From patchwork Thu Jul 20 10:19:58 2017 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vladimir Murzin X-Patchwork-Id: 9854533 Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork.web.codeaurora.org (Postfix) with ESMTP id C547C602BA for ; Thu, 20 Jul 2017 10:25:29 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id AE34D2872D for ; Thu, 20 Jul 2017 10:25:29 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id A2A9028784; Thu, 20 Jul 2017 10:25:29 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on pdx-wl-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-1.9 required=2.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID autolearn=unavailable version=3.3.1 Received: from bombadil.infradead.org (bombadil.infradead.org [65.50.211.133]) (using TLSv1.2 with cipher AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.wl.linuxfoundation.org (Postfix) with ESMTPS id C18162872D for ; Thu, 20 Jul 2017 10:25:28 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=lists.infradead.org; s=bombadil.20170209; h=Sender: Content-Transfer-Encoding:Content-Type:MIME-Version:Cc:List-Subscribe: List-Help:List-Post:List-Archive:List-Unsubscribe:List-Id:References: In-Reply-To:Message-Id:Date:Subject:To:From:Reply-To:Content-ID: Content-Description:Resent-Date:Resent-From:Resent-Sender:Resent-To:Resent-Cc :Resent-Message-ID:List-Owner; bh=mjJJnJ6fpoJXcaXSJiF3kPYk/KoGtlOlYK29EccZj04=; b=Vlud7F/SctLikOmA3C+Oq5CKW5 uxMPUZiWx6q5vJARNE1HsMGDyoi8pw4t5Rh/T/Gpj1Atw76ADQIgWEMcq+n8PRPgwD9W03crHasi1 nlH+w/25x2QZrekmzPS4rbX36Ud4i0TmLx5vd3TY8eA1BcWEkErLJ6MU06ZZfhT2Emt0Ngpo2Nqw8 RWQ0CdFRdhMibmDtx5kEQuat3Z7b7p3PLBXf6ozwwustuPEQICSGF63mPMxKW5kUVQjCGLKbMrxZz YphHcp7QUEPVGeAvnTteetzec0oYC195fIermA8PAsOxWwFTWC9q8TaH0W2NmchukPoJyKY45RvLE yGuB8YMQ==; Received: from localhost ([127.0.0.1] helo=bombadil.infradead.org) by bombadil.infradead.org with esmtp (Exim 4.87 #1 (Red Hat Linux)) id 1dY8e4-0001e5-GR; Thu, 20 Jul 2017 10:25:24 +0000 Received: from usa-sjc-mx-foss1.foss.arm.com ([217.140.101.70] helo=foss.arm.com) by bombadil.infradead.org with esmtp (Exim 4.87 #1 (Red Hat Linux)) id 1dY8Zb-0005hY-Ip for linux-arm-kernel@lists.infradead.org; Thu, 20 Jul 2017 10:21:02 +0000 Received: from usa-sjc-imap-foss1.foss.arm.com (unknown [10.72.51.249]) by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id 8B24B13D5; Thu, 20 Jul 2017 03:20:30 -0700 (PDT) Received: from bc-e11-3-13.euhpc.arm.com. (bc-e11-3-13.euhpc.arm.com [10.6.16.164]) by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPA id 93C1B3F578; Thu, 20 Jul 2017 03:20:27 -0700 (PDT) From: Vladimir Murzin To: linux-kernel@vger.kernel.org Subject: [PATCH 1/2] drivers: dma-coherent: Introduce interface for default DMA pool Date: Thu, 20 Jul 2017 11:19:58 +0100 Message-Id: <1500545999-17177-2-git-send-email-vladimir.murzin@arm.com> X-Mailer: git-send-email 2.0.0 In-Reply-To: <1500545999-17177-1-git-send-email-vladimir.murzin@arm.com> References: <1500545999-17177-1-git-send-email-vladimir.murzin@arm.com> X-CRM114-Version: 20100106-BlameMichelson ( TRE 0.8.0 (BSD) ) MR-646709E3 X-CRM114-CacheID: sfid-20170720_032047_736989_0742B282 X-CRM114-Status: GOOD ( 20.45 ) X-BeenThere: linux-arm-kernel@lists.infradead.org X-Mailman-Version: 2.1.21 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: vitaly_kuzmichev@mentor.com, Catalin Marinas , sza@esh.hu, arnd@arndb.de, gregkh@linuxfoundation.org, Will Deacon , linux@armlinux.org.uk, Ralf Baechle , george_davis@mentor.com, kbuild-all@01.org, benjamin.gaignard@linaro.org, Vineet Gupta , akpm@linux-foundation.org, m.szyprowski@samsung.com, robin.murphy@arm.com, hch@lst.de, linux-arm-kernel@lists.infradead.org, alexandre.torgue@st.com MIME-Version: 1.0 Sender: "linux-arm-kernel" Errors-To: