[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20170428005615.GN28800@wotan.suse.de>
Date: Fri, 28 Apr 2017 02:56:15 +0200
From: "Luis R. Rodriguez" <mcgrof@...nel.org>
To: Luca Coelho <luca@...lho.fi>
Cc: "Luis R. Rodriguez" <mcgrof@...nel.org>,
gregkh@...uxfoundation.org, wagi@...om.org, dwmw2@...radead.org,
rafal@...ecki.pl, arend.vanspriel@...adcom.com, rjw@...ysocki.net,
yi1.li@...ux.intel.com, atull@...nsource.altera.com,
moritz.fischer@...us.com, pmladek@...e.com,
johannes.berg@...el.com, emmanuel.grumbach@...el.com,
kvalo@...eaurora.org, luto@...nel.org, takahiro.akashi@...aro.org,
dhowells@...hat.com, pjones@...hat.com,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH v6 4/5] iwlwifi: convert to use driver data API
On Mon, Apr 10, 2017 at 04:19:12PM +0300, Luca Coelho wrote:
> On Wed, 2017-03-29 at 20:25 -0700, Luis R. Rodriguez wrote:
> > The driver data API provides support for looking for firmware
> > from a specific set of API ranges, so just use that. Since we
> > free the firmware on the callback immediately after consuming it,
> > this also takes avantage of that feature.
> >
> > Signed-off-by: Luis R. Rodriguez <mcgrof@...nel.org>
> > ---
>
> Looks fine, with one nitpick.
>
>
> > drivers/net/wireless/intel/iwlwifi/iwl-drv.c | 67 ++++++++++------------------
> > 1 file changed, 23 insertions(+), 44 deletions(-)
> >
> > diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
> > index be466a074c1d..b6643aa5b344 100644
> > --- a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
> > +++ b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
>
> [...]
>
> > @@ -1541,11 +1522,9 @@ struct iwl_drv *iwl_drv_start(struct iwl_trans *trans)
> > }
> > #endif
> >
> > - ret = iwl_request_firmware(drv, true);
> > - if (ret) {
> > - IWL_ERR(trans, "Couldn't request the fw\n");
> > + ret = iwl_request_firmware(drv);
> > + if (ret)
> > goto err_fw;
> > - }
>
> Why remove the error message here?
The driver data API now has enough semantics even for async requests so that
an error is either always issued or supressed (optional is true), driver errors
then are superfluous on error now.
Let me know if this is OK.
Luis
Powered by blists - more mailing lists