From: David Hildenbrand <david@redhat.com>
To: Anshuman Khandual <anshuman.khandual@arm.com>,
David Hildenbrand <dhildenb@redhat.com>
Cc: mark.rutland@arm.com, catalin.marinas@arm.com,
linux-mm@kvack.org, arunks@codeaurora.org,
cpandya@codeaurora.org, ira.weiny@intel.com, will@kernel.org,
steven.price@arm.com, valentin.schneider@arm.com,
suzuki.poulose@arm.com, robin.murphy@arm.com, broonie@kernel.org,
cai@lca.pw, ard.biesheuvel@arm.com, dan.j.williams@intel.com,
linux-arm-kernel@lists.infradead.org, osalvador@suse.de,
steve.capper@arm.com, logang@deltatee.com,
linux-kernel@vger.kernel.org, akpm@linux-foundation.org,
mgorman@techsingularity.net
Subject: Re: [PATCH V11 1/5] mm/hotplug: Introduce arch callback validating the hot remove range
Date: Tue, 14 Jan 2020 13:30:43 +0100 [thread overview]
Message-ID: <2570152d-47b4-c9c0-6a40-18952355069d@redhat.com> (raw)
In-Reply-To: <2c4b04d6-6d86-e87a-9b09-b931133a0d9c@arm.com>
On 14.01.20 12:09, Anshuman Khandual wrote:
>
>
> On 01/14/2020 07:43 AM, Anshuman Khandual wrote:
>>
>>
>> On 01/13/2020 04:07 PM, David Hildenbrand wrote:
>>> On 13.01.20 10:50, Anshuman Khandual wrote:
>>>>
>>>>
>>>> On 01/13/2020 02:44 PM, David Hildenbrand wrote:
>>>>>
>>>>>
>>>>>> Am 13.01.2020 um 10:10 schrieb Anshuman Khandual <anshuman.khandual@arm.com>:
>>>>>>
>>>>>>
>>>>>>
>>>>>>> On 01/10/2020 02:12 PM, David Hildenbrand wrote:
>>>>>>>> On 10.01.20 04:09, Anshuman Khandual wrote:
>>>>>>>> Currently there are two interfaces to initiate memory range hot removal i.e
>>>>>>>> remove_memory() and __remove_memory() which then calls try_remove_memory().
>>>>>>>> Platform gets called with arch_remove_memory() to tear down required kernel
>>>>>>>> page tables and other arch specific procedures. But there are platforms
>>>>>>>> like arm64 which might want to prevent removal of certain specific memory
>>>>>>>> ranges irrespective of their present usage or movability properties.
>>>>>>>
>>>>>>> Why? Is this only relevant for boot memory? I hope so, otherwise the
>>>>>>> arch code needs fixing IMHO.
>>>>>>
>>>>>> Right, it is relevant only for the boot memory on arm64 platform. But this
>>>>>> new arch callback makes it flexible to reject any given memory range.
>>>>>>
>>>>>>>
>>>>>>> If it's only boot memory, we should disallow offlining instead via a
>>>>>>> memory notifier - much cleaner.
>>>>>>
>>>>>> Dont have much detail understanding of MMU notifier mechanism but from some
>>>>>> initial reading, it seems like we need to have a mm_struct for a notifier
>>>>>> to monitor various events on the page table. Just wondering how a physical
>>>>>> memory range like boot memory can be monitored because it can be used both
>>>>>> for for kernel (init_mm) or user space process at same time. Is there some
>>>>>> mechanism we could do this ?
>>>>>>
>>>>>>>
>>>>>>>>
>>>>>>>> Current arch call back arch_remove_memory() is too late in the process to
>>>>>>>> abort memory hot removal as memory block devices and firmware memory map
>>>>>>>> entries would have already been removed. Platforms should be able to abort
>>>>>>>> the process before taking the mem_hotplug_lock with mem_hotplug_begin().
>>>>>>>> This essentially requires a new arch callback for memory range validation.
>>>>>>>
>>>>>>> I somewhat dislike this very much. Memory removal should never fail if
>>>>>>> used sanely. See e.g., __remove_memory(), it will BUG() whenever
>>>>>>> something like that would strike.
>>>>>>>
>>>>>>>>
>>>>>>>> This differentiates memory range validation between memory hot add and hot
>>>>>>>> remove paths before carving out a new helper check_hotremove_memory_range()
>>>>>>>> which incorporates a new arch callback. This call back provides platforms
>>>>>>>> an opportunity to refuse memory removal at the very onset. In future the
>>>>>>>> same principle can be extended for memory hot add path if required.
>>>>>>>>
>>>>>>>> Platforms can choose to override this callback in order to reject specific
>>>>>>>> memory ranges from removal or can just fallback to a default implementation
>>>>>>>> which allows removal of all memory ranges.
>>>>>>>
>>>>>>> I suspect we want really want to disallow offlining instead. E.g., I
>>>>>>
>>>>>> If boot memory pages can be prevented from being offlined for sure, then it
>>>>>> would indirectly definitely prevent hot remove process as well.
>>>>>>
>>>>>>> remember s390x does that with certain areas needed for dumping/kexec.
>>>>>>
>>>>>> Could not find any references to mmu_notifier in arch/s390 or any other arch
>>>>>> for that matter apart from KVM (which has an user space component), could you
>>>>>> please give some pointers ?
>>>>>
>>>>> Memory (hotplug) notifier, not MMU notifier :)
>>>>
>>>> They are so similarly named :)
>>>>
>>>>>
>>>>> Not on my notebook right now, grep for MEM_GOING_OFFLINE, that should be it.
