<![CDATA[reposting for the day crowd: I ran into a memcmp implementation that only compared 25% of the bytes, and the issue wasn't caught in the build because the vendor toolchain failed to emit a warning.]]>reposting for the day crowd: I ran into a memcmp implementation that only compared 25% of the bytes, and the issue wasn't caught in the build because the vendor toolchain failed to emit a warning.

Watch out for missed warnings on vendor C++ toolchains - Graham Sutherland's Blog

favicon

(blog.poly.nomial.co.uk)

]]>https://board.circlewithadot.net/topic/f8d3da56-6738-4939-a102-88d6ea35903b/reposting-for-the-day-crowd-i-ran-into-a-memcmp-implementation-that-only-compared-25-of-the-bytes-and-the-issue-wasn-t-caught-in-the-build-because-the-vendor-toolchain-failed-to-emit-a-warning.RSS for NodeThu, 09 Apr 2026 10:56:55 GMTTue, 31 Mar 2026 13:20:01 GMT60<![CDATA[Reply to reposting for the day crowd: I ran into a memcmp implementation that only compared 25% of the bytes, and the issue wasn't caught in the build because the vendor toolchain failed to emit a warning. on Tue, 31 Mar 2026 20:45:30 GMT]]>@gsuberland that seems not good

]]>https://board.circlewithadot.net/post/https://infosec.exchange/users/ryanc/statuses/116325740082821417https://board.circlewithadot.net/post/https://infosec.exchange/users/ryanc/statuses/116325740082821417Tue, 31 Mar 2026 20:45:30 GMT<![CDATA[Reply to reposting for the day crowd: I ran into a memcmp implementation that only compared 25% of the bytes, and the issue wasn't caught in the build because the vendor toolchain failed to emit a warning. on Tue, 31 Mar 2026 19:10:59 GMT]]>@gsuberland Hmm, interesting, haven't seen those!

]]>https://board.circlewithadot.net/post/https://mastodon.social/users/WAHa_06x36/statuses/116325368431299866https://board.circlewithadot.net/post/https://mastodon.social/users/WAHa_06x36/statuses/116325368431299866Tue, 31 Mar 2026 19:10:59 GMT<![CDATA[Reply to reposting for the day crowd: I ran into a memcmp implementation that only compared 25% of the bytes, and the issue wasn't caught in the build because the vendor toolchain failed to emit a warning. on Tue, 31 Mar 2026 18:52:00 GMT]]>@WAHa_06x36 quite a few people! there are even coverage tools specifically for doing this.

]]>https://board.circlewithadot.net/post/https://chaos.social/users/gsuberland/statuses/116325293765008274https://board.circlewithadot.net/post/https://chaos.social/users/gsuberland/statuses/116325293765008274Tue, 31 Mar 2026 18:52:00 GMT<![CDATA[Reply to reposting for the day crowd: I ran into a memcmp implementation that only compared 25% of the bytes, and the issue wasn't caught in the build because the vendor toolchain failed to emit a warning. on Tue, 31 Mar 2026 18:49:13 GMT]]>@gsuberland I mean, if you set up a special test harness against a known-good implementation and used something like afl that actually instruments the code itself, maybe, but, who would ever do that?

]]>https://board.circlewithadot.net/post/https://mastodon.social/users/WAHa_06x36/statuses/116325282869558225https://board.circlewithadot.net/post/https://mastodon.social/users/WAHa_06x36/statuses/116325282869558225Tue, 31 Mar 2026 18:49:13 GMT<![CDATA[Reply to reposting for the day crowd: I ran into a memcmp implementation that only compared 25% of the bytes, and the issue wasn't caught in the build because the vendor toolchain failed to emit a warning. on Tue, 31 Mar 2026 17:31:33 GMT]]>@uecker @rjmccall I'll update the blog post later tonight if I get time. annoyingly today is extremely busy >_<

