[PATCH v2 0/2] docs: admin-guide: EFI Boot Stub

[PATCH v2 0/2] docs: admin-guide: EFI Boot Stub

[Heinrich Schuchardt]

Describe the passing of the device tree and the initial RAM disk to the
EFI Boot Stub.

Describe the handoff to the main Linux kernel. The links given here
provide the necessary information about structures like boot_params that
have to be filled by the EFI stub. So we don't have to describe them in
the admin-guide.

Heinrich Schuchardt (2):
  docs: admin-guide: fdt and initrd load in EFI stub
  docs: admin-guide: EFI stub to main Linux handoff

 Documentation/admin-guide/efi-stub.rst | 47 ++++++++++++++++++++++++++
 1 file changed, 47 insertions(+)

--
2.28.0


_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

[PATCH v2 1/2] docs: admin-guide: fdt and initrd load in EFI stub

[Heinrich Schuchardt]

Describe how a device tree and an initial RAM disk can be passed to the EFI
Boot Stub.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
---
v2:
	mention EFI_GENERIC_STUB_INITRD_CMDLINE_LOADER (thx Atish)
---
 Documentation/admin-guide/efi-stub.rst | 35 ++++++++++++++++++++++++++
 1 file changed, 35 insertions(+)

diff --git a/Documentation/admin-guide/efi-stub.rst b/Documentation/admin-guide/efi-stub.rst
index 833edb0d0bc4..4965dec48af4 100644
--- a/Documentation/admin-guide/efi-stub.rst
+++ b/Documentation/admin-guide/efi-stub.rst
@@ -38,6 +38,34 @@ arch/arm/boot/zImage should be copied to the system partition, and it
 may not need to be renamed. Similarly for arm64, arch/arm64/boot/Image
 should be copied but not necessarily renamed.

+Passing an initial RAM disk to the EFI Boot Stub
+------------------------------------------------
+
+The following means sorted by decreasing priority can be used to provide an
+initial RAM disk to the EFI Boot Stub:
+
+* The firmware may provide a UEFI Load File 2 Protocol. The stub will try to
+  load the RAM disk by calling the LoadFile() service of the protocol using
+  a vendor device path with the vendor GUID
+  5568e427-0x68fc-4f3d-ac74-ca555231cc68.
+* Next the EFI stub will try to load the file indicated by the "initrd=" command
+  line parameter if CONFIG_EFI_GENERIC_STUB_INITRD_CMDLINE_LOADER is enabled.
+* The prior boot stage may pass the location of the initial RAM disk via the
+  "linux,initrd-start" and "linux,initrd-end" properties of the "/chosen" node
+  of the device-tree.
+
+The first two items are inhibited by the "noinitrd" command line parameter.
+
+Passing a device-tree to the EFI Boot Stub
+------------------------------------------
+
+A device-tree can be passed to the EFI Boot Stub in decreasing priority using
+
+* command line option dtb=
+* a UEFI configuration table with GUID b1b621d5-f19c-41a5-830b-d9152c69aae0.
+
+The command line option is only available if CONFIG_EFI_ARMSTUB_DTB_LOADER=y
+and secure boot is disabled.

 Passing kernel parameters from the EFI shell
 --------------------------------------------
@@ -46,6 +74,10 @@ Arguments to the kernel can be passed after bzImage.efi, e.g.::

 	fs0:> bzImage.efi console=ttyS0 root=/dev/sda4

+The "noinitrd" option
+---------------------
+
+The "noinitrd" option stops the EFI stub from loading an initial RAM disk.

 The "initrd=" option
 --------------------
@@ -98,3 +130,6 @@ CONFIGURATION TABLE.

 "dtb=" is processed in the same manner as the "initrd=" option that is
 described above.
+
+This option is only available if CONFIG_EFI_ARMSTUB_DTB_LOADER=y and secure
+boot is disabled.
--
2.28.0


_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

[PATCH v2 2/2] docs: admin-guide: EFI stub to main Linux handoff

[Heinrich Schuchardt]

Describe the handoff from the EFI stub to the main Linux kernel.

Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
---
v2:
	new patch
---
 Documentation/admin-guide/efi-stub.rst | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/Documentation/admin-guide/efi-stub.rst b/Documentation/admin-guide/efi-stub.rst
index 4965dec48af4..669602ff4515 100644
--- a/Documentation/admin-guide/efi-stub.rst
+++ b/Documentation/admin-guide/efi-stub.rst
@@ -133,3 +133,15 @@ described above.

 This option is only available if CONFIG_EFI_ARMSTUB_DTB_LOADER=y and secure
 boot is disabled.
+
+Handoff to main Linux kernel
+----------------------------
+
+Currently the EFI stub enters the main Linux kernel via the same entry points
+that are used when booting without UEFI. This might change in future versions.
+For a description of these entry points see:
+
+* for arm :doc:`/arm/booting`
+* for arm64 :doc:`/arm64/booting`
+* for riscv :doc:`/riscv/index`
+* for x86 :doc:`/x86/boot`
--
2.28.0


_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

Re: [PATCH v2 1/2] docs: admin-guide: fdt and initrd load in EFI stub

[Atish Patra]

On Fri, Oct 2, 2020 at 11:38 PM Heinrich Schuchardt <xypron.glpk@gmx.de> wrote:
>
> Describe how a device tree and an initial RAM disk can be passed to the EFI
> Boot Stub.
>
> Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
> ---
> v2:
>         mention EFI_GENERIC_STUB_INITRD_CMDLINE_LOADER (thx Atish)
> ---
>  Documentation/admin-guide/efi-stub.rst | 35 ++++++++++++++++++++++++++
>  1 file changed, 35 insertions(+)
>
> diff --git a/Documentation/admin-guide/efi-stub.rst b/Documentation/admin-guide/efi-stub.rst
> index 833edb0d0bc4..4965dec48af4 100644
> --- a/Documentation/admin-guide/efi-stub.rst
> +++ b/Documentation/admin-guide/efi-stub.rst
> @@ -38,6 +38,34 @@ arch/arm/boot/zImage should be copied to the system partition, and it
>  may not need to be renamed. Similarly for arm64, arch/arm64/boot/Image
>  should be copied but not necessarily renamed.
>
> +Passing an initial RAM disk to the EFI Boot Stub
> +------------------------------------------------
> +
> +The following means sorted by decreasing priority can be used to provide an
> +initial RAM disk to the EFI Boot Stub:
> +
> +* The firmware may provide a UEFI Load File 2 Protocol. The stub will try to
> +  load the RAM disk by calling the LoadFile() service of the protocol using
> +  a vendor device path with the vendor GUID
> +  5568e427-0x68fc-4f3d-ac74-ca555231cc68.
> +* Next the EFI stub will try to load the file indicated by the "initrd=" command
> +  line parameter if CONFIG_EFI_GENERIC_STUB_INITRD_CMDLINE_LOADER is enabled.
> +* The prior boot stage may pass the location of the initial RAM disk via the
> +  "linux,initrd-start" and "linux,initrd-end" properties of the "/chosen" node
> +  of the device-tree.
> +

Should we also specify which method is enabled by default for which
ARCH and recommended methods?

For example, It's recommended to use the LoadFile method for RISC-V
and new ARM systems.
Existing ARM ones will continue to use the initrd argument as that's
the method enabled by default.

> +The first two items are inhibited by the "noinitrd" command line parameter.
> +
> +Passing a device-tree to the EFI Boot Stub
> +------------------------------------------
> +
> +A device-tree can be passed to the EFI Boot Stub in decreasing priority using
> +
> +* command line option dtb=
> +* a UEFI configuration table with GUID b1b621d5-f19c-41a5-830b-d9152c69aae0.
> +

I am just curious. Is there any specific reason why efistub tries to
load the dtb from the command line first
and loads from the config table only if it fails from the first approach ?

