From mboxrd@z Thu Jan 1 00:00:00 1970 From: subashrp@gmail.com (Subash Patel) Date: Thu, 20 Dec 2012 14:19:48 -0800 Subject: [PATCH] ARM: dts: fix split memory bank for SSDK5440 In-Reply-To: <1464138.NEt7mMkYJc@flatron> References: <1356030217-5472-1-git-send-email-kgene.kim@samsung.com> <1464138.NEt7mMkYJc@flatron> Message-ID: <50D38F04.8020307@gmail.com> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org I would like to ask a question here. Do we need to have sparse even if the physical memory is contiguous? All the recent exynos machines come with physical banks without any holes, and I am thinking why not drop it and use flat mem instead. With LPAE these sections sizes wont be useful, and I dont like to keep different section sizes for different configurations. Any suggestions/opinions are very much helpful to me. Regards, Subash On Thursday 20 December 2012 01:14 PM, Tomasz Figa wrote: > Hi Olof, > > On Thursday 20 of December 2012 11:56:59 Olof Johansson wrote: >> Hi, >> >> On Thu, Dec 20, 2012 at 11:03 AM, Kukjin Kim > wrote: >>> The size of memory bank should be under 256MB, because current >>> section size is 256MB on EXYNOS SoCs. This patch fixes it. >> >> This makes no sense. You don't have to split up memory ranges, the >> code should be made to handle it instead. > > It's not Exynos code which causes the problem. Sparsemem initialization > relies on the fact that initial amount of structures to described memory > equals to maximum section size which is defined per arch (e.g. > ARCH_EXYNOS). > >> What's the actual bug caused by this? The description is vague. > > The kernel panics early on NULL pointer dereference in memory > initialization. > > Best regards, > Tomasz Figa > > -- > To unsubscribe from this list: send the line "unsubscribe linux-samsung-soc" in > the body of a message to majordomo at vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html >