]]>https://board.circlewithadot.net/post/https://chaos.social/users/gsuberland/statuses/116324977419440678https://board.circlewithadot.net/post/https://chaos.social/users/gsuberland/statuses/116324977419440678Tue, 31 Mar 2026 17:31:33 GMT<![CDATA[Reply to reposting for the day crowd: I ran into a memcmp implementation that only compared 25% of the bytes, and the issue wasn't caught in the build because the vendor toolchain failed to emit a warning. on Tue, 31 Mar 2026 16:14:09 GMT]]>@gsuberland @rjmccall It seems it is under the language dialects options and explanation is not really clear. https://gcc.gnu.org/onlinedocs/gcc-15.2.0/gcc/C_002b_002b-Dialect-Options.html

]]>https://board.circlewithadot.net/post/https://mastodon.social/users/uecker/statuses/116324673063946130https://board.circlewithadot.net/post/https://mastodon.social/users/uecker/statuses/116324673063946130Tue, 31 Mar 2026 16:14:09 GMT<![CDATA[Reply to reposting for the day crowd: I ran into a memcmp implementation that only compared 25% of the bytes, and the issue wasn't caught in the build because the vendor toolchain failed to emit a warning. on Tue, 31 Mar 2026 15:48:55 GMT]]>@rjmccall @uecker gcc's docs don't even have a paragraph explaining what Wnarrowing does, as far as I can see.

]]>https://board.circlewithadot.net/post/https://chaos.social/users/gsuberland/statuses/116324573853297429https://board.circlewithadot.net/post/https://chaos.social/users/gsuberland/statuses/116324573853297429Tue, 31 Mar 2026 15:48:55 GMT<![CDATA[Reply to reposting for the day crowd: I ran into a memcmp implementation that only compared 25% of the bytes, and the issue wasn't caught in the build because the vendor toolchain failed to emit a warning. on Tue, 31 Mar 2026 15:47:20 GMT]]>@gsuberland @uecker I won’t defend Clang’s naming choices in every case, but I believe this specific one is all GCC; Clang originally called this -Wc++0x-narrowing (eventually -Wc++11-narrowing) and only added the -Wnarrowing alias for GCC compatibility. In any case, the documentation should really suggest -Wconversion, and on that front I can definitely accept blame for Clang, because our warning group documentation is awful

]]>https://board.circlewithadot.net/post/https://hachyderm.io/users/rjmccall/statuses/116324567625941320https://board.circlewithadot.net/post/https://hachyderm.io/users/rjmccall/statuses/116324567625941320Tue, 31 Mar 2026 15:47:20 GMT<![CDATA[Reply to reposting for the day crowd: I ran into a memcmp implementation that only compared 25% of the bytes, and the issue wasn't caught in the build because the vendor toolchain failed to emit a warning. on Tue, 31 Mar 2026 15:21:11 GMT]]>@WAHa_06x36 @gsuberland i think „only one byte differs“ kind of tests would probably find it, right? And these seem like something you’d write to test that

]]>https://board.circlewithadot.net/post/https://icosahedron.website/users/halcy/statuses/116324464798563046https://board.circlewithadot.net/post/https://icosahedron.website/users/halcy/statuses/116324464798563046Tue, 31 Mar 2026 15:21:11 GMT<![CDATA[Reply to reposting for the day crowd: I ran into a memcmp implementation that only compared 25% of the bytes, and the issue wasn't caught in the build because the vendor toolchain failed to emit a warning. on Tue, 31 Mar 2026 15:19:59 GMT]]>@WAHa_06x36 of course. fuzz testing would quickly find memcmp("aaaa", "Aaaa") == 0 or memcmp("aaaa", "aaaA") == 0 as a violation of the contract (depending on endianness)

]]>https://board.circlewithadot.net/post/https://chaos.social/users/gsuberland/statuses/116324460134235426https://board.circlewithadot.net/post/https://chaos.social/users/gsuberland/statuses/116324460134235426Tue, 31 Mar 2026 15:19:59 GMT<![CDATA[Reply to reposting for the day crowd: I ran into a memcmp implementation that only compared 25% of the bytes, and the issue wasn't caught in the build because the vendor toolchain failed to emit a warning. on Tue, 31 Mar 2026 15:14:09 GMT]]>@gsuberland Hmm, would even fuzz testing find it? That seems tricky to set up in a way that a) would actually find the bug and b) would occur to you before seeing the bug.

I guess for very short inputs you might find it more easily by chance...

