From patchwork Fri Jul 17 19:01:03 2015 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: Jerome Glisse X-Patchwork-Id: 6818891 Return-Path: X-Original-To: patchwork-linux-rdma@patchwork.kernel.org Delivered-To: patchwork-parsemail@patchwork2.web.kernel.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.136]) by patchwork2.web.kernel.org (Postfix) with ESMTP id 23DC7C05AC for ; Fri, 17 Jul 2015 19:02:22 +0000 (UTC) Received: from mail.kernel.org (localhost [127.0.0.1]) by mail.kernel.org (Postfix) with ESMTP id A55D7207D3 for ; Fri, 17 Jul 2015 19:02:20 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id EC162207DE for ; Fri, 17 Jul 2015 19:02:18 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755535AbbGQTCB (ORCPT ); Fri, 17 Jul 2015 15:02:01 -0400 Received: from mx1.redhat.com ([209.132.183.28]:37498 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755383AbbGQTBY (ORCPT ); Fri, 17 Jul 2015 15:01:24 -0400 Received: from int-mx14.intmail.prod.int.phx2.redhat.com (int-mx14.intmail.prod.int.phx2.redhat.com [10.5.11.27]) by mx1.redhat.com (Postfix) with ESMTPS id CF9803674CE; Fri, 17 Jul 2015 19:01:23 +0000 (UTC) Received: from localhost.localdomain.com (vpn-56-84.rdu2.redhat.com [10.10.56.84]) by int-mx14.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with ESMTP id t6HJ1AjK032515; Fri, 17 Jul 2015 15:01:22 -0400 From: =?UTF-8?q?J=C3=A9r=C3=B4me=20Glisse?= To: , Cc: Christophe Harle , Duncan Poole , Sherry Cheung , Subhash Gutti , John Hubbard , Mark Hairgrove , Lucien Dunning , Cameron Buschardt , Arvind Gopalakrishnan , Haggai Eran , Shachar Raindel , Liran Liss , =?UTF-8?q?J=C3=A9r=C3=B4me=20Glisse?= Subject: [PATCH 6/8] IB/mlx5/hmm: add mlx5 HMM device initialization and callback v3. Date: Fri, 17 Jul 2015 15:01:03 -0400 Message-Id: <1437159665-6612-7-git-send-email-jglisse@redhat.com> In-Reply-To: <1437159665-6612-1-git-send-email-jglisse@redhat.com> References: <1437159665-6612-1-git-send-email-jglisse@redhat.com> MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.68 on 10.5.11.27 Sender: linux-rdma-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-rdma@vger.kernel.org X-Spam-Status: No, score=-8.1 required=5.0 tests=BAYES_00, RCVD_IN_DNSWL_HI, RP_MATCHES_RCVD, UNPARSEABLE_RELAY autolearn=unavailable version=3.3.1 X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on mail.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP This add the core HMM callback for mlx5 device driver and initialize the HMM device for the mlx5 infiniband device driver. Changed since v1: - Adapt to new hmm_mirror lifetime rules. - HMM_ISDIRTY no longer exist. Changed since v2: - Adapt to HMM page table changes. Signed-off-by: Jérôme Glisse Signed-off-by: John Hubbard --- drivers/infiniband/core/umem_odp.c | 12 ++- drivers/infiniband/hw/mlx5/main.c | 5 + drivers/infiniband/hw/mlx5/mem.c | 38 +++++++- drivers/infiniband/hw/mlx5/mlx5_ib.h | 18 +++- drivers/infiniband/hw/mlx5/mr.c | 8 +- drivers/infiniband/hw/mlx5/odp.c | 178 ++++++++++++++++++++++++++++++++++- include/rdma/ib_umem_odp.h | 20 +++- 7 files changed, 269 insertions(+), 10 deletions(-) diff --git a/drivers/infiniband/core/umem_odp.c b/drivers/infiniband/core/umem_odp.c index ac87ac6..c5e7461 100644 --- a/drivers/infiniband/core/umem_odp.c +++ b/drivers/infiniband/core/umem_odp.c @@ -134,7 +134,7 @@ int ib_umem_odp_get(struct ib_ucontext *context, struct ib_umem *umem) return -ENOMEM; } kref_init(&ib_mirror->kref); - init_rwsem(&ib_mirror->hmm_mr_rwsem); + init_rwsem(&ib_mirror->umem_rwsem); ib_mirror->umem_tree = RB_ROOT; ib_mirror->ib_device = ib_device; @@ -151,10 +151,11 @@ int ib_umem_odp_get(struct ib_ucontext *context, struct ib_umem *umem) context->ib_mirror = ib_mirror_ref(ib_mirror); } mutex_unlock(&ib_device->hmm_mutex); - umem->odp_data.ib_mirror = ib_mirror; + umem->odp_data->ib_mirror = ib_mirror; down_write(&ib_mirror->umem_rwsem); - rbt_ib_umem_insert(&umem->odp_data->interval_tree, &mirror->umem_tree); + rbt_ib_umem_insert(&umem->odp_data->interval_tree, + &ib_mirror->umem_tree); up_write(&ib_mirror->umem_rwsem); mmput(mm); @@ -163,7 +164,7 @@ int ib_umem_odp_get(struct ib_ucontext *context, struct ib_umem *umem) void ib_umem_odp_release(struct ib_umem *umem) { - struct ib_mirror *ib_mirror = umem->odp_data; + struct ib_mirror *ib_mirror = umem->odp_data->ib_mirror; /* * Ensure that no more pages are mapped in the umem. @@ -180,7 +181,8 @@ void ib_umem_odp_release(struct ib_umem *umem) * range covered by one and only one umem while holding the umem rwsem. */ down_write(&ib_mirror->umem_rwsem); - rbt_ib_umem_remove(&umem->odp_data->interval_tree, &mirror->umem_tree); + rbt_ib_umem_remove(&umem->odp_data->interval_tree, + &ib_mirror->umem_tree); up_write(&ib_mirror->umem_rwsem); ib_mirror_unref(ib_mirror); diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c index 5593f22..b731d06 100644 --- a/drivers/infiniband/hw/mlx5/main.c +++ b/drivers/infiniband/hw/mlx5/main.c @@ -1531,6 +1531,9 @@ static void *mlx5_ib_add(struct mlx5_core_dev *mdev) if (err) goto err_rsrc; + /* If HMM initialization fails we just do not enable odp. */ + mlx5_dev_init_odp_hmm(&dev->ib_dev, &mdev->pdev->dev); + err = ib_register_device(&dev->ib_dev, NULL); if (err) goto err_odp; @@ -1555,6 +1558,7 @@ err_umrc: err_dev: ib_unregister_device(&dev->ib_dev); + mlx5_dev_fini_odp_hmm(&dev->ib_dev); err_odp: mlx5_ib_odp_remove_one(dev); @@ -1574,6 +1578,7 @@ static void mlx5_ib_remove(struct mlx5_core_dev *mdev, void *context) ib_unregister_device(&dev->ib_dev); destroy_umrc_res(dev); + mlx5_dev_fini_odp_hmm(&dev->ib_dev); mlx5_ib_odp_remove_one(dev); destroy_dev_resources(&dev->devr); ib_dealloc_device(&dev->ib_dev); diff --git a/drivers/infiniband/hw/mlx5/mem.c b/drivers/infiniband/hw/mlx5/mem.c index 21084c7..8b11d30 100644 --- a/drivers/infiniband/hw/mlx5/mem.c +++ b/drivers/infiniband/hw/mlx5/mem.c @@ -154,6 +154,8 @@ void __mlx5_ib_populate_pas(struct mlx5_ib_dev *dev, struct ib_umem *umem, __be64 *pas, int access_flags, void *data) { unsigned long umem_page_shift = ilog2(umem->page_size); + unsigned long start = ib_umem_start(umem) + (offset << PAGE_SHIFT); + unsigned long end = start + (num_pages << PAGE_SHIFT); int shift = page_shift - umem_page_shift; int mask = (1 << shift) - 1; int i, k; @@ -164,7 +166,41 @@ void __mlx5_ib_populate_pas(struct mlx5_ib_dev *dev, struct ib_umem *umem, int entry; #ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING #ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING_HMM -#error "CONFIG_INFINIBAND_ON_DEMAND_PAGING_HMM not supported at this stage !" + if (umem->odp_data) { + struct