Todo List
Documantation
- Usage (Interface, etc....)
User Memory Hotplug
- Removable attribute (by Yasunori Goto)
To remove memory easily,
definition of
removable area is effective. For example, if kernel memory scattered to
each node, it will become very difficult that each node will be
removed. If kernel memory is collected on same place, hotremove of
other place will be easier than nothing to do.
- Improvement of performance
- Avoiding performance deterioration with memory hotplug code.
Many of memory hotplug code will
be
cause of performance deterioration.
It should be less as much as possible.
- Performance of memory hot-remove
If there is sticky memory which
can't be migrate(and swap out), the time of hot remove will be long. At
least, memory removing time should be less than the rebooting time of
huge machine.
- Case of trouble of Network
In NFS case, if there is no
response from server, the file system keeps HIGHMEM page and waits
eternally. Probably, there are many cases like this problem when
network trouble occur.
Some drivers like Infiniband
use user
memory as a DMA buffer. The drivers have to support page migration.
This interfaces it to call driver's specific work.
Mem_section
- Specifing mem_section size by boot option.
Kernel configuration and compiling
isn't convenient for user.
Page migration
- Handle VM_NONLINEARed pages.
Sys_remap_file_pages() system
call make VM_NONLINEARed pages. Mmigrate code can't find PTEs of these
pages, because objmap rmap doesn't treat them.
- Implementation of page_migration for each filesystem.
Each filesystem uses
page->private for each reason. So, when page migration occurs, each
filesystem has to move their private data.
Sysfs and API
- Correlation between ACPI memory object and mem_section
Mem section is just logical
definition. To remove physically, ACPI information is also necessary.
- Hotplug agent of Memory hotplug
Hotplug whole of the memory on the Node
- Arrangement of pgdat and mem_map
Pgdat and mem_map should be
arranged on each node because of performance and size. They have to be
allocated before initialization and have to be freed after freed pages.
And if a
mem_section includes pgdat area, the mem_section must not be removed
before other mem_section which is on the same node is removed.
- Disable page allocation from mem_section on the removing node.
And graceful removing
There will be many pages which
must be migrated at node hotplug time. And its time must be short as
much as possible. In scheduled hotplug case for capacity on demand, if
the mem_sections which is removed are disabled of allocation
beforehand, hotremove will be graceful.
- Selection of destination of migration.
Destination of migration should
be appropriate area, because memory access speed should be kept as much
as possible. If one mem_section on the node is removed, the content
should be moved to the same node. However, if whole of
mem_sections on the node have to be removed, contents on the removing
mem_section have to be moved to other node.
Architecture/CPU depending code
IA64 (by Fujitsu & HP)
Appropriate size of mem_section
for IA64 should be researched.
To treat memory hole easily,
mem_section size should be small
than its hole size. However, if it is too small, mem_section array will
be big to treat TB class memory. And if its size is smaller than
GRANULE_SIZE or MAX_ORDER, memory hotplug will become more
difficult.
- TLB fault handler with Mem_section for region 6/7.
Virtual address of region 6/7 is
calculated by just register in current implementation. (No
memory access). But kernel with mem_section must read memory at
tlb handler. If mem_section size is too small, section array area will
be bigl, and TLB fault will occur. It become cause of performance
deterioration.
ppc (by IBM)
x86_64 (EM64T) (by Intel)
etc ?
Kernel memory Hotplug (Future Work)
- Some hardware can change destination of DMA to support memory
migration. When memory migration, the driver has to set this
destination too.
- PTE, PMD, PGD, and page table must moved theirown (if it is
necessary).
- vmalloc, kernel text need rmap for migration.
- and so on.... (many)
Related development
- Memory defragmentation (by Marcelo Tosatti)
Now implementing migration cache.
(Hugetlbpage migration will be used this implementation.)