All of lore.kernel.org
 help / color / mirror / Atom feed
* [Buildroot] COG crashes on Pi4 (wpe_webkit)
@ 2020-05-19 13:16 usul27
  2020-05-19 17:44 ` Peter Seiderer
  0 siblings, 1 reply; 10+ messages in thread
From: usul27 @ 2020-05-19 13:16 UTC (permalink / raw)
  To: buildroot

I'm using the latest release candidate on the Raspberry Pi 3 and 4. While the
Pi3 version works well, the wpe_webkit browser crashes. I don't have a
version with debug symbols running, but maybe somebody have seen this
already and has some hint:

(gdb) backtrace
#0  0x00000000 in ?? ()
#1  0xb26382cc in ?? () from /lib/libffi.so.7
#2  0xb2637c94 in ?? () from /lib/libffi.so.7
#3  0xb2637f24 in ffi_call () from /lib/libffi.so.7
#4  0xb2380c30 in ?? () from /lib/libwayland-server.so.0
Backtrace stopped: previous frame identical to this frame (corrupt stack?)

Weston is using the DRM backend:
/usr/bin/weston --backend=drm-backend.so --modules=systemd-notify.so





--
Sent from: http://buildroot-busybox.2317881.n4.nabble.com/

^ permalink raw reply	[flat|nested] 10+ messages in thread

* [Buildroot] COG crashes on Pi4 (wpe_webkit)
  2020-05-19 13:16 [Buildroot] COG crashes on Pi4 (wpe_webkit) usul27
@ 2020-05-19 17:44 ` Peter Seiderer
  2020-05-20  7:07   ` usul27
  0 siblings, 1 reply; 10+ messages in thread
From: Peter Seiderer @ 2020-05-19 17:44 UTC (permalink / raw)
  To: buildroot

Hello Daniel,

On Tue, 19 May 2020 08:16:20 -0500 (CDT), usul27 <daniel@hifiberry.com> wrote:

> I'm using the latest release candidate on the Raspberry Pi 3 and 4. While the
> Pi3 version works well, the wpe_webkit browser crashes. I don't have a
> version with debug symbols running, but maybe somebody have seen this
> already and has some hint:
>
> (gdb) backtrace
> #0  0x00000000 in ?? ()
> #1  0xb26382cc in ?? () from /lib/libffi.so.7
> #2  0xb2637c94 in ?? () from /lib/libffi.so.7
> #3  0xb2637f24 in ffi_call () from /lib/libffi.so.7
> #4  0xb2380c30 in ?? () from /lib/libwayland-server.so.0
> Backtrace stopped: previous frame identical to this frame (corrupt stack?)
>
> Weston is using the DRM backend:
> /usr/bin/weston --backend=drm-backend.so --modules=systemd-notify.so

No known failure (at least for me), can you share your config (or defconfig)?

Regards,
Peter

>
>
>
>
>
> --
> Sent from: http://buildroot-busybox.2317881.n4.nabble.com/
> _______________________________________________
> buildroot mailing list
> buildroot at busybox.net
> http://lists.busybox.net/mailman/listinfo/buildroot

^ permalink raw reply	[flat|nested] 10+ messages in thread

* [Buildroot] COG crashes on Pi4 (wpe_webkit)
  2020-05-19 17:44 ` Peter Seiderer
@ 2020-05-20  7:07   ` usul27
  2020-05-22 14:17     ` usul27
  0 siblings, 1 reply; 10+ messages in thread
From: usul27 @ 2020-05-20  7:07 UTC (permalink / raw)
  To: buildroot

Peter Seiderer wrote
> No known failure (at least for me), can you share your config (or
> defconfig)?
> 
> Regards,
> Peter

Good to hear this. I'll do some more research using debug symbols to figure
out what's crashing this. 
Interesting to me is that it works on the Pi3, but crashes on the Pi4.

Best regards,
Daniel




--
Sent from: http://buildroot-busybox.2317881.n4.nabble.com/

^ permalink raw reply	[flat|nested] 10+ messages in thread

* [Buildroot] COG crashes on Pi4 (wpe_webkit)
  2020-05-20  7:07   ` usul27
@ 2020-05-22 14:17     ` usul27
  2020-05-23 20:29       ` Peter Seiderer
  0 siblings, 1 reply; 10+ messages in thread
From: usul27 @ 2020-05-22 14:17 UTC (permalink / raw)
  To: buildroot

Ok, finally got a build with debug symbols:

Thread 1 "cog" received signal SIGSEGV, Segmentation fault.
0x00000000 in ?? ()
(gdb) backtrace
#0  0x00000000 in ?? ()
#1  0xb1799f84 in ffi_call_SYSV () from /lib/libffi.so.7
#2  0xb17993fc in ffi_call_int () from /lib/libffi.so.7
#3  0xb1799b1c in ffi_call () from /lib/libffi.so.7
#4  0xb149b28c in wl_closure_invoke (closure=closure at entry=0xadb03a00,
flags=flags at entry=2, target=<optimized out>, opcode=3025639260,
data=<optimized out>, data at entry=0x74c60) at src/connection.c:1014
#5  0xb1497db0 in wl_client_connection_data (fd=<optimized out>,
mask=<optimized out>, data=0x74c60) at src/wayland-server.c:429
#6  0xb1499608 in wl_event_loop_dispatch (loop=0x5bfc8, timeout=<optimized
out>) at src/event-loop.c:640
#7  0xb0c59400 in WS::ServerSource::{lambda(_GSource*, int (*)(void*),
void*)#3}::_FUN(_GSource*, int (*)(void*), void*) () from
/lib/libWPEBackend-fdo-1.0.so.1
Backtrace stopped: previous frame identical to this frame (corrupt stack?)

Somebody has any idea?




--
Sent from: http://buildroot-busybox.2317881.n4.nabble.com/

^ permalink raw reply	[flat|nested] 10+ messages in thread

* [Buildroot] COG crashes on Pi4 (wpe_webkit)
  2020-05-22 14:17     ` usul27
@ 2020-05-23 20:29       ` Peter Seiderer
  2020-05-24  8:59         ` Daniel HiFiBerry
  0 siblings, 1 reply; 10+ messages in thread
From: Peter Seiderer @ 2020-05-23 20:29 UTC (permalink / raw)
  To: buildroot

Hello Daniel,

On Fri, 22 May 2020 09:17:18 -0500 (CDT), usul27 <daniel@hifiberry.com> wrote:

> Ok, finally got a build with debug symbols:
>
> Thread 1 "cog" received signal SIGSEGV, Segmentation fault.
> 0x00000000 in ?? ()
> (gdb) backtrace
> #0  0x00000000 in ?? ()
> #1  0xb1799f84 in ffi_call_SYSV () from /lib/libffi.so.7
> #2  0xb17993fc in ffi_call_int () from /lib/libffi.so.7
> #3  0xb1799b1c in ffi_call () from /lib/libffi.so.7
> #4  0xb149b28c in wl_closure_invoke (closure=closure at entry=0xadb03a00,
> flags=flags at entry=2, target=<optimized out>, opcode=3025639260,
> data=<optimized out>, data at entry=0x74c60) at src/connection.c:1014
> #5  0xb1497db0 in wl_client_connection_data (fd=<optimized out>,
> mask=<optimized out>, data=0x74c60) at src/wayland-server.c:429
> #6  0xb1499608 in wl_event_loop_dispatch (loop=0x5bfc8, timeout=<optimized
> out>) at src/event-loop.c:640
> #7  0xb0c59400 in WS::ServerSource::{lambda(_GSource*, int (*)(void*),
> void*)#3}::_FUN(_GSource*, int (*)(void*), void*) () from
> /lib/libWPEBackend-fdo-1.0.so.1
> Backtrace stopped: previous frame identical to this frame (corrupt stack?)
>
> Somebody has any idea?

