From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932177AbdDQWi2 (ORCPT ); Mon, 17 Apr 2017 18:38:28 -0400 Received: from mx1.redhat.com ([209.132.183.28]:53700 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755404AbdDQWiX (ORCPT ); Mon, 17 Apr 2017 18:38:23 -0400 DMARC-Filter: OpenDMARC Filter v1.3.2 mx1.redhat.com 5EBBFC05973C Authentication-Results: ext-mx08.extmail.prod.ext.phx2.redhat.com; dmarc=none (p=none dis=none) header.from=redhat.com Authentication-Results: ext-mx08.extmail.prod.ext.phx2.redhat.com; spf=pass smtp.mailfrom=alex.williamson@redhat.com DKIM-Filter: OpenDKIM Filter v2.11.0 mx1.redhat.com 5EBBFC05973C Subject: [PATCH v5 3/3] vfio/type1: Reduce repetitive calls in vfio_pin_pages_remote() From: Alex Williamson To: alex.williamson@redhat.com, kvm@vger.kernel.org Cc: eric.auger@redhat.com, kwankhede@nvidia.com, peterx@redhat.com, linux-kernel@vger.kernel.org, slp@redhat.com Date: Mon, 17 Apr 2017 16:38:10 -0600 Message-ID: <20170417223810.25720.68385.stgit@gimli.home> In-Reply-To: <20170417223158.25720.59453.stgit@gimli.home> References: <20170417223158.25720.59453.stgit@gimli.home> User-Agent: StGit/0.17.1-dirty MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.32]); Mon, 17 Apr 2017 22:38:18 +0000 (UTC) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org vfio_pin_pages_remote() is typically called to iterate over a range of memory. Testing CAP_IPC_LOCK is relatively expensive, so it makes sense to push it up to the caller, which can then repeatedly call vfio_pin_pages_remote() using that value. This can show nearly a 20% improvement on the worst case path through VFIO_IOMMU_MAP_DMA with contiguous page mapping disabled. Testing RLIMIT_MEMLOCK is much more lightweight, but we bring it along on the same principle and it does seem to show a marginal improvement. Signed-off-by: Alex Williamson --- drivers/vfio/vfio_iommu_type1.c | 17 +++++++++++------ 1 file changed, 11 insertions(+), 6 deletions(-) diff --git a/drivers/vfio/vfio_iommu_type1.c b/drivers/vfio/vfio_iommu_type1.c index 372e4f626138..8549cb111627 100644 --- a/drivers/vfio/vfio_iommu_type1.c +++ b/drivers/vfio/vfio_iommu_type1.c @@ -380,10 +380,10 @@ static int vaddr_get_pfn(struct mm_struct *mm, unsigned long vaddr, * first page and all consecutive pages with the same locking. */ static long vfio_pin_pages_remote(struct vfio_dma *dma, unsigned long vaddr, - long npage, unsigned long *pfn_base) + long npage, unsigned long *pfn_base, + bool lock_cap, unsigned long limit) { - unsigned long pfn = 0, limit = rlimit(RLIMIT_MEMLOCK) >> PAGE_SHIFT; - bool lock_cap = capable(CAP_IPC_LOCK); + unsigned long pfn = 0; long ret, pinned = 0, lock_acct = 0; bool rsvd; dma_addr_t iova = vaddr - dma->vaddr + dma->iova; @@ -924,13 +924,15 @@ static int vfio_pin_map_dma(struct vfio_iommu *iommu, struct vfio_dma *dma, unsigned long vaddr = dma->vaddr; size_t size = map_size; long npage; - unsigned long pfn; + unsigned long pfn, limit = rlimit(RLIMIT_MEMLOCK) >> PAGE_SHIFT; + bool lock_cap = capable(CAP_IPC_LOCK); int ret = 0; while (size) { /* Pin a contiguous chunk of memory */ npage = vfio_pin_pages_remote(dma, vaddr + dma->size, - size >> PAGE_SHIFT, &pfn); + size >> PAGE_SHIFT, &pfn, + lock_cap, limit); if (npage <= 0) { WARN_ON(!npage); ret = (int)npage; @@ -1040,6 +1042,8 @@ static int vfio_iommu_replay(struct vfio_iommu *iommu, { struct vfio_domain *d; struct rb_node *n; + unsigned long limit = rlimit(RLIMIT_MEMLOCK) >> PAGE_SHIFT; + bool lock_cap = capable(CAP_IPC_LOCK); int ret; /* Arbitrarily pick the first domain in the list for lookups */ @@ -1086,7 +1090,8 @@ static int vfio_iommu_replay(struct vfio_iommu *iommu, npage = vfio_pin_pages_remote(dma, vaddr, n >> PAGE_SHIFT, - &pfn); + &pfn, lock_cap, + limit); if (npage <= 0) { WARN_ON(!npage); ret = (int)npage;