> +The command line option is only available if CONFIG_EFI_ARMSTUB_DTB_LOADER=y
> +and secure boot is disabled.
>
>  Passing kernel parameters from the EFI shell
>  --------------------------------------------
> @@ -46,6 +74,10 @@ Arguments to the kernel can be passed after bzImage.efi, e.g.::
>
>         fs0:> bzImage.efi console=ttyS0 root=/dev/sda4
>
> +The "noinitrd" option
> +---------------------
> +
> +The "noinitrd" option stops the EFI stub from loading an initial RAM disk.
>
>  The "initrd=" option
>  --------------------
> @@ -98,3 +130,6 @@ CONFIGURATION TABLE.
>
>  "dtb=" is processed in the same manner as the "initrd=" option that is
>  described above.
> +
> +This option is only available if CONFIG_EFI_ARMSTUB_DTB_LOADER=y and secure
> +boot is disabled.
> --
> 2.28.0
>
>
> _______________________________________________
> linux-riscv mailing list
> linux-riscv@lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-riscv



-- 
Regards,
Atish

_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

Re: [PATCH v2 1/2] docs: admin-guide: fdt and initrd load in EFI stub

[Heinrich Schuchardt]

On 03.10.20 09:34, Atish Patra wrote:
> On Fri, Oct 2, 2020 at 11:38 PM Heinrich Schuchardt <xypron.glpk@gmx.de> wrote:
>>
>> Describe how a device tree and an initial RAM disk can be passed to the EFI
>> Boot Stub.
>>
>> Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
>> ---
>> v2:
>>         mention EFI_GENERIC_STUB_INITRD_CMDLINE_LOADER (thx Atish)
>> ---
>>  Documentation/admin-guide/efi-stub.rst | 35 ++++++++++++++++++++++++++
>>  1 file changed, 35 insertions(+)
>>
>> diff --git a/Documentation/admin-guide/efi-stub.rst b/Documentation/admin-guide/efi-stub.rst
>> index 833edb0d0bc4..4965dec48af4 100644
>> --- a/Documentation/admin-guide/efi-stub.rst
>> +++ b/Documentation/admin-guide/efi-stub.rst
>> @@ -38,6 +38,34 @@ arch/arm/boot/zImage should be copied to the system partition, and it
>>  may not need to be renamed. Similarly for arm64, arch/arm64/boot/Image
>>  should be copied but not necessarily renamed.
>>
>> +Passing an initial RAM disk to the EFI Boot Stub
>> +------------------------------------------------
>> +
>> +The following means sorted by decreasing priority can be used to provide an
>> +initial RAM disk to the EFI Boot Stub:
>> +
>> +* The firmware may provide a UEFI Load File 2 Protocol. The stub will try to
>> +  load the RAM disk by calling the LoadFile() service of the protocol using
>> +  a vendor device path with the vendor GUID
>> +  5568e427-0x68fc-4f3d-ac74-ca555231cc68.
>> +* Next the EFI stub will try to load the file indicated by the "initrd=" command
>> +  line parameter if CONFIG_EFI_GENERIC_STUB_INITRD_CMDLINE_LOADER is enabled.
>> +* The prior boot stage may pass the location of the initial RAM disk via the
>> +  "linux,initrd-start" and "linux,initrd-end" properties of the "/chosen" node
>> +  of the device-tree.
>> +
>
> Should we also specify which method is enabled by default for which
> ARCH and recommended methods?

The user relevant configuration is not the Linux' defconfig but what the
distribution maintainer has baked. I doubt mentioning Linux' defaults is
meaningful here.

>
> For example, It's recommended to use the LoadFile method for RISC-V
> and new ARM systems.

GRUB does not implement the LoadFile2 protocol yet. In U-Boot it is only
good for testing. I am not aware of usability with unmodified EDK II.
Why should we recommend anything before building the ecosystem that
makes it useful?

What is best may depend on the use case. There is nothing insecure in
passing the initrd via "linux,initrd-start" and "linux,initrd-end" if
you control the load options.

The EBBR (https://github.com/arm-software/ebbr) might be a better place
for a recommendation.

> Existing ARM ones will continue to use the initrd argument as that's
> the method enabled by default.

Only if if the LoadFile2 protocol is not available because that has a
higher priority for ARM, x86, and RISC-V.

Should I consider my i.mx6 Wandboard Quad bought in 2013 "old" while it
is running the U-Boot v2020.10-rc5, Linux v5.9-rc7, and Debian testing?
A distinction between "old" and "new" systems seems irrelevant here. All
are treated equal by the EFI stub.

