Hi, Andy, Thanks for the review. > -----Original Message----- > From: linux-media-owner@xxxxxxxxxxxxxxx [mailto:linux-media- > owner@xxxxxxxxxxxxxxx] On Behalf Of Andy Shevchenko > Sent: Saturday, June 17, 2017 12:07 PM > To: Sakari Ailus <sakari.ailus@xxxxxx> > Cc: Zhi, Yong <yong.zhi@xxxxxxxxx>; Linux Media Mailing List <linux- > media@xxxxxxxxxxxxxxx>; sakari.ailus@xxxxxxxxxxxxxxx; Zheng, Jian Xu > <jian.xu.zheng@xxxxxxxxx>; Tomasz Figa <tfiga@xxxxxxxxxxxx>; Mani, > Rajmohan <rajmohan.mani@xxxxxxxxx>; Toivonen, Tuukka > <tuukka.toivonen@xxxxxxxxx> > Subject: Re: [PATCH v2 09/12] intel-ipu3: css hardware setup > > On Sat, Jun 17, 2017 at 9:43 PM, Sakari Ailus <sakari.ailus@xxxxxx> wrote: > > On Sat, Jun 17, 2017 at 01:54:51AM +0300, Andy Shevchenko wrote: > >> On Thu, Jun 15, 2017 at 1:19 AM, Yong Zhi <yong.zhi@xxxxxxxxx> wrote: > > >> > +static void writes(void *mem, ssize_t len, void __iomem *reg) { > >> > + while (len >= 4) { > >> > + writel(*(u32 *)mem, reg); > >> > + mem += 4; > >> > + reg += 4; > >> > + len -= 4; > >> > + } > >> > +} > >> > >> Again, I just looked into patches and first what I see is reinventing the > wheel. > >> > >> memcpy_toio() > > > That doesn't quite work: the hardware only supports 32-bit access. > > > > So the answer is writesl(). > > Makes sense! > We are not able to use writesl() in the past because it's not defined for x86 platform, but now the helper function is available in 4.14, we will use it in next update. > -- > With Best Regards, > Andy Shevchenko