Did some test on RPi4 with the following defconfig:

BR2_arm=y
BR2_cortex_a72=y
BR2_ARM_FPU_NEON_VFPV4=y
BR2_ENABLE_DEBUG=y
BR2_TOOLCHAIN_BUILDROOT_GLIBC=y
BR2_PACKAGE_HOST_LINUX_HEADERS_CUSTOM_4_19=y
BR2_BINUTILS_VERSION_2_31_X=y
BR2_TOOLCHAIN_BUILDROOT_CXX=y
BR2_PACKAGE_HOST_GDB=y
BR2_GDB_VERSION_8_3=y
BR2_ROOTFS_DEVICE_CREATION_DYNAMIC_EUDEV=y
BR2_SYSTEM_DHCP="eth0"
BR2_GENERATE_LOCALE="en_US"
BR2_ROOTFS_POST_BUILD_SCRIPT="board/raspberrypi4/post-build.sh"
BR2_ROOTFS_POST_IMAGE_SCRIPT="board/raspberrypi4/post-image.sh"
BR2_ROOTFS_POST_SCRIPT_ARGS="--add-pi3-miniuart-bt-overlay"
BR2_LINUX_KERNEL=y
BR2_LINUX_KERNEL_CUSTOM_TARBALL=y
BR2_LINUX_KERNEL_CUSTOM_TARBALL_LOCATION="$(call github,raspberrypi,linux,4f2a4cc501c428c940549f39d5562e60404ac4f7)/linux-4f2a4cc501c428c940549f39d5562e60404ac4f7.tar.gz"
BR2_LINUX_KERNEL_DEFCONFIG="bcm2711"
BR2_LINUX_KERNEL_DTS_SUPPORT=y
BR2_LINUX_KERNEL_INTREE_DTS_NAME="bcm2711-rpi-4-b"
BR2_LINUX_KERNEL_NEEDS_HOST_OPENSSL=y
BR2_PACKAGE_STRACE=y
BR2_PACKAGE_DEJAVU=y
BR2_PACKAGE_COG=y
BR2_PACKAGE_MESA3D=y
BR2_PACKAGE_MESA3D_GALLIUM_DRIVER_KMSRO=y
BR2_PACKAGE_MESA3D_GALLIUM_DRIVER_V3D=y
BR2_PACKAGE_MESA3D_GALLIUM_DRIVER_VC4=y
BR2_PACKAGE_MESA3D_OPENGL_ES=y
BR2_PACKAGE_WESTON=y
BR2_PACKAGE_WESTON_DRM=y
BR2_PACKAGE_WESTON_DEMO_CLIENTS=y
BR2_PACKAGE_RPI_FIRMWARE=y
BR2_PACKAGE_RPI_FIRMWARE_VARIANT_PI4=y
BR2_PACKAGE_WPEWEBKIT=y
BR2_PACKAGE_OPENSSH=y
BR2_TARGET_ROOTFS_EXT2=y
BR2_TARGET_ROOTFS_EXT2_4=y
BR2_TARGET_ROOTFS_EXT2_LABEL=""
BR2_TARGET_ROOTFS_EXT2_SIZE="256M"
# BR2_TARGET_ROOTFS_TAR is not set
BR2_PACKAGE_HOST_DOSFSTOOLS=y
BR2_PACKAGE_HOST_GENIMAGE=y
BR2_PACKAGE_HOST_MTOOLS=y

Running:

	$ export XDG_RUNTIME_DIR=/tmp
	$ weston --tty 1 &
	$ cog --platform=fdo

leads to the 'Segmentation fault' as you described...

Running:

	$ export XDG_RUNTIME_DIR=/tmp
	$ weston --tty 1 --backend=drm-backend.so &
	$ cog --platform=fdo

works...

Regards,
Peter


>
>
>
>
> --
> Sent from: http://buildroot-busybox.2317881.n4.nabble.com/
> _______________________________________________
> buildroot mailing list
> buildroot at busybox.net
> http://lists.busybox.net/mailman/listinfo/buildroot

^ permalink raw reply	[flat|nested] 10+ messages in thread

* [Buildroot] COG crashes on Pi4 (wpe_webkit)
  2020-05-23 20:29       ` Peter Seiderer
@ 2020-05-24  8:59         ` Daniel HiFiBerry
  2020-05-24  9:42           ` Peter Seiderer
  0 siblings, 1 reply; 10+ messages in thread
From: Daniel HiFiBerry @ 2020-05-24  8:59 UTC (permalink / raw)
  To: buildroot

Hi Peter,

While I didn't test with a minimal configuration, using the 
"--backend=drm-backend.so" doesn't make a difference in my full-blown 
build.
I'm using this already. I will do some more experiments...

Best regards,
Daniel

------ Originalnachricht ------
Von: "Peter Seiderer" <ps.report@gmx.net>
An: "usul27" <daniel@hifiberry.com>
Cc: buildroot at busybox.net
Gesendet: 5/23/2020 10:29:48 PM
Betreff: Re: [Buildroot] COG crashes on Pi4 (wpe_webkit)

