* Regression: firmware/sysfb.c device path @ 2024-07-13 17:20 Tj 2024-07-14 19:46 ` Thomas Zimmermann 2024-07-15 6:44 ` Thomas Zimmermann 0 siblings, 2 replies; 7+ messages in thread From: Tj @ 2024-07-13 17:20 UTC (permalink / raw) To: Thomas Zimmermann; +Cc: linux-kernel, 1075713 The recent commits to add the parent device path broke Debian's kvm based QA workers for testing installer ISOs after a kernel upgrade from v6.8.12 to v6.9.7. For the details: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1075713 It took some time to track it down since the superficial symptom appeared to involve the QXL driver. It turned out however to be the fbdev driver; specifically the change in the parent device path in sysfs reported between the two kernels: 6.8.12: /sys/class/graphics/fb0 -> ../../devices/platform/vesa-framebuffer.0/graphics/fb0 6.9.7: /sys/class/graphics/fb0 -> ../../devices/pci0000:00/0000:00:01.0/vesa-framebuffer.0/graphics/fb0 This breaks xserver-xorg-core's libfbdevhw.so because it differentiates code-paths based on whether "devices/pci" is matched in the symlink. See hw/xfree86/fbdevhw/fbdevhw.c::fbdev_open() https://gitlab.freedesktop.org/xorg/xserver/-/blob/master/hw/xfree86/fbdevhw/fbdevhw.c?ref_type=heads#L381 ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: Regression: firmware/sysfb.c device path 2024-07-13 17:20 Regression: firmware/sysfb.c device path Tj @ 2024-07-14 19:46 ` Thomas Zimmermann 2024-07-15 6:44 ` Thomas Zimmermann 1 sibling, 0 replies; 7+ messages in thread From: Thomas Zimmermann @ 2024-07-14 19:46 UTC (permalink / raw) To: Tj, Marek Olsak; +Cc: linux-kernel, 1075713 Marek, could this also be the cause of your reported bug? Am 13.07.24 um 19:20 schrieb Tj: > The recent commits to add the parent device path broke Debian's kvm based QA workers for testing installer ISOs after a kernel upgrade from v6.8.12 to v6.9.7. For the details: > > https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1075713 > > It took some time to track it down since the superficial symptom appeared to involve the QXL driver. It turned out however to be the fbdev driver; specifically the change in the parent device path in sysfs reported between the two kernels: > > 6.8.12: > > /sys/class/graphics/fb0 -> ../../devices/platform/vesa-framebuffer.0/graphics/fb0 > > 6.9.7: > > /sys/class/graphics/fb0 -> ../../devices/pci0000:00/0000:00:01.0/vesa-framebuffer.0/graphics/fb0 > > This breaks xserver-xorg-core's libfbdevhw.so because it differentiates code-paths based on whether "devices/pci" is matched in the symlink. > > See hw/xfree86/fbdevhw/fbdevhw.c::fbdev_open() > > https://gitlab.freedesktop.org/xorg/xserver/-/blob/master/hw/xfree86/fbdevhw/fbdevhw.c?ref_type=heads#L381 -- -- Thomas Zimmermann Graphics Driver Developer SUSE Software Solutions Germany GmbH Frankenstrasse 146, 90461 Nuernberg, Germany GF: Ivo Totev, Andrew Myers, Andrew McDonald, Boudien Moerman HRB 36809 (AG Nuernberg) ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: Regression: firmware/sysfb.c device path 2024-07-13 17:20 Regression: firmware/sysfb.c device path Tj 2024-07-14 19:46 ` Thomas Zimmermann @ 2024-07-15 6:44 ` Thomas Zimmermann 2024-07-15 9:03 ` Tj 1 sibling, 1 reply; 7+ messages in thread From: Thomas Zimmermann @ 2024-07-15 6:44 