On Mon, Aug 24, 2026 at 12:25:19AM +0200, Kirill A. Korinsky wrote: > On Sun, 23 Aug 2026 16:22:53 +0200, > Theo Buehler <tb@theobuehler.org> wrote: > > > > On Sun, Aug 23, 2026 at 06:50:10AM -0400, Brad Smith wrote: > > > On Sun, Aug 23, 2026 at 10:53:25AM +0200, Kirill A. Korinsky wrote: > > > > On Thu, 13 Aug 2026 12:13:54 +0200, > > > > Kirill A. Korinsky <kirill@korins.ky> wrote: > > > > > > > > > > ports@, Brad, > > > > > > > > > > here a diff to update graphics/ffmpeg to 9.0.1, and it fixes and bumps > > > > > x11/mplayer as well. > > > > > > > > > > Bulk on amd64 with previous diffs and this one successeded, unsigned > > > > > packages available here: https://kirill.korins.ky/pub/packages/ffmpeg-9.0.1/ > > > > > > > > > > Tests and OKs are welcome. > > > > > > > > > > > > > The previous version was without subject what is fixed now. > > > > > > > > Also, it survives a few bulks on amd64 but I haven't tested BTI/IBT amd64, > > > > so any tests are welcome. > > > > > > Here is an updated diff. The configure patch SSL changes can be > > > simplified a bit and add in my MIPS fix for CABAC. > > > > It works for me in some light testing. Tests run into segfaults on both > > amd64 and arm64, which is not necessarily a blocker but a bit annoying. > > > > Quite a few tests do run with amd64/ibt without SIGILL, so that is > > reassuring. > > > > > > apple M1: > > > > NEON: > > FAILURE: float_fft_16384_neon (segmentation fault) > > - av_tx [FAILED] > > checkasm: 1 of 8 tests failed > > threads=1 > > > > on both > > > > cpu0: 12th Gen Intel(R) Core(TM) i7-1260P, 1995.55 MHz, 06-9a-03, patch 0000043b > > cpu0: AMD Ryzen 7 PRO 5850U with Radeon Graphics, 1900.00 MHz, 19-50-00, patch 0a500012 > > > > I see this: > > > > ffmpeg version 9.0.1 Copyright (c) 2000-2026 the FFmpeg developers > > built with OpenBSD clang version 22.1.6 > > configuration: --enable-shared --arch=amd64 --cc=cc --cxx=c++ --enable-debug --disable-stripping --disable-indev=jack --disable-vulkan --enable-fontconfig --enable-frei0r --enable-gpl --enable-ladspa --enable-libaom --enable-libass --enable-libdav1d --enable-libfontconfig --enable-libfreetype --enable-libfribidi --enable-libgsm --enable-libharfbuzz --enable-libmp3lame --enable-libopus --enable-libspeex --enable-libsvtav1 --enable-libtheora --enable-libv4l2 --enable-libvorbis --enable-libvpx --enable-libwebp --enable-libx264 --enable-libx265 --enable-libxml2 --enable-libxvid --enable-libzimg --enable-nonfree --enable-openssl --enable-libvidstab --extra-cflags='-I/usr/local/include -I/usr/X11R6/include' --extra-libs='-L/usr/local/lib -L/usr/X11R6/lib' --extra-ldsoflags= --mandir=/usr/local/man --objcc=/usr/bin/false --optflags='-O2 -pipe -g -Wno-redundant-decls' > > libavutil 61. 1.101 / 61. 1.101 > > libavcodec 63. 1.101 / 63. 1.101 > > libavformat 63. 1.101 / 63. 1.101 > > libavdevice 63. 1.101 / 63. 1.101 > > libavfilter 12. 1.101 / 12. 1.101 > > libswscale 10. 1.101 / 10. 1.101 > > libswresample 7. 1.101 / 7. 1.101 > > Stream mapping: > > crop:default -> Stream #0:0 (rawvideo) > > Segmentation fault (core dumped) > > threads=1 > > > > Program terminated with signal SIGSEGV, Segmentation fault. > > #0 memcpy (dst0=0xf3c943c2000, src0=0xf3c87d941c4 <ff_init_hscaler_mmxext+180>, length=44) > > at /usr/src/lib/libc/string/memcpy.c:97 > > 97 TLOOP1(*dst++ = *src++); > > [Current thread is 1 (process 148644)] > > (gdb) bt > > #0 memcpy (dst0=0xf3c943c2000, src0=0xf3c87d941c4 <ff_init_hscaler_mmxext+180>, length=44) > > at /usr/src/lib/libc/string/memcpy.c:97 > > #1 0x00000f3c87d943b5 in ff_init_hscaler_mmxext (dstW=40032, xInc=65504, > > filterCode=0xf3c943c2000 "", filter=0xf3bda2a8000, filterPos=0xf3c966f6000, > > numSplits=<optimized out>) at src/libswscale/x86/hscale_fast_bilinear_simd.c:157 > > #2 0x00000f3c87d90eb0 in ff_sws_init_single_context (sws=0xf3c09903000, > > srcFilter=0xf3c227dd770, dstFilter=0xf3c227dd770) at src/libswscale/utils.c:1705 > > #3 0x00000f3c87c9c8ab in add_legacy_sws_pass (graph=0xf3c2131d400, src=0xf3c227dd988, > > dst=0xf3c227dd900, input=0x0, output=0xf3c227dd8e8) at src/libswscale/graph.c:584 > > > > > > Thanks for testing amd64, here updated diff which fixes a bug. Thanks. Makes sense and passes regress on my IBT machine.
No comments:
Post a Comment