From mboxrd@z Thu Jan 1 00:00:00 1970 From: Boris Brezillon Date: Mon, 05 Oct 2015 15:25:16 +0000 Subject: Re: [PATCH] Revert "backlight: pwm: Handle EPROBE_DEFER while requesting the PWM" Message-Id: <20151005172516.59b3ba55@bbrezillon> List-Id: References: <1443295482-18687-1-git-send-email-robert.jarzmik@free.fr> <87bncjemdk.fsf@belgarion.home> <560CE816.1060307@atmel.com> <20151005093543.GB30219@ulmo> <20151005131912.4da38169@bbrezillon> <20151005125803.GA30025@ulmo> <20151005153024.3a245b0b@bbrezillon> <20151005140741.GA10625@ulmo> In-Reply-To: <20151005140741.GA10625@ulmo> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable To: Thierry Reding Cc: Nicolas Ferre , Robert Jarzmik , linux-pwm@vger.kernel.org, Alexandre Belloni , Jingoo Han , Lee Jones , Jean-Christophe Plagniol-Villard , Tomi Valkeinen , linux-fbdev@vger.kernel.org, linux-kernel@vger.kernel.org On Mon, 5 Oct 2015 16:07:41 +0200 Thierry Reding wrote: > On Mon, Oct 05, 2015 at 03:30:24PM +0200, Boris Brezillon wrote: > > On Mon, 5 Oct 2015 14:58:03 +0200 > > Thierry Reding wrote: > >=20 > > > On Mon, Oct 05, 2015 at 01:19:12PM +0200, Boris Brezillon wrote: > > > > Hi Thierry, > > > >=20 > > > > On Mon, 5 Oct 2015 11:35:43 +0200 > > > > Thierry Reding wrote: > > > >=20 > > > > > On Thu, Oct 01, 2015 at 10:00:22AM +0200, Nicolas Ferre wrote: > > > > > > Le 30/09/2015 21:29, Robert Jarzmik a =C3=A9crit : > > > > > > > Robert Jarzmik writes: > > > > > > >=20 > > > > > > >> This reverts commit 68feaca0b13e453aa14ee064c1736202b48b342f. > > > > > > >> This commit breaks legacy platforms, for which : > > > > > > >> (a) no pwm table is added (legacy platforms) > > > > > > >> (b) in this case, in pwm_get(), pmw_lookup_list is empty, a= nd therefore > > > > > > >> chosen =3D NULL, and therefore pwm_get() returns NULL, = and pwm_get() > > > > > > >> returns -EPROBE_DEFER > > > > > > >> (c) as a consequence, this code is unreachable in pwm_bl.c : > > > > > > >> if (IS_ERR(pb->pwm)) { > > > > > > >> ret =3D PTR_ERR(pb->pwm); > > > > > > >> dev_info(&pdev->dev, "%s:%d(): %d\n", __func__, __LINE__, = ret); > > > > > > >> if (ret =3D -EPROBE_DEFER) > > > > > > >> goto err_alloc; > > > > > > >> > > > > > > >> dev_err(&pdev->dev, "unable to request PWM, trying legacy = API\n"); > > > > > > >> pb->legacy =3D true; > > > > > > >> pb->pwm =3D pwm_request(data->pwm_id, "pwm-backlight"); > > > > > > >> > > > > > > >> As this code is unreachable, all legacy platforms relying on= pwm_id are > > > > > > >> broken, amongst which pxa have been tested as broken. > > > > > > >> > > > > > > >> Signed-off-by: Robert Jarzmik > > > > > > > Thierry, would you have a look please ? > > > > > > > As I said before, all legacy platform relying on pwm_id are b= roken. I'd like to > > > > > > > be sure this lands in the next -rc series. > > > > > >=20 > > > > > > Well, as I answered on the linux-pwm mailing-list (I was not in= copy) here: > > > > > > http://article.gmane.org/gmane.linux.pwm/2744 > > > > > > I wonder if it's not easier to fix the platforms and add the pw= m tables... > > > > > >=20 > > > > > > Otherwise, Boris proposed this fix: > > > > > > 8<----------------------------------------------------------- > > > > > > diff --git a/drivers/video/backlight/pwm_bl.c b/drivers/video/b= acklight/pwm_bl.c > > > > > > index eff379b..00483d4 100644 > > > > > > --- a/drivers/video/backlight/pwm_bl.c > > > > > > +++ b/drivers/video/backlight/pwm_bl.c > > > > > > @@ -273,15 +273,15 @@ static int pwm_backlight_probe(struct pla= tform_device *pdev) > > > > > > pb->pwm =3D devm_pwm_get(&pdev->dev, NULL); > > > > > > if (IS_ERR(pb->pwm)) { > > > > > > ret =3D PTR_ERR(pb->pwm); > > > > > > - if (ret =3D -EPROBE_DEFER) > > > > > > - goto err_alloc; > > > > > > =20 > > > > > > dev_err(&pdev->dev, "unable to request PWM, trying legacy AP= I\n"); > > > > > > pb->legacy =3D true; > > > > > > pb->pwm =3D pwm_request(data->pwm_id, "pwm-backlight"); > > > > > > if (IS_ERR(pb->pwm)) { > > > > > > dev_err(&pdev->dev, "unable to request legacy PWM\n"); > > > > > > - ret =3D PTR_ERR(pb->pwm); > > > > > > + if (ret !