UTC (permalink / raw) To: Tj, Marek Olsak; +Cc: linux-kernel, 1075713 Hi Am 13.07.24 um 19:20 schrieb Tj: > The recent commits to add the parent device path broke Debian's kvm based QA workers for testing installer ISOs after a kernel upgrade from v6.8.12 to v6.9.7. For the details: > > https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1075713 > > It took some time to track it down since the superficial symptom appeared to involve the QXL driver. It turned out however to be the fbdev driver; specifically the change in the parent device path in sysfs reported between the two kernels: > > 6.8.12: > > /sys/class/graphics/fb0 -> ../../devices/platform/vesa-framebuffer.0/graphics/fb0 > > 6.9.7: > > /sys/class/graphics/fb0 -> ../../devices/pci0000:00/0000:00:01.0/vesa-framebuffer.0/graphics/fb0 > > This breaks xserver-xorg-core's libfbdevhw.so because it differentiates code-paths based on whether "devices/pci" is matched in the symlink. > > See hw/xfree86/fbdevhw/fbdevhw.c::fbdev_open() > > https://gitlab.freedesktop.org/xorg/xserver/-/blob/master/hw/xfree86/fbdevhw/fbdevhw.c?ref_type=heads#L381 Amazing debugging skills! The patch that causes the regression in X sets the PCI device as parent for the VESA framebuffer. That means that the PCI device won't be unplugged or suspended while the VESA framebuffer is still in use. Without, results are undefined. Therefore, could this be fixed in X' fbdev driver? It currently tests against the sysfs path of the fbdev device. A fix could look like this: 1) readlink /sys/class/graphics/fb0/device/subsystem 2) strcmp to "bus/pci" fb0 is the fbdev device fb0/device/ is the Linux device (the VESA/EFI framebuffer or native PCI graphics device) fb0/device/subsystem is the Linux devices' subsystem. The subsystem string is the path to the underlying bus. For PCI graphics cards it's something like ../../../bus/pci and for VESA/EFI framebuffer devices it's ../../../bus/platform That's more correct, reliable and also keeps the correct organization of the devices within the kernel. Best regards Thomas -- -- Thomas Zimmermann Graphics Driver Developer SUSE Software Solutions Germany GmbH Frankenstrasse 146, 90461 Nuernberg, Germany GF: Ivo Totev, Andrew Myers, Andrew McDonald, Boudien Moerman HRB 36809 (AG Nuernberg) ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: Regression: firmware/sysfb.c device path 2024-07-15 6:44 ` Thomas Zimmermann @ 2024-07-15 9:03 ` Tj 2024-07-15 9:22 ` Thomas Zimmermann 0 siblings, 1 reply; 7+ messages in thread From: Tj @ 2024-07-15 9:03 UTC (permalink / raw) To: Thomas Zimmermann; +Cc: Marek Olsak, linux-kernel, 1075713 On Monday, 15 July 2024 at 07:44, Thomas Zimmermann <tzimmermann@suse.de> wrote: > > See hw/xfree86/fbdevhw/fbdevhw.c::fbdev_open() > > > > https://gitlab.freedesktop.org/xorg/xserver/-/blob/master/hw/xfree86/fbdevhw/fbdevhw.c?ref_type=heads#L381 > > > Amazing debugging skills! > > The patch that causes the regression in X sets the PCI device as parent > for the VESA framebuffer. That means that the PCI device won't be > unplugged or suspended while the VESA framebuffer is still in use. > Without, results are undefined. > > Therefore, could this be fixed in X' fbdev driver? I was in two minds about this one. On the 'perfection' side I agree Xorg should