From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dejin Zheng Date: Thu, 23 Apr 2020 15:51:16 +0000 Subject: Re: [PATCH v2] console: console: Complete exception handling in newport_probe() Message-Id: <20200423155116.GA2247@nuc8i5> List-Id: References: <20200423142627.1820-1-zhengdejin5@gmail.com> <081f8192-1708-80ff-6eef-885d72bdf5c5@samsung.com> In-Reply-To: <081f8192-1708-80ff-6eef-885d72bdf5c5@samsung.com> MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 8bit To: Bartlomiej Zolnierkiewicz Cc: linux-fbdev@vger.kernel.org, tsbogend@alpha.franken.de, FlorianSchandinat@gmx.de, gregkh@linuxfoundation.org, linux-kernel@vger.kernel.org, dri-devel@lists.freedesktop.org, Andy Shevchenko , ralf@linux-mips.org, tglx@linutronix.de On Thu, Apr 23, 2020 at 04:55:35PM +0200, Bartlomiej Zolnierkiewicz wrote: > > Hi, > > I believe that the patch summary line should be: > > "[PATCH v2] console: newport_con: ..." > OK, thanks! > On 4/23/20 4:26 PM, Dejin Zheng wrote: > > A call of the function ¡°do_take_over_console¡± can fail here. > > The corresponding system resources were not released then. > > Thus add a call of the function ¡°iounmap¡± together with the check > > of a failure predicate. > > > > Fixes: e84de0c6190503 ("MIPS: GIO bus support for SGI IP22/28") > > I cannot see how this patch fixes commit e84de0c6190503 > (AFAICS npregs has also been leaked on error before)? > yes, you are right, the commit should be e86bb8acc0fdca82d2 ("[PATCH] VT binding: Make newport_con support binding") - move register ioremap from newport_startup() to newport_console_init() > > CC: Andy Shevchenko > > Signed-off-by: Dejin Zheng > > --- > > v1 -> v2: > > - modify the commit comments by Markus'suggestion. > > > > drivers/video/console/newport_con.c | 3 +++ > > 1 file changed, 3 insertions(+) > > > > diff --git a/drivers/video/console/newport_con.c b/drivers/video/console/newport_con.c > > index 00dddf6e08b0..6bfc8e3ffd4a 100644 > > --- a/drivers/video/console/newport_con.c > > +++ b/drivers/video/console/newport_con.c > > @@ -720,6 +720,9 @@ static int newport_probe(struct gio_device *dev, > > console_lock(); > > err = do_take_over_console(&newport_con, 0, MAX_NR_CONSOLES - 1, 1); > > console_unlock(); > > + > > + if (err) > > + iounmap((void *)npregs); > > Looks OK but while you are at it, could you please also add missing > release_mem_region() on error and on device removal: > Ok, Thanks, I will send the patch v3 for it. > newport_addr = dev->resource.start + 0xF0000; > if (!request_mem_region(newport_addr, 0x10000, "Newport")) > return -ENODEV; > > npregs = (struct newport_regs *)/* ioremap cannot fail */ > ioremap(newport_addr, sizeof(struct newport_regs)); > console_lock(); > err = do_take_over_console(&newport_con, 0, MAX_NR_CONSOLES - 1, 1); > console_unlock(); > return err; > } > > static void newport_remove(struct gio_device *dev) > { > give_up_console(&newport_con); > iounmap((void *)npregs); > } > > ? > > > return err; > > } > > > > Best regards, > -- > Bartlomiej Zolnierkiewicz > Samsung R&D Institute Poland > Samsung Electronics