Date: Mon, 23 Jun 2014 04:12:56 +0200 From: Daniel Lovasko <daniel.lovasko@gmail.com> To: soc-status@freebsd.org Subject: [libctf and DDB improvements] report #3 and #4 Message-ID: <CA%2BcSnN02=mSBjb5WMdxNHiWZfizVb=345cKrmVci8RL4JTgViQ@mail.gmail.com>
next in thread | raw e-mail | index | archive | help
Hello, Right now this things are done: - reading of the whole CTF format (labels, types, ...) - parent reference solving - decompressing - public API to query the CTF data - error reporting from public API functions - ctfdump rewrite (still in beta stage) - everything seems to be working for the /boot/kernel/kernel file The public API should not change very much, but the implementation may get some polishing. I have done some basic performance and memory utilisation tests and currently it takes ~0.2 seconds for the ctfdump to print all information to a file (basic stdout redirection) and the maximal resident set is ~20MB. Both tests were done with the kernel file /boot/kernel/kernel (which is zipped). Near future updates may include using valgrind to find some memleaks or lint to check for potential bugs and to make the code more straight. Also, the documentation seems to be few steps behind the code and algorithms, I should fix that ASAP. After that, I need to consult with George whether to continue with the library to provide writing API or leave it for now and work on the DDB part of the project. The ctfdump code can be found here [1] Any requests or questions are, of course, highly welcome. Best, Daniel [1] https://github.com/lovasko/ctfdump
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CA%2BcSnN02=mSBjb5WMdxNHiWZfizVb=345cKrmVci8RL4JTgViQ>