not rely on a symbolic link at all - though that is easy to say; I wasn't the one that had to find a way to do what fbdevhw has to do at the time! But the other me argues that kernel ought not to break userspace and as it is a pretty common scenario across distros that userspace stays relatively stable (sans security bugs) but kernels can and do move on quite rapidly and frequently the kernel ought to play nice here :) With this in mind I can foresee many systems that work perfectly fine will break when only the kernel is upgraded and thus it'll have the finger of blame pointed at it. Since this is more likely to strike in automated test harness scenarios (due to it being relatively obscure) as well there may well be circumstances where changing a userspace library immediately is extremely problematical. I wonder if there's some half-way deprecation measure here that would allow a period of transition? I can imagine one way would be a custom udev rule (for systemd-udevd hosts - not sure about non-udevd) that replaces the new symlink with the old - but if a distro does that it's almost as easy to patch and rebuild fbdevhw - and far more brittle. > A fix could look like this: > > 1) readlink /sys/class/graphics/fb0/device/subsystem > 2) strcmp to "bus/pci" This works and doesn't cause a regression in v6.8.12: ~ # uname -r; grep fbdev /var/log/Xorg.0.log 6.9.7-amd64 [ 31.376] (==) Matched fbdev as autoconfigured driver 2 [ 31.377] (II) LoadModule: "fbdev" [ 31.377] (II) Loading /usr/lib/xorg/modules/drivers/fbdev_drv.so [ 31.377] (II) Module fbdev: vendor="X.Org Foundation" [ 31.377] (II) FBDEV: driver for framebuffer: fbdev [ 31.410] fbdev trace: FBDevPciProbe() [ 31.410] (II) Loading sub module "fbdevhw" [ 31.410] (II) LoadModule: "fbdevhw" [ 31.410] (II) Loading /usr/lib/xorg/modules/libfbdevhw.so [ 31.410] (II) Module fbdevhw: vendor="X.Org Foundation" [ 31.410] fbdev trace: FBDevPciProbe() return [ 31.410] (WW) Falling back to old probe method for fbdev [ 31.410] fbdev trace: FBDevProbe() [ 31.410] (II) Loading sub module "fbdevhw" [ 31.410] (II) LoadModule: "fbdevhw" [ 31.410] (II) Loading /usr/lib/xorg/modules/libfbdevhw.so [ 31.410] (II) Module fbdevhw: vendor="X.Org Foundation" [ 31.410] fbdev: FBDevProbe() for() numDevSection=0 [ 31.410] fbdev: FBDevProbe() isPci0 isISA=0 [ 31.410] fbdev: FBDevProbe() calling fbdevHWProbe(NULL, '(null)', NULL) [ 31.410] (II) fbdev_open(scrnIndex=-1, dev=(null), namep=(nil)) [ 31.410] (II) fbdev_open() using dev from env FRAMEBUFFER=(null) [ 31.410] (II) fbdev_open() using default dev=/dev/fb0 [ 31.410] (II) fbdev_open() sysfs_path=/sys/class/graphics/fb0/device/subsystem [ 31.410] (II) fbdev_open() buf=../../../../bus/platform [ 31.410] (II) fbdev_open() returning file descriptor 11 [ 31.410] fbdev trace: FBDevProbe() fbdevHWProbe() [ 31.410] fbdev trace: FBDevProbe() else xf86ClaimFbSlot() [ 31.410] fbdev trace: FBDevProbe() return [ 31.410] (II) UnloadModule: "fbdev" [ 31.410] (II) UnloadSubModule: "fbdevhw" [ 31.410] fbdev: PreInit 0 [ 31.410] (II) FBDEV(0): fbdev_open(scrnIndex=0, dev=(null), namep=(nil)) [ 31.410] (II) FBDEV(0): fbdev_open() using dev from env FRAMEBUFFER=(null) [ 31.410] (II) FBDEV(0): fbdev_open() using default dev=/dev/fb0 [ 