---
name: "lam_rtrget"
section: "2"
platform: "linux"
package: "lam2-dev"
version: "6.3.2-3"
release: "Debian 2.2 (Potato)"
platforms: ["linux"]
url: "https://new.carta.tech/lam_rtrget"
---

# lam\_rtrget(2)

Unload LAM trace data.

## NAME

lam\_rtrget, lam\_rtrforget, lam\_rtrfget, lam\_rtrfforget - Unload LAM trace data.

## SYNOPSIS

int lam\_rtrget (int nodeid, int listno, int pid, char \*\*buffer);

int lam\_rtrforget (int nodeid, int listno, int pid, char \*\*buffer);

int lam\_rtrfget (int nodeid, int listno, int pid, int fd);

int lam\_rtrfforget (int nodeid, int listno, int pid, int fd);

## DESCRIPTION

The client function, *lam\_rtrget()*, gets a copy of selected trace data held by a remote trace daemon and stores it in space allocated and assigned to the *buffer* argument. It does not obtain trace data still stored with traced processes and not found in the daemon. A traced process can be forced to flush its trace data by sending it a SIGTRACE signal. See [MPIL\_Signal(2)](/MPIL_Signal/2.md).

Trace data is unloaded from the selected node and list (see [lam\_rtrstore(2)](/lam_rtrstore/2.md)) that was produced by the selected process ID. A value of -1 for *listno* and/or *pid* selects all lists and/or all trace producing clients, respectively. The combination effectively unloads all stored trace data from the given node.

Trace data is read non-destructively by *lam\_rtrget()* and destructively by *lam\_rtrforget()*. The trace data is stored in an allocated buffer whose address is passed back to the caller through the *buffer* argument. The caller is responsible for calling [free(3)](/free/3.md) on this pointer.

Trace data is read into a file, instead of an allocated buffer, by *lam\_rtrfget()* and *lam\_rtrfforget()*. The file is given as a descriptor, returned by [open(2)](/open/2.md) and opened for write.

## SEE ALSO

[lam\_rtrstore(2)](/lam_rtrstore/2.md), [MPIL\_Signal(2)](/MPIL_Signal/2.md)