| Message ID | CAOMZO5Be5Y_81g8F1to4kRX7e3TgOqRa6-kVXti0C5pchzCW1w@mail.gmail.com |
|---|---|
| State | Not Applicable |
| Headers | show |
Hi Fabio, On Sat, Mar 4, 2017 at 8:41 PM, Fabio Estevam <festevam@gmail.com> wrote: > On Sat, Mar 4, 2017 at 11:47 AM, Fabio Estevam <festevam@gmail.com> wrote: >> Hi Gary, >> >> On Sat, Mar 4, 2017 at 9:33 AM, Fabio Estevam <festevam@gmail.com> wrote: >> >>> Not yet, still stuck at the "drmModeGetResources failed" error. >> >> I can also get this same error with the new kmscube package from: >> https://cgit.freedesktop.org/mesa/kmscube/log/ >> >> # kmscube >> drmModeGetResources failed: Invalid argument >> failed to initialize legacy DRM >> >> It works if I explicitly pass the /dev/dri device: >> >> kmscube -D /dev/dri/card1 >> >> I think QT5 is trying to access /dev/dri/card0, but it should be using >> /dev/dri/card1 instead. > > Yes, this is the problem. > > If I do an ugly hack like this: > > --- a/src/plugins/platforms/eglfs/deviceintegration/eglfs_kms/qeglfskmsdevice.cpp > +++ b/src/plugins/platforms/eglfs/deviceintegration/eglfs_kms/qeglfskmsdevice.cpp > @@ -344,7 +344,7 @@ bool QEglFSKmsDevice::open() > Q_ASSERT(m_gbm_device == Q_NULLPTR); > > qCDebug(qLcEglfsKmsDebug) << "Opening device" << m_path; > - m_dri_fd = qt_safe_open(m_path.toLocal8Bit().constData(), O_RDWR > | O_CLOEXEC); > + m_dri_fd = qt_safe_open("/dev/dri/card1", O_RDWR | O_CLOEXEC); > if (m_dri_fd == -1) { > qErrnoWarning("Could not open DRM device %s", qPrintable(m_path)); > return false; > -- > 2.7.4 > > Then the demo is launched. It still do not run very well, but at least > it starts. > > Does anyone have an idea on what would be a proper fix? That's strange I didn't have to do that. Worst part is that my demo image is using Qt5.6.2 (same as your image) and that I checked no special env variable is set. Maybe you can try your kernel on my image, see if that works. Regards, Gary
On Mon, Mar 6, 2017 at 3:49 PM, Gary Bisson <gary.bisson@boundarydevices.com> wrote: > Hi Fabio, > > On Sat, Mar 4, 2017 at 8:41 PM, Fabio Estevam <festevam@gmail.com> wrote: >> On Sat, Mar 4, 2017 at 11:47 AM, Fabio Estevam <festevam@gmail.com> wrote: >>> Hi Gary, >>> >>> On Sat, Mar 4, 2017 at 9:33 AM, Fabio Estevam <festevam@gmail.com> wrote: >>> >>>> Not yet, still stuck at the "drmModeGetResources failed" error. >>> >>> I can also get this same error with the new kmscube package from: >>> https://cgit.freedesktop.org/mesa/kmscube/log/ >>> >>> # kmscube >>> drmModeGetResources failed: Invalid argument >>> failed to initialize legacy DRM >>> >>> It works if I explicitly pass the /dev/dri device: >>> >>> kmscube -D /dev/dri/card1 >>> >>> I think QT5 is trying to access /dev/dri/card0, but it should be using >>> /dev/dri/card1 instead. >> >> Yes, this is the problem. >> >> If I do an ugly hack like this: >> >> --- a/src/plugins/platforms/eglfs/deviceintegration/eglfs_kms/qeglfskmsdevice.cpp >> +++ b/src/plugins/platforms/eglfs/deviceintegration/eglfs_kms/qeglfskmsdevice.cpp >> @@ -344,7 +344,7 @@ bool QEglFSKmsDevice::open() >> Q_ASSERT(m_gbm_device == Q_NULLPTR); >> >> qCDebug(qLcEglfsKmsDebug) << "Opening device" << m_path; >> - m_dri_fd = qt_safe_open(m_path.toLocal8Bit().constData(), O_RDWR >> | O_CLOEXEC); >> + m_dri_fd = qt_safe_open("/dev/dri/card1", O_RDWR | O_CLOEXEC); >> if (m_dri_fd == -1) { >> qErrnoWarning("Could not open DRM device %s", qPrintable(m_path)); >> return false; >> -- >> 2.7.4 >> >> Then the demo is launched. It still do not run very well, but at least >> it starts. >> >> Does anyone have an idea on what would be a proper fix? > > That's strange I didn't have to do that. Worst part is that my demo > image is using Qt5.6.2 (same as your image) and that I checked no > special env variable is set. > > Maybe you can try your kernel on my image, see if that works. I just used rootfs, from the link, and kernel 4.9.13 mainline. Able to see the demo nicely. Order: - kmscube - Boundary devices Tux - jellyfish - ideas in motion - 28 fps (1. Lord of the Rings) So, I'm investigating config options on buildroot will let you know for further update. thanks!