]]>https://board.circlewithadot.net/post/https://mastodon.social/users/WAHa_06x36/statuses/116324437134447512https://board.circlewithadot.net/post/https://mastodon.social/users/WAHa_06x36/statuses/116324437134447512Tue, 31 Mar 2026 15:14:09 GMT<![CDATA[Reply to reposting for the day crowd: I ran into a memcmp implementation that only compared 25% of the bytes, and the issue wasn't caught in the build because the vendor toolchain failed to emit a warning. on Tue, 31 Mar 2026 14:55:35 GMT]]>@WAHa_06x36 this is why fuzz testing is a thing!

]]>https://board.circlewithadot.net/post/https://chaos.social/users/gsuberland/statuses/116324364159672094https://board.circlewithadot.net/post/https://chaos.social/users/gsuberland/statuses/116324364159672094Tue, 31 Mar 2026 14:55:35 GMT<![CDATA[Reply to reposting for the day crowd: I ran into a memcmp implementation that only compared 25% of the bytes, and the issue wasn't caught in the build because the vendor toolchain failed to emit a warning. on Tue, 31 Mar 2026 14:46:46 GMT]]>@gsuberland Pretty sure this would have passed the unit tests that anyone would have been likely to write anyway.

]]>https://board.circlewithadot.net/post/https://mastodon.social/users/WAHa_06x36/statuses/116324329491306956https://board.circlewithadot.net/post/https://mastodon.social/users/WAHa_06x36/statuses/116324329491306956Tue, 31 Mar 2026 14:46:46 GMT<![CDATA[Reply to reposting for the day crowd: I ran into a memcmp implementation that only compared 25% of the bytes, and the issue wasn't caught in the build because the vendor toolchain failed to emit a warning. on Tue, 31 Mar 2026 14:11:37 GMT]]>@Doomed_Daniel @gsuberland Obviously. The problem is there are too many people with broken code that do not want to fix it. For example, implicit int in C was disallowed in C99, GCC made it a hard error in 2024 (GCC 14) - 25 years later.

]]>https://board.circlewithadot.net/post/https://mastodon.social/users/uecker/statuses/116324191297335789https://board.circlewithadot.net/post/https://mastodon.social/users/uecker/statuses/116324191297335789Tue, 31 Mar 2026 14:11:37 GMT<![CDATA[Reply to reposting for the day crowd: I ran into a memcmp implementation that only compared 25% of the bytes, and the issue wasn't caught in the build because the vendor toolchain failed to emit a warning. on Tue, 31 Mar 2026 14:02:31 GMT]]>@gsuberland Fair. You should add clang as well... and please add that you need to use -Wconversion

]]>https://board.circlewithadot.net/post/https://mastodon.social/users/uecker/statuses/116324155505161861https://board.circlewithadot.net/post/https://mastodon.social/users/uecker/statuses/116324155505161861Tue, 31 Mar 2026 14:02:31 GMT<![CDATA[Reply to reposting for the day crowd: I ran into a memcmp implementation that only compared 25% of the bytes, and the issue wasn't caught in the build because the vendor toolchain failed to emit a warning. on Tue, 31 Mar 2026 14:02:10 GMT]]>@uecker @gsuberland
shouldn't things that are disallowed be errors, while things that are allowed but probably a bad idea warnings?

]]>https://board.circlewithadot.net/post/https://mastodon.gamedev.place/users/Doomed_Daniel/statuses/116324154088821069https://board.circlewithadot.net/post/https://mastodon.gamedev.place/users/Doomed_Daniel/statuses/116324154088821069Tue, 31 Mar 2026 14:02:10 GMT<![CDATA[Reply to reposting for the day crowd: I ran into a memcmp implementation that only compared 25% of the bytes, and the issue wasn't caught in the build because the vendor toolchain failed to emit a warning. on Tue, 31 Mar 2026 13:59:22 GMT]]>@uecker if -Wnarrowing doesn't catch narrowing conversions then I will edit the post to say "also gcc is terrible at naming things and encourages bugs as a result"

