Flag the checkpoint descriptor so on restore we can identify it and use the
commite757f81e26b0fa9fb2b6a1192151ba6ceb4002d2
authorMatthew Dillon <dillon@dragonflybsd.org>
Fri, 29 Jun 2007 23:40:00 +0000 (29 23:40 +0000)
committerMatthew Dillon <dillon@dragonflybsd.org>
Fri, 29 Jun 2007 23:40:00 +0000 (29 23:40 +0000)
treeda55f3512b20c672bec88896b265a9d9cb18040a
parent56a123de105c1e2e42ce4df9b7e46983cad53b0c
Flag the checkpoint descriptor so on restore we can identify it and use the
descriptor for the restore rather then trying to look up the original
checkpoint file.  This issue occurs when a program calls sys_checkpoint()
manually.

This allows a checkpoint-resume to be done on a copied checkpoint file,
or a gzipped (then gunzipped) checkpoint file, etc.  The original checkpoint
file no longer needs to remain intact.

Requested-by: _why <why@ruby-lang.org>
sys/kern/imgact_elf.c
sys/kern/kern_checkpoint.c
sys/sys/ckpt.h