From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: [PATCH] block: fix q->max_segment_size checking in blk_recalc_rq_segments about VMERGE Date: Thu, 24 Jul 2008 22:21:52 -0700 (PDT) Message-ID: <20080724.222152.193709632.davem@davemloft.net> References: <20080724.145336.41899163.davem@davemloft.net> <1216957670.4524.87.camel@localhost.localdomain> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <1216957670.4524.87.camel@localhost.localdomain> Sender: linux-parisc-owner@vger.kernel.org To: James.Bottomley@HansenPartnership.com Cc: mpatocka@redhat.com, fujita.tomonori@lab.ntt.co.jp, jens.axboe@oracle.com, linux-kernel@vger.kernel.org, linux-scsi@vger.kernel.org, linux-parisc@vger.kernel.org List-Id: linux-scsi@vger.kernel.org From: James Bottomley Date: Thu, 24 Jul 2008 23:47:50 -0400 > I'm sorry sparc broke, really I am ... but you changed your iommu code > from one with a working vmerge to one without and failed to turn off > vmerging. Partly it wasn't noticed because at 2*PAGE_SIZE you have a > strange vmerge boundary, so it's harder to notice. However, I can't > extrapolate that just because this happened on sparc it will inevitably > happen on all other architectures. The vmerge boundary on sparc64 was 8K which is equal to sparc64's base PAGE_SIZE. I don't know where you get that 2*PAGE_SIZE from.