On Mon, Mar 6, 2017 at 4:33 PM, Jagan Teki <jagannadh.teki@gmail.com> wrote: > On Mon, Mar 6, 2017 at 3:49 PM, Gary Bisson > <gary.bisson@boundarydevices.com> wrote: >> Hi Fabio, >> >> On Sat, Mar 4, 2017 at 8:41 PM, Fabio Estevam <festevam@gmail.com> wrote: >>> On Sat, Mar 4, 2017 at 11:47 AM, Fabio Estevam <festevam@gmail.com> wrote: >>>> Hi Gary, >>>> >>>> On Sat, Mar 4, 2017 at 9:33 AM, Fabio Estevam <festevam@gmail.com> wrote: >>>> >>>>> Not yet, still stuck at the "drmModeGetResources failed" error. >>>> >>>> I can also get this same error with the new kmscube package from: >>>> https://cgit.freedesktop.org/mesa/kmscube/log/ >>>> >>>> # kmscube >>>> drmModeGetResources failed: Invalid argument >>>> failed to initialize legacy DRM >>>> >>>> It works if I explicitly pass the /dev/dri device: >>>> >>>> kmscube -D /dev/dri/card1 >>>> >>>> I think QT5 is trying to access /dev/dri/card0, but it should be using >>>> /dev/dri/card1 instead. >>> >>> Yes, this is the problem. >>> >>> If I do an ugly hack like this: >>> >>> --- a/src/plugins/platforms/eglfs/deviceintegration/eglfs_kms/qeglfskmsdevice.cpp >>> +++ b/src/plugins/platforms/eglfs/deviceintegration/eglfs_kms/qeglfskmsdevice.cpp >>> @@ -344,7 +344,7 @@ bool QEglFSKmsDevice::open() >>> Q_ASSERT(m_gbm_device == Q_NULLPTR); >>> >>> qCDebug(qLcEglfsKmsDebug) << "Opening device" << m_path; >>> - m_dri_fd = qt_safe_open(m_path.toLocal8Bit().constData(), O_RDWR >>> | O_CLOEXEC); >>> + m_dri_fd = qt_safe_open("/dev/dri/card1", O_RDWR | O_CLOEXEC); >>> if (m_dri_fd == -1) { >>> qErrnoWarning("Could not open DRM device %s", qPrintable(m_path)); >>> return false; >>> -- >>> 2.7.4 >>> >>> Then the demo is launched. It still do not run very well, but at least >>> it starts. >>> >>> Does anyone have an idea on what would be a proper fix? >> >> That's strange I didn't have to do that. Worst part is that my demo >> image is using Qt5.6.2 (same as your image) and that I checked no >> special env variable is set. >> >> Maybe you can try your kernel on my image, see if that works. > > I just used rootfs, from the link, and kernel 4.9.13 mainline. > > Able to see the demo nicely. > > Order: > - kmscube > - Boundary devices Tux > - jellyfish > - ideas in motion > - 28 fps (1. Lord of the Rings) > > So, I'm investigating config options on buildroot will let you know > for further update. And by the way, did anyone find this issue. In file included from /workspace/buildroot/output/host/usr/arm-buildroot-linux-gnueabihf/sysroot/usr/include/EGL/egl.h:39:0, from ../../../../include/QtEglSupport/5.8.0/QtEglSupport/private/../../../../../src/platformsupport/eglconvenience/qt_egl_p.h:63, from ../../../../include/QtEglSupport/5.8.0/QtEglSupport/private/qt_egl_p.h:1, from api/qeglfsglobal_p.h:56, from api/qeglfsintegration_p.h:54, from qeglfsmain.cpp:41: /workspace/buildroot/output/host/usr/arm-buildroot-linux-gnueabihf/sysroot/usr/include/EGL/eglplatform.h:119:22: fatal error: X11/Xlib.h: No such file or directory #include <X11/Xlib.h> ^ compilation terminated. Look like Qt5.8.0 is unable to build with current tool-chain, 5.6.2 succeeded thanks!