31.410] (II) FBDEV(0): fbdev_open() sysfs_path=/sys/class/graphics/fb0/device/subsystem [ 31.410] (II) FBDEV(0): fbdev_open() buf=../../../../bus/platform [ 31.410] (II) FBDEV(0): fbdev_open() returning file descriptor 12 ~ # uname -r; grep fbdev /var/log/Xorg.0.log 6.8.12-amd64 [ 14.225] (==) Matched fbdev as autoconfigured driver 2 [ 14.225] (II) LoadModule: "fbdev" [ 14.225] (II) Loading /usr/lib/xorg/modules/drivers/fbdev_drv.so [ 14.225] (II) Module fbdev: vendor="X.Org Foundation" [ 14.225] (II) FBDEV: driver for framebuffer: fbdev [ 14.252] fbdev trace: FBDevPciProbe() [ 14.252] (II) Loading sub module "fbdevhw" [ 14.252] (II) LoadModule: "fbdevhw" [ 14.252] (II) Loading /usr/lib/xorg/modules/libfbdevhw.so [ 14.252] (II) Module fbdevhw: vendor="X.Org Foundation" [ 14.253] fbdev trace: FBDevPciProbe() return [ 14.253] (WW) Falling back to old probe method for fbdev [ 14.253] fbdev trace: FBDevProbe() [ 14.253] (II) Loading sub module "fbdevhw" [ 14.253] (II) LoadModule: "fbdevhw" [ 14.253] (II) Loading /usr/lib/xorg/modules/libfbdevhw.so [ 14.253] (II) Module fbdevhw: vendor="X.Org Foundation" [ 14.253] fbdev: FBDevProbe() for() numDevSection=0 [ 14.253] fbdev: FBDevProbe() isPci0 isISA=0 [ 14.253] fbdev: FBDevProbe() calling fbdevHWProbe(NULL, '(null)', NULL) [ 14.253] (II) fbdev_open(scrnIndex=-1, dev=(null), namep=(nil)) [ 14.253] (II) fbdev_open() using dev from env FRAMEBUFFER=(null) [ 14.253] (II) fbdev_open() using default dev=/dev/fb0 [ 14.253] (II) fbdev_open() sysfs_path=/sys/class/graphics/fb0/device/subsystem [ 14.253] (II) fbdev_open() buf=../../../bus/platform [ 14.253] (II) fbdev_open() returning file descriptor 11 [ 14.253] fbdev trace: FBDevProbe() fbdevHWProbe() [ 14.253] fbdev trace: FBDevProbe() else xf86ClaimFbSlot() [ 14.253] fbdev trace: FBDevProbe() return [ 14.253] (II) UnloadModule: "fbdev" [ 14.253] (II) UnloadSubModule: "fbdevhw" [ 14.253] fbdev: PreInit 0 [ 14.253] (II) FBDEV(0): fbdev_open(scrnIndex=0, dev=(null), namep=(nil)) [ 14.253] (II) FBDEV(0): fbdev_open() using dev from env FRAMEBUFFER=(null) [ 14.253] (II) FBDEV(0): fbdev_open() using default dev=/dev/fb0 [ 14.253] (II) FBDEV(0): fbdev_open() sysfs_path=/sys/class/graphics/fb0/device/subsystem [ 14.253] (II) FBDEV(0): fbdev_open() buf=../../../bus/platform [ 14.253] (II) FBDEV(0): fbdev_open() returning file descriptor 12 I'll propose this patch in Debian Xorg for Unstable/Testing but not sure if it'll be accepted for Stable. ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: Regression: firmware/sysfb.c device path 2024-07-15 9:03 ` Tj @ 2024-07-15 9:22 ` Thomas Zimmermann 2024-07-15 9:29 ` Tj 0 siblings, 1 reply; 7+ messages in thread From: Thomas Zimmermann @ 2024-07-15 9:22 UTC (permalink / raw) To: Tj; +Cc: Marek Olsak, linux-kernel, 1075713 Hi Am 15.07.24 um 11:03 schrieb Tj: > On Monday, 15 July 2024 at 07:44, Thomas Zimmermann <tzimmermann@suse.de> wrote: > >>> See hw/xfree86/fbdevhw/fbdevhw.c::fbdev_open() >>> >>> https://gitlab.freedesktop.org/xorg/xserver/-/blob/master/hw/xfree86/fbdevhw/fbdevhw.c?ref_type=heads#L381 >> >> Amazing debugging skills! >> >> The patch that causes the regression in X sets the PCI device as parent >> for the VESA framebuffer. That means that the PCI device won't be >> unplugged or suspended while the VESA framebuffer is still in