>Hello Daniel,
>
>On Fri, 22 May 2020 09:17:18 -0500 (CDT), usul27 <daniel@hifiberry.com> wrote:
>
>>  Ok, finally got a build with debug symbols:
>>
>>  Thread 1 "cog" received signal SIGSEGV, Segmentation fault.
>>  0x00000000 in ?? ()
>>  (gdb) backtrace
>>  #0  0x00000000 in ?? ()
>>  #1  0xb1799f84 in ffi_call_SYSV () from /lib/libffi.so.7
>>  #2  0xb17993fc in ffi_call_int () from /lib/libffi.so.7
>>  #3  0xb1799b1c in ffi_call () from /lib/libffi.so.7
>>  #4  0xb149b28c in wl_closure_invoke (closure=closure at entry=0xadb03a00,
>>  flags=flags at entry=2, target=<optimized out>, opcode=3025639260,
>>  data=<optimized out>, data at entry=0x74c60) at src/connection.c:1014
>>  #5  0xb1497db0 in wl_client_connection_data (fd=<optimized out>,
>>  mask=<optimized out>, data=0x74c60) at src/wayland-server.c:429
>>  #6  0xb1499608 in wl_event_loop_dispatch (loop=0x5bfc8, timeout=<optimized
>>  out>) at src/event-loop.c:640
>>  #7  0xb0c59400 in WS::ServerSource::{lambda(_GSource*, int (*)(void*),
>>  void*)#3}::_FUN(_GSource*, int (*)(void*), void*) () from
>>  /lib/libWPEBackend-fdo-1.0.so.1
>>  Backtrace stopped: previous frame identical to this frame (corrupt stack?)
>>
>>  Somebody has any idea?
>
>Did some test on RPi4 with the following defconfig:
>
>BR2_arm=y
>BR2_cortex_a72=y
>BR2_ARM_FPU_NEON_VFPV4=y
>BR2_ENABLE_DEBUG=y
>BR2_TOOLCHAIN_BUILDROOT_GLIBC=y
>BR2_PACKAGE_HOST_LINUX_HEADERS_CUSTOM_4_19=y
>BR2_BINUTILS_VERSION_2_31_X=y
>BR2_TOOLCHAIN_BUILDROOT_CXX=y
>BR2_PACKAGE_HOST_GDB=y
>BR2_GDB_VERSION_8_3=y
>BR2_ROOTFS_DEVICE_CREATION_DYNAMIC_EUDEV=y
>BR2_SYSTEM_DHCP="eth0"
>BR2_GENERATE_LOCALE="en_US"
>BR2_ROOTFS_POST_BUILD_SCRIPT="board/raspberrypi4/post-build.sh"
>BR2_ROOTFS_POST_IMAGE_SCRIPT="board/raspberrypi4/post-image.sh"
>BR2_ROOTFS_POST_SCRIPT_ARGS="--add-pi3-miniuart-bt-overlay"
>BR2_LINUX_KERNEL=y
>BR2_LINUX_KERNEL_CUSTOM_TARBALL=y
>BR2_LINUX_KERNEL_CUSTOM_TARBALL_LOCATION="$(call github,raspberrypi,linux,4f2a4cc501c428c940549f39d5562e60404ac4f7)/linux-4f2a4cc501c428c940549f39d5562e60404ac4f7.tar.gz"
>BR2_LINUX_KERNEL_DEFCONFIG="bcm2711"
>BR2_LINUX_KERNEL_DTS_SUPPORT=y
>BR2_LINUX_KERNEL_INTREE_DTS_NAME="bcm2711-rpi-4-b"
>BR2_LINUX_KERNEL_NEEDS_HOST_OPENSSL=y
>BR2_PACKAGE_STRACE=y
>BR2_PACKAGE_DEJAVU=y
>BR2_PACKAGE_COG=y
>BR2_PACKAGE_MESA3D=y
>BR2_PACKAGE_MESA3D_GALLIUM_DRIVER_KMSRO=y
>BR2_PACKAGE_MESA3D_GALLIUM_DRIVER_V3D=y
>BR2_PACKAGE_MESA3D_GALLIUM_DRIVER_VC4=y
>BR2_PACKAGE_MESA3D_OPENGL_ES=y
>BR2_PACKAGE_WESTON=y
>BR2_PACKAGE_WESTON_DRM=y
>BR2_PACKAGE_WESTON_DEMO_CLIENTS=y
>BR2_PACKAGE_RPI_FIRMWARE=y
>BR2_PACKAGE_RPI_FIRMWARE_VARIANT_PI4=y
>BR2_PACKAGE_WPEWEBKIT=y
>BR2_PACKAGE_OPENSSH=y
>BR2_TARGET_ROOTFS_EXT2=y
>BR2_TARGET_ROOTFS_EXT2_4=y
>BR2_TARGET_ROOTFS_EXT2_LABEL=""
>BR2_TARGET_ROOTFS_EXT2_SIZE="256M"
># BR2_TARGET_ROOTFS_TAR is not set
>BR2_PACKAGE_HOST_DOSFSTOOLS=y
>BR2_PACKAGE_HOST_GENIMAGE=y
>BR2_PACKAGE_HOST_MTOOLS=y
>
>Running:
>
>	$ export XDG_RUNTIME_DIR=/tmp
>	$ weston --tty 1 &
>	$ cog --platform=fdo
>
>leads to the 'Segmentation fault' as you described...
>
>Running:
>
>	$ export XDG_RUNTIME_DIR=/tmp
>	$ weston --tty 1 --backend=drm-backend.so &
>	$ cog --platform=fdo
>
>works...
>
>Regards,
>Peter
>
>
>>
>>
>>
>>
>>  --
>>  Sent from: http://buildroot-busybox.2317881.n4.nabble.com/
>>  _______________________________________________
>>  buildroot mailing list
>>buildroot at busybox.net
>>http://lists.busybox.net/mailman/listinfo/buildroot
>

^ permalink raw reply	[flat|nested] 10+ messages in thread

* [Buildroot] COG crashes on Pi4 (wpe_webkit)
  2020-05-24  8:59         ` Daniel HiFiBerry
@ 2020-05-24  9:42           ` Peter Seiderer
  2020-05-24  9:43             ` Daniel HiFiBerry
  0 siblings, 1 reply; 10+ messages in thread
From: Peter Seiderer @ 2020-05-24  9:42 UTC (permalink / raw)
  To: buildroot

Hello Daniel,

On Sun, 24 May 2020 08:59:53 +0000, "Daniel HiFiBerry" <daniel@hifiberry.com> wrote:

> Hi Peter,
>
> While I didn't test with a minimal configuration, using the
> "--backend=drm-backend.so" doesn't make a difference in my full-blown
> build.
> I'm using this already. I will do some more experiments...

The trick for hifiberry-os should do (untested) the missing

	BR2_PACKAGE_MESA3D_GALLIUM_DRIVER_KMSRO=y

in your configuration (see the 'hidden' hint in package/mesa3d/Config.in
([1])...

