|
From: | Markus Doppelbauer |
Subject: | Re: [libmicrohttpd] GNU libmicrohttpd 0.9.74 released |
Date: | Tue, 08 Mar 2022 11:05:02 +0100 |
User-agent: | Evolution 3.36.5-0ubuntu1 |
Dear Evgeny, All problems are gone. Thank you for all your efforts. Best wishes Markus -------- Ursprüngliche Nachricht -------- Von: Evgeny Grin <k2k@yandex.ru> Antwort an: libmicrohttpd development and user mailinglist <libmicrohttpd@gnu.org> Betreff: Re: [libmicrohttpd] GNU libmicrohttpd 0.9.74 released Datum: Mon, 7 Mar 2022 21:56:43 +0300 Hi Markus, Thanks for the report. Unfortunatelly ASAN documentation is not very precise about manual poisoning/unpoisoning, however I think memory must be *unpoisoned* before freeing, just because memory is not poisoned when allocated. Should be fixed by https://git.gnunet.org/libmicrohttpd.git/commit/?id=1a46bfd6e0929e2004ffda318282cbf21d9f5299 Please check whether it fixed the issue for you. -- Evgeny On 07.03.2022 20:11, Markus Doppelbauer wrote: > + Added '--enable-sanitizers[=address,undefined,leak,user-poison]'> configure parameter (instead of '--enable-sanitizer'),> implemented custom memory poisoning for memory pools.We get strange asan aborts (maybe once per 30min):ERROR: AddressSanitizer: use-after-poison on address 0xdeadbeefMy question (sorry I can't find any docs on this issue):MHD calls '_MHD_POISON_MEMORY()' before 'free()'.Other code calls 'ASAN_UNPOISON_MEMORY_REGION()' before free, e.g.:https://github.com/AlxHnr/CRegion/blob/1347314d953e8b5e26eed3b0fe3151e482958c39/src/mempool.c#L69<https://github.com/AlxHnr/CRegion/blob/1347314d953e8b5e26eed3b0fe3151e482958c39/src/mempool.c#L69> |
[Prev in Thread] | Current Thread | [Next in Thread] |