[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <1358836066.2172.45.camel@6-core>
Date: Tue, 22 Jan 2013 07:27:46 +0100
From: pierre <pierre@...entlife.com>
To: musl@...ts.openwall.com, Rich Felker <dalias@...ifal.cx>
Subject: Re: dladdr()
Rich,
> I don't see how any of these could crash without
> memory already having been corrupted.
That's what I wrote previously: my code populates
backtraces which could have been damaged by stack
smashing.
Backtraces are used to identify the cause of a crash,
which, by definition, may not be clean.
The information I wanted you guys to have is that,
for the same application code, both GLIBC and musl
compiled in DEBUG mode do not crash (while musl in
release mode crashes).
> Do you have a test case I could run,
In my previous email (the one you are answering)
I wrote:
"I did not have time to isolate the problem in a
reproducible code snippet but I will let you
know when I have done that."
> have you tried building libc with -g and examining
> the crash in gdb to see where it happens?
In my previous email (the one you are answering)
I wrote:
"for my main application (which populates backtraces
with dladdr), musl dladdr() works in DEBUG mode and
crashes in RELEASE mode."
The musl "DEBUG mode" tested here used -g2 and it did
not crash.
I am not trying to argue with you - there's no point
at doing that - I was merely providing the information
I had available when you asked "does it work for you?".
Pierre
Powered by blists - more mailing lists
Confused about mailing lists and their use?
Read about mailing lists on Wikipedia
and check out these
guidelines on proper formatting of your messages.
