Re: [RFC v2 PATCH] reserve_mem: add support for static memory

Re: [RFC v2 PATCH] reserve_mem: add support for static memory

[Mike Rapoport]

Hi Shyam,

On Wed, Jun 24, 2026 at 06:22:33PM -0700, Shyam Saini wrote:
> On 21 Jun 2026 13:36, Mike Rapoport wrote:
> > On Thu, Jun 18, 2026 at 11:23:31PM -0700, Shyam Saini wrote:
> > > reserve_mem relies on dynamic memory allocation, this limits the
> > > usecase where memory is required to be preserved across the boots.
> > > Eg: ramoops memory reservation on ACPI platforms
> > >
> > > So add support to pass a pre-determined static address and reserve
> > > memory at a specified location. This enables use case like ramoops
> > > on ACPI platforms to reliably access ramoops region with previous
> > > boot logs.
> > > 
> > > Also skip the parsing of <align> when static address is passed.
> > > 
> > > Example syntax for static address
> > >  reserve_mem=4M@0x1E0000000:oops
> > 
> > reserve_mem is best effort by design because such hacks as well as memmap=
> > cannot guarantee this memory is actually free.
> > 
> > If you want to preserve ramoops reliably, use KHO with reserve_mem.
> > The first kernel will allocate memory, this memory will be preserved by KHO
> > and could be picked up by the second kernel.
> 
> ok, On ARM64 DTS systems, we can reserve ramoops memory in the device tree during
> the warm reboot.

The cc list actually implied x86 ;-)
Added arm64 folks now.

> For an equivalent ARM64 ACPI platform, what is the recommended way to reserve
> and preserve that memory across the boots? 

I don't think it exists, but a command line option (be it memmap= or
reserve_mem=) does not seem the right way to me.

Most of the arguments that were made against adding memmap= to arm64 [1]
apply here.

If kexec is an option, KHO provides a reliable way to preserve memory
across boots.

If kexec is not an option, we should look for a generic way to specify
something like DT's reserved_mem for ACPI/EFI systems.

[1] https://lkml.kernel.org/lkml/20201118063314.22940-1-song.bao.hua@hisilicon.com/T/

> Thanks,
> Shyam

-- 
Sincerely yours,
Mike.

Re: [RFC v2 PATCH] reserve_mem: add support for static memory

[Shyam Saini]

On 25 Jun 2026 11:37, Mike Rapoport wrote:
> Hi Shyam,
> 
> On Wed, Jun 24, 2026 at 06:22:33PM -0700, Shyam Saini wrote:
> > On 21 Jun 2026 13:36, Mike Rapoport wrote:
> > > On Thu, Jun 18, 2026 at 11:23:31PM -0700, Shyam Saini wrote:
> > > > reserve_mem relies on dynamic memory allocation, this limits the
> > > > usecase where memory is required to be preserved across the boots.
> > > > Eg: ramoops memory reservation on ACPI platforms
> > > >
> > > > So add support to pass a pre-determined static address and reserve
> > > > memory at a specified location. This enables use case like ramoops
> > > > on ACPI platforms to reliably access ramoops region with previous
> > > > boot logs.
> > > > 
> > > > Also skip the parsing of <align> when static address is passed.
> > > > 
> > > > Example syntax for static address
> > > >  reserve_mem=4M@0x1E0000000:oops
> > > 
> > > reserve_mem is best effort by design because such hacks as well as memmap=
> > > cannot guarantee this memory is actually free.
> > > 
> > > If you want to preserve ramoops reliably, use KHO with reserve_mem.
> > > The first kernel will allocate memory, this memory will be preserved by KHO
> > > and could be picked up by the second kernel.
> > 
> > ok, On ARM64 DTS systems, we can reserve ramoops memory in the device tree during
> > the warm reboot.
> 
> The cc list actually implied x86 ;-)
> Added arm64 folks now.