Regards,
Peter

[1] https://git.buildroot.net/buildroot/tree/package/mesa3d/Config.in#n265

>
> Best regards,
> Daniel
>
> ------ Originalnachricht ------
> Von: "Peter Seiderer" <ps.report@gmx.net>
> An: "usul27" <daniel@hifiberry.com>
> Cc: buildroot at busybox.net
> Gesendet: 5/23/2020 10:29:48 PM
> Betreff: Re: [Buildroot] COG crashes on Pi4 (wpe_webkit)
>
> >Hello Daniel,
> >
> >On Fri, 22 May 2020 09:17:18 -0500 (CDT), usul27 <daniel@hifiberry.com> wrote:
> >
> >>  Ok, finally got a build with debug symbols:
> >>
> >>  Thread 1 "cog" received signal SIGSEGV, Segmentation fault.
> >>  0x00000000 in ?? ()
> >>  (gdb) backtrace
> >>  #0  0x00000000 in ?? ()
> >>  #1  0xb1799f84 in ffi_call_SYSV () from /lib/libffi.so.7
> >>  #2  0xb17993fc in ffi_call_int () from /lib/libffi.so.7
> >>  #3  0xb1799b1c in ffi_call () from /lib/libffi.so.7
> >>  #4  0xb149b28c in wl_closure_invoke (closure=closure at entry=0xadb03a00,
> >>  flags=flags at entry=2, target=<optimized out>, opcode=3025639260,
> >>  data=<optimized out>, data at entry=0x74c60) at src/connection.c:1014
> >>  #5  0xb1497db0 in wl_client_connection_data (fd=<optimized out>,
> >>  mask=<optimized out>, data=0x74c60) at src/wayland-server.c:429
> >>  #6  0xb1499608 in wl_event_loop_dispatch (loop=0x5bfc8, timeout=<optimized
> >>  out>) at src/event-loop.c:640
> >>  #7  0xb0c59400 in WS::ServerSource::{lambda(_GSource*, int (*)(void*),
> >>  void*)#3}::_FUN(_GSource*, int (*)(void*), void*) () from
> >>  /lib/libWPEBackend-fdo-1.0.so.1
> >>  Backtrace stopped: previous frame identical to this frame (corrupt stack?)
> >>
> >>  Somebody has any idea?
> >
> >Did some test on RPi4 with the following defconfig:
> >
> >BR2_arm=y
> >BR2_cortex_a72=y
> >BR2_ARM_FPU_NEON_VFPV4=y
> >BR2_ENABLE_DEBUG=y
> >BR2_TOOLCHAIN_BUILDROOT_GLIBC=y
> >BR2_PACKAGE_HOST_LINUX_HEADERS_CUSTOM_4_19=y
> >BR2_BINUTILS_VERSION_2_31_X=y
> >BR2_TOOLCHAIN_BUILDROOT_CXX=y
> >BR2_PACKAGE_HOST_GDB=y
> >BR2_GDB_VERSION_8_3=y
> >BR2_ROOTFS_DEVICE_CREATION_DYNAMIC_EUDEV=y
> >BR2_SYSTEM_DHCP="eth0"
> >BR2_GENERATE_LOCALE="en_US"
> >BR2_ROOTFS_POST_BUILD_SCRIPT="board/raspberrypi4/post-build.sh"
> >BR2_ROOTFS_POST_IMAGE_SCRIPT="board/raspberrypi4/post-image.sh"
> >BR2_ROOTFS_POST_SCRIPT_ARGS="--add-pi3-miniuart-bt-overlay"
> >BR2_LINUX_KERNEL=y
> >BR2_LINUX_KERNEL_CUSTOM_TARBALL=y
> >BR2_LINUX_KERNEL_CUSTOM_TARBALL_LOCATION="$(call github,raspberrypi,linux,4f2a4cc501c428c940549f39d5562e60404ac4f7)/linux-4f2a4cc501c428c940549f39d5562e60404ac4f7.tar.gz"
> >BR2_LINUX_KERNEL_DEFCONFIG="bcm2711"
> >BR2_LINUX_KERNEL_DTS_SUPPORT=y
> >BR2_LINUX_KERNEL_INTREE_DTS_NAME="bcm2711-rpi-4-b"
> >BR2_LINUX_KERNEL_NEEDS_HOST_OPENSSL=y
> >BR2_PACKAGE_STRACE=y
> >BR2_PACKAGE_DEJAVU=y
> >BR2_PACKAGE_COG=y
> >BR2_PACKAGE_MESA3D=y
> >BR2_PACKAGE_MESA3D_GALLIUM_DRIVER_KMSRO=y
> >BR2_PACKAGE_MESA3D_GALLIUM_DRIVER_V3D=y
> >BR2_PACKAGE_MESA3D_GALLIUM_DRIVER_VC4=y
> >BR2_PACKAGE_MESA3D_OPENGL_ES=y
> >BR2_PACKAGE_WESTON=y
> >BR2_PACKAGE_WESTON_DRM=y
> >BR2_PACKAGE_WESTON_DEMO_CLIENTS=y
> >BR2_PACKAGE_RPI_FIRMWARE=y
> >BR2_PACKAGE_RPI_FIRMWARE_VARIANT_PI4=y
> >BR2_PACKAGE_WPEWEBKIT=y
> >BR2_PACKAGE_OPENSSH=y
> >BR2_TARGET_ROOTFS_EXT2=y
> >BR2_TARGET_ROOTFS_EXT2_4=y
> >BR2_TARGET_ROOTFS_EXT2_LABEL=""
> >BR2_TARGET_ROOTFS_EXT2_SIZE="256M"
> ># BR2_TARGET_ROOTFS_TAR is not set
> >BR2_PACKAGE_HOST_DOSFSTOOLS=y
> >BR2_PACKAGE_HOST_GENIMAGE=y
> >BR2_PACKAGE_HOST_MTOOLS=y
> >
> >Running:
> >
> >	$ export XDG_RUNTIME_DIR=/tmp
> >	$ weston --tty 1 &
> >	$ cog --platform=fdo
> >
> >leads to the 'Segmentation fault' as you described...
> >
> >Running:
> >
> >	$ export XDG_RUNTIME_DIR=/tmp
> >	$ weston --tty 1 --backend=drm-backend.so &
> >	$ cog --platform=fdo
> >
> >works...
> >
> >Regards,
> >Peter
> >
> >
> >>
> >>
> >>
> >>
> >>  --
> >>  Sent from: http://buildroot-busybox.2317881.n4.nabble.com/
> >>  _______________________________________________
> >>  buildroot mailing list
> >>buildroot at busybox.net
> >>http://lists.busybox.net/mailman/listinfo/buildroot
> >
>
> _______________________________________________
> buildroot mailing list
> buildroot at busybox.net
> http://lists.busybox.net/mailman/listinfo/buildroot

^ permalink raw reply	[flat|nested] 10+ messages in thread

* [Buildroot] COG crashes on Pi4 (wpe_webkit)
  2020-05-24  9:42           ` Peter Seiderer
@ 2020-05-24  9:43             ` Daniel HiFiBerry
  2020-05-24 12:43               ` Peter Seiderer
  0 siblings, 1 reply; 10+ messages in thread
From: Daniel HiFiBerry @ 2020-05-24  9:43 UTC (permalink / raw)
  To: buildroot

I have this integrated (it's just not on Github).

------ Originalnachricht ------
Von: "Peter Seiderer" <ps.report@gmx.net>
An: "Daniel HiFiBerry" <daniel@hifiberry.com>
Cc: buildroot at busybox.net
Gesendet: 5/24/2020 11:42:31 AM
Betreff: Re: [Buildroot] COG crashes on Pi4 (wpe_webkit)

