From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-16.1 required=3.0 tests=BAYES_00,DKIMWL_WL_HIGH, DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,INCLUDES_PATCH,MAILING_LIST_MULTI, SIGNED_OFF_BY,SPF_HELO_NONE,SPF_PASS,URIBL_BLOCKED,USER_AGENT_GIT autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 0EE16C2BBD1 for ; Thu, 17 Sep 2020 11:22:46 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id D081F208DB for ; Thu, 17 Sep 2020 11:22:45 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=default; t=1600341765; bh=iIo/IYh8FlR2GKP24ap+k8mQZvD6y6VaaMdf+PQfaOQ=; h=From:To:Cc:Subject:Date:In-Reply-To:References:List-ID:From; b=cPCU2fhHF+OD4jL+MxMagjRcZQnHHfhL3AxurwmZmUzjcvzoq7LCLGagP/lpS5Acv QKXlhA+se1ZxfWBOEb8MShGNKRNmh+RhArXpQUKleTtdjC6D48p0HJ9sPASEPKDmlg VsXIpu1S6+ysU5j2YfoMFpXC4gJ6dtl5gphN3dkQ= Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1726740AbgIQLW2 (ORCPT ); Thu, 17 Sep 2020 07:22:28 -0400 Received: from mail.kernel.org ([198.145.29.99]:45864 "EHLO mail.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726744AbgIQLWH (ORCPT ); Thu, 17 Sep 2020 07:22:07 -0400 Received: from localhost (unknown [193.47.165.251]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPSA id 2111321974; Thu, 17 Sep 2020 11:22:03 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=default; t=1600341725; bh=iIo/IYh8FlR2GKP24ap+k8mQZvD6y6VaaMdf+PQfaOQ=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=ZpwM8ZeywpmkfujmO515g7a2y9SHOyriN1P/ihpKtPjC3PS4dhh7FA6KL9ufnA7T/ 31nksQZRsE/EXUreCnCt98QZsG6cuN2iuNFiZ89+eKKeSc7USI7QMrzolwruQR51ks w2s+GoHccKn9+3zKS0ZOkMpzJnHKzHGzv8D02jUw= From: Leon Romanovsky To: Doug Ledford , Jason Gunthorpe Cc: Yishai Hadas , linux-rdma@vger.kernel.org, Christoph Hellwig Subject: [PATCH rdma-next v1 2/4] IB/core: Enable ODP sync without faulting Date: Thu, 17 Sep 2020 14:21:50 +0300 Message-Id: <20200917112152.1075974-3-leon@kernel.org> X-Mailer: git-send-email 2.26.2 In-Reply-To: <20200917112152.1075974-1-leon@kernel.org> References: <20200917112152.1075974-1-leon@kernel.org> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Precedence: bulk List-ID: X-Mailing-List: linux-rdma@vger.kernel.org From: Yishai Hadas Enable ODP sync without faulting, this improves performance by reducing the number of page faults in the system. The gain from this option is that the device page table can be aligned with the presented pages in the CPU page table without causing page faults. As of that, the overhead on data path from hardware point of view to trigger a fault which end-up by calling the driver to bring the pages will be dropped. Signed-off-by: Yishai Hadas Signed-off-by: Leon Romanovsky --- drivers/infiniband/core/umem_odp.c | 35 +++++++++++++++++++++--------- drivers/infiniband/hw/mlx5/odp.c | 2 +- include/rdma/ib_umem_odp.h | 2 +- 3 files changed, 27 insertions(+), 12 deletions(-) diff --git a/drivers/infiniband/core/umem_odp.c b/drivers/infiniband/core/umem_odp.c index 3d702aa66d6d..2ba1b46b4c7d 100644 --- a/drivers/infiniband/core/umem_odp.c +++ b/drivers/infiniband/core/umem_odp.c @@ -340,9 +340,10 @@ static int ib_umem_odp_map_dma_single_page( * the return value. * @access_mask: bit mask of the requested access permissions for