Thanks for adding ARM folks, I had just included whatever get_maintainer script
suggested, sorry. 
> > For an equivalent ARM64 ACPI platform, what is the recommended way to reserve
> > and preserve that memory across the boots? 
> 
> I don't think it exists, but a command line option (be it memmap= or
> reserve_mem=) does not seem the right way to me.
> 
> Most of the arguments that were made against adding memmap= to arm64 [1]
> apply here.
> 
> If kexec is an option, KHO provides a reliable way to preserve memory
> across boots.
> 
> If kexec is not an option, we should look for a generic way to specify
> something like DT's reserved_mem for ACPI/EFI systems.
> 
> [1] https://lkml.kernel.org/lkml/20201118063314.22940-1-song.bao.hua@hisilicon.com/T/
> 
Well, kexec is one of the option for my use case, it also requires
memory reservation during warm reboot, I think memory can be reserved in
the firmware but this will create a dependency on firmware for Linux
reservation.

It would be great to have a in kernel memory reservation mechanism for
ARM64 ACPI platforms. I believe some other use cases like PMEM
reservation would also benefit from this.

Thanks,
Shyam

Re: [RFC v2 PATCH] reserve_mem: add support for static memory

[Shyam Saini]

Hi Everyone,

> On 25 Jun 2026 11:37, Mike Rapoport wrote:
> > Hi Shyam,
> > 
> > On Wed, Jun 24, 2026 at 06:22:33PM -0700, Shyam Saini wrote:
> > > On 21 Jun 2026 13:36, Mike Rapoport wrote:
> > > > On Thu, Jun 18, 2026 at 11:23:31PM -0700, Shyam Saini wrote:
> > > > > reserve_mem relies on dynamic memory allocation, this limits the
> > > > > usecase where memory is required to be preserved across the boots.
> > > > > Eg: ramoops memory reservation on ACPI platforms
> > > > >
> > > > > So add support to pass a pre-determined static address and reserve
> > > > > memory at a specified location. This enables use case like ramoops
> > > > > on ACPI platforms to reliably access ramoops region with previous
> > > > > boot logs.
> > > > > 
> > > > > Also skip the parsing of <align> when static address is passed.
> > > > > 
> > > > > Example syntax for static address
> > > > >  reserve_mem=4M@0x1E0000000:oops
> > > > 
> > > > reserve_mem is best effort by design because such hacks as well as memmap=
> > > > cannot guarantee this memory is actually free.
> > > > 
> > > > If you want to preserve ramoops reliably, use KHO with reserve_mem.
> > > > The first kernel will allocate memory, this memory will be preserved by KHO
> > > > and could be picked up by the second kernel.
> > > 
> > > ok, On ARM64 DTS systems, we can reserve ramoops memory in the device tree during
> > > the warm reboot.
> > 
> > The cc list actually implied x86 ;-)
> > Added arm64 folks now.
> 
> Thanks for adding ARM folks, I had just included whatever get_maintainer script
> suggested, sorry. 
> > > For an equivalent ARM64 ACPI platform, what is the recommended way to reserve
> > > and preserve that memory across the boots? 
> > 
> > I don't think it exists, but a command line option (be it memmap= or
> > reserve_mem=) does not seem the right way to me.
> > 
> > Most of the arguments that were made against adding memmap= to arm64 [1]
> > apply here.
> > 
> > If kexec is an option, KHO provides a reliable way to preserve memory
> > across boots.
> > 
> > If kexec is not an option, we should look for a generic way to specify
> > something like DT's reserved_mem for ACPI/EFI systems.
> > 
> > [1] https://lkml.kernel.org/lkml/20201118063314.22940-1-song.bao.hua@hisilicon.com/T/
> > 
> Well, kexec is one of the option for my use case, it also requires
> memory reservation during warm reboot, I think memory can be reserved in
> the firmware but this will create a dependency on firmware for Linux
> reservation.
> 
> It would be great to have a in kernel memory reservation mechanism for
> ARM64 ACPI platforms. I believe some other use cases like PMEM
> reservation would also benefit from this.
>

Following up on this, As Mike pointed that reserve_mem is best effort
reservation mechanism, so what is the recommended reliable Linux
mechanism, if any, to reserve a predetermined memory range during
early boot on ARM64/ACPI platforms for warm boot scenarios? KHO is
one option, but I'm specifically looking for a solution that preserves
the region across warm reboots.