On Mon, Mar 6, 2017 at 8:44 AM, Jagan Teki <jagannadh.teki@gmail.com> wrote: > > And by the way, did anyone find this issue. > > In file included from > /workspace/buildroot/output/host/usr/arm-buildroot-linux-gnueabihf/sysroot/usr/include/EGL/egl.h:39:0, > from > ../../../../include/QtEglSupport/5.8.0/QtEglSupport/private/../../../../../src/platformsupport/eglconvenience/qt_egl_p.h:63, > from > ../../../../include/QtEglSupport/5.8.0/QtEglSupport/private/qt_egl_p.h:1, > from api/qeglfsglobal_p.h:56, > from api/qeglfsintegration_p.h:54, > from qeglfsmain.cpp:41: > /workspace/buildroot/output/host/usr/arm-buildroot-linux-gnueabihf/sysroot/usr/include/EGL/eglplatform.h:119:22: > fatal error: X11/Xlib.h: No such file or directory > #include <X11/Xlib.h> > ^ > compilation terminated. > > Look like Qt5.8.0 is unable to build with current tool-chain, 5.6.2 succeeded Yes, that's why I needed to switch to QT 5.6.2 on my tests. I would be interested to know if you are able to run the Qt5 Cinematic demo well when building it from Buildroot master. Thanks
On Mon, Mar 6, 2017 at 5:16 PM, Fabio Estevam <festevam@gmail.com> wrote: > On Mon, Mar 6, 2017 at 8:44 AM, Jagan Teki <jagannadh.teki@gmail.com> wrote: >> >> And by the way, did anyone find this issue. >> >> In file included from >> /workspace/buildroot/output/host/usr/arm-buildroot-linux-gnueabihf/sysroot/usr/include/EGL/egl.h:39:0, >> from >> ../../../../include/QtEglSupport/5.8.0/QtEglSupport/private/../../../../../src/platformsupport/eglconvenience/qt_egl_p.h:63, >> from >> ../../../../include/QtEglSupport/5.8.0/QtEglSupport/private/qt_egl_p.h:1, >> from api/qeglfsglobal_p.h:56, >> from api/qeglfsintegration_p.h:54, >> from qeglfsmain.cpp:41: >> /workspace/buildroot/output/host/usr/arm-buildroot-linux-gnueabihf/sysroot/usr/include/EGL/eglplatform.h:119:22: >> fatal error: X11/Xlib.h: No such file or directory >> #include <X11/Xlib.h> >> ^ >> compilation terminated. >> >> Look like Qt5.8.0 is unable to build with current tool-chain, 5.6.2 succeeded > > Yes, that's why I needed to switch to QT 5.6.2 on my tests. > > I would be interested to know if you are able to run the Qt5 Cinematic > demo well when building it from Buildroot master. - Jelly fish: working ------------------------- # glmark2-es2-drm -b jellyfish MESA-LOADER: device is not located on the PCI bus MESA-LOADER: device is not located on the PCI bus MESA-LOADER: device is not located on the PCI bus ** Failed to set swap interval. Results may be bounded above by refresh rate. ======================================================= glmark2 2014.03 ======================================================= OpenGL Information GL_VENDOR: etnaviv GL_RENDERER: Gallium 0.4 on Vivante GC2000 rev 5108 GL_VERSION: OpenGL ES 2.0 Mesa 17.0.1 ======================================================= ** Failed to set swap interval. Results may be bounded above by refresh rate. [jellyfish] <default>: FPS: 25 FrameTime: 40.000 ms ================================ - CinematicExperience: ----------------------------- - demo launched but 'Black' object on list like 'Lord of the Rings' etc # CinematicExperience-demo MESA-LOADER: device is not located on the PCI bus MESA-LOADER: device is not located on the PCI bus MESA-LOADER: device is not located on the PCI bus Failed to move cursor on screen LVDS1: -14 libpng warning: iCCP: known incorrect sRGB profile libpng warning: iCCP: known incorrect sRGB profile QOpenGLFramebufferObject: Unsupported framebuffer format. QOpenGLFramebufferObject: Framebuffer incomplete, missing attachment. QOpenGLFramebufferObject: Unsupported framebuffer format. QOpenGLFramebufferObject: Framebuffer incomplete, missing attachment. QOpenGLFramebufferObject: Unsupported framebuffer format. QOpenGLFramebufferObject: Framebuffer incomplete, missing attachment. QOpenGLFramebufferObject: Unsupported framebuffer format. QOpenGLFramebufferObject: Framebuffer incomplete, missing attachment. thanks!
On Mon, Mar 6, 2017 at 9:08 AM, Jagan Teki <jagannadh.teki@gmail.com> wrote: > - CinematicExperience: > ----------------------------- > - demo launched but 'Black' object on list like 'Lord of the Rings' etc > > # CinematicExperience-demo > MESA-LOADER: device is not located on the PCI bus > MESA-LOADER: device is not located on the PCI bus > MESA-LOADER: device is not located on the PCI bus > Failed to move cursor on screen LVDS1: -14 > libpng warning: iCCP: known incorrect sRGB profile > libpng warning: iCCP: known incorrect sRGB profile > QOpenGLFramebufferObject: Unsupported framebuffer format. > QOpenGLFramebufferObject: Framebuffer incomplete, missing attachment. > QOpenGLFramebufferObject: Unsupported framebuffer format. > QOpenGLFramebufferObject: Framebuffer incomplete, missing attachment. > QOpenGLFramebufferObject: Unsupported framebuffer format. > QOpenGLFramebufferObject: Framebuffer incomplete, missing attachment. > QOpenGLFramebufferObject: Unsupported framebuffer format. > QOpenGLFramebufferObject: Framebuffer incomplete, missing attachment. I also saw this defect here. Would be nice to understand where it comes from. Gary, Do you have the Buildroot hash that your demo was generated from? Thanks
Hello Jagan, On Mon, 6 Mar 2017 17:14:23 +0530, Jagan Teki <jagannadh.teki@gmail.com> wrote: > On Mon, Mar 6, 2017 at 4:33 PM, Jagan Teki <jagannadh.teki@gmail.com> wrote: > > On Mon, Mar 6, 2017 at 3:49 PM, Gary Bisson > > <gary.bisson@boundarydevices.com> wrote: > >> Hi Fabio, > >> > >> On Sat, Mar 4, 2017 at 8:41 PM, Fabio Estevam <festevam@gmail.com> wrote: > >>> On Sat, Mar 4, 2017 at 11:47 AM, Fabio Estevam <festevam@gmail.com> wrote: > >>>> Hi Gary, > >>>> > >>>> On Sat, Mar 4, 2017 at 9:33 AM, Fabio Estevam <festevam@gmail.com> wrote: > >>>> > >>>>> Not yet, still stuck at the "drmModeGetResources failed" error. > >>>> > >>>> I can also get this same error with the new kmscube package from: > >>>> https://cgit.freedesktop.org/mesa/kmscube/log/ > >>>> > >>>> # kmscube > >>>> drmModeGetResources failed: Invalid argument > >>>> failed to initialize legacy DRM > >>>> > >>>> It works if I explicitly pass the /dev/dri device: > >>>> > >>>> kmscube -D /dev/dri/card1 > >>>> > >>>> I think QT5 is trying to access /dev/dri/card0, but it should be using > >>>> /dev/dri/card1 instead. > >>> > >>> Yes, this is