>>>>>
>>>>
>>>> Got it, thanks ! But we will still need boot memory enumeration via MEMBLOCK_BOOT
>>>> to reject affected offline requests in the callback.
>>>
>>> Do you really need that?
>>>
>>> We have SECTION_IS_EARLY. You could iterate all involved sections (for
>>> which you are getting notified) and check if any one of these is marked
>>> SECTION_IS_EARLY. then, it was added during boot and not via add_memory().
>>
>> Seems to be a better approach than adding a new memblock flag.
>
> These additional changes do the trick and prevent boot memory removal.
> Hope this is in line with your earlier suggestion.
>
> diff --git a/arch/arm64/mm/mmu.c b/arch/arm64/mm/mmu.c
> index 00f3e1836558..3b59e6a29dea 100644
> --- a/arch/arm64/mm/mmu.c
> +++ b/arch/arm64/mm/mmu.c
> @@ -17,6 +17,7 @@
> +++ b/arch/arm64/mm/mmu.c
> @@ -17,6 +17,7 @@
> #include <linux/mman.h>
> #include <linux/nodemask.h>
> #include <linux/memblock.h>
> +#include <linux/memory.h>
> #include <linux/fs.h>
> #include <linux/io.h>
> #include <linux/mm.h>
> @@ -1365,4 +1366,37 @@ void arch_remove_memory(int nid, u64 start, u64 size,
> __remove_pages(start_pfn, nr_pages, altmap);
> __remove_pgd_mapping(swapper_pg_dir, __phys_to_virt(start), size);
> }
> +
> +static int boot_mem_remove_notifier(struct notifier_block *nb,
> + unsigned long action, void *data)
> +{
> + unsigned long start_pfn, end_pfn, pfn, section_nr;
> + struct mem_section *ms;
> + struct memory_notify *arg = data;
> +
> + start_pfn =
> + end_pfn = start_pfn + arg->nr_pages;
You can initialize some of these directly
struct memory_notify *arg = data;
const unsigned long end_pfn = arg->start_pfn; + arg->nr_pages;
unsigned long pfn = arg->start_pfn;
and avoid start_pfn.
> +
> + if (action != MEM_GOING_OFFLINE)
> + return NOTIFY_OK;
> +
> + for (pfn = start_pfn; pfn < end_pfn; pfn += PAGES_PER_SECTION) {
> + section_nr = ;
> + ms = __nr_to_section(section_nr);
Also, I think you can avoid section_nr.
ms = __nr_to_section(pfn_to_section_nr(pfn));
> +
> + if (early_section(ms))
> + return NOTIFY_BAD;
> + }
> + return NOTIFY_OK;
> +}
> +
> +static struct notifier_block boot_mem_remove_nb = {
> + .notifier_call = boot_mem_remove_notifier,
> +};
> +
> +static int __init boot_mem_remove_init(void)
> +{
> + return register_memory_notifier(&boot_mem_remove_nb);
> +}
> +device_initcall(boot_mem_remove_init);
> #endif
Exactly what I was suggesting :)
If we ever need to offline+re-online boot memory (e.g., to a different
zone), we can think of something else.
--
Thanks,
David / dhildenb
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
next prev parent reply other threads:[~2020-01-14 12:31 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-01-10 3:09 [PATCH V11 0/5] arm64/mm: Enable memory hot remove Anshuman Khandual
2020-01-10 3:09 ` [PATCH V11 1/5] mm/hotplug: Introduce arch callback validating the hot remove range Anshuman Khandual
2020-01-10 8:42 ` David Hildenbrand
2020-01-13 9:11 ` Anshuman Khandual
2020-01-13 9:14 ` David Hildenbrand
2020-01-13 9:50 ` Anshuman Khandual
2020-01-13 10:37 ` David Hildenbrand
2020-01-14 2:13 ` Anshuman Khandual
2020-01-14 11:09 ` Anshuman Khandual
2020-01-14 12:30 ` David Hildenbrand [this message]
2020-01-11 14:11 ` kbuild test robot
2020-01-13 4:06 ` Anshuman Khandual
2020-01-11 19:49 ` kbuild test robot
2020-01-10 3:09 ` [PATCH V11 2/5] mm/memblock: Introduce MEMBLOCK_BOOT flag Anshuman Khandual
2020-01-13 7:37 ` Mike Rapoport
2020-01-13 8:43 ` Anshuman Khandual
2020-01-13 8:57 ` David Hildenbrand
2020-01-10 3:09 ` [PATCH V11 3/5] of/fdt: Mark boot memory with MEMBLOCK_BOOT Anshuman Khandual
2020-01-10 3:09 ` [PATCH V11 4/5] arm64/mm: Hold memory hotplug lock while walking for kernel page table dump Anshuman Khandual
2020-01-10 3:09 ` [PATCH V11 5/5] arm64/mm: Enable memory hot remove Anshuman Khandual
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=2570152d-47b4-c9c0-6a40-18952355069d@redhat.com \
--to=david@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=anshuman.khandual@arm.com \
--cc=ard.biesheuvel@arm.com \
--cc=arunks@codeaurora.org \
--cc=broonie@kernel.org \
--cc=cai@lca.pw \
--cc=catalin.marinas@arm.com \
--cc=cpandya@codeaurora.org \
--cc=dan.j.williams@intel.com \
--cc=dhildenb@redhat.com \
--cc=ira.weiny@intel.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=logang@deltatee.com \
--cc=mark.rutland@arm.com \
--cc=mgorman@techsingularity.net \
--cc=osalvador@suse.de \
--cc=robin.murphy@arm.com \
--cc=steve.capper@arm.com \
--cc=steven.price@arm.com \
--cc=suzuki.poulose@arm.com \
--cc=valentin.schneider@arm.com \
--cc=will@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).