>
>> +The first two items are inhibited by the "noinitrd" command line parameter.
>> +
>> +Passing a device-tree to the EFI Boot Stub
>> +------------------------------------------
>> +
>> +A device-tree can be passed to the EFI Boot Stub in decreasing priority using
>> +
>> +* command line option dtb=
>> +* a UEFI configuration table with GUID b1b621d5-f19c-41a5-830b-d9152c69aae0.
>> +
>
> I am just curious. Is there any specific reason why efistub tries to
> load the dtb from the command line first
> and loads from the config table only if it fails from the first approach ?

As we disable dtb= in secure boot it would make sense to turn the
priorities around for non-secure boot too.

But this is beyond the scope of a documentation patch.

Best regards

Heinrich

>
>> +The command line option is only available if CONFIG_EFI_ARMSTUB_DTB_LOADER=y
>> +and secure boot is disabled.
>>
>>  Passing kernel parameters from the EFI shell
>>  --------------------------------------------
>> @@ -46,6 +74,10 @@ Arguments to the kernel can be passed after bzImage.efi, e.g.::
>>
>>         fs0:> bzImage.efi console=ttyS0 root=/dev/sda4
>>
>> +The "noinitrd" option
>> +---------------------
>> +
>> +The "noinitrd" option stops the EFI stub from loading an initial RAM disk.
>>
>>  The "initrd=" option
>>  --------------------
>> @@ -98,3 +130,6 @@ CONFIGURATION TABLE.
>>
>>  "dtb=" is processed in the same manner as the "initrd=" option that is
>>  described above.
>> +
>> +This option is only available if CONFIG_EFI_ARMSTUB_DTB_LOADER=y and secure
>> +boot is disabled.
>> --
>> 2.28.0
>>
>>
>> _______________________________________________
>> linux-riscv mailing list
>> linux-riscv@lists.infradead.org
>> http://lists.infradead.org/mailman/listinfo/linux-riscv
>
>
>


_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

Re: [PATCH v2 1/2] docs: admin-guide: fdt and initrd load in EFI stub

[Atish Patra]

On Sat, Oct 3, 2020 at 1:29 AM Heinrich Schuchardt <xypron.glpk@gmx.de> wrote:
>
> On 03.10.20 09:34, Atish Patra wrote:
> > On Fri, Oct 2, 2020 at 11:38 PM Heinrich Schuchardt <xypron.glpk@gmx.de> wrote:
> >>
> >> Describe how a device tree and an initial RAM disk can be passed to the EFI
> >> Boot Stub.
> >>
> >> Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
> >> ---
> >> v2:
> >>         mention EFI_GENERIC_STUB_INITRD_CMDLINE_LOADER (thx Atish)
> >> ---
> >>  Documentation/admin-guide/efi-stub.rst | 35 ++++++++++++++++++++++++++
> >>  1 file changed, 35 insertions(+)
> >>
> >> diff --git a/Documentation/admin-guide/efi-stub.rst b/Documentation/admin-guide/efi-stub.rst
> >> index 833edb0d0bc4..4965dec48af4 100644
> >> --- a/Documentation/admin-guide/efi-stub.rst
> >> +++ b/Documentation/admin-guide/efi-stub.rst
> >> @@ -38,6 +38,34 @@ arch/arm/boot/zImage should be copied to the system partition, and it
> >>  may not need to be renamed. Similarly for arm64, arch/arm64/boot/Image
> >>  should be copied but not necessarily renamed.
> >>
> >> +Passing an initial RAM disk to the EFI Boot Stub
> >> +------------------------------------------------
> >> +
> >> +The following means sorted by decreasing priority can be used to provide an
> >> +initial RAM disk to the EFI Boot Stub:
> >> +
> >> +* The firmware may provide a UEFI Load File 2 Protocol. The stub will try to
> >> +  load the RAM disk by calling the LoadFile() service of the protocol using
> >> +  a vendor device path with the vendor GUID
> >> +  5568e427-0x68fc-4f3d-ac74-ca555231cc68.
> >> +* Next the EFI stub will try to load the file indicated by the "initrd=" command
> >> +  line parameter if CONFIG_EFI_GENERIC_STUB_INITRD_CMDLINE_LOADER is enabled.
> >> +* The prior boot stage may pass the location of the initial RAM disk via the
> >> +  "linux,initrd-start" and "linux,initrd-end" properties of the "/chosen" node
> >> +  of the device-tree.
> >> +
> >
> > Should we also specify which method is enabled by default for which
> > ARCH and recommended methods?
>
> The user relevant configuration is not the Linux' defconfig but what the
> distribution maintainer has baked. I doubt mentioning Linux' defaults is
> meaningful here.
>

