Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753369AbcD1PXD (ORCPT ); Thu, 28 Apr 2016 11:23:03 -0400 Received: from asav22.altibox.net ([109.247.116.9]:55608 "EHLO asav22.altibox.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752840AbcD1PS6 (ORCPT ); Thu, 28 Apr 2016 11:18:58 -0400 From: =?UTF-8?q?Noralf=20Tr=C3=B8nnes?= To: dri-devel@lists.freedesktop.org, linux-fbdev@vger.kernel.org Cc: daniel@ffwll.ch, laurent.pinchart@ideasonboard.com, tomi.valkeinen@ti.com, linux-kernel@vger.kernel.org, =?UTF-8?q?Noralf=20Tr=C3=B8nnes?= Subject: [PATCH v4 4/7] fbdev: fb_defio: Export fb_deferred_io_mmap Date: Thu, 28 Apr 2016 17:18:34 +0200 Message-Id: <1461856717-6476-5-git-send-email-noralf@tronnes.org> X-Mailer: git-send-email 2.2.2 In-Reply-To: <1461856717-6476-1-git-send-email-noralf@tronnes.org> References: <1461856717-6476-1-git-send-email-noralf@tronnes.org> MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-CMAE-Score: 0 X-CMAE-Analysis: v=2.1 cv=H7l7u7si c=1 sm=1 tr=0 a=gFHx44SYZz5JQKQKbGEAEQ==:117 a=gFHx44SYZz5JQKQKbGEAEQ==:17 a=L9H7d07YOLsA:10 a=9cW_t1CCXrUA:10 a=s5jvgZ67dGcA:10 a=IkcTkHD0fZMA:10 a=SJz97ENfAAAA:8 a=-4c9a0APCKQ1du7qJrgA:9 a=QEXdDO2ut3YA:10 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2644 Lines: 70 Export fb_deferred_io_mmap so drivers can change vma->vm_page_prot. When the framebuffer memory is allocated using dma_alloc_writecombine() instead of vmalloc(), I get cache syncing problems on ARM. This solves it: static int drm_fbdev_cma_deferred_io_mmap(struct fb_info *info, struct vm_area_struct *vma) { fb_deferred_io_mmap(info, vma); vma->vm_page_prot = pgprot_writecombine(vma->vm_page_prot); return 0; } Could this have been done in the core? Drivers that don't set (struct fb_ops *)->fb_mmap, gets a call to fb_pgprotect() at the end of the default fb_mmap implementation (drivers/video/fbdev/core/fbmem.c). This is an architecture specific function that on many platforms uses pgprot_writecombine(), but not on all. And looking at some of the fb_mmap implementations, some of them sets vm_page_prot to nocache for instance, so I think the safest bet is to do this in the driver and not in the fbdev core. And we can't call fb_pgprotect() from fb_deferred_io_mmap() either because we don't have access to the file pointer that powerpc needs. Signed-off-by: Noralf Trønnes --- Changes since v1: - Expand commit message drivers/video/fbdev/core/fb_defio.c | 3 ++- include/linux/fb.h | 1 + 2 files changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/video/fbdev/core/fb_defio.c b/drivers/video/fbdev/core/fb_defio.c index 57721c7..74b5bca 100644 --- a/drivers/video/fbdev/core/fb_defio.c +++ b/drivers/video/fbdev/core/fb_defio.c @@ -164,7 +164,7 @@ static const struct address_space_operations fb_deferred_io_aops = { .set_page_dirty = fb_deferred_io_set_page_dirty, }; -static int fb_deferred_io_mmap(struct fb_info *info, struct vm_area_struct *vma) +int fb_deferred_io_mmap(struct fb_info *info, struct vm_area_struct *vma) { vma->vm_ops = &fb_deferred_io_vm_ops; vma->vm_flags |= VM_DONTEXPAND | VM_DONTDUMP; @@ -173,6 +173,7 @@ static int fb_deferred_io_mmap(struct fb_info *info, struct vm_area_struct *vma) vma->vm_private_data = info; return 0; } +EXPORT_SYMBOL(fb_deferred_io_mmap); /* workqueue callback */ static void fb_deferred_io_work(struct work_struct *work) diff --git a/include/linux/fb.h b/include/linux/fb.h index dfe8835..a964d07 100644 --- a/include/linux/fb.h +++ b/include/linux/fb.h @@ -673,6 +673,7 @@ static inline void __fb_pad_aligned_buffer(u8 *dst, u32 d_pitch, } /* drivers/video/fb_defio.c */ +int fb_deferred_io_mmap(struct fb_info *info, struct vm_area_struct *vma); extern void fb_deferred_io_init(struct fb_info *info); extern void fb_deferred_io_open(struct fb_info *info, struct inode *inode, -- 2.2.2