Please let me know.

Thanks,
Shyam

Re: [RFC v2 PATCH] reserve_mem: add support for static memory

[Catalin Marinas]

Hi Shyam,

Adding Lorenzo, Hanjun, Sudeep.

On Tue, Jun 30, 2026 at 10:09:11AM -0700, Shyam Saini wrote:
> Following up on this, As Mike pointed that reserve_mem is best effort
> reservation mechanism, so what is the recommended reliable Linux
> mechanism, if any, to reserve a predetermined memory range during
> early boot on ARM64/ACPI platforms for warm boot scenarios? KHO is
> one option, but I'm specifically looking for a solution that preserves
> the region across warm reboots.

If you need ramoops-like support, I think ACPI ERST is a better fit and
we don't need any memory reservation tricks. I'll let the arm ACPI
maintainers comment further.

-- 
Catalin

Re: [RFC v2 PATCH] reserve_mem: add support for static memory

[Shyam Saini]

Hi Catalin,

On 08 Jul 2026 11:00, Catalin Marinas wrote:
> Hi Shyam,
> 
> Adding Lorenzo, Hanjun, Sudeep.
> 
> On Tue, Jun 30, 2026 at 10:09:11AM -0700, Shyam Saini wrote:
> > Following up on this, As Mike pointed that reserve_mem is best effort
> > reservation mechanism, so what is the recommended reliable Linux
> > mechanism, if any, to reserve a predetermined memory range during
> > early boot on ARM64/ACPI platforms for warm boot scenarios? KHO is
> > one option, but I'm specifically looking for a solution that preserves
> > the region across warm reboots.
> 
> If you need ramoops-like support, I think ACPI ERST is a better fit and
> we don't need any memory reservation tricks. I'll let the arm ACPI
> maintainers comment further.
> 

while this patch mentions use case of ramoops, there are other uses cases that
requires predetermined memory reservation at boot time, one of them is PMEM.
Like Mike pointed out in this [1] thread, we don't seem to have an
equivalent mechanism for reserving predetermined memory on ARM/ACPI
platforms, as is available on ARM/DTS based platforms.

Thanks,
Shyam

[1] https://lore.kernel.org/lkml/ak3v6dt_VIZTynT8@kernel.org/

Re: [RFC v2 PATCH] reserve_mem: add support for static memory

[Sudeep Holla]

On Wed, Jul 08, 2026 at 05:37:33PM -0700, Shyam Saini wrote:
> Hi Catalin,
> 
> On 08 Jul 2026 11:00, Catalin Marinas wrote:
> > Hi Shyam,
> > 
> > Adding Lorenzo, Hanjun, Sudeep.
> > 
> > On Tue, Jun 30, 2026 at 10:09:11AM -0700, Shyam Saini wrote:
> > > Following up on this, As Mike pointed that reserve_mem is best effort
> > > reservation mechanism, so what is the recommended reliable Linux
> > > mechanism, if any, to reserve a predetermined memory range during
> > > early boot on ARM64/ACPI platforms for warm boot scenarios? KHO is
> > > one option, but I'm specifically looking for a solution that preserves
> > > the region across warm reboots.
> > 
> > If you need ramoops-like support, I think ACPI ERST is a better fit and
> > we don't need any memory reservation tricks. I'll let the arm ACPI
> > maintainers comment further.
> > 
> 
> while this patch mentions use case of ramoops, there are other uses cases that
> requires predetermined memory reservation at boot time, one of them is PMEM.
> Like Mike pointed out in this [1] thread, we don't seem to have an
> equivalent mechanism for reserving predetermined memory on ARM/ACPI
> platforms, as is available on ARM/DTS based platforms.
> 

Indeed, it was been discussed in several forums in the past but AFAIK there
is no conclusion or a direction towards solving this yet.

-- 
Regards,
Sudeep