Weekly Edition
Return to the Development page
| |||||||||||||||||||
GDB 7.5 released
GDB 7.5 released!
Release 7.5 of GDB, the GNU Debugger, is now available via anonymous
FTP. GDB is a source-level debugger for Ada, C, C++, Objective-C,
Pascal and many other languages. GDB can target (i.e., debug programs
running on) more than a dozen different processor architectures, and GDB
itself can run on most popular GNU/Linux, Unix and Microsoft Windows
variants.
You can download GDB from the GNU FTP server in the directory:
ftp://ftp.gnu.org/gnu/gdb
The vital stats:
Size md5sum Name
21MB 24a6779a9fe0260667710de1b082ef61 gdb-7.5.tar.bz2
28MB c9f5ed81008194f8f667f131234f3ef0 gdb-7.5.tar.gz
There is a web page for GDB at:
http://www.gnu.org/software/gdb/
That page includes information about GDB mailing lists (an announcement
mailing list, developers discussion lists, etc.), details on how to
access GDB's CVS repository, locations for development snapshots,
preformatted documentation, and links to related information around
the net. We will put errata notes and host-specific tips for this release
on-line as any problems come up. All mailing lists archives are also
browsable via the web.
GDB 7.5 brings new targets, features and improvements, including:
* Go language support.
* New targets (x32 ABI, microMIPS, Renesas RL78, HP OpenVMS ia64).
* More Python scripting improvements.
* SDT (Static Defined Tracing) probes support with SystemTap probes.
* GDBserver improvements (stdio connections, target-side evaluation
of breakpoint conditions, remote protocol improvements).
* Other miscellaneous improvements (ability to stop when a shared
library is loaded/unloaded, dynamic printf, etc).
* Reverse debugging on ARM.
* The binary "gdbtui" has been abandoned and can no longer be built.
Use "gdb -tui" instead.
For a complete list and more details on each item, please see the
gdb/NEWS file.
--
Joel
(Log in to post comments)
GDB 7.5 released Posted Aug 18, 2012 21:24 UTC (Sat) by scientes (guest, #83068) [Link]
The go support in gdb 7.5 is for programs compiled with gc, not gccgo. Gccgo support is still very rudimentary, as it knows nothing more than to treat gccgo programs as if they were C programs.
Emacs 24 Posted Aug 19, 2012 17:39 UTC (Sun) by ncm (subscriber, #165) [Link] I hope this gdb works better with emacs24. I had to type "set interactive-mode on" after starting gdb 7.4 (M-x gdb RET, "gdb -i=mi") before its prompt would work right. Maybe there was a better way?
Emacs 24 Posted Aug 19, 2012 21:09 UTC (Sun) by theophrastus (guest, #80847) [Link]
what's up with emacs24? it's not available even for 'testing' on debian; the reasons that i've read is that it's got some notable bugs (comments like: "you don't want to upgrade - yet"). emacs24 must be trying to incorporate some major new functionality to break such a venerable (well-tested) package..?
Emacs 24 Posted Aug 19, 2012 21:28 UTC (Sun) by nix (subscriber, #2304) [Link]
Yeah, it has major changes (I consider them improvements): a proper package system, bidirectional text support, lexical binding and closures in elisp proper...
... for all of that, I've been using trunk Emacs in anger for real work since early last year (long before the 24.1 branch) and have hardly ever seen problems: all those I did have were fixed within a week. There are known problems (e.g. lexical binding and the debugger don't work together perfectly yet), but none serious enough to affect me. I don't use every Emacs package, of course, so my experience may be unrepresentative, but little lightweight packages like CEDET, Gnus, and emacs-w3m work perfectly well :)
It is certainly no worse than 23.x: I would call it a heck of a lot better. You probably do want to upgrade. I wonder why Debian is holding off...
Emacs 24 in Debian testing Posted Aug 19, 2012 23:04 UTC (Sun) by rleigh (subscriber, #14622) [Link]
Because it was uploaded just one day before the codefreeze, and so has effectively zero testing in Debian, let alone not yet having all the other emacs-related packages updated to work with it. I'm sure that the core Emacs 24 is solid, it's just bad timing.
Emacs 24 in Debian testing Posted Aug 20, 2012 22:20 UTC (Mon) by nix (subscriber, #2304) [Link]
Oh! It's not tracking the 24.1 branch or anything? Using anything frozen in time with known bugs like that is probably a bad idea.