use. >> Without, results are undefined. >> >> Therefore, could this be fixed in X' fbdev driver? > I was in two minds about this one. On the 'perfection' side I agree Xorg should not rely on a symbolic link at all - though that is easy to say; I wasn't the one that had to find a way to do what fbdevhw has to do at the time! > > But the other me argues that kernel ought not to break userspace and as it is a pretty common scenario across distros that userspace stays relatively stable (sans security bugs) but kernels can and do move on quite rapidly and frequently the kernel ought to play nice here :) > > With this in mind I can foresee many systems that work perfectly fine will break when only the kernel is upgraded and thus it'll have the finger of blame pointed at it. > > Since this is more likely to strike in automated test harness scenarios (due to it being relatively obscure) as well there may well be circumstances where changing a userspace library immediately is extremely problematical. > > I wonder if there's some half-way deprecation measure here that would allow a period of transition? > > I can imagine one way would be a custom udev rule (for systemd-udevd hosts - not sure about non-udevd) that replaces the new symlink with the old - but if a distro does that it's almost as easy to patch and rebuild fbdevhw - and far more brittle. I agree with the assessment. Luckily not too many systems use fbdevhw these days and updating X isn't much more effort than updating the kernel. And OTOH the kernel patch fixes a real problem. If the PCI device suspends or does a hotunplug, the state of the VESA/EFI framebuffers becomes undefined (with undefined results). So reverting it is not a good option either. > >> A fix could look like this: >> >> 1) readlink /sys/class/graphics/fb0/device/subsystem >> 2) strcmp to "bus/pci" > This works and doesn't cause a regression in v6.8.12: These device/subsystem paths have been stable since forever. That should most likely not change. > > ~ # uname -r; grep fbdev /var/log/Xorg.0.log > 6.9.7-amd64 > [ 31.376] (==) Matched fbdev as autoconfigured driver 2 > [ 31.377] (II) LoadModule: "fbdev" > [ 31.377] (II) Loading /usr/lib/xorg/modules/drivers/fbdev_drv.so > [ 31.377] (II) Module fbdev: vendor="X.Org Foundation" > [ 31.377] (II) FBDEV: driver for framebuffer: fbdev > [ 31.410] fbdev trace: FBDevPciProbe() > [ 31.410] (II) Loading sub module "fbdevhw" > [ 31.410] (II) LoadModule: "fbdevhw" > [ 31.410] (II) Loading /usr/lib/xorg/modules/libfbdevhw.so > [ 31.410] (II) Module fbdevhw: vendor="X.Org Foundation" > [ 31.410] fbdev trace: FBDevPciProbe() return > [ 31.410] (WW) Falling back to old probe method for fbdev > [ 31.410] fbdev trace: FBDevProbe() > [ 31.410] (II) Loading sub module "fbdevhw" > [ 31.410] (II) LoadModule: "fbdevhw" > [ 31.410] (II) Loading /usr/lib/xorg/modules/libfbdevhw.so > [ 31.410] (II) Module fbdevhw: vendor="X.Org Foundation" > [ 31.410] fbdev: FBDevProbe() for() numDevSection=0 > [ 31.410] fbdev: FBDevProbe() isPci0 isISA=0 > [ 31.410] fbdev: FBDevProbe() calling fbdevHWProbe(NULL, '(null)', NULL) > [ 31.410] (II) fbdev_open(scrnIndex=-1, dev=(null), namep=(nil)) > [ 31.410] (II) fbdev_open() using dev from env FRAMEBUFFER=(null) > [ 31.410] (II) fbdev_open() using default