>Hello Daniel,
>
>On Sun, 24 May 2020 08:59:53 +0000, "Daniel HiFiBerry" <daniel@hifiberry.com> wrote:
>
>>  Hi Peter,
>>
>>  While I didn't test with a minimal configuration, using the
>>  "--backend=drm-backend.so" doesn't make a difference in my full-blown
>>  build.
>>  I'm using this already. I will do some more experiments...
>
>The trick for hifiberry-os should do (untested) the missing
>
>	BR2_PACKAGE_MESA3D_GALLIUM_DRIVER_KMSRO=y
>
>in your configuration (see the 'hidden' hint in package/mesa3d/Config.in
>([1])...
>
>Regards,
>Peter
>
>[1] https://git.buildroot.net/buildroot/tree/package/mesa3d/Config.in#n265
>
>>
>>  Best regards,
>>  Daniel
>>
>>  ------ Originalnachricht ------
>>  Von: "Peter Seiderer" <ps.report@gmx.net>
>>  An: "usul27" <daniel@hifiberry.com>
>>  Cc: buildroot at busybox.net
>>  Gesendet: 5/23/2020 10:29:48 PM
>>  Betreff: Re: [Buildroot] COG crashes on Pi4 (wpe_webkit)
>>
>>  >Hello Daniel,
>>  >
>>  >On Fri, 22 May 2020 09:17:18 -0500 (CDT), usul27 <daniel@hifiberry.com> wrote:
>>  >
>>  >>  Ok, finally got a build with debug symbols:
>>  >>
>>  >>  Thread 1 "cog" received signal SIGSEGV, Segmentation fault.
>>  >>  0x00000000 in ?? ()
>>  >>  (gdb) backtrace
>>  >>  #0  0x00000000 in ?? ()
>>  >>  #1  0xb1799f84 in ffi_call_SYSV () from /lib/libffi.so.7
>>  >>  #2  0xb17993fc in ffi_call_int () from /lib/libffi.so.7
>>  >>  #3  0xb1799b1c in ffi_call () from /lib/libffi.so.7
>>  >>  #4  0xb149b28c in wl_closure_invoke (closure=closure at entry=0xadb03a00,
>>  >>  flags=flags at entry=2, target=<optimized out>, opcode=3025639260,
>>  >>  data=<optimized out>, data at entry=0x74c60) at src/connection.c:1014
>>  >>  #5  0xb1497db0 in wl_client_connection_data (fd=<optimized out>,
>>  >>  mask=<optimized out>, data=0x74c60) at src/wayland-server.c:429
>>  >>  #6  0xb1499608 in wl_event_loop_dispatch (loop=0x5bfc8, timeout=<optimized
>>  >>  out>) at src/event-loop.c:640
>>  >>  #7  0xb0c59400 in WS::ServerSource::{lambda(_GSource*, int (*)(void*),
>>  >>  void*)#3}::_FUN(_GSource*, int (*)(void*), void*) () from
>>  >>  /lib/libWPEBackend-fdo-1.0.so.1
>>  >>  Backtrace stopped: previous frame identical to this frame (corrupt stack?)
>>  >>
>>  >>  Somebody has any idea?
>>  >
>>  >Did some test on RPi4 with the following defconfig:
>>  >
>>  >BR2_arm=y
>>  >BR2_cortex_a72=y
>>  >BR2_ARM_FPU_NEON_VFPV4=y
>>  >BR2_ENABLE_DEBUG=y
>>  >BR2_TOOLCHAIN_BUILDROOT_GLIBC=y
>>  >BR2_PACKAGE_HOST_LINUX_HEADERS_CUSTOM_4_19=y
>>  >BR2_BINUTILS_VERSION_2_31_X=y
>>  >BR2_TOOLCHAIN_BUILDROOT_CXX=y
>>  >BR2_PACKAGE_HOST_GDB=y
>>  >BR2_GDB_VERSION_8_3=y
>>  >BR2_ROOTFS_DEVICE_CREATION_DYNAMIC_EUDEV=y
>>  >BR2_SYSTEM_DHCP="eth0"
>>  >BR2_GENERATE_LOCALE="en_US"
>>  >BR2_ROOTFS_POST_BUILD_SCRIPT="board/raspberrypi4/post-build.sh"
>>  >BR2_ROOTFS_POST_IMAGE_SCRIPT="board/raspberrypi4/post-image.sh"
>>  >BR2_ROOTFS_POST_SCRIPT_ARGS="--add-pi3-miniuart-bt-overlay"
>>  >BR2_LINUX_KERNEL=y
>>  >BR2_LINUX_KERNEL_CUSTOM_TARBALL=y
>>  >BR2_LINUX_KERNEL_CUSTOM_TARBALL_LOCATION="$(call github,raspberrypi,linux,4f2a4cc501c428c940549f39d5562e60404ac4f7)/linux-4f2a4cc501c428c940549f39d5562e60404ac4f7.tar.gz"
>>  >BR2_LINUX_KERNEL_DEFCONFIG="bcm2711"
>>  >BR2_LINUX_KERNEL_DTS_SUPPORT=y
>>  >BR2_LINUX_KERNEL_INTREE_DTS_NAME="bcm2711-rpi-4-b"
>>  >BR2_LINUX_KERNEL_NEEDS_HOST_OPENSSL=y
>>  >BR2_PACKAGE_STRACE=y
>>  >BR2_PACKAGE_DEJAVU=y
>>  >BR2_PACKAGE_COG=y
>>  >BR2_PACKAGE_MESA3D=y
>>  >BR2_PACKAGE_MESA3D_GALLIUM_DRIVER_KMSRO=y
>>  >BR2_PACKAGE_MESA3D_GALLIUM_DRIVER_V3D=y
>>  >BR2_PACKAGE_MESA3D_GALLIUM_DRIVER_VC4=y
>>  >BR2_PACKAGE_MESA3D_OPENGL_ES=y
>>  >BR2_PACKAGE_WESTON=y
>>  >BR2_PACKAGE_WESTON_DRM=y
>>  >BR2_PACKAGE_WESTON_DEMO_CLIENTS=y
>>  >BR2_PACKAGE_RPI_FIRMWARE=y
>>  >BR2_PACKAGE_RPI_FIRMWARE_VARIANT_PI4=y
>>  >BR2_PACKAGE_WPEWEBKIT=y
>>  >BR2_PACKAGE_OPENSSH=y
>>  >BR2_TARGET_ROOTFS_EXT2=y
>>  >BR2_TARGET_ROOTFS_EXT2_4=y
>>  >BR2_TARGET_ROOTFS_EXT2_LABEL=""
>>  >BR2_TARGET_ROOTFS_EXT2_SIZE="256M"
>>  ># BR2_TARGET_ROOTFS_TAR is not set
>>  >BR2_PACKAGE_HOST_DOSFSTOOLS=y
>>  >BR2_PACKAGE_HOST_GENIMAGE=y
>>  >BR2_PACKAGE_HOST_MTOOLS=y
>>  >
>>  >Running:
>>  >
>>  >	$ export XDG_RUNTIME_DIR=/tmp
>>  >	$ weston --tty 1 &
>>  >	$ cog --platform=fdo
>>  >
>>  >leads to the 'Segmentation fault' as you described...
>>  >
>>  >Running:
>>  >
>>  >	$ export XDG_RUNTIME_DIR=/tmp
>>  >	$ weston --tty 1 --backend=drm-backend.so &
>>  >	$ cog --platform=fdo
>>  >
>>  >works...
>>  >
>>  >Regards,
>>  >Peter
>>  >
>>  >
>>  >>
>>  >>
>>  >>
>>  >>
>>  >>  --
>>  >>  Sent from: http://buildroot-busybox.2317881.n4.nabble.com/
>>  >>  _______________________________________________
>>  >>  buildroot mailing list
>>  >>buildroot at busybox.net
>>  >>http://lists.busybox.net/mailman/listinfo/buildroot
>>  >
>>
>>  _______________________________________________
>>  buildroot mailing list
>>buildroot at busybox.net
>>http://lists.busybox.net/mailman/listinfo/buildroot
>

^ permalink raw reply	[flat|nested] 10+ messages in thread

* [Buildroot] COG crashes on Pi4 (wpe_webkit)
  2020-05-24  9:43             ` Daniel HiFiBerry
@ 2020-05-24 12:43               ` Peter Seiderer
  2020-05-25  7:18                 ` Daniel HiFiBerry
  0 siblings, 1 reply; 10+ messages in thread
From: Peter Seiderer @ 2020-05-24 12:43 UTC (permalink / raw)
  To: buildroot

Hello Daniel,

On Sun, 24 May 2020 09:43:36 +0000, "Daniel HiFiBerry" <daniel@hifiberry.com> wrote:

> I have this integrated (it's just not on Github).

Does this mean it works for you? Or do you have
BR2_PACKAGE_MESA3D_GALLIUM_DRIVER_KMSRO=y integrated
and it does not work for you (take care to not mix it with
BR2_PACKAGE_MESA3D_GALLIUM_KMSRO)?

I have a hifiberry-os running with the local gui on hdmi:

	$ ps aux
[...]
root       962  0.5  2.5  32780 23444 tty7     Ss+  12:38   0:01 /usr/bin/weston --backend=drm-backend.so --modules=systemd-notify.so
root       963  0.0  0.8  13972  8192 tty7     S+   12:38   0:00 /usr/libexec/weston-keyboard
root       964  0.0  0.6  12356  5964 tty7     S+   12:38   0:00 /usr/libexec/weston-desktop-shell
root       969  0.3  3.1 209332 29236 ?        Ssl  12:38   0:01 /usr/bin/cog -P fdo http://localhost:80/#now-playing
root       975 40.6 13.3 355752 122804 ?       SLl  12:38   2:18 /usr/libexec/wpe-webkit-1.0/WPEWebProcess 7 16
root       976  0.2  3.3 259100 31016 ?        SLl  12:38   0:01 /usr/libexec/wpe-webkit-1.0/WPENetworkProcess 8 16
[...]