Emacs 24 Posted Aug 20, 2012 6:54 UTC (Mon) by SiB (subscriber, #4048) [Link]
I've had no luck with emacs24 gnus nnimap from Debian sid. Unfortunately. I did not have time to find out why.
Emacs 24 Posted Aug 20, 2012 19:39 UTC (Mon) by daglwn (subscriber, #65432) [Link]
I am also having trouble with nnimap in a custom-built emacs24. It seems to work with some servers and not others. There's one folder on one server where all new mail is immediately marked read. It's really annoying.
Emacs 24 Posted Aug 20, 2012 22:21 UTC (Mon) by nix (subscriber, #2304) [Link]
Yeah, this is definitely not an all-server problem: I rely on Gnus+nnimap for access to my work mailbox, and it works fine.
Someone with access to both a server that works and one that doesn't probably needs to dig into the code and see what is causing them to be treated differently. It pretty much has to be something the server is doing that we don't expect.
Emacs 24 Posted Aug 21, 2012 14:26 UTC (Tue) by daglwn (subscriber, #65432) [Link]
I have access to both but I'm hardly an Emacs/Gnus wizard. If someone can help me through the debug process I'd be very happy to do this.
Emacs 24 Posted Aug 21, 2012 22:59 UTC (Tue) by nix (subscriber, #2304) [Link]
(setq gnus-verbose 9) is generally the first stage.
The next stage involves the elisp debugger or copious scattering of (message ...) calls...
Emacs 24 Posted Aug 20, 2012 13:56 UTC (Mon) by Thanatopsis (guest, #14019) [Link]
>I hope this gdb works better with emacs24....
It works just fine with vi. Perhaps it's your choice of editors. :-P
Emacs 24 Posted Aug 20, 2012 16:56 UTC (Mon) by daglwn (subscriber, #65432) [Link]
I've run into similar problems and resorted to M-x gud-gdb to get back to the emacs23 gud interface. I didn't know about "set interactive-mode on." I'll have to try that!
GDB 7.5 released Posted Aug 20, 2012 21:37 UTC (Mon) by aegl (subscriber, #37581) [Link]
If "debugging" is the process of finding bugs so that they can be removed, what is "reverse debugging"? The process of inserting (and hiding) bugs? Why is this just for ARM? Do the other architectures already have it?
GDB 7.5 released Posted Aug 20, 2012 22:07 UTC (Mon) by daney (subscriber, #24551) [Link]
Some interesting articles about it can be found here:
GDB 7.5 released Posted Aug 21, 2012 1:28 UTC (Tue) by wahern (subscriber, #37304) [Link]
Those answers aren't nearly as entertaining or interesting as the question and hypothesis.
LMGTFY considered harmful. Posted Aug 24, 2012 0:24 UTC (Fri) by mikemol (subscriber, #83507) [Link]
Please don't simply point at Google; Google gives customized results on a per-user basis, based on what that user has looked up in the past. You simply can't assume that what you see when you search for a particular set of keywords is what someone else will see when they search for that same set of keywords. Further, if you haven't already successfully executed searches on closely related subjects, the weights and biases aren't going to work in your favor, and might even work against you; that's why I ultimately switched away from Google as my search engine of choice.
If you know the answer, or a good subject-specific place to ask (such as, *gasp* a threaded forum where the question isn't offtopic), you can stand to be a lot more forthcoming.
Providing a link to a Google does little better than recommend a set of keywords; anything else it might suggest is really unflattering of your character.
GDB 7.5 released Posted Aug 20, 2012 22:23 UTC (Mon) by nix (subscriber, #2304) [Link]
This is 'reverse execution', i.e. running execution forward until a problem is hit, then backward to see its cause. It uses lots of memory and requires per-arch support, so only a few arches (the popular ones) support it to date. ARM is now one such.
PL/I Flashback Posted Aug 20, 2012 22:45 UTC (Mon) by cry_regarder (subscriber, #50545) [Link]
We had reverse debugging on IBM ES/9000 back in the early 90s for PL/I. IIRC it did cost us over $100K for the debugger.
It was extremely handy a couple of times a year.
Cry
|
|||||||||||||||||||