Kernel crash during video transcoding

2020-08-09 Thread Alexandre Levy
Hi,

I installed the port drm-devel-kmod for Plex to be able to transcode videos
using the integrated GPU of my Intel Celeron G5900.

I'm running r364031 and the kernel is compiled with GENERIC-NODEBUG profile.

Transcoding has been working fine for quite a while now but one video
transcoding is causing a kernel panic that is reproducible all the time
with that particular video. It seems like it's caused by the i915kms module
(call of i915_gms_fault() in the stack) :

Fatal trap 12: page fault while in kernel mode
cpuid = 0; apic id = 00
fault virtual address   = 0xdf
fault code  = supervisor read data, page not present
instruction pointer = 0x20:0x80bdd2b4
stack pointer   = 0x0:0xfe00d2be56d0
frame pointer   = 0x0:0xfe00d2be56d0
code segment= base 0x0, limit 0xf, type 0x1b
= DPL 0, pres 1, long 1, def32 0, gran 1
processor eflags= interrupt enabled, resume, IOPL = 0
current process = 4611 (Plex Transcoder)
trap number = 12
panic: page fault
cpuid = 0
time = 1596976796
KDB: stack backtrace:
db_trace_self_wrapper() at db_trace_self_wrapper+0x2b/frame
0xfe00d2be5390
vpanic() at vpanic+0x182/frame 0xfe00d2be53e0
panic() at panic+0x43/frame 0xfe00d2be5440
trap_fatal() at trap_fatal+0x387/frame 0xfe00d2be54a0
trap_pfault() at trap_pfault+0x4f/frame 0xfe00d2be54f0
trap() at trap+0x271/frame 0xfe00d2be5600
calltrap() at calltrap+0x8/frame 0xfe00d2be5600
--- trap 0xc, rip = 0x80bdd2b4, rsp = 0xfe00d2be56d0, rbp =
0xfe00d2be56d0 ---
_rw_wowned() at _rw_wowned+0x4/frame 0xfe00d2be56d0
vm_page_busy_acquire() at vm_page_busy_acquire+0x141/frame
0xfe00d2be5710
remap_io_mapping() at remap_io_mapping+0x120/frame 0xfe00d2be5760
i915_gem_fault() at i915_gem_fault+0x25f/frame 0xfe00d2be57d0
linux_cdev_pager_populate() at linux_cdev_pager_populate+0x11b/frame
0xfe00d2be5840
vm_fault() at vm_fault+0x3d1/frame 0xfe00d2be5950
vm_fault_trap() at vm_fault_trap+0x60/frame 0xfe00d2be5990
trap_pfault() at trap_pfault+0x19c/frame 0xfe00d2be59e0
trap() at trap+0x3f1/frame 0xfe00d2be5af0
calltrap() at calltrap+0x8/frame 0xfe00d2be5af0
--- trap 0xc, rip = 0x80296659a, rsp = 0x7fffbd38, rbp = 0x80fc0 ---
KDB: enter: panic

I don't see any crash dump in /var/crash despite having the right
configuration and I should have enough space on my swap device (128GB USB
drive) :

$ cat /etc/rc.conf | grep dump
dumpdev="AUTO"

$ swapinfo
Device  1K-blocks UsedAvail Capacity
/dev/gpt/crash0 1213070960 121307096 0%

$ cat /etc/fstab
/dev/gpt/crash0 noneswapsw  0   0

$ dumpon -l
gpt/crash0

Not sure why no dump was generated, is it because the kernel was compiled
with the GENERIC-NODEBUG profile ? However I see various KDB options in the
GENERIC profile that are inherited by GENERIC-NODEBUG.

Happy to recompile the kernel with GENERIC profile if it's required.

Thank you.
___
freebsd-current@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"


cross-build failure on objcopy

2020-08-09 Thread Michael Butler
When cross-compiling for i386 on amd64 (which has 2 by 4 cores), I get
the error below after a previously successful build.

