> From: Krzysztof Helt <krzysztof.h1@xxxxxxxxx> > Date: Sun, 24 Aug 2008 08:44:34 +0200 > > > Before using the patch below, the rtc m48t59 driver need to be > > updated with two patches from the -mm tree: > > > > rtc-m48t59-reduce-structure-m48t59_private.patch > > rtc-m48t59-add-support-for-m48t02-and-m48t59-chips-2nd-rev.patch > > Actually these patches aren't needed, the first just removes an unused > structure member and the second adds support for chips that sparc64 > systems don't even have :) > I didn't know what rtc chips are available on sparc64 so I did patches to support all types handled in the previous code. > All Sparc64 systems have m48t59 chips, only older 32-bit sparcs have > m48t02 and m48t08. BTW there is a typo in the changelog for that > second patch: > > Add support for two compatible RTC: > - M48T08 which does not have alarm part, > - M48T08 which does not have alarm part and has > only 2KB of NVRAM > > At least one of those lines should say M48T02 :-) > Right. The one with 2KB of NVRAM is M48T02. > Anyways, this means this code on the sparc64 side can be simplified > even further. > I suspected that after conversion to platform drivers the sparc code can be simplified further. > The following revised version of your patch should work. In > particular we don't need a fixed static resource, we can just point > the platform device at the of_device resource object. We also don't > need the silly mostek reg pointers, we can deduce this from the > resource the platform_device object has. > I'll test it on Monday (I am leaving for the weekend). > Actually, this mostek register access inteface is incomplete, the > mostek rtc driver ioremap()'s the resource for MEM resources but > doesn't provide the __iomem pointer to the register accessors because > the mostek driver private structure type is not available in the > header file, it's private to the RTC driver source. > It is available in readb/writeb functions: struct platform_device *pdev = to_platform_device(dev); struct m48t59_private *m48t59 = platform_get_drvdata(pdev); void __iomem *regs = m48t59->ioaddr; The patch can use the private ioaddr instead of just casting resource.start if it is a better solution. > But anyways, ioremap() does nothing but cast pointers on sparc64 so > this happens to work :-) > Ok. Thank you for your review and tips. Regards, Krzysztof --------------------------------------------------------------- Nasilaja sie kradzieze. Mieszkancy osiedli zaniepokojeni. Prosimy o pomoc w tej sprawie >>> http://link.interia.pl/f1eef -- To unsubscribe from this list: send the line "unsubscribe sparclinux" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