the given * range. + * @fault: is faulting required for the given range */ int ib_umem_odp_map_dma_and_lock(struct ib_umem_odp *umem_odp, u64 user_virt, - u64 bcnt, u64 access_mask) + u64 bcnt, u64 access_mask, bool fault) __acquires(&umem_odp->umem_mutex) { struct task_struct *owning_process = NULL; @@ -378,10 +379,12 @@ int ib_umem_odp_map_dma_and_lock(struct ib_umem_odp *umem_odp, u64 user_virt, range.end = ALIGN(user_virt + bcnt, 1UL << page_shift); pfn_start_idx = (range.start - ib_umem_start(umem_odp)) >> PAGE_SHIFT; num_pfns = (range.end - range.start) >> PAGE_SHIFT; - range.default_flags = HMM_PFN_REQ_FAULT; + if (fault) { + range.default_flags = HMM_PFN_REQ_FAULT; - if (access_mask & ODP_WRITE_ALLOWED_BIT) - range.default_flags |= HMM_PFN_REQ_WRITE; + if (access_mask & ODP_WRITE_ALLOWED_BIT) + range.default_flags |= HMM_PFN_REQ_WRITE; + } range.hmm_pfns = &(umem_odp->pfn_list[pfn_start_idx]); timeout = jiffies + msecs_to_jiffies(HMM_RANGE_DEFAULT_TIMEOUT); @@ -410,12 +413,24 @@ int ib_umem_odp_map_dma_and_lock(struct ib_umem_odp *umem_odp, u64 user_virt, for (pfn_index = 0; pfn_index < num_pfns; pfn_index += 1 << (page_shift - PAGE_SHIFT), dma_index++) { - /* - * Since we asked for hmm_range_fault() to populate pages, - * it shouldn't return an error entry on success. - */ - WARN_ON(range.hmm_pfns[pfn_index] & HMM_PFN_ERROR); - WARN_ON(!(range.hmm_pfns[pfn_index] & HMM_PFN_VALID)); + + if (fault) { + /* + * Since we asked for hmm_range_fault() to populate + * pages it shouldn't return an error entry on success. + */ + WARN_ON(range.hmm_pfns[pfn_index] & HMM_PFN_ERROR); + WARN_ON(!(range.hmm_pfns[pfn_index] & HMM_PFN_VALID)); + } else { + if (!(range.hmm_pfns[pfn_index] & HMM_PFN_VALID)) { + WARN_ON(umem_odp->dma_list[dma_index]); + continue; + } + access_mask = ODP_READ_ALLOWED_BIT; + if (range.hmm_pfns[pfn_index] & HMM_PFN_WRITE) + access_mask |= ODP_WRITE_ALLOWED_BIT; + } + hmm_order = hmm_pfn_to_map_order(range.hmm_pfns[pfn_index]); /* If a hugepage was detected and ODP wasn't set for, the umem * page_shift will be used, the opposite case is an error. diff --git a/drivers/infiniband/hw/mlx5/odp.c b/drivers/infiniband/hw/mlx5/odp.c index 0f203141a6ad..5bd5e19d76a2 100644 --- a/drivers/infiniband/hw/mlx5/odp.c +++ b/drivers/infiniband/hw/mlx5/odp.c @@ -681,7 +681,7 @@ static int pagefault_real_mr(struct mlx5_ib_mr *mr, struct ib_umem_odp *odp, if (odp->umem.writable && !downgrade) access_mask |= ODP_WRITE_ALLOWED_BIT; - np = ib_umem_odp_map_dma_and_lock(odp, user_va, bcnt, access_mask); + np = ib_umem_odp_map_dma_and_lock(odp, user_va, bcnt, access_mask, true); if (np < 0) return np; diff --git a/include/rdma/ib_umem_odp.h b/include/rdma/ib_umem_odp.h index a53b62ac8a9d..0844c1d05ac6 100644 --- a/include/rdma/ib_umem_odp.h +++ b/include/rdma/ib_umem_odp.h @@ -94,7 +94,7 @@ ib_umem_odp_alloc_child(struct ib_umem_odp *root_umem, unsigned long addr, void ib_umem_odp_release(struct ib_umem_odp *umem_odp); int ib_umem_odp_map_dma_and_lock(struct ib_umem_odp *umem_odp, u64 start_offset, - u64 bcnt, u64 access_mask); + u64 bcnt, u64 access_mask, bool fault); void ib_umem_odp_unmap_dma_pages(struct ib_umem_odp *umem_odp, u64 start_offset, u64 bound); -- 2.26.2