From 135332f34ba2662bc1e32b5c612e06a8cc41a053 Mon Sep 17 00:00:00 2001 From: Javier Martinez Canillas Date: Wed, 4 May 2022 13:59:17 +0200 Subject: Revert "fbdev: Make fb_release() return -ENODEV if fbdev was unregistered" This reverts commit aafa025c76dcc7d1a8c8f0bdefcbe4eb480b2f6a. That commit attempted to fix a NULL pointer dereference, caused by the struct fb_info associated with a framebuffer device to not longer be valid when the file descriptor was closed. The issue was exposed by commit 27599aacbaef ("fbdev: Hot-unplug firmware fb devices on forced removal"), which added a new path that goes through the struct device removal instead of directly unregistering the fb. Most fbdev drivers have issues with the fb_info lifetime, because call to framebuffer_release() from their driver's .remove callback, rather than doing from fbops.fb_destroy callback. This meant that due to this switch, the fb_info was now destroyed too early, while references still existed, while before it was simply leaked. The patch we're reverting here reinstated that leak, hence "fixed" the regression. But the proper solution is to fix the drivers to not release the fb_info too soon. Suggested-by: Daniel Vetter Signed-off-by: Javier Martinez Canillas Reviewed-by: Daniel Vetter Link: https://patchwork.freedesktop.org/patch/msgid/20220504115917.758787-1-javierm@redhat.com --- drivers/video/fbdev/core/fbmem.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/drivers/video/fbdev/core/fbmem.c b/drivers/video/fbdev/core/fbmem.c index 97eb0dee411c..a6bb0e438216 100644 --- a/drivers/video/fbdev/core/fbmem.c +++ b/drivers/video/fbdev/core/fbmem.c @@ -1434,10 +1434,7 @@ fb_release(struct inode *inode, struct file *file) __acquires(&info->lock) __releases(&info->lock) { - struct fb_info * const info = file_fb_info(file); - - if (!info) - return -ENODEV; + struct fb_info * const info = file->private_data; lock_fb_info(info); if (info->fbops->fb_release) -- cgit v1.2.3-58-ga151 From 89bfd4017e58faaf70411555e7f508495114e90b Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Fri, 6 May 2022 00:04:13 +0200 Subject: fbdev: Prevent possible use-after-free in fb_release() Most fbdev drivers have issues with the fb_info lifetime, because call to framebuffer_release() from their driver's .remove callback, rather than doing from fbops.fb_destroy callback. Doing that will destroy the fb_info too early, while references to it may still exist, leading to a use-after-free error. To prevent this, check the fb_info reference counter when attempting to kfree the data structure in framebuffer_release(). That will leak it but at least will prevent the mentioned error. Signed-off-by: Daniel Vetter Signed-off-by: Javier Martinez Canillas Reviewed-by: Thomas Zimmermann Link: https://patchwork.freedesktop.org/patch/msgid/20220505220413.365977-1-javierm@redhat.com --- drivers/video/fbdev/core/fbsysfs.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/video/fbdev/core/fbsysfs.c b/drivers/video/fbdev/core/fbsysfs.c index 26892940c213..82e31a2d845e 100644 --- a/drivers/video/fbdev/core/fbsysfs.c +++ b/drivers/video/fbdev/core/fbsysfs.c @@ -80,6 +80,10 @@ void framebuffer_release(struct fb_info *info) { if (!info) return; + + if (WARN_ON(refcount_read(&info->count))) + return; + kfree(info->apertures); kfree(info); } -- cgit v1.2.3-58-ga151 From 666b90b3ce9e4aac1e1deba266c3a230fb3913b0 Mon Sep 17 00:00:00 2001 From: Javier Martinez Canillas Date: Fri, 6 May 2022 00:04:56 +0200 Subject: fbdev: simplefb: Cleanup fb_info in .fb_destroy rather than .remove The driver is calling framebuffer_release() in its .remove callback, but this will cause the struct fb_info to be freed too early. Since it could be that a reference is still hold to it if user-space opened the fbdev. This would lead to a use-after-free error if the framebuffer device was unregistered but later a user-space process tries to close the fbdev fd. To prevent this, move the