dev=/dev/fb0 > [ 31.410] (II) fbdev_open() sysfs_path=/sys/class/graphics/fb0/device/subsystem > [ 31.410] (II) fbdev_open() buf=../../../../bus/platform > [ 31.410] (II) fbdev_open() returning file descriptor 11 > [ 31.410] fbdev trace: FBDevProbe() fbdevHWProbe() > [ 31.410] fbdev trace: FBDevProbe() else xf86ClaimFbSlot() > [ 31.410] fbdev trace: FBDevProbe() return > [ 31.410] (II) UnloadModule: "fbdev" > [ 31.410] (II) UnloadSubModule: "fbdevhw" > [ 31.410] fbdev: PreInit 0 > [ 31.410] (II) FBDEV(0): fbdev_open(scrnIndex=0, dev=(null), namep=(nil)) > [ 31.410] (II) FBDEV(0): fbdev_open() using dev from env FRAMEBUFFER=(null) > [ 31.410] (II) FBDEV(0): fbdev_open() using default dev=/dev/fb0 > [ 31.410] (II) FBDEV(0): fbdev_open() sysfs_path=/sys/class/graphics/fb0/device/subsystem > [ 31.410] (II) FBDEV(0): fbdev_open() buf=../../../../bus/platform > [ 31.410] (II) FBDEV(0): fbdev_open() returning file descriptor 12 > > ~ # uname -r; grep fbdev /var/log/Xorg.0.log > 6.8.12-amd64 > [ 14.225] (==) Matched fbdev as autoconfigured driver 2 > [ 14.225] (II) LoadModule: "fbdev" > [ 14.225] (II) Loading /usr/lib/xorg/modules/drivers/fbdev_drv.so > [ 14.225] (II) Module fbdev: vendor="X.Org Foundation" > [ 14.225] (II) FBDEV: driver for framebuffer: fbdev > [ 14.252] fbdev trace: FBDevPciProbe() > [ 14.252] (II) Loading sub module "fbdevhw" > [ 14.252] (II) LoadModule: "fbdevhw" > [ 14.252] (II) Loading /usr/lib/xorg/modules/libfbdevhw.so > [ 14.252] (II) Module fbdevhw: vendor="X.Org Foundation" > [ 14.253] fbdev trace: FBDevPciProbe() return > [ 14.253] (WW) Falling back to old probe method for fbdev > [ 14.253] fbdev trace: FBDevProbe() > [ 14.253] (II) Loading sub module "fbdevhw" > [ 14.253] (II) LoadModule: "fbdevhw" > [ 14.253] (II) Loading /usr/lib/xorg/modules/libfbdevhw.so > [ 14.253] (II) Module fbdevhw: vendor="X.Org Foundation" > [ 14.253] fbdev: FBDevProbe() for() numDevSection=0 > [ 14.253] fbdev: FBDevProbe() isPci0 isISA=0 > [ 14.253] fbdev: FBDevProbe() calling fbdevHWProbe(NULL, '(null)', NULL) > [ 14.253] (II) fbdev_open(scrnIndex=-1, dev=(null), namep=(nil)) > [ 14.253] (II) fbdev_open() using dev from env FRAMEBUFFER=(null) > [ 14.253] (II) fbdev_open() using default dev=/dev/fb0 > [ 14.253] (II) fbdev_open() sysfs_path=/sys/class/graphics/fb0/device/subsystem > [ 14.253] (II) fbdev_open() buf=../../../bus/platform > [ 14.253] (II) fbdev_open() returning file descriptor 11 > [ 14.253] fbdev trace: FBDevProbe() fbdevHWProbe() > [ 14.253] fbdev trace: FBDevProbe() else xf86ClaimFbSlot() > [ 14.253] fbdev trace: FBDevProbe() return > [ 14.253] (II) UnloadModule: "fbdev" > [ 14.253] (II) UnloadSubModule: "fbdevhw" > [ 14.253] fbdev: PreInit 0 > [ 14.253] (II) FBDEV(0): fbdev_open(scrnIndex=0, dev=(null), namep=(nil)) > [ 14.253] (II) FBDEV(0): fbdev_open() using dev from env FRAMEBUFFER=(null) > [ 14.253] (II) FBDEV(0): fbdev_open() using default dev=/dev/fb0 > [ 14.253] (II) FBDEV(0): fbdev_open() sysfs_path=/sys/class/graphics/fb0/device/subsystem > [ 14.253] (II) FBDEV(0): fbdev_open() buf=../../../bus/platform > [ 14.253] (II) FBDEV(0): fbdev_open() returning file descriptor 12 > > I'll propose this patch in Debian Xorg for Unstable/Testing but not sure if it'll be accepted for Stable. It should not be a problem, as