Difference between revisions of "Memory dumping and restoring"
Jump to navigation
Jump to search
(Categorize) |
|||
| Line 9: | Line 9: | ||
** shared memory "identifier" to detect the sharing | ** shared memory "identifier" to detect the sharing | ||
* proc pagemap file says which pages are to be dumped and which are not (to be fixed as well) | * proc pagemap file says which pages are to be dumped and which are not (to be fixed as well) | ||
| − | * Ptrace SEIZE is used to | + | * Ptrace SEIZE is used to grab pages from task's VM into pipe (with vmsplice) |
=== Restoring === | === Restoring === | ||
This one depends only on the /proc/pid/map_files/ to restore the shmem regions -- tasks just open some other's link and map it to create shmem region. The pages restoration just writes page data "in place". | This one depends only on the /proc/pid/map_files/ to restore the shmem regions -- tasks just open some other's link and map it to create shmem region. The pages restoration just writes page data "in place". | ||
| − | |||
| − | |||
=== Non linear mappings === | === Non linear mappings === | ||
Revision as of 21:15, 14 May 2013
How it works now
Dumping
Currently memory dumping depends on 3 big technologies:
- /proc/pid/map_files/ directory with links is used to determine
- which file is mapped
- shared memory "identifier" to detect the sharing
- proc pagemap file says which pages are to be dumped and which are not (to be fixed as well)
- Ptrace SEIZE is used to grab pages from task's VM into pipe (with vmsplice)
Restoring
This one depends only on the /proc/pid/map_files/ to restore the shmem regions -- tasks just open some other's link and map it to create shmem region. The pages restoration just writes page data "in place".
Non linear mappings
Currently we don't support non-linear mappings (fail dump if present)