the problem. > >>> > >>> If I do an ugly hack like this: > >>> > >>> --- a/src/plugins/platforms/eglfs/deviceintegration/eglfs_kms/qeglfskmsdevice.cpp > >>> +++ b/src/plugins/platforms/eglfs/deviceintegration/eglfs_kms/qeglfskmsdevice.cpp > >>> @@ -344,7 +344,7 @@ bool QEglFSKmsDevice::open() > >>> Q_ASSERT(m_gbm_device == Q_NULLPTR); > >>> > >>> qCDebug(qLcEglfsKmsDebug) << "Opening device" << m_path; > >>> - m_dri_fd = qt_safe_open(m_path.toLocal8Bit().constData(), O_RDWR > >>> | O_CLOEXEC); > >>> + m_dri_fd = qt_safe_open("/dev/dri/card1", O_RDWR | O_CLOEXEC); > >>> if (m_dri_fd == -1) { > >>> qErrnoWarning("Could not open DRM device %s", qPrintable(m_path)); > >>> return false; > >>> -- > >>> 2.7.4 > >>> > >>> Then the demo is launched. It still do not run very well, but at least > >>> it starts. > >>> > >>> Does anyone have an idea on what would be a proper fix? > >> > >> That's strange I didn't have to do that. Worst part is that my demo > >> image is using Qt5.6.2 (same as your image) and that I checked no > >> special env variable is set. > >> > >> Maybe you can try your kernel on my image, see if that works. > > > > I just used rootfs, from the link, and kernel 4.9.13 mainline. > > > > Able to see the demo nicely. > > > > Order: > > - kmscube > > - Boundary devices Tux > > - jellyfish > > - ideas in motion > > - 28 fps (1. Lord of the Rings) > > > > So, I'm investigating config options on buildroot will let you know > > for further update. > > And by the way, did anyone find this issue. > > In file included from > /workspace/buildroot/output/host/usr/arm-buildroot-linux-gnueabihf/sysroot/usr/include/EGL/egl.h:39:0, > from > ../../../../include/QtEglSupport/5.8.0/QtEglSupport/private/../../../../../src/platformsupport/eglconvenience/qt_egl_p.h:63, > from > ../../../../include/QtEglSupport/5.8.0/QtEglSupport/private/qt_egl_p.h:1, > from api/qeglfsglobal_p.h:56, > from api/qeglfsintegration_p.h:54, > from qeglfsmain.cpp:41: > /workspace/buildroot/output/host/usr/arm-buildroot-linux-gnueabihf/sysroot/usr/include/EGL/eglplatform.h:119:22: > fatal error: X11/Xlib.h: No such file or directory > #include <X11/Xlib.h> > ^ > compilation terminated. > > Look like Qt5.8.0 is unable to build with current tool-chain, 5.6.2 succeeded Should not happen ;-), mind to share the .config file? Regards, Peter > > thanks!
Hi Peter,
On Mon, Mar 6, 2017 at 6:05 PM, Peter Seiderer <ps.report@gmx.net> wrote:
> Should not happen ;-), mind to share the .config file?
I am also able to reproduce this build error. Just shared my defconfig
on a new thread, so that we can discuss about it.
Thanks
--- a/src/plugins/platforms/eglfs/deviceintegration/eglfs_kms/qeglfskmsdevice.cpp +++ b/src/plugins/platforms/eglfs/deviceintegration/eglfs_kms/qeglfskmsdevice.cpp @@ -344,7 +344,7 @@ bool QEglFSKmsDevice::open() Q_ASSERT(m_gbm_device == Q_NULLPTR); qCDebug(qLcEglfsKmsDebug) << "Opening device" << m_path; - m_dri_fd = qt_safe_open(m_path.toLocal8Bit().constData(), O_RDWR | O_CLOEXEC); + m_dri_fd = qt_safe_open("/dev/dri/card1", O_RDWR | O_CLOEXEC); if (m_dri_fd == -1) { qErrnoWarning("Could not open DRM device %s", qPrintable(m_path));