From patchwork Thu Jan 16 20:38:49 2014 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Thomas Hellstrom X-Patchwork-Id: 3500401 Return-Path: X-Original-To: patchwork-dri-devel@patchwork.kernel.org Delivered-To: patchwork-parsemail@patchwork1.web.kernel.org Received: from mail.kernel.org (mail.kernel.org [198.145.19.201]) by patchwork1.web.kernel.org (Postfix) with ESMTP id 35AB89F169 for ; Thu, 16 Jan 2014 20:40:53 +0000 (UTC) Received: from mail.kernel.org (localhost [127.0.0.1]) by mail.kernel.org (Postfix) with ESMTP id 181862015E for ; Thu, 16 Jan 2014 20:40:52 +0000 (UTC) Received: from gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) by mail.kernel.org (Postfix) with ESMTP id 1690020170 for ; Thu, 16 Jan 2014 20:40:51 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id AD0F7105CFB; Thu, 16 Jan 2014 12:40:37 -0800 (PST) X-Original-To: dri-devel@lists.freedesktop.org Delivered-To: dri-devel@lists.freedesktop.org Received: from smtp-outbound-1.vmware.com (smtp-outbound-1.vmware.com [208.91.2.12]) by gabe.freedesktop.org (Postfix) with ESMTP id 7EE60FC823 for ; Thu, 16 Jan 2014 12:39:51 -0800 (PST) Received: from sc9-mailhost3.vmware.com (sc9-mailhost3.vmware.com [10.113.161.73]) by smtp-outbound-1.vmware.com (Postfix) with ESMTP id F2162283DE; Thu, 16 Jan 2014 12:39:50 -0800 (PST) Received: from zcs-prod-mta-2.vmware.com (zcs-prod-mta-2.vmware.com [10.113.163.64]) by sc9-mailhost3.vmware.com (Postfix) with ESMTP id E9C3840439; Thu, 16 Jan 2014 12:39:50 -0800 (PST) Received: from zcs-prod-mta-2.vmware.com (localhost.localdomain [127.0.0.1]) by zcs-prod-mta-2.vmware.com (Postfix) with ESMTP id B9C22E0041; Thu, 16 Jan 2014 12:39:50 -0800 (PST) Received: from ubuntu.localdomain (unknown [10.113.160.14]) by zcs-prod-mta-2.vmware.com (Postfix) with ESMTPSA; Thu, 16 Jan 2014 12:39:48 -0800 (PST) From: Thomas Hellstrom To: dri-devel@lists.freedesktop.org Subject: [PATCH 01/33] drm/vmwgfx: Fix the driver for large dma addresses Date: Thu, 16 Jan 2014 21:38:49 +0100 Message-Id: <1389904761-16163-2-git-send-email-thellstrom@vmware.com> X-Mailer: git-send-email 1.8.3.2 In-Reply-To: <1389904761-16163-1-git-send-email-thellstrom@vmware.com> References: <1389904761-16163-1-git-send-email-thellstrom@vmware.com> Cc: Thomas Hellstrom , linux-graphics-maintainer@vmware.com, stable@vger.kernel.org X-BeenThere: dri-devel@lists.freedesktop.org X-Mailman-Version: 2.1.13 Precedence: list List-Id: Direct Rendering Infrastructure - Development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , MIME-Version: 1.0 Sender: dri-devel-bounces@lists.freedesktop.org Errors-To: dri-devel-bounces@lists.freedesktop.org X-Spam-Status: No, score=-4.5 required=5.0 tests=BAYES_00, RCVD_IN_DNSWL_MED, 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 With dma compliance / IOMMU support added to the driver in kernel 3.13, the dma addresses can exceed 44 bits, which is what we support in 32-bit mode and with GMR1. So in 32-bit mode and optionally in 64-bit mode, restrict the dma addresses to 44 bits, and strip the old GMR1 code. Signed-off-by: Thomas Hellstrom Reviewed-by: Jakob Bornecrantz Cc: stable@vger.kernel.org --- drivers/gpu/drm/vmwgfx/vmwgfx_drv.c | 47 ++++++++--- drivers/gpu/drm/vmwgfx/vmwgfx_drv.h | 1 - drivers/gpu/drm/vmwgfx/vmwgfx_gmr.c | 160 +----------------------------------- 3 files changed, 39 insertions(+), 169 deletions(-) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c index c7a5496..6c792f7 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c @@ -189,6 +189,7 @@ static int enable_fbdev = IS_ENABLED(CONFIG_DRM_VMWGFX_FBCON); static int vmw_force_iommu; static int vmw_restrict_iommu; static int vmw_force_coherent; +static int vmw_restrict_dma_mask; static int vmw_probe(struct pci_dev *, const struct pci_device_id *); static void vmw_master_init(struct