Configuring the Lisp Process

The command used to start the Lisp process can be configured from the SLIMA settings page. The basic setup takes a Lisp executable and a directory containing SLIME's files. In this case, SLIMA automatically chooses the command line arguments and the port to communicate over. This will work for most lisp executables and workflows.

Fine-grain control of the lisp process is also available for advanced users. First, the network address used by SLIMA can be adjusted in the advanced settings if the default port is in use or when connecting to remote processes. Next, if the slime path field points to a file instead of a directory, that file will be passed to the Lisp process instead of SLIMA's generated configuration file; however, the user must ensure the Swank server and SLIMA are configured to use the same network port. Finally, the command to start the lisp process can be manually entered. If the Custom Swank Command field is not empty, it is evaluated by cmd or /bin/sh, depending on Operating System, when the process is started. The lisp process always inherits the environment variables, including PATH, that are present when Atom is run.

Example Commands

Linedit

The linedit package recommends running sbcl with the --no-linedit flag in Slime. So, setting the command to sbcl --no-linedit --load /path/to/slime/start-swank.lisp will allow this.

More Heap Size

Some applications require a larger heap than default. On SBCL, this can be configured with sbcl --dynamic-space-size 2560 --load /path/to/slime/start-swank.lisp

Qlot

Qlot allows for project specific Quicklisp distributions and provides a executable. So, you can run qlot with the current directory using qlot exec ros -S . run /path/to/slime/start-swank.lisp. Note that the root directory of the current project is used as the current working directory, with the directory of the active file as a fall back.