Running the build again (a 3rd time) completes successfully :-(

This is the output from

cd /usr/src/release; ./release.sh -c release-i386.conf

 [ .. ]

===> usr.bin/cxxfilt (all)
===> usr.bin/objcopy (all)
===> usr.sbin/bsnmpd/tools/bsnmptools (all)
===> usr.bin/file2c (all)
===> usr.bin/gprof (all)
===> usr.bin/indent (all)
===> usr.bin/lex (all)
===> usr.bin/mkstr (all)
===> usr.bin/nm (all)
objcopy: open objcopy failed: Text file busy
--- objcopy ---
*** [objcopy] Error code 1
make[4]: *** objcopy removed

make[4]: stopped in /usr/local/release-builds/i386/usr/src/usr.bin/objcopy
.ERROR_TARGET='objcopy'
.ERROR_META_FILE='/usr/local/release-builds/i386/tmp/obj/usr/local/release-builds/i386/usr/src/amd64.amd64/usr.bin/objcopy/objcopy.meta'
.MAKE.LEVEL='4'
MAKEFILE=''
.MAKE.MODE='meta missing-filemon=yes missing-meta=yes silent=yes'
_ERROR_CMD='objcopy --strip-debug --add-gnu-debuglink=objcopy.debug
objcopy.full objcopy;'
.CURDIR='/usr/local/release-builds/i386/usr/src/usr.bin/objcopy'
.MAKE='make'
.OBJDIR='/usr/local/release-builds/i386/tmp/obj/usr/local/release-builds/i386/usr/src/amd64.amd64/usr.bin/objcopy'
.TARGETS='all'
DESTDIR='/usr/local/release-builds/i386/tmp/obj/usr/local/release-builds/i386/usr/src/amd64.amd64/tmp'
LD_LIBRARY_PATH=''
MACHINE='amd64'
MACHINE_ARCH='amd64'
MAKEOBJDIRPREFIX=''
MAKESYSPATH='/usr/local/release-builds/i386/usr/src/share/mk'

imb

___
freebsd-current@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"


Re: Problem compiling chromium

2020-08-09 Thread John-Mark Gurney
Filippo Moretti wrote this message on Sun, Aug 02, 2020 at 08:23 +:
> This is the last eror I get attempting to compile chromium on amd64 arch 
> R363570[263/38159] python ../../mojo/public/tools/mojom/mojom_parser.py 
> --input-root /usr/ports/www/chromium/work/chromium-84.0.4147.105/ 
> --input-root 
> /usr/ports/www/chromium/work/chromium-84.0.4147.105/out/Release/gen 
> --output-root 
> /usr/ports/www/chromium/work/chromium-84.0.4147.105/out/Release/gen 
> --mojom-file-list=__mojo_public_mojom_base_base__parser___build_toolchain_linux_clang_x64__rule..rsp
>  --check-imports 
> /usr/ports/www/chromium/work/chromium-84.0.4147.105/out/Release/gen/mojo/public/mojom/base/base.build_metadata
>  --enable-feature file_path_is_string --enable-feature is_posix 
> --enable-feature is_linux
> [264/38159] python ../../tools/licenses.py --target-os=freebsd --depfile 
> gen/components/resources/about_credits.d credits 
> gen/components/resources/about_credits.html
> FAILED: gen/components/resources/about_credits.html 
> python ../../tools/licenses.py --target-os=freebsd --depfile 
> gen/components/resources/about_credits.d credits 
> gen/components/resources/about_credits.html
> Traceback (most recent call last):
>   File "../../tools/licenses.py", line 807, in 
>     sys.exit(main())
>   File "../../tools/licenses.py", line 792, in main
>     args.gn_target, args.depfile): File "../../tools/licenses.py", line 720, 
> in GenerateCredits
>     license_file_list + ['build.ninja'])
>   File 
> "/usr/ports/www/chromium/work/chromium-84.0.4147.105/build/android/gyp/util/build_utils.py",
>  line 619, in WriteDepfile
>     inputs = ComputePythonDependencies() + inputs
>   File 
> "/usr/ports/www/chromium/work/chromium-84.0.4147.105/build/android/gyp/util/build_utils.py",
>  line 557, in ComputePythonDependencies
>     p for p in abs_module_paths if p.startswith(abs_dir_source_root)
>   File 
> "/usr/ports/www/chromium/work/chromium-84.0.4147.105/build/android/gyp/util/build_utils.py",
>  line 557, in 
>     p for p in abs_module_paths if p.startswith(abs_dir_source_root)
>   File 
> "/usr/ports/www/chromium/work/.bin/../../../../../local/lib/python3.7/posixpath.py",
>  line 378, in abspath
>     path = os.fspath(path)
> TypeError: expected str, bytes or os.PathLike object, not NoneType
> ninja: build stopped: subcommand failed.
> *** Error code 1
> 
> Stop.
> make[1]: stopped in /usr/ports/www/chromium

Are you sure that chromium is Python 3 clean?  Python 2 was recently
deprecated, so likely this is fallout from that.

-- 
  John-Mark Gurney  Voice: +1 415 225 5579

 "All that I will do, has been done, All that I have, has not."
___
freebsd-current@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"