>This patch modifies the fcoemon signal handler so that it >can call a different routine for new signals. It also has >fcoemon register for the SIGUSR1 signal and when it catches >that signal it cruedly dumps the state of fcoemon's fcoe_port >and fcm_netif lists. > >No documentation is provided with this feature as it is a >developer only feature. > >Signed-off-by: Robert Love <robert.w.l...@intel.com> >--- > fcoemon.c | 68 >++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++- > 1 files changed, 67 insertions(+), 1 deletions(-) > >diff --git a/fcoemon.c b/fcoemon.c >index fd8cd1e..68b7281 100644 >--- a/fcoemon.c >+++ b/fcoemon.c >@@ -2781,9 +2781,70 @@ static void fcm_usage(void) > exit(1); > } > >+static enum fcoe_status fcm_dump(void)
This can be void... Originally I had created a message from fcoeadm to fcoemon to trigger the dump. When I started thinking about if I wanted to document it in the man page I started thinking that I didn't really want to make it an "official" option to fcoeadm. An option to fcoemon doesn't make sense, since you really want to trigger it at runtime. SIGUSR1 seemed to be the most appropriate alternative. I'll clean up the function signature, but I just wanted to see if anyone had any heartburn with this approach. Thanks, //Rob _______________________________________________ devel mailing list devel@open-fcoe.org https://lists.open-fcoe.org/mailman/listinfo/devel