The only thing I have to hack was the resize-partitions for the extra large image ;-)

Regards,
Peter

>
> ------ Originalnachricht ------
> Von: "Peter Seiderer" <ps.report@gmx.net>
> An: "Daniel HiFiBerry" <daniel@hifiberry.com>
> Cc: buildroot at busybox.net
> Gesendet: 5/24/2020 11:42:31 AM
> Betreff: Re: [Buildroot] COG crashes on Pi4 (wpe_webkit)
>
> >Hello Daniel,
> >
> >On Sun, 24 May 2020 08:59:53 +0000, "Daniel HiFiBerry" <daniel@hifiberry.com> wrote:
> >
> >>  Hi Peter,
> >>
> >>  While I didn't test with a minimal configuration, using the
> >>  "--backend=drm-backend.so" doesn't make a difference in my full-blown
> >>  build.
> >>  I'm using this already. I will do some more experiments...
> >
> >The trick for hifiberry-os should do (untested) the missing
> >
> >	BR2_PACKAGE_MESA3D_GALLIUM_DRIVER_KMSRO=y
> >
> >in your configuration (see the 'hidden' hint in package/mesa3d/Config.in
> >([1])...
> >
> >Regards,
> >Peter
> >
> >[1] https://git.buildroot.net/buildroot/tree/package/mesa3d/Config.in#n265
> >
> >>
> >>  Best regards,
> >>  Daniel
> >>
> >>  ------ Originalnachricht ------
> >>  Von: "Peter Seiderer" <ps.report@gmx.net>
> >>  An: "usul27" <daniel@hifiberry.com>
> >>  Cc: buildroot at busybox.net
> >>  Gesendet: 5/23/2020 10:29:48 PM
> >>  Betreff: Re: [Buildroot] COG crashes on Pi4 (wpe_webkit)
> >>
> >>  >Hello Daniel,
> >>  >
> >>  >On Fri, 22 May 2020 09:17:18 -0500 (CDT), usul27 <daniel@hifiberry.com> wrote:
> >>  >
> >>  >>  Ok, finally got a build with debug symbols:
> >>  >>
> >>  >>  Thread 1 "cog" received signal SIGSEGV, Segmentation fault.
> >>  >>  0x00000000 in ?? ()
> >>  >>  (gdb) backtrace
> >>  >>  #0  0x00000000 in ?? ()
> >>  >>  #1  0xb1799f84 in ffi_call_SYSV () from /lib/libffi.so.7
> >>  >>  #2  0xb17993fc in ffi_call_int () from /lib/libffi.so.7
> >>  >>  #3  0xb1799b1c in ffi_call () from /lib/libffi.so.7
> >>  >>  #4  0xb149b28c in wl_closure_invoke (closure=closure at entry=0xadb03a00,
> >>  >>  flags=flags at entry=2, target=<optimized out>, opcode=3025639260,
> >>  >>  data=<optimized out>, data at entry=0x74c60) at src/connection.c:1014
> >>  >>  #5  0xb1497db0 in wl_client_connection_data (fd=<optimized out>,
> >>  >>  mask=<optimized out>, data=0x74c60) at src/wayland-server.c:429
> >>  >>  #6  0xb1499608 in wl_event_loop_dispatch (loop=0x5bfc8, timeout=<optimized
> >>  >>  out>) at src/event-loop.c:640
> >>  >>  #7  0xb0c59400 in WS::ServerSource::{lambda(_GSource*, int (*)(void*),
> >>  >>  void*)#3}::_FUN(_GSource*, int (*)(void*), void*) () from
> >>  >>  /lib/libWPEBackend-fdo-1.0.so.1
> >>  >>  Backtrace stopped: previous frame identical to this frame (corrupt stack?)
> >>  >>
> >>  >>  Somebody has any idea?
> >>  >
> >>  >Did some test on RPi4 with the following defconfig:
> >>  >
> >>  >BR2_arm=y
> >>  >BR2_cortex_a72=y
> >>  >BR2_ARM_FPU_NEON_VFPV4=y
> >>  >BR2_ENABLE_DEBUG=y
> >>  >BR2_TOOLCHAIN_BUILDROOT_GLIBC=y
> >>  >BR2_PACKAGE_HOST_LINUX_HEADERS_CUSTOM_4_19=y
> >>  >BR2_BINUTILS_VERSION_2_31_X=y
> >>  >BR2_TOOLCHAIN_BUILDROOT_CXX=y
> >>  >BR2_PACKAGE_HOST_GDB=y
> >>  >BR2_GDB_VERSION_8_3=y
> >>  >BR2_ROOTFS_DEVICE_CREATION_DYNAMIC_EUDEV=y
> >>  >BR2_SYSTEM_DHCP="eth0"
> >>  >BR2_GENERATE_LOCALE="en_US"
> >>  >BR2_ROOTFS_POST_BUILD_SCRIPT="board/raspberrypi4/post-build.sh"
> >>  >BR2_ROOTFS_POST_IMAGE_SCRIPT="board/raspberrypi4/post-image.sh"
> >>  >BR2_ROOTFS_POST_SCRIPT_ARGS="--add-pi3-miniuart-bt-overlay"
> >>  >BR2_LINUX_KERNEL=y
> >>  >BR2_LINUX_KERNEL_CUSTOM_TARBALL=y
> >>  >BR2_LINUX_KERNEL_CUSTOM_TARBALL_LOCATION="$(call github,raspberrypi,linux,4f2a4cc501c428c940549f39d5562e60404ac4f7)/linux-4f2a4cc501c428c940549f39d5562e60404ac4f7.tar.gz"
> >>  >BR2_LINUX_KERNEL_DEFCONFIG="bcm2711"
> >>  >BR2_LINUX_KERNEL_DTS_SUPPORT=y
> >>  >BR2_LINUX_KERNEL_INTREE_DTS_NAME="bcm2711-rpi-4-b"
> >>  >BR2_LINUX_KERNEL_NEEDS_HOST_OPENSSL=y
> >>  >BR2_PACKAGE_STRACE=y
> >>  >BR2_PACKAGE_DEJAVU=y
> >>  >BR2_PACKAGE_COG=y
> >>  >BR2_PACKAGE_MESA3D=y
> >>  >BR2_PACKAGE_MESA3D_GALLIUM_DRIVER_KMSRO=y
> >>  >BR2_PACKAGE_MESA3D_GALLIUM_DRIVER_V3D=y
> >>  >BR2_PACKAGE_MESA3D_GALLIUM_DRIVER_VC4=y
> >>  >BR2_PACKAGE_MESA3D_OPENGL_ES=y
> >>  >BR2_PACKAGE_WESTON=y
> >>  >BR2_PACKAGE_WESTON_DRM=y
> >>  >BR2_PACKAGE_WESTON_DEMO_CLIENTS=y
> >>  >BR2_PACKAGE_RPI_FIRMWARE=y
> >>  >BR2_PACKAGE_RPI_FIRMWARE_VARIANT_PI4=y
> >>  >BR2_PACKAGE_WPEWEBKIT=y
> >>  >BR2_PACKAGE_OPENSSH=y
> >>  >BR2_TARGET_ROOTFS_EXT2=y
> >>  >BR2_TARGET_ROOTFS_EXT2_4=y
> >>  >BR2_TARGET_ROOTFS_EXT2_LABEL=""
> >>  >BR2_TARGET_ROOTFS_EXT2_SIZE="256M"
> >>  ># BR2_TARGET_ROOTFS_TAR is not set
> >>  >BR2_PACKAGE_HOST_DOSFSTOOLS=y
> >>  >BR2_PACKAGE_HOST_GENIMAGE=y
> >>  >BR2_PACKAGE_HOST_MTOOLS=y
> >>  >
> >>  >Running:
> >>  >
> >>  >	$ export XDG_RUNTIME_DIR=/tmp
> >>  >	$ weston --tty 1 &
> >>  >	$ cog --platform=fdo
> >>  >
> >>  >leads to the 'Segmentation fault' as you described...
> >>  >
> >>  >Running:
> >>  >
> >>  >	$ export XDG_RUNTIME_DIR=/tmp
> >>  >	$ weston --tty 1 --backend=drm-backend.so &
> >>  >	$ cog --platform=fdo
> >>  >
> >>  >works...
> >>  >
> >>  >Regards,
> >>  >Peter
> >>  >
> >>  >
> >>  >>
> >>  >>
> >>  >>
> >>  >>
> >>  >>  --
> >>  >>  Sent from: http://buildroot-busybox.2317881.n4.nabble.com/
> >>  >>  _______________________________________________
> >>  >>  buildroot mailing list
> >>  >>buildroot at busybox.net
> >>  >>http://lists.busybox.net/mailman/listinfo/buildroot
> >>  >
> >>
> >>  _______________________________________________
> >>  buildroot mailing list
> >>buildroot at busybox.net
> >>http://lists.busybox.net/mailman/listinfo/buildroot
> >
>

