* CVS User trittweiler [2010-03-18 16:59+0100] writes:
Update of /project/slime/cvsroot/slime In directory cl-net:/tmp/cvs-serv19810
Modified Files: ChangeLog slime.el Log Message: Add M-x slime-attach-gdb as an interactive function.
The ATTACH-GDB restart is nice because it's convenient and the backends can specify customized gdb configuration.
Is there a technical reason that the restart must be there or is it a time limited advertising campaign?
Helmut
Helmut Eller heller@common-lisp.net writes:
- CVS User trittweiler [2010-03-18 16:59+0100] writes:
Update of /project/slime/cvsroot/slime In directory cl-net:/tmp/cvs-serv19810
Modified Files: ChangeLog slime.el Log Message: Add M-x slime-attach-gdb as an interactive function.
The ATTACH-GDB restart is nice because it's convenient and the backends can specify customized gdb configuration.
Is there a technical reason that the restart must be there or is it a time limited advertising campaign?
I use the restart in ECL where the Lisp backtrace is often not enough, so I want to get at the C backtrace conveniently and quickly.
Also the bit about enabling backends to specify default configuration is handy so you get to a gdb prompt which is ready to use. In principle, M-x slime-attach-gdb could send a request to the inferior Lisp but then that would violate its lightweight purpose.
Any problem with the restart being present? It's pretty down the restart list.
-T.
* Tobias C. Rittweiler [2010-03-18 17:24+0100] writes:
Helmut Eller heller@common-lisp.net writes:
- CVS User trittweiler [2010-03-18 16:59+0100] writes:
Update of /project/slime/cvsroot/slime In directory cl-net:/tmp/cvs-serv19810
Modified Files: ChangeLog slime.el Log Message: Add M-x slime-attach-gdb as an interactive function.
The ATTACH-GDB restart is nice because it's convenient and the backends can specify customized gdb configuration.
Is there a technical reason that the restart must be there or is it a time limited advertising campaign?
I use the restart in ECL where the Lisp backtrace is often not enough, so I want to get at the C backtrace conveniently and quickly.
It could be a key in sldb-mode-map.
Also the bit about enabling backends to specify default configuration is handy so you get to a gdb prompt which is ready to use. In principle, M-x slime-attach-gdb could send a request to the inferior Lisp but then that would violate its lightweight purpose.
Requesting the args for gdb is just as lightweight as invoking a restart. I would even say that it's simpler than invoking a restart which tells Emacs to start gud; the latter requires one more packet on the wire.
Any problem with the restart being present? It's pretty down the restart list.
It takes a lot of space and clutters swank.lisp considerably.
Helmut
Helmut Eller heller@common-lisp.net writes:
- Tobias C. Rittweiler [2010-03-18 17:24+0100] writes:
Helmut Eller heller@common-lisp.net writes:
- CVS User trittweiler [2010-03-18 16:59+0100] writes:
Update of /project/slime/cvsroot/slime In directory cl-net:/tmp/cvs-serv19810
Modified Files: ChangeLog slime.el Log Message: Add M-x slime-attach-gdb as an interactive function.
The ATTACH-GDB restart is nice because it's convenient and the backends can specify customized gdb configuration.
Is there a technical reason that the restart must be there or is it a time limited advertising campaign?
I use the restart in ECL where the Lisp backtrace is often not enough, so I want to get at the C backtrace conveniently and quickly.
It could be a key in sldb-mode-map.
Also the bit about enabling backends to specify default configuration is handy so you get to a gdb prompt which is ready to use. In principle, M-x slime-attach-gdb could send a request to the inferior Lisp but then that would violate its lightweight purpose.
Requesting the args for gdb is just as lightweight as invoking a restart. I would even say that it's simpler than invoking a restart which tells Emacs to start gud; the latter requires one more packet on the wire.
M-x slime-attach-gdb as of now does not involve any Swank request, that's what I compared it, too.
Sldb shortcut sounds fine, too. Any recommendation? Both g and G seem to be available. I'll make the the shortcut get the default config from the inferior-lisp unless called with prefix arg.
-T.
"Tobias C. Rittweiler" tcr@freebits.de writes:
Helmut Eller heller@common-lisp.net writes:
- Tobias C. Rittweiler [2010-03-18 17:24+0100] writes:
Helmut Eller heller@common-lisp.net writes:
- CVS User trittweiler [2010-03-18 16:59+0100] writes:
Update of /project/slime/cvsroot/slime In directory cl-net:/tmp/cvs-serv19810
Modified Files: ChangeLog slime.el Log Message: Add M-x slime-attach-gdb as an interactive function.
The ATTACH-GDB restart is nice because it's convenient and the backends can specify customized gdb configuration.
Is there a technical reason that the restart must be there or is it a time limited advertising campaign?
I use the restart in ECL where the Lisp backtrace is often not enough, so I want to get at the C backtrace conveniently and quickly.
It could be a key in sldb-mode-map.
Also the bit about enabling backends to specify default configuration is handy so you get to a gdb prompt which is ready to use. In principle, M-x slime-attach-gdb could send a request to the inferior Lisp but then that would violate its lightweight purpose.
Requesting the args for gdb is just as lightweight as invoking a restart. I would even say that it's simpler than invoking a restart which tells Emacs to start gud; the latter requires one more packet on the wire.
M-x slime-attach-gdb as of now does not involve any Swank request, that's what I compared it, too.
Sldb shortcut sounds fine, too. Any recommendation? Both g and G seem to be available. I'll make the the shortcut get the default config from the inferior-lisp unless called with prefix arg.
g is usually used for updating.
* Stas Boukarev [2010-03-18 18:43+0100] writes:
Sldb shortcut sounds fine, too. Any recommendation? Both g and G seem to be available. I'll make the the shortcut get the default config from the inferior-lisp unless called with prefix arg.
g is usually used for updating.
Right, g is not a good choice. I would also prefer a shifted key to reduce the risk of pressing it accidentally. G or A for "attach" are better.
Helmut
Helmut Eller heller@common-lisp.net writes:
Any problem with the restart being present? It's pretty down the restart list.
It takes a lot of space and clutters swank.lisp considerably.
Implemented as sldb command now. The implementation is indeed much simpler. Thanks for the eyeing.
-T.
-
Helmut Eller -
Stas Boukarev -
Tobias C. Rittweiler