framebuffer_release() call to fb_ops.fb_destroy instead of doing it in the driver's .remove callback. Strictly speaking, the code flow in the driver is still wrong because all the hardware cleanupd (i.e: iounmap) should be done in .remove while the software cleanup (i.e: releasing the framebuffer) should be done in the .fb_destroy handler. But this at least makes to match the behavior before commit 27599aacbaef ("fbdev: Hot-unplug firmware fb devices on forced removal"). Fixes: 27599aacbaef ("fbdev: Hot-unplug firmware fb devices on forced removal") Suggested-by: Daniel Vetter Signed-off-by: Javier Martinez Canillas Reviewed-by: Thomas Zimmermann Reviewed-by: Daniel Vetter Link: https://patchwork.freedesktop.org/patch/msgid/20220505220456.366090-1-javierm@redhat.com --- drivers/video/fbdev/simplefb.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/video/fbdev/simplefb.c b/drivers/video/fbdev/simplefb.c index 94fc9c6d0411..2c198561c338 100644 --- a/drivers/video/fbdev/simplefb.c +++ b/drivers/video/fbdev/simplefb.c @@ -84,6 +84,10 @@ struct simplefb_par { static void simplefb_clocks_destroy(struct simplefb_par *par); static void simplefb_regulators_destroy(struct simplefb_par *par); +/* + * fb_ops.fb_destroy is called by the last put_fb_info() call at the end + * of unregister_framebuffer() or fb_release(). Do any cleanup here. + */ static void simplefb_destroy(struct fb_info *info) { struct simplefb_par *par = info->par; @@ -94,6 +98,8 @@ static void simplefb_destroy(struct fb_info *info) if (info->screen_base) iounmap(info->screen_base); + framebuffer_release(info); + if (mem) release_mem_region(mem->start, resource_size(mem)); } @@ -545,8 +551,8 @@ static int simplefb_remove(struct platform_device *pdev) { struct fb_info *info = platform_get_drvdata(pdev); + /* simplefb_destroy takes care of info cleanup */ unregister_framebuffer(info); - framebuffer_release(info); return 0; } -- cgit v1.2.3-58-ga151 From d258d00fb9c7c0cdf9d10c1ded84f10339d2d349 Mon Sep 17 00:00:00 2001 From: Javier Martinez Canillas Date: Fri, 6 May 2022 00:05:40 +0200 Subject: fbdev: efifb: Cleanup fb_info in .fb_destroy rather than .remove The driver is calling framebuffer_release() in its .remove callback, but this will cause the struct fb_info to be freed too early. Since it could be that a reference is still hold to it if user-space opened the fbdev. This would lead to a use-after-free error if the framebuffer device was unregistered but later a user-space process tries to close the fbdev fd. To prevent this, move the framebuffer_release() call to fb_ops.fb_destroy instead of doing it in the driver's .remove callback. Strictly speaking, the code flow in the driver is still wrong because all the hardware cleanupd (i.e: iounmap) should be done in .remove while the software cleanup (i.e: releasing the framebuffer) should be done in the .fb_destroy handler. But this at least makes to match the behavior before commit 27599aacbaef ("fbdev: Hot-unplug firmware fb devices on forced removal"). Fixes: 27599aacbaef ("fbdev: Hot-unplug firmware fb devices on forced removal") Suggested-by: Daniel Vetter Signed-off-by: Javier Martinez Canillas Reviewed-by: Thomas Zimmermann Reviewed-by: Daniel Vetter Link: https://patchwork.freedesktop.org/patch/msgid/20220505220540.366218-1-javierm@redhat.com --- drivers/video/fbdev/efifb.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/drivers/video/fbdev/efifb.c b/drivers/video/fbdev/efifb.c index ea42ba6445b2..cfa3dc0b4eee 100644 --- a/drivers/video/fbdev/efifb.c +++ b/drivers/video/fbdev/efifb.c @@ -243,6 +243,10 @@ error: static inline void efifb_show_boot_graphics(struct fb_info *info) {} #endif +/* + * fb_ops.fb_destroy is called by the last put_fb_info() call at the end + * of unregister_framebuffer() or fb_release(). Do any cleanup here. + */ static void efifb_destroy(struct fb_info *info) { if (efifb_pci_dev) @@ -254,6 +258,9 @@ static void efifb_destroy(struct fb_info *info) else memunmap(info->screen_base); } + + framebuffer_release(info); + if (request_mem_succeeded) release_mem_region(info->apertures->ranges[0].base, info->apertures->ranges[0].size); @@ -620,9 +627,9 @@ static int efifb_remove(struct platform_device *pdev) { struct fb_info *info = platform_get_drvdata(pdev); + /* efifb_destroy takes care of info cleanup */ unregister_framebuffer(info); sysfs_remove_groups(&pdev->dev.kobj, efifb_groups); - framebuffer_release(info); return 0; } -- cgit v1.2.3-58-ga151 From b3c9a924aab61adbc29df110006aa03afe1a78ba Mon Sep 17 00:00:00 2001 From: Javier Martinez Canillas Date: Fri, 6 May 2022 00:06:31 +0200 Subject: fbdev: vesafb: Cleanup fb_info in .fb_destroy rather than .remove The driver is calling framebuffer_release() in its .remove callback, but this will cause the struct fb_info to be freed too early. Since it could be that a reference is still hold to it if user-space opened the fbdev. This would lead to a use-after-free error if the framebuffer device was unregistered but later a user-space process tries to close the fbdev fd. To prevent this, move the framebuffer_release() call to fb_ops.fb_destroy instead of doing it in the driver's .remove callback. Strictly speaking, the code flow in the driver is still wrong because all the hardware cleanupd (i.e: iounmap) should be done in .remove while the software cleanup (i.e: releasing the framebuffer) should be done in the .fb_destroy handler. But this at least makes to match the behavior before commit 27599aacbaef ("fbdev: Hot-unplug firmware fb devices on forced removal"). Fixes: 27599aacbaef ("fbdev: Hot-unplug firmware fb devices on forced removal") Suggested-by: Daniel Vetter Signed-off-by: Javier Martinez Canillas Reviewed-by: Thomas Zimmermann Reviewed-by: Daniel Vetter Link: https://patchwork.freedesktop.org/patch/msgid/20220505220631.366371-1-javierm@redhat.com --- drivers/video/fbdev/vesafb.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/video/fbdev/vesafb.c b/drivers/video/fbdev/vesafb.c index df6de5a9dd4c..e25e8de5ff67 100644 --- a/drivers/video/fbdev/vesafb.c +++ b/drivers/video/fbdev/vesafb.c @@ -179,6 +179,10 @@ static int vesafb_setcolreg(unsigned regno, unsigned red, unsigned green, return err; } +/* + * fb_ops.fb_destroy is called by the last put_fb_info() call at the end + * of unregister_framebuffer() or fb_release(). Do any cleanup here. + */ static void vesafb_destroy(struct fb_info *info) { struct vesafb_par *par = info->par; @@ -188,6 +192,8 @@ static void vesafb_destroy(struct fb_info *info) if (info->screen_base) iounmap(info->screen_base); release_mem_region(info->apertures->ranges[0].base, info->apertures->ranges[0].size); + + framebuffer_release(info); } static struct fb_ops vesafb_ops = { @@ -484,10 +490,10 @@ static int vesafb_remove(struct platform_device *pdev) { struct fb_info *info = platform_get_drvdata(pdev); + /* vesafb_destroy takes care of info cleanup */ unregister_framebuffer(info); if (((struct vesafb_par *)(info->par))->region) release_region(0x3c0, 32); - framebuffer_release(info); return 0; } -- cgit v1.2.3-58-ga151 From 87fd2b091fb33871a7f812658a0971e8e26f903f Mon Sep 17 00:00:00 2001 From: Robin Murphy Date: Tue, 5 Apr 2022 15:21:34 +0100 Subject: drm/nouveau/tegra: Stop using iommu_present() Even if some IOMMU has registered itself on the platform "bus", that doesn't necessarily mean it provides translation for the device we care about. Replace iommu_present() with a more appropriate check. Signed-off-by: Robin Murphy Reviewed-by: Lyude Paul [added cc for stable] Signed-off-by: Lyude Paul Cc: stable@vger.kernel.org # v5.0+ Link: https://patchwork.freedesktop.org/patch/msgid/70d40ea441da3663c2824d54102b471e9a621f8a.1649168494.git.robin.murphy@arm.com --- drivers/gpu/drm/nouveau/nvkm/engine/device/tegra.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/tegra.