ib_mirror *ib_mirror = umem->odp_data->ib_mirror; + struct hmm_mirror *mirror = &ib_mirror->base; + struct hmm_pt_iter *iter = data, local_iter; + unsigned long addr; + + if (iter == NULL) { + iter = &local_iter; + hmm_pt_iter_init(iter, &mirror->pt); + } + + for (i=0, addr=start; i < num_pages; ++i, addr+=PAGE_SIZE) { + unsigned long next = end; + dma_addr_t *ptep, pte; + + /* Get and lock pointer to mirror page table. */ + ptep = hmm_pt_iter_lookup(iter, addr, &next); + pte = ptep ? *ptep : 0; + /* HMM will not have any page tables set up, if this + * function is called before page faults have happened + * on the MR. In that case, we don't have PA's yet, so + * just set each one to zero and continue on. The hw + * will trigger a page fault. + */ + if (hmm_pte_test_valid_dma(&pte)) + pas[i] = cpu_to_be64(umem_dma_to_mtt(pte)); + else + pas[i] = (__be64)0; + } + + if (iter == &local_iter) + hmm_pt_iter_fini(iter); + + return; + } #else /* CONFIG_INFINIBAND_ON_DEMAND_PAGING_HMM */ const bool odp = umem->odp_data != NULL; diff --git a/drivers/infiniband/hw/mlx5/mlx5_ib.h b/drivers/infiniband/hw/mlx5/mlx5_ib.h index 5f1d0dd..83b832e 100644 --- a/drivers/infiniband/hw/mlx5/mlx5_ib.h +++ b/drivers/infiniband/hw/mlx5/mlx5_ib.h @@ -635,6 +635,7 @@ int mlx5_ib_check_mr_status(struct ib_mr *ibmr, u32 check_mask, struct ib_mr_status *mr_status); #ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING + extern struct workqueue_struct *mlx5_ib_page_fault_wq; void mlx5_ib_internal_fill_odp_caps(struct mlx5_ib_dev *dev); @@ -649,12 +650,16 @@ void mlx5_ib_qp_disable_pagefaults(struct mlx5_ib_qp *qp); void mlx5_ib_qp_enable_pagefaults(struct mlx5_ib_qp *qp); #ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING_HMM -#error "CONFIG_INFINIBAND_ON_DEMAND_PAGING_HMM not supported at this stage !" +void mlx5_dev_init_odp_hmm(struct ib_device *ib_dev, struct device *dev); +void mlx5_dev_fini_odp_hmm(struct ib_device *ib_dev); +int mlx5_ib_umem_invalidate(struct ib_umem *umem, u64 start, + u64 end, void *cookie); #else /* CONFIG_INFINIBAND_ON_DEMAND_PAGING_HMM */ void mlx5_ib_invalidate_range(struct ib_umem *umem, unsigned long start, unsigned long end); #endif /* CONFIG_INFINIBAND_ON_DEMAND_PAGING_HMM */ + #else /* CONFIG_INFINIBAND_ON_DEMAND_PAGING */ static inline void mlx5_ib_internal_fill_odp_caps(struct mlx5_ib_dev *dev) { @@ -691,4 +696,15 @@ static inline u8 convert_access(int acc) #define MLX5_MAX_UMR_SHIFT 16 #define MLX5_MAX_UMR_PAGES (1 << MLX5_MAX_UMR_SHIFT) +#ifndef CONFIG_INFINIBAND_ON_DEMAND_PAGING_HMM +static inline void mlx5_dev_init_odp_hmm(struct ib_device *ib_dev, + struct device *dev) +{ +} + +static inline void mlx5_dev_fini_odp_hmm(struct ib_device *ib_dev) +{ +} +#endif /* ! CONFIG_INFINIBAND_ON_DEMAND_PAGING_HMM */ + #endif /* MLX5_IB_H */ diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/infiniband/hw/mlx5/mr.c index 6e9e117..3f3a339 100644 --- a/drivers/infiniband/hw/mlx5/mr.c +++ b/drivers/infiniband/hw/mlx5/mr.c @@ -1209,7 +1209,13 @@ int mlx5_ib_dereg_mr(struct ib_mr *ibmr) /* Wait for all running page-fault handlers to finish. */ synchronize_srcu(&dev->mr_srcu); #ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING_HMM -#error "CONFIG_INFINIBAND_ON_DEMAND_PAGING_HMM not supported at this stage !" + if (mlx5_ib_umem_invalidate(umem, ib_umem_start(umem), + ib_umem_end(umem), NULL)) + /* + * FIXME do something to kill all mr and umem + * in use by this process. + */ + pr_err("killing all mr with odp due to mtt update failure\n"); #else /* CONFIG_INFINIBAND_ON_DEMAND_PAGING_HMM */ /* Destroy all page mappings */ mlx5_ib_invalidate_range(umem, ib_umem_start(umem), diff --git a/drivers/infiniband/hw/mlx5/odp.c b/drivers/infiniband/hw/mlx5/odp.c index b2bf671..d8156ec 100644 --- a/drivers/infiniband/hw/mlx5/odp.c +++ b/drivers/infiniband/hw/mlx5/odp.c @@ -52,8 +52,184 @@ static struct mlx5_ib_mr *mlx5_ib_odp_find_mr_lkey(struct mlx5_ib_dev *dev, return container_of(mmr, struct mlx5_ib_mr, mmr); } + #ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING_HMM -#error "CONFIG_INFINIBAND_ON_DEMAND_PAGING_HMM not supported at this stage !" + + +int mlx5_ib_umem_invalidate(struct ib_umem *umem, u64 start, + u64 end, void *cookie) +{ + const u64 umr_block_mask = (MLX5_UMR_MTT_ALIGNMENT / sizeof(u64)) - 1; + u64 idx = 0, blk_start_idx = 0; + struct hmm_pt_iter iter; + struct mlx5_ib_mr *mlx5_ib_mr; + struct hmm_mirror *mirror; + unsigned long addr; + int in_block = 0; + int ret = 0; + + if (!umem || !umem->odp_data) { + pr_err("invalidation called on NULL umem or non-ODP umem\n"); + return -EINVAL; + } + + /* Is this ib_mr active and registered yet ? */ + if (umem->odp_data->private == NULL) + return 0; + + mlx5_ib_mr = umem->odp_data->private; + if (!mlx5_ib_mr->ibmr.pd) + return 0; + + mirror = &umem->odp_data->ib_mirror->base; + start = max_t(u64, ib_umem_start(umem), start); + end = min_t(u64, ib_umem_end(umem), end); + hmm_pt_iter_init(&iter, &mirror->pt); + + /* + * Iteration one - zap the HW's MTTs. HMM ensures that while we are + * doing the invalidation, no page fault will attempt to overwrite the + * same MTTs. Concurent invalidations might race us, but they will + * write 0s as well, so no difference in the end result. + */ + for (addr = start; addr < end; addr += (u64)umem->page_size) { + unsigned long next = end; + dma_addr_t *ptep; + + /* Get and lock pointer to mirror page table. */ + ptep = hmm_pt_iter_walk(&iter, &addr, &next); + for (; ptep && addr < next; addr += PAGE_SIZE, ptep++) { + idx = (addr - ib_umem_start(umem)) / PAGE_SIZE; + /* + * Strive to write the MTTs in chunks, but avoid + * overwriting non-existing MTTs. The huristic here can + * be improved to estimate the cost of another UMR vs. + * the cost of bigger UMR. + */ + if ((*ptep) & (ODP_READ_ALLOWED_BIT | + ODP_WRITE_ALLOWED_BIT)) { + if ((*ptep) & ODP_WRITE_ALLOWED_BIT) + hmm_pte_set_dirty(ptep); + /* + * Because there can not be concurrent overlapping + * munmap, page migrate, page write protect then it + * is safe here to clear those bits. + */ + hmm_pte_clear_bit(ptep, ODP_READ_ALLOWED_SHIFT); + hmm_pte_clear_bit(ptep, ODP_WRITE_ALLOWED_SHIFT); + if (!in_block) { + blk_start_idx = idx; + in_block = 1; + } + } else { + u64 umr_offset = idx & umr_block_mask; + + if (in_block && umr_offset == 0) { + ret = mlx5_ib_update_mtt(mlx5_ib_mr, + blk_start_idx, + idx - blk_start_idx, + 1, &iter) || ret; + in_block = 0; + } + } + } + } + if (in_block) + ret = mlx5_ib_update_mtt(mlx5_ib_mr, blk_start_idx, + idx - blk_start_idx + 1, 1, + &iter) || ret; + hmm_pt_iter_fini(&iter); + return ret; +} + +static