To guarantee successful hot-plugging of memory blocks, these blocks
should belong to ZONE_MOVABLE which has migratable pages. Linux kernel
during boot initializes all the available memory blocks to ZONE_DMA and
ZONE_NORMAL. Hence, boot the system with less memory whose percentage
is given by mem-offline DT property 'mem-percent' and override any memory
limit imposed by kernel boot params and let mem-offline driver initialize
the remaining with ZONE_MOVABLE. The memory blocks in this zone are
applicable for memory offlining.
Change-Id: Ia7b804522fe9425101ee83d1ee354ed09483df13
Signed-off-by: Sudarshan Rajagopalan <sudaraja@codeaurora.org>
[swatsrid@codeaurora.org: Fix merge conflicts]
Signed-off-by: Swathi Sridhar <swatsrid@codeaurora.org>
Signed-off-by: Isaac J. Manjarres <isaacm@codeaurora.org>
This is a snapshot of the mem-offline driver as of msm-4.19
commit ef9b421ba16b ("sched: walt: Improve the scheduler").
Change-Id: I81be12ce85033d5d5ab6be8b10aa76a2685d5720
Signed-off-by: Isaac J. Manjarres <isaacm@codeaurora.org>
Add configuration changes in lahaina to support tsens driver.
Change-Id: Ib6f6a1684bb62355741e826419e179b159cb3d31
Signed-off-by: Ajay Prathi <aprathi@codeaurora.org>
Enable CPUFREQ-HW for enabling cpu frequency scaling.
Change-Id: Ieb299b5e253070b6c5ff18c65547f9ead429cb37
Signed-off-by: Naveen Yadav <naveenky@codeaurora.org>
Section mapping is good for TLB. But sometimes, when
system is low on physically continuous memory, this can
cause memory allocation failures. Add a fallback option to
allocate vmemmap pages from discontinuous pages.
Change-Id: Ide921c36c1e8980a8af047e2c991310405804858
Signed-off-by: Arun KS <arunks@codeaurora.org>
Signed-off-by: Vinayak Menon <vinmenon@codeaurora.org>
Signed-off-by: Sudarshan Rajagopalan <sudaraja@codeaurora.org>
[swatsrid@codeaurora.org: Fix merge conflicts]
Signed-off-by: Swathi Sridhar <swatsrid@codeaurora.org>
[isaacm@codeaurora.org: resolve build errors due to API changes]
Signed-off-by: Isaac J. Manjarres <isaacm@codeaurora.org>
Add arm64 to the list of architectures which supports
memory hotplug.
Change-Id: Iefeb8294bf06eaebb17a3b3aa8b33bb3b7133099
Signed-off-by: Arun KS <arunks@codeaurora.org>
Signed-off-by: Srivatsa Vaddagiri <vatsa@codeaurora.org>
Signed-off-by: Sudarshan Rajagopalan <sudaraja@codeaurora.org>
Signed-off-by: Isaac J. Manjarres <isaacm@codeaurora.org>
Introduce a variable to save bootloader enforced memory limits and
restricts adding beyond this boundary during a memory hotplug. Also,
export this symbol so that other kernel module have access to it.
Change-Id: I28c100644b7287ec4625c4c018b5fffc865e2e72
Signed-off-by: Arun KS <arunks@codeaurora.org>
[sudaraja@codeaurora.org: check limit with physical address of page]
Signed-off-by: Sudarshan Rajagopalan <sudaraja@codeaurora.org>
[swatsrid@codeaurora.org: Fix merge conflicts]
Signed-off-by: Swathi Sridhar <swatsrid@codeaurora.org>
[isaacm@codeaurora.org: Fix merge conflicts/use generic page onlining]
Signed-off-by: Isaac J. Manjarres <isaacm@codeaurora.org>
Allows to remove sections of memory that have been previously offlined
from userspace.
This is symmetric to the "memory probe" interface, as described in
Documentation/memory-hotplug.txt. It can be used to manually notify the
OS that one memory section has been removed.
Please, remind that a memory section can only be removed after it has
been logically off-lined; trying to remove a section which has not been
previously off-lined is not supported and will have undefined (but most
likely very bad) behaviour.
To offline a section one can:
To use the remove interface to remove the section:
where 0xYYYYYY is the physical address of the memory section to remove.
Change-Id: I1ab2fafe17b2697a5c667feae25c4ac0655f394e
Signed-off-by: Andrea Reale <ar@linux.vnet.ibm.com>
Signed-off-by: Maciej Bielski <m.bielski@virtualopensystems.com>
Patch-mainline: linux-kernel @ 11 Apr 2017, 18:25
Signed-off-by: Srivatsa Vaddagiri <vatsa@codeaurora.org>
Signed-off-by: Sudarshan Rajagopalan <sudaraja@codeaurora.org>
[isaacm@codeaurora.org: resolved merge conflicts due to API changes]
Signed-off-by: Isaac J. Manjarres <isaacm@codeaurora.org>
Add memory-hotplug support for ARM64 platform.