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/tegra.c index 992cc285f2fe..2ed528c065fa 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/device/tegra.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/tegra.c @@ -123,7 +123,7 @@ nvkm_device_tegra_probe_iommu(struct nvkm_device_tegra *tdev) mutex_init(&tdev->iommu.mutex); - if (iommu_present(&platform_bus_type)) { + if (device_iommu_mapped(dev)) { tdev->iommu.domain = iommu_domain_alloc(&platform_bus_type); if (!tdev->iommu.domain) goto error; -- cgit v1.2.3-58-ga151 From ab244be47a8f111bc82496a8a20c907236e37f95 Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Wed, 9 Feb 2022 07:03:11 +0100 Subject: drm/nouveau: Fix a potential theorical leak in nouveau_get_backlight_name() If successful ida_simple_get() calls are not undone when needed, some additional memory may be allocated and wasted. Here, an ID between 0 and MAX_INT is required. If this ID is >=100, it is not taken into account and is wasted. It should be released. Instead of calling ida_simple_remove(), take advantage of the 'max' parameter to require the ID not to be too big. Should it be too big, it is not allocated and don't need to be freed. While at it, use ida_alloc_xxx()/ida_free() instead to ida_simple_get()/ida_simple_remove(). The latter is deprecated and more verbose. Fixes: db1a0ae21461 ("drm/nouveau/bl: Assign different names to interfaces") Signed-off-by: Christophe JAILLET Reviewed-by: Lyude Paul [Fixed formatting warning from checkpatch] Signed-off-by: Lyude Paul Link: https://patchwork.freedesktop.org/patch/msgid/9ba85bca59df6813dc029e743a836451d5173221.1644386541.git.christophe.jaillet@wanadoo.fr --- drivers/gpu/drm/nouveau/nouveau_backlight.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nouveau_backlight.c b/drivers/gpu/drm/nouveau/nouveau_backlight.c index daf9f87477ba..a2141d3d9b1d 100644 --- a/drivers/gpu/drm/nouveau/nouveau_backlight.c +++ b/drivers/gpu/drm/nouveau/nouveau_backlight.c @@ -46,8 +46,9 @@ static bool nouveau_get_backlight_name(char backlight_name[BL_NAME_SIZE], struct nouveau_backlight *bl) { - const int nb = ida_simple_get(&bl_ida, 0, 0, GFP_KERNEL); - if (nb < 0 || nb >= 100) + const int nb = ida_alloc_max(&bl_ida, 99, GFP_KERNEL); + + if (nb < 0) return false; if (nb > 0) snprintf(backlight_name, BL_NAME_SIZE, "nv_backlight%d", nb); @@ -414,7 +415,7 @@ nouveau_backlight_init(struct drm_connector *connector) nv_encoder, ops, &props); if (IS_ERR(bl->dev)) { if (bl->id >= 0) - ida_simple_remove(&bl_ida, bl->id); + ida_free(&bl_ida, bl->id); ret = PTR_ERR(bl->dev); goto fail_alloc; } @@ -442,7 +443,7 @@ nouveau_backlight_fini(struct drm_connector *connector) return; if (bl->id >= 0) - ida_simple_remove(&bl_ida, bl->id); + ida_free(&bl_ida, bl->id); backlight_device_unregister(bl->dev); nv_conn->backlight = NULL; -- cgit v1.2.3-58-ga151 From 1b5853dfab7fdde450f00f145327342238135c8a Mon Sep 17 00:00:00 2001 From: Javier Martinez Canillas Date: Fri, 6 May 2022 15:22:25 +0200 Subject: fbdev: efifb: Fix a use-after-free due early fb_info cleanup MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Commit d258d00fb9c7 ("fbdev: efifb: Cleanup fb_info in .fb_destroy rather than .remove") attempted to fix a use-after-free error due driver freeing the fb_info in the .remove handler instead of doing it in .fb_destroy. But ironically that change introduced yet another use-after-free since the fb_info was still used after the free. This should fix for good by freeing the fb_info at the end of the handler. Fixes: d258d00fb9c7 ("fbdev: efifb: Cleanup fb_info in .fb_destroy rather than .remove") Reported-by: Ville Syrjälä Reported-by: Andrzej Hajda Signed-off-by: Javier Martinez Canillas Reviewed-by: Andi Shyti Reviewed-by: Andrzej Hajda Reviewed-by: Thomas Zimmermann Signed-off-by: Lucas De Marchi Link: https://patchwork.freedesktop.org/patch/msgid/20220506132225.588379-1-javierm@redhat.com --- drivers/video/fbdev/efifb.