Yes. But some distribution admin may think that one of these two configs
(initrd or dtb) are not enabled for RISC-V by mistake and enable it in
the distro config.
Ard had suggested that it is best if RISC-V doesn't inherit the legacy options.

> >
> > For example, It's recommended to use the LoadFile method for RISC-V
> > and new ARM systems.
>
> GRUB does not implement the LoadFile2 protocol yet. In U-Boot it is only
> good for testing. I am not aware of usability with unmodified EDK II.
> Why should we recommend anything before building the ecosystem that
> makes it useful?
>
> What is best may depend on the use case. There is nothing insecure in
> passing the initrd via "linux,initrd-start" and "linux,initrd-end" if
> you control the load options.
>
> The EBBR (https://github.com/arm-software/ebbr) might be a better place
> for a recommendation.
>

Agreed.

> > Existing ARM ones will continue to use the initrd argument as that's
> > the method enabled by default.
>
> Only if if the LoadFile2 protocol is not available because that has a
> higher priority for ARM, x86, and RISC-V.
>
> Should I consider my i.mx6 Wandboard Quad bought in 2013 "old" while it
> is running the U-Boot v2020.10-rc5, Linux v5.9-rc7, and Debian testing?
> A distinction between "old" and "new" systems seems irrelevant here. All
> are treated equal by the EFI stub.
>
> >
> >> +The first two items are inhibited by the "noinitrd" command line parameter.
> >> +
> >> +Passing a device-tree to the EFI Boot Stub
> >> +------------------------------------------
> >> +
> >> +A device-tree can be passed to the EFI Boot Stub in decreasing priority using
> >> +
> >> +* command line option dtb=
> >> +* a UEFI configuration table with GUID b1b621d5-f19c-41a5-830b-d9152c69aae0.
> >> +
> >
> > I am just curious. Is there any specific reason why efistub tries to
> > load the dtb from the command line first
> > and loads from the config table only if it fails from the first approach ?
>
> As we disable dtb= in secure boot it would make sense to turn the
> priorities around for non-secure boot too.
>
> But this is beyond the scope of a documentation patch.
>

Yes. I was just using the context to ask the question. I will send a
separate patch
for that.

> Best regards
>
> Heinrich
>
> >
> >> +The command line option is only available if CONFIG_EFI_ARMSTUB_DTB_LOADER=y
> >> +and secure boot is disabled.
> >>
> >>  Passing kernel parameters from the EFI shell
> >>  --------------------------------------------
> >> @@ -46,6 +74,10 @@ Arguments to the kernel can be passed after bzImage.efi, e.g.::
> >>
> >>         fs0:> bzImage.efi console=ttyS0 root=/dev/sda4
> >>
> >> +The "noinitrd" option
> >> +---------------------
> >> +
> >> +The "noinitrd" option stops the EFI stub from loading an initial RAM disk.
> >>
> >>  The "initrd=" option
> >>  --------------------
> >> @@ -98,3 +130,6 @@ CONFIGURATION TABLE.
> >>
> >>  "dtb=" is processed in the same manner as the "initrd=" option that is
> >>  described above.
> >> +
> >> +This option is only available if CONFIG_EFI_ARMSTUB_DTB_LOADER=y and secure
> >> +boot is disabled.
> >> --
> >> 2.28.0
> >>
> >>
> >> _______________________________________________
> >> linux-riscv mailing list
> >> linux-riscv@lists.infradead.org
> >> http://lists.infradead.org/mailman/listinfo/linux-riscv
> >
> >
> >
>


-- 
Regards,
Atish

_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv