Skip to contentSkip to conversation

kernel-fitimage/uboot-sign: support 64bits address

PATCH In review Jamin Lin 3 participants 6 messages Apr 13, 2023 – May 4, 2023

lore
Share & tools
lore

Jamin Lin

author
#0 lore raw

cover kernel-fitimage/uboot-sign: support 64bits address

v1:
kernel-fitimage.bbclass: allow users to change address-cells to support 64bits address
uboot-sign.bbclass: allow users to change address-cells to support 64bits address.

Jamin Lin (2):
kernel-fitimage: support 64 bits address
uboot-sign: support 64bits address

 meta/classes-recipe/kernel-fitimage.bbclass | 6 +++++-
 meta/classes-recipe/uboot-sign.bbclass      | 6 +++++-
 2 files changed, 10 insertions(+), 2 deletions(-)
Signature · 1 line
2.17.1
author
#1 lore raw

PATCH 1/2 kernel-fitimage: support 64 bits address

The default value of address-cells is "1", so the generated "its" file only supports 32bits address for kernel FIT image.
However, some platforms may want to support 64bits address
of UBOOT_LOADADDRESS and UBOOT_ENTRYPOINT.
Therefore, adds a variable to support both 64bits and 32bits address cells.

By default, the address-cell is "1" which is used for 32bits load address. If users would like to use 64bits load address, users are able to set as following for "0x400000000" 64bits load address.
1. FIT_ADDRESS_CELLS = "2"
2. UBOOT_LOADADDRESS= "0x04 0x00000000"

(From OE-Core rev: f79046d082f3a7752e6780b2990470dcd36cbba4)

---
 meta/classes-recipe/kernel-fitimage.bbclass | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)
meta/classes-recipe/kernel-fitimage.bbclass +5 -1
index b77747404f..d294f28d17 100644
@@ -92,6 +92,10 @@ FIT_SUPPORTED_INITRAMFS_FSTYPES ?= "cpio.lz4 cpio.lzo cpio.lzma cpio.xz cpio.zst
 # Allow user to select the default DTB for FIT image when multiple dtb's exists.
 FIT_CONF_DEFAULT_DTB ?= ""
 
+# length of address in number of <u32> cells
+# ex: 1 32bits address, 2 64bits address
+FIT_ADDRESS_CELLS ?= "1"
+
 # Keys used to sign individually image nodes.
 # The keys to sign image nodes must be different from those used to sign
 # configuration nodes, otherwise the "required" property, from
@@ -110,7 +114,7 @@ fitimage_emit_fit_header() {
 
 / {
         description = "${FIT_DESC}";
-        #address-cells = <1>;
+        #address-cells = <${FIT_ADDRESS_CELLS}>;
 EOF
 }
 
Signature · 1 line
2.17.1
author
#4 lore raw

PATCH 2/2 uboot-sign: support 64bits address

In reply to Jamin Lin

The default value of address-cells is "1", so the generated "its" file only support 32bits address for uboot FIT image.
However, some platforms may want to support 64bits address of UBOOT_LOADADDRESS and UBOOT_ENTRYPOINT.
Therefore, add variables to support both 64bits and 32bits address.

By default, the address-cell is 1 which is used for 32bits load address. If users would like to use 64bits load address,
users are able to set as following for "0x400000000" 64bits load address.
1. FIT_ADDRESS_CELLS = "2"
2. UBOOT_LOADADDRESS= "0x04 0x00000000"

(From OE-Core rev: f79046d082f3a7752e6780b2990470dcd36cbba4)

---
 meta/classes-recipe/uboot-sign.bbclass | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)
meta/classes-recipe/uboot-sign.bbclass +5 -1
index 3065c54577..ad04c82378 100644
@@ -81,6 +81,10 @@ UBOOT_FIT_KEY_REQ_ARGS ?= "-batch -new"
 # Standard format for public key certificate
 UBOOT_FIT_KEY_SIGN_PKCS ?= "-x509"
 
+# length of address in number of <u32> cells
+# ex: 1 32bits address, 2 64bits address
+UBOOT_FIT_ADDRESS_CELLS ?= "1"
+
 # This is only necessary for determining the signing configuration
 KERNEL_PN = "${PREFERRED_PROVIDER_virtual/kernel}"
 
@@ -234,7 +238,7 @@ uboot_fitimage_assemble() {
 
 / {
     description = "${UBOOT_FIT_DESC}";
-    #address-cells = <1>;
+    #address-cells = <${UBOOT_FIT_ADDRESS_CELLS}>;
 
     images {
         uboot {
Signature · 1 line
2.17.1
#2 lore raw

Re: [OE-core] [PATCH 1/2] kernel-fitimage: support 64 bits address

In reply to Jamin Lin

Hello Jamin,

thank you for your contribution!

Quoted 6 lines · On Thu, 13 Apr 2023 18:31:08 +0800 "Jamin Lin" <[email protected]> wrote:

The default value of address-cells is "1", so the generated "its" file only supports 32bits address for kernel FIT image.
However, some platforms may want to support 64bits address
of UBOOT_LOADADDRESS and UBOOT_ENTRYPOINT.
Therefore, adds a variable to support both 64bits and 32bits address cells.

Quoted 7 lines

By default, the address-cell is "1" which is used for 32bits load address. If users would like to use 64bits load address, users are able to set as following for "0x400000000" 64bits load address.
1. FIT_ADDRESS_CELLS = "2"
2. UBOOT_LOADADDRESS= "0x04 0x00000000"

(From OE-Core rev: f79046d082f3a7752e6780b2990470dcd36cbba4)

This line should not be there. It looks like you generated your patches from a poky repository where this patch landed via combo-layers, but it just doesn't make sense as an oe-core patch. Also, that SHA-1 does not exist in the public oe-core repository.

I removed that line while applying the patch for testing, so you don't need to resend for this time, but please make sure you generate your patches from an oe-core repository in the future.

The same applies to patch 2/2.

Best regards,
Luca

Signature · 3 lines
Luca Ceresoli, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com
author
#3 lore raw

Re: [PATCH 1/2] kernel-fitimage: support 64 bits address

Thanks for your review and correct me.

I should create a patch based on an openembedded-core repository rather than a poky repository.

#5 lore raw
In reply to Jamin Lin
Quoted 6 lines · Jamin Lin

v1:
kernel-fitimage.bbclass: allow users to change address-cells to support 64bits address
uboot-sign.bbclass: allow users to change address-cells to support 64bits address.

Quoted 7 lines

Jamin Lin (2):
kernel-fitimage: support 64 bits address
uboot-sign: support 64bits address

 meta/classes-recipe/kernel-fitimage.bbclass | 6 +++++-
 meta/classes-recipe/uboot-sign.bbclass      | 6 +++++-
 2 files changed, 10 insertions(+), 2 deletions(-)

Since this adds new variables, could you send patches to the documentation please to document them?

I did also wonder if there was some way we could combine and share some of this fit image information since we appear to have two variables with different names encoding the same info here :/. Maybe that is something we could fix or improve in a later patch?

I'm kind of living in hope we can simplify some of these different places to share code ultimately.

Cheers,

Richard