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/video/fbdev/efifb.c b/drivers/video/fbdev/efifb.c index cfa3dc0b4eee..b3d5f884c544 100644 --- a/drivers/video/fbdev/efifb.c +++ b/drivers/video/fbdev/efifb.c @@ -259,12 +259,12 @@ static void efifb_destroy(struct fb_info *info) memunmap(info->screen_base); } - framebuffer_release(info); - if (request_mem_succeeded) release_mem_region(info->apertures->ranges[0].base, info->apertures->ranges[0].size); fb_dealloc_cmap(&info->cmap); + + framebuffer_release(info); } static const struct fb_ops efifb_ops = { -- cgit v1.2.3-58-ga151 From ef3a6b70507a2add2cd2e01f5eb9b54d561bacb9 Mon Sep 17 00:00:00 2001 From: Charan Teja Reddy Date: Tue, 10 May 2022 01:19:57 +0530 Subject: dma-buf: call dma_buf_stats_setup after dmabuf is in valid list MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When dma_buf_stats_setup() fails, it closes the dmabuf file which results into the calling of dma_buf_file_release() where it does list_del(&dmabuf->list_node) with out first adding it to the proper list. This is resulting into panic in the below path: __list_del_entry_valid+0x38/0xac dma_buf_file_release+0x74/0x158 __fput+0xf4/0x428 ____fput+0x14/0x24 task_work_run+0x178/0x24c do_notify_resume+0x194/0x264 work_pending+0xc/0x5f0 Fix it by moving the dma_buf_stats_setup() after dmabuf is added to the list. Fixes: bdb8d06dfefd ("dmabuf: Add the capability to expose DMA-BUF stats in sysfs") Signed-off-by: Charan Teja Reddy Tested-by: T.J. Mercier Acked-by: T.J. Mercier Cc: # 5.15.x+ Reviewed-by: Christian König Signed-off-by: Christian König Link: https://patchwork.freedesktop.org/patch/msgid/1652125797-2043-1-git-send-email-quic_charante@quicinc.com --- drivers/dma-buf/dma-buf.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/dma-buf/dma-buf.c b/drivers/dma-buf/dma-buf.c index df23239b04fc..b1e25ae98302 100644 --- a/drivers/dma-buf/dma-buf.c +++ b/drivers/dma-buf/dma-buf.c @@ -543,10 +543,6 @@ struct dma_buf *dma_buf_export(const struct dma_buf_export_info *exp_info) file->f_mode |= FMODE_LSEEK; dmabuf->file = file; - ret = dma_buf_stats_setup(dmabuf); - if (ret) - goto err_sysfs; - mutex_init(&dmabuf->lock); INIT_LIST_HEAD(&dmabuf->attachments); @@ -554,6 +550,10 @@ struct dma_buf *dma_buf_export(const struct dma_buf_export_info *exp_info) list_add(&dmabuf->list_node, &db_list.head); mutex_unlock(&db_list.lock); + ret = dma_buf_stats_setup(dmabuf); + if (ret) + goto err_sysfs; + return dmabuf; err_sysfs: -- cgit v1.2.3-58-ga151 From 6fed53de560768bde6d701a7c79c253b45b259e3 Mon Sep 17 00:00:00 2001 From: Hui Tang Date: Tue, 10 May 2022 21:51:48 +0800 Subject: drm/vc4: hdmi: Fix build error for implicit function declaration MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit drivers/gpu/drm/vc4/vc4_hdmi.c: In function ‘vc4_hdmi_connector_detect’: drivers/gpu/drm/vc4/vc4_hdmi.c:228:7: error: implicit declaration of function ‘gpiod_get_value_cansleep’; did you mean ‘gpio_get_value_cansleep’? [-Werror=implicit-function-declaration] if (gpiod_get_value_cansleep(vc4_hdmi->hpd_gpio)) ^~~~~~~~~~~~~~~~~~~~~~~~ gpio_get_value_cansleep CC [M] drivers/gpu/drm/vc4/vc4_validate.o CC [M] drivers/gpu/drm/vc4/vc4_v3d.o CC [M] drivers/gpu/drm/vc4/vc4_validate_shaders.o CC [M] drivers/gpu/drm/vc4/vc4_debugfs.o drivers/gpu/drm/vc4/vc4_hdmi.c: In function ‘vc4_hdmi_bind’: drivers/gpu/drm/vc4/vc4_hdmi.c:2883:23: error: implicit declaration of function ‘devm_gpiod_get_optional’; did you mean ‘devm_clk_get_optional’? [-Werror=implicit-function-declaration] vc4_hdmi->hpd_gpio = devm_gpiod_get_optional(dev, "hpd", GPIOD_IN); ^~~~~~~~~~~~~~~~~~~~~~~ devm_clk_get_optional drivers/gpu/drm/vc4/vc4_hdmi.c:2883:59: error: ‘GPIOD_IN’ undeclared (first use in this function); did you mean ‘GPIOF_IN’? vc4_hdmi->hpd_gpio = devm_gpiod_get_optional(dev, "hpd", GPIOD_IN); ^~~~~~~~ GPIOF_IN drivers/gpu/drm/vc4/vc4_hdmi.c:2883:59: note: each undeclared identifier is reported only once for each function it appears in cc1: all warnings being treated as errors Fixes: 6800234ceee0 ("drm/vc4: hdmi: Convert to gpiod") Signed-off-by: Hui Tang Signed-off-by: Maxime Ripard Link: https://patchwork.freedesktop.org/patch/msgid/20220510135148.247719-1-tanghui20@huawei.com --- drivers/gpu/drm/vc4/vc4_hdmi.