From: Greg Kroah-Hartman Date: Wed, 26 Oct 2022 14:46:51 +0000 (+0200) Subject: 6.0-stable patches X-Git-Tag: v5.10.151~60 X-Git-Url: http://git.ipfire.org/gitweb.cgi?a=commitdiff_plain;h=a94157b761a943d1a88352d11c4578e791ed894d;p=thirdparty%2Fkernel%2Fstable-queue.git 6.0-stable patches added patches: video-aperture-call-sysfb_disable-before-removing-pci-devices.patch --- diff --git a/queue-6.0/series b/queue-6.0/series new file mode 100644 index 00000000000..78407f2cd1e --- /dev/null +++ b/queue-6.0/series @@ -0,0 +1 @@ +video-aperture-call-sysfb_disable-before-removing-pci-devices.patch diff --git a/queue-6.0/video-aperture-call-sysfb_disable-before-removing-pci-devices.patch b/queue-6.0/video-aperture-call-sysfb_disable-before-removing-pci-devices.patch new file mode 100644 index 00000000000..37ca469f3c3 --- /dev/null +++ b/queue-6.0/video-aperture-call-sysfb_disable-before-removing-pci-devices.patch @@ -0,0 +1,130 @@ +From tzimmermann@suse.de Wed Oct 26 16:46:12 2022 +From: Thomas Zimmermann +Date: Wed, 26 Oct 2022 16:44:48 +0200 +Subject: [PATCH v2] video/aperture: Call sysfb_disable() before removing PCI devices +To: javierm@redhat.com, deller@gmx.de, sashal@kernel.org, gregkh@linuxfoundation.org +Cc: stable@vger.kernel.org, Thomas Zimmermann , Andreas Thalhammer , Thorsten Leemhuis , Zack Rusin , Daniel Vetter , Daniel Vetter , Sam Ravnborg , Alex Deucher , Zhen Lei , Changcheng Deng , Maarten Lankhorst , Maxime Ripard , dri-devel@lists.freedesktop.org, linux-fbdev@vger.kernel.org +Message-ID: <20221026144448.424-1-tzimmermann@suse.de> + +From: Thomas Zimmermann + +Call sysfb_disable() from aperture_remove_conflicting_pci_devices() +before removing PCI devices. Without, simpledrm can still bind to +simple-framebuffer devices after the hardware driver has taken over +the hardware. Both drivers interfere with each other and results are +undefined. + +Reported modesetting errors [1] are shown below. + +---- snap ---- +rcu: INFO: rcu_sched detected expedited stalls on CPUs/tasks: { 13-.... } 7 jiffies s: 165 root: 0x2000/. +rcu: blocking rcu_node structures (internal RCU debug): +Task dump for CPU 13: +task:X state:R running task stack: 0 pid: 4242 ppid: 4228 flags:0x00000008 +Call Trace: + + ? commit_tail+0xd7/0x130 + ? drm_atomic_helper_commit+0x126/0x150 + ? drm_atomic_commit+0xa4/0xe0 + ? drm_plane_get_damage_clips.cold+0x1c/0x1c + ? drm_atomic_helper_dirtyfb+0x19e/0x280 + ? drm_mode_dirtyfb_ioctl+0x10f/0x1e0 + ? drm_mode_getfb2_ioctl+0x2d0/0x2d0 + ? drm_ioctl_kernel+0xc4/0x150 + ? drm_ioctl+0x246/0x3f0 + ? drm_mode_getfb2_ioctl+0x2d0/0x2d0 + ? __x64_sys_ioctl+0x91/0xd0 + ? do_syscall_64+0x60/0xd0 + ? entry_SYSCALL_64_after_hwframe+0x4b/0xb5 + +... +rcu: INFO: rcu_sched detected expedited stalls on CPUs/tasks: { 13-.... } 30 jiffies s: 169 root: 0x2000/. +rcu: blocking rcu_node structures (internal RCU debug): +Task dump for CPU 13: +task:X state:R running task stack: 0 pid: 4242 ppid: 4228 flags:0x0000400e +Call