#11537: ffmpeg 6.1.3 please? -------------------------------------+------------------------------------- Reporter: Artem S. | Owner: (none) Tashkinov | Type: defect | Status: new Priority: critical | Component: | undetermined Version: 6.1.1 | Resolution: Keywords: | Blocked By: Blocking: | Reproduced by developer: 0 Analyzed by developer: 0 | -------------------------------------+------------------------------------- Changes (by MasterQuestionable): * cc: MasterQuestionable (added) Comment: ͏    Quite wondering that, why would some mere arithmetic overflow of limited influence: ͏    Be exaggerated as that "high risk security vulnerability" that may enable arbitrary remote code execution..? ͏ https://github.com/FFmpeg/FFmpeg/blob/86f73277bf014e2ce36dd2594f1e0fb8b3bd66... ͏ https://github.com/FFmpeg/FFmpeg/blob/65ddc74988245a01421a63c5cffa4d900c4711... ͏    . ͏    Arithmetic overflow itself won't cause any real issue. ͏    That causes issues is the misuse of overflowed values. ͏    Also: ͏ https://github.com/FFmpeg/FFmpeg/commit/7a089ed8e049e3bfcb22de1250b86f210606... ![1] ͏    The premise (of "ULONG_MAX") holds only in atypical compiling environment. ͏    Also tagged alike? ͏    [ Refer also: https://github.com/MasterInQuestion/talk/discussions/15#C-absurdity ] ͏    Is it "医之好治不病以为功"..? +    "Doctors favored curing non-ill as accomplishment"? [ ![1] ͏    The type casting here would be actually no-op. ͏    For the previous definition: https://github.com/FFmpeg/FFmpeg/blob/7a089ed8e049e3bfcb22de1250b86f21060608... ͏    `int64_t min_pos, pos;` ͏    . ͏    "pos" cannot hold anything larger than "int64_t" permits. ͏    Demonstrable alike: [[ {{{#!c #include <stdio.h> #include <stdint.h> int main () { int32_t x = 2147483647; int32_t _ = x + (int64_t) 1; printf( "%d", _ ); }; }}} ]] ͏    Unsure what it really addresses: ͏ https://github.com/FFmpeg/FFmpeg/commit/108957c661f9e2dc35dea8d55e5e5b1776f4... ] -- Ticket URL: <https://trac.ffmpeg.org/ticket/11537#comment:1> FFmpeg <https://ffmpeg.org> FFmpeg issue tracker