]]>https://board.circlewithadot.net/post/https://chaos.social/users/gsuberland/statuses/116324143102360054https://board.circlewithadot.net/post/https://chaos.social/users/gsuberland/statuses/116324143102360054Tue, 31 Mar 2026 13:59:22 GMT<![CDATA[Reply to reposting for the day crowd: I ran into a memcmp implementation that only compared 25% of the bytes, and the issue wasn't caught in the build because the vendor toolchain failed to emit a warning. on Tue, 31 Mar 2026 13:59:02 GMT]]>@gsuberland It is a narrowing conversion, but it seems C++ only disallows this in initializer lists and this is when compiler warn:
https://eel.is/c++draft/dcl.init.list#def:conversion,narrowing

]]>https://board.circlewithadot.net/post/https://mastodon.social/users/uecker/statuses/116324141802082866https://board.circlewithadot.net/post/https://mastodon.social/users/uecker/statuses/116324141802082866Tue, 31 Mar 2026 13:59:02 GMT<![CDATA[Reply to reposting for the day crowd: I ran into a memcmp implementation that only compared 25% of the bytes, and the issue wasn't caught in the build because the vendor toolchain failed to emit a warning. on Tue, 31 Mar 2026 13:50:24 GMT]]>@uecker I might've bungled the flags in the post, 'cos I was tired, but the actual flags they were using in the build did generate the warning in gcc.

I would expect -Wnarrowing to catch implicit narrowing conversions, though.

]]>https://board.circlewithadot.net/post/https://chaos.social/users/gsuberland/statuses/116324107837027363https://board.circlewithadot.net/post/https://chaos.social/users/gsuberland/statuses/116324107837027363Tue, 31 Mar 2026 13:50:24 GMT<![CDATA[Reply to reposting for the day crowd: I ran into a memcmp implementation that only compared 25% of the bytes, and the issue wasn't caught in the build because the vendor toolchain failed to emit a warning. on Tue, 31 Mar 2026 13:47:57 GMT]]>@gsuberland But the other huge problem are the strict aliasing violations.

]]>https://board.circlewithadot.net/post/https://mastodon.social/users/uecker/statuses/116324098240018924https://board.circlewithadot.net/post/https://mastodon.social/users/uecker/statuses/116324098240018924Tue, 31 Mar 2026 13:47:57 GMT<![CDATA[Reply to reposting for the day crowd: I ran into a memcmp implementation that only compared 25% of the bytes, and the issue wasn't caught in the build because the vendor toolchain failed to emit a warning. on Tue, 31 Mar 2026 13:46:29 GMT]]>@gsuberland Do you get a warning with GCC? If I am not mistaken, this is not catched by GCC nor clang with the warning flags shown. You need -Wconversion.

]]>https://board.circlewithadot.net/post/https://mastodon.social/users/uecker/statuses/116324092471176632https://board.circlewithadot.net/post/https://mastodon.social/users/uecker/statuses/116324092471176632Tue, 31 Mar 2026 13:46:29 GMT<![CDATA[Reply to reposting for the day crowd: I ran into a memcmp implementation that only compared 25% of the bytes, and the issue wasn't caught in the build because the vendor toolchain failed to emit a warning. on Tue, 31 Mar 2026 13:37:58 GMT]]>to be fair it should also have been unit tested but I'm gonna cut the devs some slack here because the toolchain vendor rugpulling a whole warning category is a significantly worse offense.

]]>https://board.circlewithadot.net/post/https://chaos.social/users/gsuberland/statuses/116324058954438835https://board.circlewithadot.net/post/https://chaos.social/users/gsuberland/statuses/116324058954438835Tue, 31 Mar 2026 13:37:58 GMT<![CDATA[Reply to reposting for the day crowd: I ran into a memcmp implementation that only compared 25% of the bytes, and the issue wasn't caught in the build because the vendor toolchain failed to emit a warning. on Tue, 31 Mar 2026 13:23:25 GMT]]>@gsuberland oh yeah i saw this earlier and thought surely the compiler would yell at you for that comparison

oopsie

]]>https://board.circlewithadot.net/post/https://icosahedron.website/users/halcy/statuses/116324001748192322https://board.circlewithadot.net/post/https://icosahedron.website/users/halcy/statuses/116324001748192322Tue, 31 Mar 2026 13:23:25 GMT