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.c b/drivers/gpu/drm/vc4/vc4_hdmi.c index 6c58b0fd13fb..98b78ec6b37d 100644 --- a/drivers/gpu/drm/vc4/vc4_hdmi.c +++ b/drivers/gpu/drm/vc4/vc4_hdmi.c @@ -38,6 +38,7 @@ #include #include #include +#include #include #include #include -- cgit v1.2.3-58-ga151 From 1d6595b4cd47acfd824550f48f10b54a6f0e93ee Mon Sep 17 00:00:00 2001 From: Zack Rusin Date: Wed, 2 Mar 2022 10:24:22 -0500 Subject: drm/vmwgfx: Fix fencing on SVGAv3 Port of the vmwgfx to SVGAv3 lacked support for fencing. SVGAv3 removed FIFO's and replaced them with command buffers and extra registers. The initial version of SVGAv3 lacked support for most advanced features (e.g. 3D) which made fences unnecessary. That is no longer the case, especially as 3D support is being turned on. Switch from FIFO commands and capabilities to command buffers and extra registers to enable fences on SVGAv3. Fixes: 2cd80dbd3551 ("drm/vmwgfx: Add basic support for SVGA3") Signed-off-by: Zack Rusin Reviewed-by: Martin Krastev Reviewed-by: Maaz Mombasawala Link: https://patchwork.freedesktop.org/patch/msgid/20220302152426.885214-5-zack@kde.org --- drivers/gpu/drm/vmwgfx/vmwgfx_cmd.c | 2 +- drivers/gpu/drm/vmwgfx/vmwgfx_drv.h | 8 ++++++++ drivers/gpu/drm/vmwgfx/vmwgfx_fence.c | 28 +++++++++++++++++++++------- drivers/gpu/drm/vmwgfx/vmwgfx_irq.c | 26 ++++++++++++++++++-------- drivers/gpu/drm/vmwgfx/vmwgfx_kms.c | 8 +++++--- 5 files changed, 53 insertions(+), 19 deletions(-) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_cmd.c b/drivers/gpu/drm/vmwgfx/vmwgfx_cmd.c index a3bfbb6c3e14..bf1b394753da 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_cmd.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_cmd.c @@ -528,7 +528,7 @@ int vmw_cmd_send_fence(struct vmw_private *dev_priv, uint32_t *seqno) *seqno = atomic_add_return(1, &dev_priv->marker_seq); } while (*seqno == 0); - if (!(vmw_fifo_caps(dev_priv) & SVGA_FIFO_CAP_FENCE)) { + if (!vmw_has_fences(dev_priv)) { /* * Don't request hardware to send a fence. The diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h index ea3ecdda561d..6de0b9ef5c77 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h @@ -1679,4 +1679,12 @@ static inline void vmw_irq_status_write(struct vmw_private *vmw, outl(status, vmw->io_start + SVGA_IRQSTATUS_PORT); } +static inline bool vmw_has_fences(struct vmw_private *vmw) +{ + if ((vmw->capabilities & (SVGA_CAP_COMMAND_BUFFERS | + SVGA_CAP_CMD_BUFFERS_2)) != 0) + return true; + return (vmw_fifo_caps(vmw) & SVGA_FIFO_CAP_FENCE) != 0; +} + #endif diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c b/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c index 59d6a2dd4c2e..66cc35dc223e 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c @@ -82,6 +82,22 @@ fman_from_fence(struct vmw_fence_obj *fence) return container_of(fence->base.lock, struct vmw_fence_manager, lock); } +static u32 vmw_fence_goal_read(struct vmw_private *vmw) +{ + if ((vmw->capabilities2 & SVGA_CAP2_EXTRA_REGS) != 0) + return vmw_read(vmw, SVGA_REG_FENCE_GOAL); + else + return vmw_fifo_mem_read(vmw, SVGA_FIFO_FENCE_GOAL); +} + +static void vmw_fence_goal_write(struct vmw_private *vmw, u32 value) +{ + if ((vmw->capabilities2 & SVGA_CAP2_EXTRA_REGS) != 0) + vmw_write(vmw, SVGA_REG_FENCE_GOAL, value); + else + vmw_fifo_mem_write(vmw, SVGA_FIFO_FENCE_GOAL, value); +} + /* * Note on fencing subsystem usage of irqs: * Typically the vmw_fences_update function is called @@ -392,7 +408,7 @@ static bool vmw_fence_goal_new_locked(struct vmw_fence_manager *fman, if (likely(!fman->seqno_valid)) return false; - goal_seqno = vmw_fifo_mem_read(fman->dev_priv, SVGA_FIFO_FENCE_GOAL); + goal_seqno = vmw_fence_goal_read(fman->dev_priv); if (likely(passed_seqno - goal_seqno >= VMW_FENCE_WRAP)) return false; @@ -400,9 +416,8 @@ static bool vmw_fence_goal_new_locked(struct vmw_fence_manager *fman, list_for_each_entry(fence, &fman->fence_list, head) { if (!list_empty(&fence->seq_passed_actions)) { fman->seqno_valid = true; - vmw_fifo_mem_write(fman->dev_priv, - SVGA_FIFO_FENCE_GOAL, - fence->base.seqno); + vmw_fence_goal_write(fman->dev_priv, + fence->base.seqno); break; } } @@ -434,13 +449,12 @@ static bool vmw_fence_goal_check_locked(struct vmw_fence_obj *fence) if (dma_fence_is_signaled_locked(&fence->base)) return false; - goal_seqno = vmw_fifo_mem_read(fman->dev_priv, SVGA_FIFO_FENCE_GOAL); + goal_seqno = vmw_fence_goal_read(fman->dev_priv); if (likely(fman->seqno_valid && goal_seqno - fence->base.seqno < VMW_FENCE_WRAP)) return false; - vmw_fifo_mem_write(fman->dev_priv, SVGA_FIFO_FENCE_GOAL, - fence->base.seqno); + vmw_fence_goal_write(fman->dev_priv, fence->base.seqno); fman->seqno_valid = true; return true; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_irq.c b/drivers/gpu/drm/vmwgfx/vmwgfx_irq.c index c5191de365ca..fe4732bf2c9d 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_irq.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_irq.c @@ -32,6 +32,14 @@ #define VMW_FENCE_WRAP (1 << 24) +static u32 vmw_irqflag_fence_goal(struct vmw_private *vmw) +{ + if ((vmw->capabilities2 & SVGA_CAP2_EXTRA_REGS) != 0) + return SVGA_IRQFLAG_REG_FENCE_GOAL; + else + return SVGA_IRQFLAG_FENCE_GOAL; +} + /** * vmw_thread_fn - Deferred (process context) irq handler * @@ -96,7 +104,7 @@ static irqreturn_t vmw_irq_handler(int irq, void *arg) wake_up_all(&dev_priv->fifo_queue); if ((masked_status & (SVGA_IRQFLAG_ANY_FENCE | - SVGA_IRQFLAG_FENCE_GOAL)) && + vmw_irqflag_fence_goal(dev_priv))) && !test_and_set_bit(VMW_IRQTHREAD_FENCE, dev_priv->irqthread_pending)) ret = IRQ_WAKE_THREAD; @@ -137,8 +145,7 @@ bool vmw_seqno_passed(struct vmw_private *dev_priv, if (likely(dev_priv->last_read_seqno - seqno < VMW_FENCE_WRAP)) return true; - if (!(vmw_fifo_caps(dev_priv) & SVGA_FIFO_CAP_FENCE) && - vmw_fifo_idle(dev_priv, seqno)) + if (!vmw_has_fences(dev_priv) && vmw_fifo_idle(dev_priv, seqno)) return true; /** @@ -160,6 +167,7 @@ int vmw_fallback_wait(struct vmw_private *dev_priv, unsigned long timeout) { struct vmw_fifo_state *fifo_state = dev_priv->fifo; + bool fifo_down = false; uint32_t count = 0; uint32_t signal_seq; @@ -176,12 +184,14 @@ int vmw_fallback_wait(struct vmw_private *dev_priv, */ if (fifo_idle) { - down_read(&fifo_state->rwsem); if (dev_priv->cman) { ret = vmw_cmdbuf_idle(dev_priv->cman, interruptible, 10*HZ); if (ret) goto out_err; + } else if (fifo_state) { + down_read(&fifo_state->rwsem); + fifo_down = true; } } @@ -218,12 +228,12 @@ int vmw_fallback_wait(struct vmw_private *dev_priv, } } finish_wait(&dev_priv->fence_queue, &__wait); - if (ret == 0 && fifo_idle) + if (ret == 0 && fifo_idle && fifo_state) vmw_fence_write(dev_priv, signal_seq); wake_up_all(&dev_priv->fence_queue); out_err: - if (fifo_idle) + if (fifo_down) up_read(&fifo_state->rwsem); return ret; @@ -266,13 +276,13 @@ void vmw_seqno_waiter_remove(struct vmw_private *dev_priv) void vmw_goal_waiter_add(struct vmw_private *dev_priv) { - vmw_generic_waiter_add(dev_priv, SVGA_IRQFLAG_FENCE_GOAL, + vmw_generic_waiter_add(dev_priv, vmw_irqflag_fence_goal(dev_priv), &dev_priv->goal_queue_waiters); } void vmw_goal_waiter_remove(struct vmw_private *dev_priv) { - vmw_generic_waiter_remove(dev_priv, SVGA_IRQFLAG_FENCE_GOAL, + vmw_generic_waiter_remove(dev_priv, vmw_irqflag_fence_goal(dev_priv), &dev_priv->goal_queue_waiters); } diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c index bbd2f4ec08ec..93431e8f6606 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c @@ -1344,7 +1344,6 @@ vmw_kms_new_framebuffer(struct vmw_private *dev_priv, ret = vmw_kms_new_framebuffer_surface(dev_priv, surface, &vfb, mode_cmd, is_bo_proxy); - /* * vmw_create_bo_proxy() adds a reference that is no longer * needed @@ -1385,13 +1384,16 @@ static struct drm_framebuffer *vmw_kms_fb_create(struct drm_device *dev, ret = vmw_user_lookup_handle(dev_priv, file_priv, mode_cmd->handles[0], &surface, &bo); - if (ret) + if (ret) { + DRM_ERROR("Invalid buffer object handle %u (0x%x).