=3D -EPROBE_DEFER) > > > > > > + ret =3D PTR_ERR(pb->pwm); > > > > > > + > > > > > > goto err_alloc; > > > > > > } > > > > > > } > > > > > >=20 > > > > > > which is not tested and may add an extra non-valid error log. > > > > >=20 > > > > > This is a little risky in my opinion. Not only does it print two = error > > > > > messages for non-legacy platforms (that would be another regressi= on if > > > > > you want to be nit-picking), but it is subtly buggy. If you have a > > > > > system with multiple PWM providers, you could end up failing the = first > > > > > pwm_get() with -EPROBE_DEFER but then continue to the legacy case= , and > > > > > this could succeed because data->pwm_id =3D 0, and that other pro= vider > > > > > could be exporting the PWM with this ID. If I remember correctly = this > > > > > was one of the reasons why the offending commit was merged in the= first > > > > > place. > > > >=20 > > > > Just for the record, when I proposed this fix to Nicolas, I clearly > > > > stated that this was not the way to go, and that fixing the offendi= ng > > > > platforms to use PWM lookup table was the only sane solution, thoug= h I > > > > didn't thought about the invalid PWM id case leading to buggy behav= ior. > > >=20 > > > As chance would have it, this bubbled to the top of my inbox today: > > >=20 > > > http://patchwork.ozlabs.org/patch/483993/ > >=20 > > AFAICT, this is not valid either. This patch is assuming -EPROBE_DEFER > > can only be returned in the DT case, which is not the case: it is also > > returned if the PWMs were declared with a lookup table but the driver > > is not registered yet (module not loaded, or driver registration > > taking place after the PWM backlight driver). >=20 > Right, the non-DT, slightly less legacy case... >=20 > > If we were about to differentiate the missing PWM definition from > > the missing driver case, we should do something like this [1]. > >=20 > > Best Regards, > >=20 > > Boris > >=20 > > [1]http://code.bulix.org/2oozbq-89125 >=20 > Haha, I came up with exactly this earlier and I've been trying to think > of ways in which it could potentially break. >From a quick glance, I don't see any obvious problem in this approach. >=20 > Thierry >=20 > --- >8 --- > From f7fee34e0c414b4268c59e97937c51e0c91a74cf Mon Sep 17 00:00:00 2001 > From: Thierry Reding > Date: Mon, 5 Oct 2015 14:38:32 +0200 > Subject: [PATCH] pwm: Return -ENODEV if no PWM lookup match is found >=20 > When looking up a PWM using the lookup table, assume that all entries > will have been added already, so failure to find a match means that no > corresponding entry has been registered. >=20 > This fixes an issue where -EPROBE_DEFER would be returned if the PWM > lookup table is empty. After this fix, -EPROBE_DEFER is reserved for > situations where no provider has yet registered for a matching entry. >=20 > Signed-off-by: Thierry Reding Not sure it has any value since I proposed the same patch, but here is my Reviewed-by: Boris Brezillon > --- > drivers/pwm/core.c | 4 +++- > 1 file changed, 3 insertions(+), 1 deletion(-) >=20 > diff --git a/drivers/pwm/core.c b/drivers/pwm/core.c > index 3f9df3ea3350..94e5af123660 100644 > --- a/drivers/pwm/core.c > +++ b/drivers/pwm/core.c > @@ -719,8 +719,10 @@ struct pwm_device *pwm_get(struct device *dev, const= char *con_id) > } > } > =20 > - if (!chosen) > + if (!chosen) { > + pwm =3D ERR_PTR(-ENODEV); > goto out; > + } > =20 > chip =3D pwmchip_find_by_name(chosen->provider); > if (!chip) --=20 Boris Brezillon, Free Electrons Embedded Linux and Kernel engineering http://free-electrons.com