linux-arm-kernel-bounces+patchwork-linux-arm=patchwork.kernel.org@lists.infradead.org X-Virus-Scanned: ClamAV using ClamSMTP Christoph noticed [1] that default DMA pool in current form overload the DMA coherent infrastructure. In reply, Robin suggested [2] to split the per-device vs. global pool interfaces, so allocation/release from default DMA pool is driven by dma ops implementation. This patch implements Robin's idea and provide interface to allocate/release/mmap the default (aka global) DMA pool. To make it clear that existing *_from_coherent routines work on per-device pool rename them to *_from_dev_coherent. [1] https://lkml.org/lkml/2017/7/7/370 [2] https://lkml.org/lkml/2017/7/7/431 Cc: Vineet Gupta Cc: Russell King Cc: Catalin Marinas Cc: Will Deacon Cc: Ralf Baechle Suggested-by: Robin Murphy Tested-by: Andras Szemzo Signed-off-by: Vladimir Murzin Reviewed-by: Robin Murphy --- arch/arc/mm/dma.c | 2 +- arch/arm/mm/dma-mapping.c | 2 +- arch/arm64/mm/dma-mapping.c | 4 +- arch/mips/mm/dma-default.c | 2 +- drivers/base/dma-coherent.c | 169 +++++++++++++++++++++++++++++--------------- drivers/base/dma-mapping.c | 2 +- include/linux/dma-mapping.h | 40 ++++++++--- 7 files changed, 149 insertions(+), 72 deletions(-) diff --git a/arch/arc/mm/dma.c b/arch/arc/mm/dma.c index 2a07e6e..71d3eff 100644 --- a/arch/arc/mm/dma.c +++ b/arch/arc/mm/dma.c @@ -117,7 +117,7 @@ static int arc_dma_mmap(struct device *dev, struct vm_area_struct *vma, vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot); - if (dma_mmap_from_coherent(dev, vma, cpu_addr, size, &ret)) + if (dma_mmap_from_dev_coherent(dev, vma, cpu_addr, size, &ret)) return ret; if (off < count && user_count <= (count - off)) { diff --git a/arch/arm/mm/dma-mapping.c b/arch/arm/mm/dma-mapping.c index e7380ba..fcf1473 100644 --- a/arch/arm/mm/dma-mapping.c +++ b/arch/arm/mm/dma-mapping.c @@ -851,7 +851,7 @@ static int __arm_dma_mmap(struct device *dev, struct vm_area_struct *vma, unsigned long pfn = dma_to_pfn(dev, dma_addr); unsigned long off = vma->vm_pgoff; - if (dma_mmap_from_coherent(dev, vma, cpu_addr, size, &ret)) + if (dma_mmap_from_dev_coherent(dev, vma, cpu_addr, size, &ret)) return ret; if (off < nr_pages && nr_vma_pages <= (nr_pages - off)) { diff --git a/arch/arm64/mm/dma-mapping.c b/arch/arm64/mm/dma-mapping.c index e90cd1d..f27d4dd 100644 --- a/arch/arm64/mm/dma-mapping.c +++ b/arch/arm64/mm/dma-mapping.c @@ -329,7 +329,7 @@ static int __swiotlb_mmap(struct device *dev, vma->vm_page_prot = __get_dma_pgprot(attrs, vma->vm_page_prot, is_device_dma_coherent(dev)); - if (dma_mmap_from_coherent(dev, vma, cpu_addr, size, &ret)) + if (dma_mmap_from_dev_coherent(dev, vma, cpu_addr, size, &ret)) return ret; return __swiotlb_mmap_pfn(vma, pfn, size); @@ -706,7 +706,7 @@ static int __iommu_mmap_attrs(struct device *dev, struct vm_area_struct *vma, vma->vm_page_prot = __get_dma_pgprot(attrs, vma->vm_page_prot, is_device_dma_coherent(dev)); - if (dma_mmap_from_coherent(dev, vma, cpu_addr, size, &ret)) + if (dma_mmap_from_dev_coherent(dev, vma, cpu_addr, size, &ret)) return ret; if (attrs & DMA_ATTR_FORCE_CONTIGUOUS) { diff --git a/arch/mips/mm/dma-default.c b/arch/mips/mm/dma-default.c index e08598c..8e78251 100644 --- a/arch/mips/mm/dma-default.c +++ b/arch/mips/mm/dma-default.c @@ -232,7 +232,7 @@ static int mips_dma_mmap(struct device *dev, struct vm_area_struct *vma, else vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot); - if (dma_mmap_from_coherent(dev, vma, cpu_addr, size, &ret)) + if (dma_mmap_from_dev_coherent(dev, vma, cpu_addr, size, &ret)) return ret; if (off < count && user_count <= (count - off)) { diff --git a/drivers/base/dma-coherent.c b/drivers/base/dma-coherent.c index 2ae24c2..d6f231c 100644 --- a/drivers/base/dma-coherent.c +++ b/drivers/base/dma-coherent.c @@ -25,7 +25,7 @@ static inline struct dma_coherent_mem *dev_get_coherent_memory(struct device *de { if (dev && dev->dma_mem) return dev->dma_mem; - return dma_coherent_default_memory; + return NULL; } static inline dma_addr_t dma_get_device_base(struct device *dev, @@ -165,34 +165,15 @@ void *dma_mark_declared_memory_occupied(struct device *dev, } EXPORT_SYMBOL(dma_mark_declared_memory_occupied); -/** - * dma_alloc_from_coherent() - try to allocate memory from the per-device coherent area - * - * @dev: device from which we allocate memory - * @size: size of requested memory area - * @dma_handle: This will be filled with the correct dma handle - * @ret: This pointer will be filled with the virtual address - * to allocated area. - * - * This function should be only called from per-arch dma_alloc_coherent() - * to support allocation from per-device coherent memory pools. - * - * Returns 0 if dma_alloc_coherent should continue with allocating from - * generic memory areas, or !0 if dma_alloc_coherent should return @ret. - */ -int dma_alloc_from_coherent(struct device *dev, ssize_t size, - dma_addr_t *dma_handle, void **ret) +static void *__dma_alloc_from_dev_coherent(struct dma_coherent_mem *mem, + ssize_t size, dma_addr_t *dma_handle) { - struct dma_coherent_mem *mem = dev_get_coherent_memory(dev); int order = get_order(size); unsigned long flags; int pageno; int dma_memory_map; + void *ret; - if (!mem) - return 0; - - *ret = NULL; spin_lock_irqsave(&mem->spinlock, flags); if (unlikely(size > (mem->size << PAGE_SHIFT))) @@ -203,21 +184,51 @@ int dma_alloc_from_coherent(struct device *dev, ssize_t size, goto err; /* - * Memory was found in the per-device area. + * Memory was found in the coherent area. */ - *dma_handle = dma_get_device_base(dev, mem) + (pageno << PAGE_SHIFT); - *ret = mem->virt_base + (pageno << PAGE_SHIFT); + *dma_handle = mem->device_base + (pageno << PAGE_SHIFT); + ret = mem->virt_base + (pageno << PAGE_SHIFT); dma_memory_map = (mem->flags & DMA_MEMORY_MAP); spin_unlock_irqrestore(&mem->spinlock, flags); if (dma_memory_map) - memset(*ret, 0, size); + memset(ret, 0, size); else - memset_io(*ret, 0, size); + memset_io(ret, 0, size); - return 1; + return ret; err: spin_unlock_irqrestore(&mem->spinlock, flags); + return NULL; +} + +/** + * dma_alloc_from_dev_coherent() - try to allocate memory from the per-device coherent area + * + * @dev: device from which we allocate memory + * @size: size of requested memory area + * @dma_handle: This will be filled with the correct dma handle + * @ret: This pointer will be filled with the virtual address + * to allocated area. + * + * This function should be only called from per-arch dma_alloc_coherent() + * to support allocation from per-device coherent memory pools. + * + * Returns 0 if dma_alloc_coherent should continue with allocating from + * generic memory areas, or !0 if dma_alloc_coherent should return @ret. + */ +int dma_alloc_from_dev_coherent(struct device *dev, ssize_t size, + dma_addr_t *dma_handle, void **ret) +{ + struct dma_coherent_mem *mem = dev_get_coherent_memory(dev); + + if (!mem) + return 0; + + *ret = __dma_alloc_from_dev_coherent(mem, size, dma_handle); + if (*ret) + return 1; + /* * In the case where the allocation can not be satisfied from the * per-device area, try to fall back to generic memory if the @@ -225,25 +236,20 @@ int dma_alloc_from_coherent(struct device *dev, ssize_t size, */ return mem->flags & DMA_MEMORY_EXCLUSIVE; } -EXPORT_SYMBOL(dma_alloc_from_coherent); +EXPORT_SYMBOL(dma_alloc_from_dev_coherent); -/** - * dma_release_from_coherent() - try to free the memory allocated from per-device coherent memory pool - * @dev: device from which the memory was allocated - * @order: the order of pages allocated - * @vaddr: virtual address of allocated pages - * - * This checks whether the memory was allocated from the per-device - * coherent memory pool and if so, releases that memory. - * - * Returns 1 if we correctly released the memory, or 0 if - * dma_release_coherent() should proceed with releasing memory from - * generic pools. - */ -int dma_release_from_coherent(struct device *dev, int order, void *vaddr) +void *dma_alloc_from_global_coherent(ssize_t size, dma_addr_t *dma_handle) { - struct dma_coherent_mem *mem = dev_get_coherent_memory(dev); + if (!dma_coherent_default_memory) + return NULL; + return __dma_alloc_from_dev_coherent(dma_coherent_default_memory, size, dma_handle); +} + + +static int __dma_release_from_dev_coherent(struct dma_coherent_mem *mem, + int order, void *vaddr) +{ if (mem && vaddr >= mem->virt_base && vaddr < (mem->virt_base + (mem->size << PAGE_SHIFT))) { int page = (vaddr - mem->virt_base) >> PAGE_SHIFT; @@ -256,28 +262,42 @@ int dma_release_from_coherent(struct device *dev, int order, void *vaddr) } return 0; } -EXPORT_SYMBOL(dma_release_from_coherent); /** - * dma_mmap_from_coherent() - try to mmap the memory allocated from - * per-device coherent memory pool to userspace + * dma_release_from_dev_coherent() - try to free the memory allocated from per-device coherent memory pool * @dev: device from which the memory was allocated - * @vma: vm_area for the userspace memory - * @vaddr: cpu address returned by dma_alloc_from_coherent - * @size: size of the memory buffer allocated by dma_alloc_from_coherent - * @ret: result from remap_pfn_range() + * @order: the order of pages allocated + * @vaddr: virtual address of allocated pages * * This checks whether the memory was allocated from the per-device - * coherent memory pool and if so, maps that memory to the provided vma. + * coherent memory pool and if so, releases that memory. * - * Returns 1 if we correctly mapped the memory, or 0 if the caller should - * proceed with mapping memory from generic pools. + * Returns 1 if we correctly released the memory, or 0 if + * dma_release_coherent() should proceed with releasing memory from + * generic pools. */ -int dma_mmap_from_coherent(struct device *dev, struct vm_area_struct *vma, - void *vaddr, size_t size, int *ret) +int dma_release_from_dev_coherent(struct device *dev, int order, void *vaddr) { struct dma_coherent_mem *mem = dev_get_coherent_memory(dev); + return __dma_release_from_dev_coherent(mem, order, vaddr); +} +EXPORT_SYMBOL(dma_release_from_dev_coherent); + + +int dma_release_from_global_coherent(int order, void *vaddr) +{ + if (!dma_coherent_default_memory) + return 0; + + return __dma_release_from_dev_coherent(dma_coherent_default_memory, + order, vaddr); +} + +static int __dma_mmap_from_dev_coherent(struct dma_coherent_mem *mem, + struct vm_area_struct *vma, void *vaddr, + size_t size, int *ret) +{ if (mem && vaddr >= mem->virt_base && vaddr + size <= (mem->virt_base + (mem->size << PAGE_SHIFT))) { unsigned long off = vma->vm_pgoff; @@ -296,7 +316,40 @@ int dma_mmap_from_coherent(struct device *dev, struct vm_area_struct *vma, } return 0; } -EXPORT_SYMBOL(dma_mmap_from_coherent); + +/** + * dma_mmap_from_dev_coherent() - try to mmap the memory allocated from + * per-device coherent memory pool to userspace + * @dev: device from which the memory was allocated + * @vma: vm_area for the userspace memory + * @vaddr: cpu address returned by dma_alloc_from_dev_coherent + * @size: size of the memory buffer allocated by dma_alloc_from_dev_coherent + * @ret: result from remap_pfn_range() + * + * This checks whether the memory was allocated from the per-device + * coherent memory pool and if so, maps that memory to the provided