Trace: + + ? memcpy_toio+0x76/0xc0 + ? memcpy_toio+0x1b/0xc0 + ? drm_fb_memcpy_toio+0x76/0xb0 + ? drm_fb_blit_toio+0x75/0x2b0 + ? simpledrm_simple_display_pipe_update+0x132/0x150 + ? drm_atomic_helper_commit_planes+0xb6/0x230 + ? drm_atomic_helper_commit_tail+0x44/0x80 + ? commit_tail+0xd7/0x130 + ? drm_atomic_helper_commit+0x126/0x150 + ? drm_atomic_commit+0xa4/0xe0 + ? drm_plane_get_damage_clips.cold+0x1c/0x1c + ? drm_atomic_helper_dirtyfb+0x19e/0x280 + ? drm_mode_dirtyfb_ioctl+0x10f/0x1e0 + ? drm_mode_getfb2_ioctl+0x2d0/0x2d0 + ? drm_ioctl_kernel+0xc4/0x150 + ? drm_ioctl+0x246/0x3f0 + ? drm_mode_getfb2_ioctl+0x2d0/0x2d0 + ? __x64_sys_ioctl+0x91/0xd0 + ? do_syscall_64+0x60/0xd0 + ? entry_SYSCALL_64_after_hwframe+0x4b/0xb5 + + +The problem was added by commit 5e0137612430 ("video/aperture: Disable +and unregister sysfb devices via aperture helpers") to v6.0.3 and does +not exist in the mainline branch. + +The mainline commit 5e0137612430 ("video/aperture: Disable and +unregister sysfb devices via aperture helpers") has been backported +from v6.0-rc1 to stable v6.0.3 from a larger patch series [2] that +reworks fbdev framebuffer ownership. The backport misses a change to +aperture_remove_conflicting_pci_devices(). Mainline itself is fine, +because the function does not exist there as a result of the patch +series. + +Instead of backporting the whole series, fix the additional function. + +Reported-by: Andreas Thalhammer +Reported-by: Thorsten Leemhuis +Signed-off-by: Thomas Zimmermann +Tested-by: Andreas Thalhammer +Fixes: cfecfc98a78d ("video/aperture: Disable and unregister sysfb devices via aperture helpers") +Cc: Thomas Zimmermann +Cc: Javier Martinez Canillas +Cc: Zack Rusin +Cc: Daniel Vetter +Cc: Daniel Vetter +Cc: Sam Ravnborg +Cc: Helge Deller +Cc: Alex Deucher +Cc: Zhen Lei +Cc: Changcheng Deng +Cc: Maarten Lankhorst +Cc: Maxime Ripard +Cc: dri-devel@lists.freedesktop.org +Cc: Sasha Levin +Cc: linux-fbdev@vger.kernel.org +Cc: # v6.0.3+ +Link: https://lore.kernel.org/dri-devel/d6afe54b-f8d7-beb2-3609-186e566cbfac@gmx.net/T/#t # [1] +Link: https://patchwork.freedesktop.org/series/106040/ # [2] +Signed-off-by: Greg Kroah-Hartman +--- + drivers/video/aperture.c | 11 +++++++++++ + 1 file changed, 11 insertions(+) + +--- a/drivers/video/aperture.c ++++ b/drivers/video/aperture.c +@@ -358,6 +358,17 @@ int aperture_remove_conflicting_pci_devi + return ret; + + /* ++ * If a driver asked to unregister a platform device registered by ++ * sysfb, then can be assumed that this is a driver for a display ++ * that is set up by the system firmware and has a generic driver. ++ * ++ * Drivers for devices that don't have a generic driver will never ++ * ask for this, so let's assume that a real driver for the display ++ * was already probed and prevent sysfb to register devices later. ++ */ ++ sysfb_disable(); ++ ++ /* + * WARNING: Apparently we must kick fbdev drivers before vgacon, + * otherwise the vga fbdev driver falls over. + */