kicad-developers team mailing list archive

["Steven A. Falco" <stevenfalco@xxxxxxxxx>]

I asked on the Fedora development list about removing the "-Wp,-D_GLIBCXX_ASSERTIONS" flag as requested in https://bugs.launchpad.net/kicad/+bug/1838448, and I got some pushback (attached) stating that it is not a good idea to hide crashes caused by out-of-bounds memory accesses.

I agree with that, and I made a similar argument in comment #22 of the bug.

Therefore I'd like to have more discussion about this.  Are we really sure we want to hide memory access errors?  In some cases they could cause hard-to-find corruption bugs, as well as hard crashes.  Personally, I'd rather know there is a problem.

	Steve

--- Begin Message ---

• To: Development discussions related to Fedora <devel@xxxxxxxxxxxxxxxxxxxxxxx>, "Steven A. Falco" <stevenfalco@xxxxxxxxx>
• From: Tom Hughes <tom@xxxxxxxxxx>
• Date: Fri, 2 Aug 2019 16:09:44 +0100
• Arc-authentication-results: i=1; mx.google.com; spf=pass (google.com: domain of tom@xxxxxxxxxx designates 2001:8b0:bd:1:21c:c0ff:fede:937a as permitted sender) smtp.mailfrom=tom@xxxxxxxxxx
• Arc-message-signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=content-transfer-encoding:content-language:in-reply-to:mime-version :user-agent:date:message-id:from:references:to:subject; bh=dEJEKiIR3iks4VW8SwerX6FVxV6JrLlK6WNVUbiUj7g=; b=Jrr0z6HlqrH+7HJe+Re6h40rMPZHT3sdSIPQMSSQxkhpnTArb58D2WxAMry4R4cjfk AyXE0b5rTWgq5mORrfjEZv1yVfyKB16A4j1tzEIyy/18p/5g9+q8pZ+vRnsiWoetQ7zN YEZd5Gp4CeidqYArWwnHQzismt3UDnnkv7+nb6C2XjGfnunR4z3eTG5jKWm5OCIZLf7O Z2dYxMUrAJq4O1Dury3mgxw81uHqIPVF16bq3hJQKiX+wsAkZYGgrCm7AQ5DPMHGbiU7 8F+Cvh8zs5irdpZLHS+Yf8ovWY5zeeVpLLtxr4l/9I+XkqU0A0sIebb/fuy5zE8MosDL othg==
• Arc-seal: i=1; a=rsa-sha256; t=1564758589; cv=none; d=google.com; s=arc-20160816; b=NVc83Yu7xKiC8QzhOwm36MzAnDA3BCrwQfJVBo+nL0ElIvpFhFPxwJfJ3Leh3AJy/N uJ/tiKS4tGmAAkcR+nJ2Ppeg8mijeuIvHwP4slj3KhJEzeyDJfizCUNe7WiIdhR6QIOE X2I67zrh/IK7aJzw63uiAi20yJBrY/JhEz7/aXP5rqQzPRnxL6TZcvtD9F9ccZmZdukI yvHRy/bDptg9jba5ezTW0tuKJ4MXYv2XYHrphRAW4Id1I9qm7WK3mwbyrmRTg6OdYoNE Rlgk4Scz4YhgJm/jmjG2WZseUjuA8cjeEx51ldcR7wGunfntOc3re6YdZPRTum0js0Oq 3gCg==
• Authentication-results: mx.google.com; spf=pass (google.com: domain of tom@xxxxxxxxxx designates 2001:8b0:bd:1:21c:c0ff:fede:937a as permitted sender) smtp.mailfrom=tom@xxxxxxxxxx
• Delivered-to: stevenfalco@xxxxxxxxx
• In-reply-to: <9ab8eb29-42df-7c28-a567-1a0d63a2eb84@gmail.com>
• User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:60.0) Gecko/20100101 Thunderbird/60.7.2

On 01/08/2019 19:28, Steven A. Falco wrote:

The upstream KiCAD project has requested that I remove GLIBCXX_ASSERTIONS from the Fedora package, as described here: https://bugs.launchpad.net/kicad/+bug/1838448

What is the best way to do that?  I can add "%undefine _hardened_build" (which I am testing now) but I think that will remove other hardening features that I might want to leave enabled.

Well you just need to add -U_GLIBCXX_ASSERTIONS to the end of the
compiler flags.

But I think upstream is giving very bad advice...

That define does not "add extra crashes" in the way that they
seem to think - well I mean it does literally but those crashes
are reports of program errors on their part.

Specifically in this case they appear to be accessing a
std::vector at an index beyond the end, so they are accessing
memory that may not be allocated at all, and if it is does
not belong to the vector in question. So the program is quite
likely to crash there one day anyway, the extra assertion just
makes sure it always does.

Tom

--
Tom Hughes (tom@xxxxxxxxxx)
http://compton.nu/

--- End Message ---