int mlx5_hmm_invalidate_range(struct hmm_mirror *mirror, + unsigned long start, + unsigned long end) +{ + struct ib_mirror *ib_mirror; + int ret; + + ib_mirror = container_of(mirror, struct ib_mirror, base); + + /* Go over all memory region and invalidate them. */ + down_read(&ib_mirror->umem_rwsem); + ret = rbt_ib_umem_for_each_in_range(&ib_mirror->umem_tree, start, end, + mlx5_ib_umem_invalidate, NULL); + up_read(&ib_mirror->umem_rwsem); + return ret; +} + +static void mlx5_hmm_release(struct hmm_mirror *mirror) +{ + struct ib_mirror *ib_mirror; + + ib_mirror = container_of(mirror, struct ib_mirror, base); + + /* Go over all memory region and invalidate them. */ + mlx5_hmm_invalidate_range(mirror, 0, ULLONG_MAX); +} + +static void mlx5_hmm_free(struct hmm_mirror *mirror) +{ + struct ib_mirror *ib_mirror; + + ib_mirror = container_of(mirror, struct ib_mirror, base); + kfree(ib_mirror); +} + +static int mlx5_hmm_update(struct hmm_mirror *mirror, + struct hmm_event *event) +{ + struct device *device = mirror->device->dev; + int ret = 0; + + switch (event->etype) { + case HMM_DEVICE_RFAULT: + case HMM_DEVICE_WFAULT: + /* FIXME implement. */ + break; + case HMM_NONE: + default: + dev_warn(device, "Warning: unhandled HMM event (%d) defaulting to invalidation\n", + event->etype); + /* Fallthrough. */ + /* For write protect and fork we could only invalidate writeable mr. */ + case HMM_WRITE_PROTECT: + case HMM_MIGRATE: + case HMM_MUNMAP: + case HMM_FORK: + ret = mlx5_hmm_invalidate_range(mirror, + event->start, + event->end); + break; + } + + return ret; +} + +static const struct hmm_device_ops mlx5_hmm_ops = { + .release = &mlx5_hmm_release, + .free = &mlx5_hmm_free, + .update = &mlx5_hmm_update, +}; + +void mlx5_dev_init_odp_hmm(struct ib_device *ib_device, struct device *dev) +{ + INIT_LIST_HEAD(&ib_device->ib_mirrors); + ib_device->hmm_dev.dev = dev; + ib_device->hmm_dev.ops = &mlx5_hmm_ops; + ib_device->hmm_ready = !hmm_device_register(&ib_device->hmm_dev); + mutex_init(&ib_device->hmm_mutex); +} + +void mlx5_dev_fini_odp_hmm(struct ib_device *ib_device) +{ + if (!ib_device->hmm_ready) + return; + hmm_device_unregister(&ib_device->hmm_dev); +} + + #else /* CONFIG_INFINIBAND_ON_DEMAND_PAGING_HMM */ diff --git a/include/rdma/ib_umem_odp.h b/include/rdma/ib_umem_odp.h index c7c2670..e982fd3 100644 --- a/include/rdma/ib_umem_odp.h +++ b/include/rdma/ib_umem_odp.h @@ -133,7 +133,25 @@ struct umem_odp_node *rbt_ib_umem_iter_next(struct umem_odp_node *node, #ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING_HMM -#error "CONFIG_INFINIBAND_ON_DEMAND_PAGING_HMM not supported at this stage !" + + +/* + * HMM have few bits reserved for hardware specific bits inside the mirror page + * table. For IB we record the mapping protection per page there. + */ +#define ODP_READ_ALLOWED_SHIFT (HMM_PTE_HW_SHIFT + 0) +#define ODP_WRITE_ALLOWED_SHIFT (HMM_PTE_HW_SHIFT + 1) +#define ODP_READ_ALLOWED_BIT (1 << ODP_READ_ALLOWED_SHIFT) +#define ODP_WRITE_ALLOWED_BIT (1 << ODP_WRITE_ALLOWED_SHIFT) + +/* Make sure we are not overwritting valid address bit on target arch. */ +#if (HMM_PTE_HW_SHIFT + 2) > PAGE_SHIFT +#error (HMM_PTE_HW_SHIFT + 2) > PAGE_SHIFT +#endif + +#define ODP_DMA_ADDR_MASK HMM_PTE_DMA_MASK + + #else /* CONFIG_INFINIBAND_ON_DEMAND_PAGING_HMM */