\n", + mode_cmd->handles[0], mode_cmd->handles[0]); goto err_out; + } if (!bo && !vmw_kms_srf_ok(dev_priv, mode_cmd->width, mode_cmd->height)) { - DRM_ERROR("Surface size cannot exceed %dx%d", + DRM_ERROR("Surface size cannot exceed %dx%d\n", dev_priv->texture_max_width, dev_priv->texture_max_height); goto err_out; -- cgit v1.2.3-58-ga151 From 3059d9b9f6aa433a55b9d0d21b566396d5497c33 Mon Sep 17 00:00:00 2001 From: Zack Rusin Date: Wed, 2 Mar 2022 10:24:24 -0500 Subject: drm/vmwgfx: Initialize drm_mode_fb_cmd2 Transition to drm_mode_fb_cmd2 from drm_mode_fb_cmd left the structure unitialized. drm_mode_fb_cmd2 adds a few additional members, e.g. flags and modifiers which were never initialized. Garbage in those members can cause random failures during the bringup of the fbcon. Initializing the structure fixes random blank screens after bootup due to flags/modifiers mismatches during the fbcon bring up. Fixes: dabdcdc9822a ("drm/vmwgfx: Switch to mode_cmd2") Signed-off-by: Zack Rusin Cc: Daniel Vetter Cc: # v4.10+ Reviewed-by: Martin Krastev Reviewed-by: Maaz Mombasawala Link: https://patchwork.freedesktop.org/patch/msgid/20220302152426.885214-7-zack@kde.org --- drivers/gpu/drm/vmwgfx/vmwgfx_fb.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_fb.c b/drivers/gpu/drm/vmwgfx/vmwgfx_fb.c index 8ee34576c7d0..adf17c740656 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_fb.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_fb.c @@ -483,7 +483,7 @@ static int vmw_fb_kms_detach(struct vmw_fb_par *par, static int vmw_fb_kms_framebuffer(struct fb_info *info) { - struct drm_mode_fb_cmd2 mode_cmd; + struct drm_mode_fb_cmd2 mode_cmd = {0}; struct vmw_fb_par *par = info->par; struct fb_var_screeninfo *var = &info->var; struct drm_framebuffer *cur_fb; -- cgit v1.2.3-58-ga151 From 21d1d192890ced87f2f04f8f4dea92406e0b162a Mon Sep 17 00:00:00 2001 From: Zack Rusin Date: Fri, 18 Mar 2022 13:43:31 -0400 Subject: drm/vmwgfx: Disable command buffers on svga3 without gbobjects With very limited vram on svga3 it's difficult to handle all the surface migrations. Without gbobjects, i.e. the ability to store surfaces in guest mobs, there's no reason to support intermediate svga2 features, especially because we can fall back to fb traces and svga3 will never support those in-between features. On svga3 we wither want to use fb traces or screen targets (i.e. gbobjects), nothing in between. This fixes presentation on a lot of fusion/esxi tech previews where the exposed svga3 caps haven't been finalized yet. Signed-off-by: Zack Rusin Fixes: 2cd80dbd3551 ("drm/vmwgfx: Add basic support for SVGA3") Cc: # v5.14+ Reviewed-by: Martin Krastev Link: https://patchwork.freedesktop.org/patch/msgid/20220318174332.440068-5-zack@kde.org --- drivers/gpu/drm/vmwgfx/vmwgfx_cmd.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_cmd.c b/drivers/gpu/drm/vmwgfx/vmwgfx_cmd.c index bf1b394753da..162dfeb1cc5a 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_cmd.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_cmd.c @@ -675,11 +675,14 @@ int vmw_cmd_emit_dummy_query(struct vmw_private *dev_priv, */ bool vmw_cmd_supported(struct vmw_private *vmw) { - if ((vmw->capabilities & (SVGA_CAP_COMMAND_BUFFERS | - SVGA_CAP_CMD_BUFFERS_2)) != 0) - return true; + bool has_cmdbufs = + (vmw->capabilities & (SVGA_CAP_COMMAND_BUFFERS | + SVGA_CAP_CMD_BUFFERS_2)) != 0; + if (vmw_is_svga_v3(vmw)) + return (has_cmdbufs && + (vmw->capabilities & SVGA_CAP_GBOBJECTS) != 0); /* * We have FIFO cmd's */ - return vmw->fifo_mem != NULL; + return has_cmdbufs || vmw->fifo_mem != NULL; } -- cgit v1.2.3-58-ga151