This articles describes some intricacies of handling shared memory mappings, i.e. mappings that are shared between a few processes.
Every process has one or more mmaped files. Some mappings (for example, ones of shared libraries) are shared between a few processes. During the checkpointing, CRIU need to figure out all the mappings that are shared in order to dump them as such.
It does so by performing
fstatat() for each entry in
noting the device and inode fields of the structure returned by fstatat(). This information
is collected and sorted. Now, if any few processes have a mapping with same device and inode,
this mapping is a shared one and should be dumped as such.
It's important to note that the above mechanism works not just for the file-based mappings,
but also for the anonymous ones. For an anonymous mapping, kernel actually creates a hidden
tmpfs file, and so
fstatat() on the
works the same way as for other files. The tmpfs file itself is not visible from any tmpfs
mounts, but can be opened via its
During the restore, CRIU already knows which mappings are shared, so they need to be restored as shared. To restore file mappings, no tricks are needed, they are opened and mmaped with with a MAP_SHARED flag set.
Anonymous memory mappings, though, need some work to be restored as such. Here is how it is done.
Among all the processes sharing a mapping, the one with the lowest PID among the group (see postulates) is assigned to be a mapping creator. The creator task is to obtain a mapping file descriptor, restore the mapping data, and signal all the other process that it's ready. During this process, all the other processes are waiting.
First, the creator need to obtain a file descriptor for the mapping. To achieve it, two different approaches are used, depending on the availability.
In case memfd_create()
syscall is available (Linux kernel v3.17+), it is used to obtain a file descriptor.
ftruncate() is called to set the proper size of mapping.
memfd_create() is not available, the alternative approach is used.
First, mmap() is called to create a mapping. Next, a file in
is opened to get a file descriptor for the mapping. The limitation of this method is,
due to security concerns, /proc/$PID/map_files/ is not available for processes that
live inside a user namespace, so it is impossible to use it if there
are any user namespaces in the dump.
Once the creator have the file descriptor, it mmap()s it and restores its content from the dump (using memcpy()). The creator then unmaps the the mapping (note the file descriptor is still available). Next, it calls futex(FUTEX_WAKE) to signal all the waiting processes that the mapping file descriptor is ready.
All the other processes that need this mapping wait on futex(FUTEX_WAIT). Once the wait is over, they open the creator's /proc/$CREATOR_PID/fd/$FD file to get the mapping file descriptor.
Finally, all the processes (including the creator itself) call mmap() to create a needed mapping (note that mmap() arguments such as length, offset and flags may differ for different processes), and close() the mapping file descriptor as it is no longer needed.
For iterative migration it's very useful to track changes in memory. Until CRIU v2.5, changes were tracked for anonymous memory only, but now it is also shared memory can be tracked as well. To achieve it, CRIU scans all the shmem segment owners' pagemap (as it does for anonymous memory) and then ANDs the collected soft-dirty bits.
The changes tracking caused developers to implement memory images deduplication for shmem segments as well.
Dumping present pages
When dumping the contents of shared memory, CRIU does not dump all of the data. Instead, it determines which pages contain it, and only dumps those pages. This is done similarly to how regular memory dumping and restoring works, i.e. by looking for PRESENT or SWAPPED bits in owners' pagemap entries.
There is one particular feature of shared memory dumps worth mentioning. Sometimes, a shared memory page can exist in the kernel, but it is not mapped to any process. CRIU detects such pages by calling mincore() on the shmem segment, which reports back the page in-memory status. The mincore bitmap is when ANDed with the per-process ones.