only 6.9 is affected. Debian stable does update to such newer kernels, right? We should definitely get your patch into the Xorg upstream. Best regards Thomas -- -- Thomas Zimmermann Graphics Driver Developer SUSE Software Solutions Germany GmbH Frankenstrasse 146, 90461 Nuernberg, Germany GF: Ivo Totev, Andrew Myers, Andrew McDonald, Boudien Moerman HRB 36809 (AG Nuernberg) ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: Regression: firmware/sysfb.c device path 2024-07-15 9:22 ` Thomas Zimmermann @ 2024-07-15 9:29 ` Tj [not found] ` <DM6PR12MB473145875C16DFDC5B0DE65FF9A92@DM6PR12MB4731.namprd12.prod.outlook.com> 0 siblings, 1 reply; 7+ messages in thread From: Tj @ 2024-07-15 9:29 UTC (permalink / raw) To: Thomas Zimmermann; +Cc: Marek Olsak, linux-kernel, 1075713 On Monday, 15 July 2024 at 10:22, Thomas Zimmermann <tzimmermann@suse.de> wrote: > We should definitely get your patch into the Xorg upstream. Working on that now. ^ permalink raw reply [flat|nested] 7+ messages in thread
[parent not found: <DM6PR12MB473145875C16DFDC5B0DE65FF9A92@DM6PR12MB4731.namprd12.prod.outlook.com>]
* Re: Regression: firmware/sysfb.c device path [not found] ` <DM6PR12MB473145875C16DFDC5B0DE65FF9A92@DM6PR12MB4731.namprd12.prod.outlook.com> @ 2024-07-29 7:00 ` Thomas Zimmermann 0 siblings, 0 replies; 7+ messages in thread From: Thomas Zimmermann @ 2024-07-29 7:00 UTC (permalink / raw) To: Olsak, Marek, Tj; +Cc: linux-kernel, 1075713 Hi Am 23.07.24 um 19:38 schrieb Olsak, Marek: > > [AMD Official Use Only - AMD Internal Distribution Only] > > > Hi, > > If you have a patch, I'd like to test it. Here's the proposed change: https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1612/diffs?commit_id=a745f59bb193c7a87959fddade9b637278e5528e BTW, did you see https://lore.kernel.org/dri-devel/1ff5b407-ff81-40ef-8aa3-9f155491f497@leemhuis.info/ ? This looks similar to your bug report. Best regards Thomas > > Thanks, > Marek > ------------------------------------------------------------------------ > *From:* Tj <tj.iam.tj@proton.me> > *Sent:* July 15, 2024 05:29 > *To:* Thomas Zimmermann <tzimmermann@suse.de> > *Cc:* Olsak, Marek <Marek.Olsak@amd.com>; linux-kernel@vger.kernel.org > <linux-kernel@vger.kernel.org>; 1075713@bugs.debian.org > <1075713@bugs.debian.org> > *Subject:* Re: Regression: firmware/sysfb.c device path > On Monday, 15 July 2024 at 10:22, Thomas Zimmermann > <tzimmermann@suse.de> wrote: > > > > We should definitely get your patch into the Xorg upstream. > > Working on that now. -- -- Thomas Zimmermann Graphics Driver Developer SUSE Software Solutions Germany GmbH Frankenstrasse 146, 90461 Nuernberg, Germany GF: Ivo Totev, Andrew Myers, Andrew McDonald, Boudien Moerman HRB 36809 (AG Nuernberg) ^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2024-07-29 7:00 UTC | newest]
Thread overview: 7+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2024-07-13 17:20 Regression: firmware/sysfb.c device path Tj
2024-07-14 19:46 ` Thomas Zimmermann
2024-07-15 6:44 ` Thomas Zimmermann
2024-07-15 9:03 ` Tj
2024-07-15 9:22 ` Thomas Zimmermann
2024-07-15 9:29 ` Tj
[not found] ` <DM6PR12MB473145875C16DFDC5B0DE65FF9A92@DM6PR12MB4731.namprd12.prod.outlook.com>
2024-07-29 7:00 ` Thomas Zimmermann
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®