^ permalink raw reply	[flat|nested] 10+ messages in thread

* [Buildroot] COG crashes on Pi4 (wpe_webkit)
  2020-05-24 12:43               ` Peter Seiderer
@ 2020-05-25  7:18                 ` Daniel HiFiBerry
  0 siblings, 0 replies; 10+ messages in thread
From: Daniel HiFiBerry @ 2020-05-25  7:18 UTC (permalink / raw)
  To: buildroot

I think that was problem. With BR2_PACKAGE_MESA3D_GALLIUM_DRIVER_KMSRO 
now it seems to work.

Thank you very much,
Daniel


------ Originalnachricht ------
Von: "Peter Seiderer" <ps.report@gmx.net>
An: "Daniel HiFiBerry" <daniel@hifiberry.com>
Cc: buildroot at busybox.net
Gesendet: 5/24/2020 2:43:12 PM
Betreff: Re: [Buildroot] COG crashes on Pi4 (wpe_webkit)

>Hello Daniel,
>
>On Sun, 24 May 2020 09:43:36 +0000, "Daniel HiFiBerry" <daniel@hifiberry.com> wrote:
>
>>  I have this integrated (it's just not on Github).
>
>Does this mean it works for you? Or do you have
>BR2_PACKAGE_MESA3D_GALLIUM_DRIVER_KMSRO=y integrated
>and it does not work for you (take care to not mix it with
>BR2_PACKAGE_MESA3D_GALLIUM_KMSRO)?
>
>I have a hifiberry-os running with the local gui on hdmi:
>
>	$ ps aux
>[...]
>root       962  0.5  2.5  32780 23444 tty7     Ss+  12:38   0:01 /usr/bin/weston --backend=drm-backend.so --modules=systemd-notify.so
>root       963  0.0  0.8  13972  8192 tty7     S+   12:38   0:00 /usr/libexec/weston-keyboard
>root       964  0.0  0.6  12356  5964 tty7     S+   12:38   0:00 /usr/libexec/weston-desktop-shell
>root       969  0.3  3.1 209332 29236 ?        Ssl  12:38   0:01 /usr/bin/cog -P fdo http://localhost:80/#now-playing
>root       975 40.6 13.3 355752 122804 ?       SLl  12:38   2:18 /usr/libexec/wpe-webkit-1.0/WPEWebProcess 7 16
>root       976  0.2  3.3 259100 31016 ?        SLl  12:38   0:01 /usr/libexec/wpe-webkit-1.0/WPENetworkProcess 8 16
>[...]
>
>The only thing I have to hack was the resize-partitions for the extra large image ;-)
>
>Regards,
>Peter
>
>>
>>  ------ Originalnachricht ------
>>  Von: "Peter Seiderer" <ps.report@gmx.net>
>>  An: "Daniel HiFiBerry" <daniel@hifiberry.com>
>>  Cc: buildroot at busybox.net
>>  Gesendet: 5/24/2020 11:42:31 AM
>>  Betreff: Re: [Buildroot] COG crashes on Pi4 (wpe_webkit)
>>
>>  >Hello Daniel,
>>  >
>>  >On Sun, 24 May 2020 08:59:53 +0000, "Daniel HiFiBerry" <daniel@hifiberry.com> wrote:
>>  >
>>  >>  Hi Peter,
>>  >>
>>  >>  While I didn't test with a minimal configuration, using the
>>  >>  "--backend=drm-backend.so" doesn't make a difference in my full-blown
>>  >>  build.
>>  >>  I'm using this already. I will do some more experiments...
>>  >
>>  >The trick for hifiberry-os should do (untested) the missing
>>  >
>>  >	BR2_PACKAGE_MESA3D_GALLIUM_DRIVER_KMSRO=y
>>  >
>>  >in your configuration (see the 'hidden' hint in package/mesa3d/Config.in
>>  >([1])...
>>  >
>>  >Regards,
>>  >Peter
>>  >
>>  >[1] https://git.buildroot.net/buildroot/tree/package/mesa3d/Config.in#n265
>>  >
>>  >>
>>  >>  Best regards,
>>  >>  Daniel
>>  >>
>>  >>  ------ Originalnachricht ------
>>  >>  Von: "Peter Seiderer" <ps.report@gmx.net>
>>  >>  An: "usul27" <daniel@hifiberry.com>
>>  >>  Cc: buildroot at busybox.net
>>  >>  Gesendet: 5/23/2020 10:29:48 PM
>>  >>  Betreff: Re: [Buildroot] COG crashes on Pi4 (wpe_webkit)
>>  >>
>>  >>  >Hello Daniel,
>>  >>  >
>>  >>  >On Fri, 22 May 2020 09:17:18 -0500 (CDT), usul27 <daniel@hifiberry.com> wrote:
>>  >>  >
>>  >>  >>  Ok, finally got a build with debug symbols:
>>  >>  >>
>>  >>  >>  Thread 1 "cog" received signal SIGSEGV, Segmentation fault.
>>  >>  >>  0x00000000 in ?? ()
>>  >>  >>  (gdb) backtrace
>>  >>  >>  #0  0x00000000 in ?? ()
>>  >>  >>  #1  0xb1799f84 in ffi_call_SYSV () from /lib/libffi.so.7
>>  >>  >>  #2  0xb17993fc in ffi_call_int () from /lib/libffi.so.7
>>  >>  >>  #3  0xb1799b1c in ffi_call () from /lib/libffi.so.7
>>  >>  >>  #4  0xb149b28c in wl_closure_invoke (closure=closure at entry=0xadb03a00,
>>  >>  >>  flags=flags at entry=2, target=<optimized out>, opcode=3025639260,
>>  >>  >>  data=<optimized out>, data at entry=0x74c60) at src/connection.c:1014
>>  >>  >>  #5  0xb1497db0 in wl_client_connection_data (fd=<optimized out>,
>>  >>  >>  mask=<optimized out>, data=0x74c60) at src/wayland-server.c:429