vma. + * + * Returns 1 if we correctly mapped the memory, or 0 if the caller should + * proceed with mapping memory from generic pools. + */ +int dma_mmap_from_dev_coherent(struct device *dev, struct vm_area_struct *vma, + void *vaddr, size_t size, int *ret) +{ + struct dma_coherent_mem *mem = dev_get_coherent_memory(dev); + + return __dma_mmap_from_dev_coherent(mem, vma, vaddr, size, ret); +} +EXPORT_SYMBOL(dma_mmap_from_dev_coherent); + +int dma_mmap_from_global_coherent(struct vm_area_struct *vma, void *vaddr, + size_t size, int *ret) +{ + if (!dma_coherent_default_memory) + return 0; + + return __dma_mmap_from_dev_coherent(dma_coherent_default_memory, vma, + vaddr, size, ret); +} /* * Support for reserved memory regions defined in device tree diff --git a/drivers/base/dma-mapping.c b/drivers/base/dma-mapping.c index 5096755..b555ff9 100644 --- a/drivers/base/dma-mapping.c +++ b/drivers/base/dma-mapping.c @@ -235,7 +235,7 @@ int dma_common_mmap(struct device *dev, struct vm_area_struct *vma, vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot); - if (dma_mmap_from_coherent(dev, vma, cpu_addr, size, &ret)) + if (dma_mmap_from_dev_coherent(dev, vma, cpu_addr, size, &ret)) return ret; if (off < count && user_count <= (count - off)) { diff --git a/include/linux/dma-mapping.h b/include/linux/dma-mapping.h index 843ab86..03c0196 100644 --- a/include/linux/dma-mapping.h +++ b/include/linux/dma-mapping.h @@ -157,16 +157,40 @@ static inline int is_device_dma_capable(struct device *dev) * These three functions are only for dma allocator. * Don't use them in device drivers. */ -int dma_alloc_from_coherent(struct device *dev, ssize_t size, +int dma_alloc_from_dev_coherent(struct device *dev, ssize_t size, dma_addr_t *dma_handle, void **ret); -int dma_release_from_coherent(struct device *dev, int order, void *vaddr); +int dma_release_from_dev_coherent(struct device *dev, int order, void *vaddr); -int dma_mmap_from_coherent(struct device *dev, struct vm_area_struct *vma, +int dma_mmap_from_dev_coherent(struct device *dev, struct vm_area_struct *vma, void *cpu_addr, size_t size, int *ret); + +void *dma_alloc_from_global_coherent(ssize_t size, dma_addr_t *dma_handle); +int dma_release_from_global_coherent(int order, void *vaddr); +int dma_mmap_from_global_coherent(struct vm_area_struct *vma, void *cpu_addr, + size_t size, int *ret); + #else -#define dma_alloc_from_coherent(dev, size, handle, ret) (0) -#define dma_release_from_coherent(dev, order, vaddr) (0) -#define dma_mmap_from_coherent(dev, vma, vaddr, order, ret) (0) +#define dma_alloc_from_dev_coherent(dev, size, handle, ret) (0) +#define dma_release_from_dev_coherent(dev, order, vaddr) (0) +#define dma_mmap_from_dev_coherent(dev, vma, vaddr, order, ret) (0) + +static inline void *dma_alloc_from_global_coherent(ssize_t size, + dma_addr_t *dma_handle) +{ + return NULL; +} + +static inline int dma_release_from_global_coherent(int order, void *vaddr) +{ + return 0; +} + +static inline int dma_mmap_from_global_coherent(struct vm_area_struct *vma, + void *cpu_addr, size_t size, + int *ret) +{ + return 0; +} #endif /* CONFIG_HAVE_GENERIC_DMA_COHERENT */ #ifdef CONFIG_HAS_DMA @@ -481,7 +505,7 @@ static inline void *dma_alloc_attrs(struct device *dev, size_t size, BUG_ON(!ops); - if (dma_alloc_from_coherent(dev, size, dma_handle, &cpu_addr)) + if (dma_alloc_from_dev_coherent(dev, size, dma_handle, &cpu_addr)) return cpu_addr; if (!arch_dma_alloc_attrs(&dev, &flag)) @@ -503,7 +527,7 @@ static inline void dma_free_attrs(struct device *dev, size_t size, BUG_ON(!ops); WARN_ON(irqs_disabled()); - if (dma_release_from_coherent(dev, get_order(size), cpu_addr)) + if (dma_release_from_dev_coherent(dev, get_order(size), cpu_addr)) return; if (!ops->free || !cpu_addr)