  205          mutex_unlock(&wvif->wdev->conf_mutex);
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  206          up(&wvif->scan.lock);
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19 @207          schedule_work(&wvif->scan.work);
> > 1a61af0f8cbecd Jérôme Pouiller 2019-09-19  208  }
> 
> wvif->scan.lock is unlocked when scan finish or timeout (in
> wfx_scan_complete()). So, the current behavior works (even if it
> could/should be written more cleanly).

Yes, please rewrite this to be more "obvious" otherwise you are going to
have to be answering this question for the next 10+ years...

thanks,

greg k-h
_______________________________________________
devel mailing list
devel@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel

      reply	other threads:[~2019-10-10  8:55 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-10-09  7:38 [staging:staging-testing 57/111] drivers/staging/wfx/scan.c:207 wfx_scan_work() warn: inconsistent returns 'sem:&wvif->scan.lock' kbuild test robot
2019-10-09 15:07 ` Jerome Pouiller
2019-10-10  8:42   ` Greg Kroah-Hartman [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=20191010084247.GA365617@kroah.com \
    --to=gregkh@linuxfoundation.org \
    --cc=Jerome.Pouiller@silabs.com \
    --cc=dan.carpenter@oracle.com \
    --cc=devel@driverdev.osuosl.org \
    --cc=kbuild-all@01.org \
    --cc=kbuild@01.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).