This requires addition of
ARCH_ENABLE_MEMORY_HOTPLUG and ARCH_ENABLE_MEMORY_HOTREMOVE config options.
MEMORY_PROBE config option is added to support
/sys/devices/system/memory/probe functionality.
In addition architecture specific arch_add_memory and
arch_remove memory management functions are added.
Change-Id: I1fc4e8c93cfe1b9722b2fc8d811390e6cdfa7995
Signed-off-by: Scott Branden <scott.branden@broadcom.com>
Patch-mainline: linux-kernel @ 11 Apr 2017, 18:24
Signed-off-by: Srivatsa Vaddagiri <vatsa@codeaurora.org>
[sudaraja@codeaurora.org: remove explicit zone calculation]
Signed-off-by: Sudarshan Rajagopalan <sudaraja@codeaurora.org>
[isaacm@codeaurora.org: remove duplicate hotplug/hot remove code in init.c]
Signed-off-by: Isaac J. Manjarres <isaacm@codeaurora.org>
This is a second and improved version of the patch previously released
in [3].
It builds on the work by Scott Branden [2] and, henceforth,
it needs to be applied on top of Scott's patches [2].
Comments are very welcome.
Changes from the original patchset and known issues:
- Compared to Scott's original patchset, this work adds the mapping of
the new hotplugged pages into the kernel page tables. This is done by
copying the old swapper_pg_dir over a new page, adding the new mappings,
and then switching to the newly built pg_dir (see `hotplug_paging` in
arch/arm64/mmu.c). There might be better ways to to this: suggestions
are more than welcome.
- The stub function for `arch_remove_memory` has been removed for now; we
are working in parallel on memory hot remove, and we plan to contribute
it as a separate patch.
- Corresponding Kconfig flags have been added;
- Note that this patch does not work when NUMA is enabled; in fact,
the function `memory_add_physaddr_to_nid` does not have an
implementation when the NUMA flag is on: this function is supposed to
return the nid the hotplugged memory should be associated with. However
it is not really clear to us yet what the semantics of this function
in the context of a NUMA system should be. A quick and dirty fix would
be to always attach to the first available NUMA node.
- In arch/arm64/mm/init.c `arch_add_memory`, we are doing a hack with the
nomap memory block flags to satisfy preconditions and postconditions of
`__add_pages` and postconditions of `arch_add_memory`. Compared to
memory hotplug implementation for other architectures, the "issue"
seems to be in the implemenation of `pfn_valid`. Suggestions on how
to cleanly avoid this hack are welcome.
This patchset can be tested by starting the kernel with the `mem=X` flag, where
X is less than the total available physical memory and has to be multiple of
MIN_MEMORY_BLOCK_SIZE. We also tested it on a customised version of QEMU
capable to emulate physical hotplug on arm64 platform.
To enable the feature the CONFIG_MEMORY_HOTPLUG compilation flag
needs to be set to true. Then, after memory is physically hotplugged,
the standard two steps to make it available (as also documented in
Documentation/memory-hotplug.txt) are:
(1) Notify memory hot-add
echo '0xYY000000' > /sys/devices/system/memory/probe
where 0xYY000000 is the first physical address of the new memory section.
(2) Online new memory block(s)
echo online > /sys/devices/system/memory/memoryXXX/state
-- or --
echo online_movable > /sys/devices/system/memory/memoryXXX/state
where XXX corresponds to the ids of newly added blocks.
Onlining can optionally be automatic at hot-add notification by enabling
the global flag:
echo online > /sys/devices/system/memory/auto_online_blocks
or by setting the corresponding config flag in the kernel build.
Again, any comment is highly appreciated.
[1] https://lkml.org/lkml/2016/11/17/49
[2] https://lkml.org/lkml/2016/12/1/811
[3] https://lkml.org/lkml/2016/12/14/188
Change-Id: I545807e3121c159aaa2f917ea914ee98f38fb296
Signed-off-by: Maciej Bielski <m.bielski@virtualopensystems.com>
Signed-off-by: Andrea Reale <ar@linux.vnet.ibm.com>
Patch-mainline: linux-kernel @ 11 Apr 2017, 18:25
Signed-off-by: Srivatsa Vaddagiri <vatsa@codeaurora.org>
[arunks@codeaurora.org: fix to pass checker test]
Signed-off-by: Arun KS <arunks@codeaurora.org>
[sudaraja@codeaurora.org: fixed merge conflicts]
Signed-off-by: Sudarshan Rajagopalan <sudaraja@codeaurora.org>
[swatsrid@codeaurora.org: Fix merge conflicts]
Signed-off-by: Swathi Sridhar <swatsrid@codeaurora.org>
[isaacm@codeaurora.org: Remove now unnecessary workarounds due to MHP changes]
Signed-off-by: Isaac J. Manjarres <isaacm@codeaurora.org>