>>  >>  >>  #6  0xb1499608 in wl_event_loop_dispatch (loop=0x5bfc8, timeout=<optimized
>>  >>  >>  out>) at src/event-loop.c:640
>>  >>  >>  #7  0xb0c59400 in WS::ServerSource::{lambda(_GSource*, int (*)(void*),
>>  >>  >>  void*)#3}::_FUN(_GSource*, int (*)(void*), void*) () from
>>  >>  >>  /lib/libWPEBackend-fdo-1.0.so.1
>>  >>  >>  Backtrace stopped: previous frame identical to this frame (corrupt stack?)
>>  >>  >>
>>  >>  >>  Somebody has any idea?
>>  >>  >
>>  >>  >Did some test on RPi4 with the following defconfig:
>>  >>  >
>>  >>  >BR2_arm=y
>>  >>  >BR2_cortex_a72=y
>>  >>  >BR2_ARM_FPU_NEON_VFPV4=y
>>  >>  >BR2_ENABLE_DEBUG=y
>>  >>  >BR2_TOOLCHAIN_BUILDROOT_GLIBC=y
>>  >>  >BR2_PACKAGE_HOST_LINUX_HEADERS_CUSTOM_4_19=y
>>  >>  >BR2_BINUTILS_VERSION_2_31_X=y
>>  >>  >BR2_TOOLCHAIN_BUILDROOT_CXX=y
>>  >>  >BR2_PACKAGE_HOST_GDB=y
>>  >>  >BR2_GDB_VERSION_8_3=y
>>  >>  >BR2_ROOTFS_DEVICE_CREATION_DYNAMIC_EUDEV=y
>>  >>  >BR2_SYSTEM_DHCP="eth0"
>>  >>  >BR2_GENERATE_LOCALE="en_US"
>>  >>  >BR2_ROOTFS_POST_BUILD_SCRIPT="board/raspberrypi4/post-build.sh"
>>  >>  >BR2_ROOTFS_POST_IMAGE_SCRIPT="board/raspberrypi4/post-image.sh"
>>  >>  >BR2_ROOTFS_POST_SCRIPT_ARGS="--add-pi3-miniuart-bt-overlay"
>>  >>  >BR2_LINUX_KERNEL=y
>>  >>  >BR2_LINUX_KERNEL_CUSTOM_TARBALL=y
>>  >>  >BR2_LINUX_KERNEL_CUSTOM_TARBALL_LOCATION="$(call github,raspberrypi,linux,4f2a4cc501c428c940549f39d5562e60404ac4f7)/linux-4f2a4cc501c428c940549f39d5562e60404ac4f7.tar.gz"
>>  >>  >BR2_LINUX_KERNEL_DEFCONFIG="bcm2711"
>>  >>  >BR2_LINUX_KERNEL_DTS_SUPPORT=y
>>  >>  >BR2_LINUX_KERNEL_INTREE_DTS_NAME="bcm2711-rpi-4-b"
>>  >>  >BR2_LINUX_KERNEL_NEEDS_HOST_OPENSSL=y
>>  >>  >BR2_PACKAGE_STRACE=y
>>  >>  >BR2_PACKAGE_DEJAVU=y
>>  >>  >BR2_PACKAGE_COG=y
>>  >>  >BR2_PACKAGE_MESA3D=y
>>  >>  >BR2_PACKAGE_MESA3D_GALLIUM_DRIVER_KMSRO=y
>>  >>  >BR2_PACKAGE_MESA3D_GALLIUM_DRIVER_V3D=y
>>  >>  >BR2_PACKAGE_MESA3D_GALLIUM_DRIVER_VC4=y
>>  >>  >BR2_PACKAGE_MESA3D_OPENGL_ES=y
>>  >>  >BR2_PACKAGE_WESTON=y
>>  >>  >BR2_PACKAGE_WESTON_DRM=y
>>  >>  >BR2_PACKAGE_WESTON_DEMO_CLIENTS=y
>>  >>  >BR2_PACKAGE_RPI_FIRMWARE=y
>>  >>  >BR2_PACKAGE_RPI_FIRMWARE_VARIANT_PI4=y
>>  >>  >BR2_PACKAGE_WPEWEBKIT=y
>>  >>  >BR2_PACKAGE_OPENSSH=y
>>  >>  >BR2_TARGET_ROOTFS_EXT2=y
>>  >>  >BR2_TARGET_ROOTFS_EXT2_4=y
>>  >>  >BR2_TARGET_ROOTFS_EXT2_LABEL=""
>>  >>  >BR2_TARGET_ROOTFS_EXT2_SIZE="256M"
>>  >>  ># BR2_TARGET_ROOTFS_TAR is not set
>>  >>  >BR2_PACKAGE_HOST_DOSFSTOOLS=y
>>  >>  >BR2_PACKAGE_HOST_GENIMAGE=y
>>  >>  >BR2_PACKAGE_HOST_MTOOLS=y
>>  >>  >
>>  >>  >Running:
>>  >>  >
>>  >>  >	$ export XDG_RUNTIME_DIR=/tmp
>>  >>  >	$ weston --tty 1 &
>>  >>  >	$ cog --platform=fdo
>>  >>  >
>>  >>  >leads to the 'Segmentation fault' as you described...
>>  >>  >
>>  >>  >Running:
>>  >>  >
>>  >>  >	$ export XDG_RUNTIME_DIR=/tmp
>>  >>  >	$ weston --tty 1 --backend=drm-backend.so &
>>  >>  >	$ cog --platform=fdo
>>  >>  >
>>  >>  >works...
>>  >>  >
>>  >>  >Regards,
>>  >>  >Peter
>>  >>  >
>>  >>  >
>>  >>  >>
>>  >>  >>
>>  >>  >>
>>  >>  >>
>>  >>  >>  --
>>  >>  >>  Sent from: http://buildroot-busybox.2317881.n4.nabble.com/
>>  >>  >>  _______________________________________________
>>  >>  >>  buildroot mailing list
>>  >>  >>buildroot at busybox.net
>>  >>  >>http://lists.busybox.net/mailman/listinfo/buildroot
>>  >>  >
>>  >>
>>  >>  _______________________________________________
>>  >>  buildroot mailing list
>>  >>buildroot at busybox.net
>>  >>http://lists.busybox.net/mailman/listinfo/buildroot
>>  >
>>
>

^ permalink raw reply	[flat|nested] 10+ messages in thread

end of thread, other threads:[~2020-05-25  7:18 UTC | newest]

Thread overview: 10+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-05-19 13:16 [Buildroot] COG crashes on Pi4 (wpe_webkit) usul27
2020-05-19 17:44 ` Peter Seiderer
2020-05-20  7:07   ` usul27
2020-05-22 14:17     ` usul27
2020-05-23 20:29       ` Peter Seiderer
2020-05-24  8:59         ` Daniel HiFiBerry
2020-05-24  9:42           ` Peter Seiderer
2020-05-24  9:43             ` Daniel HiFiBerry
2020-05-24 12:43               ` Peter Seiderer
2020-05-25  7:18                 ` Daniel HiFiBerry

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.