vmw_master *); @@ -203,6 +204,8 @@ MODULE_PARM_DESC(restrict_iommu, "Try to limit IOMMU usage for TTM pages"); module_param_named(restrict_iommu, vmw_restrict_iommu, int, 0600); MODULE_PARM_DESC(force_coherent, "Force coherent TTM pages"); module_param_named(force_coherent, vmw_force_coherent, int, 0600); +MODULE_PARM_DESC(restrict_dma_mask, "Restrict DMA mask to 44 bits with IOMMU"); +module_param_named(restrict_dma_mask, vmw_restrict_dma_mask, int, 0600); static void vmw_print_capabilities(uint32_t capabilities) @@ -510,6 +513,33 @@ out_fixup: return 0; } +/** + * vmw_dma_masks - set required page- and dma masks + * + * @dev: Pointer to struct drm-device + * + * With 32-bit we can only handle 32 bit PFNs. Optionally set that + * restriction also for 64-bit systems. + */ +#ifdef CONFIG_INTEL_IOMMU +static int vmw_dma_masks(struct vmw_private *dev_priv) +{ + struct drm_device *dev = dev_priv->dev; + + if (intel_iommu_enabled && + (sizeof(unsigned long) == 4 || vmw_restrict_dma_mask)) { + DRM_INFO("Restricting DMA addresses to 44 bits.\n"); + return dma_set_mask(dev->dev, DMA_BIT_MASK(44)); + } + return 0; +} +#else +static int vmw_dma_masks(struct vmw_private *dev_priv) +{ + return 0; +} +#endif + static int vmw_driver_load(struct drm_device *dev, unsigned long chipset) { struct vmw_private *dev_priv; @@ -578,14 +608,9 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset) vmw_get_initial_size(dev_priv); - if (dev_priv->capabilities & SVGA_CAP_GMR) { - dev_priv->max_gmr_descriptors = - vmw_read(dev_priv, - SVGA_REG_GMR_MAX_DESCRIPTOR_LENGTH); + if (dev_priv->capabilities & SVGA_CAP_GMR2) { dev_priv->max_gmr_ids = vmw_read(dev_priv, SVGA_REG_GMR_MAX_IDS); - } - if (dev_priv->capabilities & SVGA_CAP_GMR2) { dev_priv->max_gmr_pages = vmw_read(dev_priv, SVGA_REG_GMRS_MAX_PAGES); dev_priv->memory_size = @@ -599,17 +624,17 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset) dev_priv->memory_size = 512*1024*1024; } + ret = vmw_dma_masks(dev_priv); + if (unlikely(ret != 0)) + goto out_err0; + mutex_unlock(&dev_priv->hw_mutex); vmw_print_capabilities(dev_priv->capabilities); - if (dev_priv->capabilities & SVGA_CAP_GMR) { + if (dev_priv->capabilities & SVGA_CAP_GMR2) { DRM_INFO("Max GMR ids is %u\n", (unsigned)dev_priv->max_gmr_ids); - DRM_INFO("Max GMR descriptors is %u\n", - (unsigned)dev_priv->max_gmr_descriptors); - } - if (dev_priv->capabilities & SVGA_CAP_GMR2) { DRM_INFO("Max number of GMR pages is %u\n", (unsigned)dev_priv->max_gmr_pages); DRM_INFO("Max dedicated hypervisor surface memory is %u kiB\n", diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h index 036629d..ba49c1d 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h @@ -290,7 +290,6 @@ struct vmw_private { __le32 __iomem *mmio_virt; int mmio_mtrr; uint32_t capabilities; - uint32_t max_gmr_descriptors; uint32_t max_gmr_ids; uint32_t max_gmr_pages; uint32_t memory_size; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_gmr.c b/drivers/gpu/drm/vmwgfx/vmwgfx_gmr.c index 6ef0b03..61d8d80 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_gmr.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_gmr.c @@ -125,181 +125,27 @@ static void vmw_gmr2_unbind(struct vmw_private *dev_priv, } -static void vmw_gmr_free_descriptors(struct device *dev, dma_addr_t desc_dma, - struct list_head *desc_pages) -{ - struct page *page, *next; - struct svga_guest_mem_descriptor *page_virtual; - unsigned int desc_per_page = PAGE_SIZE / - sizeof(struct svga_guest_mem_descriptor) - 1; - - if (list_empty(desc_pages)) - return; - - list_for_each_entry_safe(page, next, desc_pages, lru) { - list_del_init(&page->lru); - - if (likely(desc_dma != DMA_ADDR_INVALID)) { - dma_unmap_page(dev, desc_dma, PAGE_SIZE, - DMA_TO_DEVICE); - } - - page_virtual = kmap_atomic(page); - desc_dma = (dma_addr_t) - le32_to_cpu(page_virtual[desc_per_page].ppn) << - PAGE_SHIFT; - kunmap_atomic(page_virtual); - - __free_page(page); - } -} - -/** - * FIXME: Adjust to the ttm lowmem / highmem storage to minimize - * the number of used descriptors. - * - */ - -static int vmw_gmr_build_descriptors(struct device *dev, - struct list_head *desc_pages, - struct vmw_piter *iter, - unsigned long num_pages, - dma_addr_t *first_dma) -{ - struct page *page; - struct svga_guest_mem_descriptor *page_virtual = NULL; - struct svga_guest_mem_descriptor *desc_virtual = NULL; - unsigned int desc_per_page; - unsigned long prev_pfn; - unsigned long pfn; - int ret; - dma_addr_t desc_dma; - - desc_per_page = PAGE_SIZE / - sizeof(struct svga_guest_mem_descriptor) - 1; - - while (likely(num_pages != 0)) { - page = alloc_page(__GFP_HIGHMEM); - if (unlikely(page == NULL)) { - ret = -ENOMEM; - goto out_err; - } - - list_add_tail(&page->lru, desc_pages); - page_virtual = kmap_atomic(page); - desc_virtual = page_virtual - 1; - prev_pfn = ~(0UL); - - while (likely(num_pages != 0)) { - pfn = vmw_piter_dma_addr(iter) >> PAGE_SHIFT; - - if (pfn != prev_pfn + 1) { - - if (desc_virtual - page_virtual == - desc_per_page - 1) - break; - - (++desc_virtual)->ppn = cpu_to_le32(pfn); - desc_virtual->num_pages = cpu_to_le32(1); - } else { - uint32_t tmp = - le32_to_cpu(desc_virtual->num_pages); - desc_virtual->num_pages = cpu_to_le32(tmp + 1); - } - prev_pfn = pfn; - --num_pages; - vmw_piter_next(iter); - } - - (++desc_virtual)->ppn = DMA_PAGE_INVALID; - desc_virtual->num_pages = cpu_to_le32(0); - kunmap_atomic(page_virtual); - } - - desc_dma = 0; - list_for_each_entry_reverse(page, desc_pages, lru) { - page_virtual = kmap_atomic(page); - page_virtual[desc_per_page].ppn = cpu_to_le32 - (desc_dma >> PAGE_SHIFT); - kunmap_atomic(page_virtual); - desc_dma = dma_map_page(dev, page, 0, PAGE_SIZE, - DMA_TO_DEVICE); - - if (unlikely(dma_mapping_error(dev, desc_dma))) - goto out_err; - } - *first_dma = desc_dma; - - return 0; -out_err: - vmw_gmr_free_descriptors(dev, DMA_ADDR_INVALID, desc_pages); - return ret; -} - -static void vmw_gmr_fire_descriptors(struct vmw_private *dev_priv, - int gmr_id, dma_addr_t desc_dma) -{ - mutex_lock(&dev_priv->hw_mutex); - - vmw_write(dev_priv, SVGA_REG_GMR_ID, gmr_id); - wmb(); - vmw_write(dev_priv, SVGA_REG_GMR_DESCRIPTOR, desc_dma >> PAGE_SHIFT); - mb(); - - mutex_unlock(&dev_priv->hw_mutex); - -} - int vmw_gmr_bind(struct vmw_private *dev_priv, const struct vmw_sg_table *vsgt, unsigned long num_pages, int gmr_id) { - struct list_head desc_pages; - dma_addr_t desc_dma = 0; - struct device *dev = dev_priv->dev->dev; struct vmw_piter data_iter; - int ret; vmw_piter_start(&data_iter, vsgt, 0); if (unlikely(!vmw_piter_next(&data_iter))) return 0; - if (likely(dev_priv->capabilities & SVGA_CAP_GMR2)) - return vmw_gmr2_bind(dev_priv, &data_iter, num_pages, gmr_id); - - if (unlikely(!(dev_priv->capabilities & SVGA_CAP_GMR))) - return -EINVAL; - - if (vsgt->num_regions > dev_priv->max_gmr_descriptors) + if (unlikely(!(dev_priv->capabilities & SVGA_CAP_GMR2))) return -EINVAL; - INIT_LIST_HEAD(&desc_pages); - - ret = vmw_gmr_build_descriptors(dev, &desc_pages, &data_iter, - num_pages, &desc_dma); - if (unlikely(ret != 0)) - return ret; - - vmw_gmr_fire_descriptors(dev_priv, gmr_id, desc_dma); - vmw_gmr_free_descriptors(dev, desc_dma, &desc_pages); - - return 0; + return vmw_gmr2_bind(dev_priv, &data_iter, num_pages, gmr_id); } void vmw_gmr_unbind(struct vmw_private *dev_priv, int gmr_id) { - if (likely(dev_priv->capabilities & SVGA_CAP_GMR2)) { + if (likely(dev_priv->capabilities & SVGA_CAP_GMR2)) vmw_gmr2_unbind(dev_priv, gmr_id); - return; - } - - mutex_lock(&dev_priv->hw_mutex); - vmw_write(dev_priv, SVGA_REG_GMR_ID, gmr_id); - wmb(); - vmw_write(dev_priv, SVGA_REG_GMR_DESCRIPTOR, 0); - mb(); - mutex_unlock(&dev_priv->hw_mutex); }