doc: environment: Drop u-boot_addr_r
PATCH 1/3 doc: environment: Drop u-boot_addr_r
This variable is never set nor explained why it would be set, drop it.
--- doc/usage/environment.rst | 1 - 1 file changed, 1 deletion(-)
doc/usage/environment.rst +0 -1
@@ -390,7 +390,6 @@ in U-Boot code. ================= ============== ================ ============== Image File Name RAM Address Flash Location ================= ============== ================ ============== -u-boot u-boot u-boot_addr_r u-boot_addr Linux kernel bootfile kernel_addr_r kernel_addr device tree blob fdtfile fdt_addr_r fdt_addr ramdisk ramdiskfile ramdisk_addr_r ramdisk_addr
Signature · 1 line
2.25.1
PATCH 2/3 doc: environment: Expand on fdt_addr, initrd_addr and loadaddr
- Explain why fdt_addr and initrd_addr should not be set to disable relocation normally.
- Provide some advice on the typical loadaddr default value.
--- doc/usage/environment.rst | 15 ++++++++++++--- 1 file changed, 12 insertions(+), 3 deletions(-)
doc/usage/environment.rst +12 -3
@@ -204,7 +204,9 @@ fdt_high to work it must reside in writable memory, have sufficient padding on the end of it for u-boot to add the information it needs into it, and the memory - must be accessible by the kernel. + must be accessible by the kernel. This usage is strongly discouraged + however as it also stops U-Boot from ensuring the device tree starting + address is properly aligned and a misaligned tree will cause OS failures. fdtcontroladdr if set this is the address of the control flattened
@@ -240,14 +242,21 @@ initrd_high memory. In this case U-Boot will NOT COPY the ramdisk at all. This may be useful to reduce the boot time on your system, but requires that this - feature is supported by your Linux kernel. + feature is supported by your Linux kernel. This usage however requires + that the user ensure that there will be no overlap with other parts of the + iamge such as the Linux kernel BSS. It should not be enabled by default + and only done as part of optimizing a deployment. ipaddr IP address; needed for tftpboot command loadaddr Default load address for commands like "bootp", - "rarpboot", "tftpboot", "loadb" or "diskboot" + "rarpboot", "tftpboot", "loadb" or "diskboot". Note that the optimal + default values here will vary between architectures. On 32bit ARM for + example, some offset from start of memory is used as the Linux kernel + zImage has a self decompressor and it's best if we stay out of where that + will be working. loads_echo see CONFIG_LOADS_ECHO
Signature · 1 line
2.25.1
Re: [PATCH 2/3] doc: environment: Expand on fdt_addr, initrd_addr and loadaddr
Quoted 9 lines · Tom Rini
- Explain why
fdt_addrandinitrd_addrshould not be set to disable relocation normally.
- Provide some advice on the typical loadaddr default value.Signed-off-by Tom Rini <[email protected]>--- doc/usage/environment.rst | 15 ++++++++++++--- 1 file changed, 12 insertions(+), 3 deletions(-)
Re: [PATCH 2/3] doc: environment: Expand on fdt_addr, initrd_addr and loadaddr
In reply to Tom RiniQuoted 14 lines · Tom Rini
- Explain why
fdt_addrandinitrd_addrshould not be set to disable relocation normally.
- Provide some advice on the typical loadaddr default value.Signed-off-by Tom Rini <[email protected]>--- doc/usage/environment.rst | 15 ++++++++++++--- 1 file changed, 12 insertions(+), 3 deletions(-)doc/usage/environment.rst +0
-0
doc/usage/environment.rst fdt_high · line 209
Quoted 7 lines
doc/usage/environment.rst +3
-1@@ -204,7 +204,9 @@ fdt_high to work it must reside in writable memory, have sufficient padding on the end of it for u-boot to add the information it needs into it, and the memory - must be accessible by the kernel. + must be accessible by the kernel. This usage is strongly discouraged + however as it also stops U-Boot from ensuring the device tree starting + address is properly aligned and a misaligned tree will cause OS failures.
fdt_addr_r is typically 4 byte aligned. Is there a bug in some part of the code?
doc/usage/environment.rst initrd_high · line 246
Quoted 10 lines
doc/usage/environment.rst +2
-1fdtcontroladdr if set this is the address of the control flattened@@ -240,14 +242,21 @@ initrd_high memory. In this case U-Boot will NOT COPY the ramdisk at all. This may be useful to reduce the boot time on your system, but requires that this - feature is supported by your Linux kernel. + feature is supported by your Linux kernel. This usage however requires + that the user ensure that there will be no overlap with other parts of the
It is not the user but the developer who define $kernel_addr_r, $initr_addr_r, and $fdt_addr_r.
Relocating initrd does not stop the user from loading the kernel to initrd_high. The user is always responsible for avoiding overlap. In this regard there is nothing special about initrd_high=~0UL.
doc/usage/environment.rst initrd_high · line 247
Quoted 1 line
doc/usage/environment.rst +1
-0+ iamge such as the Linux kernel BSS. It should not be enabled by default
%s/iamge/image
doc/usage/environment.rst initrd_high · line 248
Quoted 1 line
doc/usage/environment.rst +1
-0+ and only done as part of optimizing a deployment.
Relocating initrd and fdt is using boot time. On many boards it is simply superfluous. On these boards I cannot see any reason not to disable it by default.
Best regards
Heinrich
Quoted 15 lines
doc/usage/environment.rst +5
-1ipaddr IP address; needed for tftpboot command loadaddr Default load address for commands like "bootp", - "rarpboot", "tftpboot", "loadb" or "diskboot" + "rarpboot", "tftpboot", "loadb" or "diskboot". Note that the optimal + default values here will vary between architectures. On 32bit ARM for + example, some offset from start of memory is used as the Linux kernel + zImage has a self decompressor and it's best if we stay out of where that + will be working. loads_echo see CONFIG_LOADS_ECHO
Re: [PATCH 2/3] doc: environment: Expand on fdt_addr, initrd_addr and loadaddr
Quoted 10 lines · Heinrich Schuchardt
Quoted 9 lines · Tom Rini
- Explain why
fdt_addrandinitrd_addrshould not be set to disable relocation normally.
- Provide some advice on the typical loadaddr default value.Signed-off-by Tom Rini <[email protected]>--- doc/usage/environment.rst | 15 ++++++++++++--- 1 file changed, 12 insertions(+), 3 deletions(-)
Quoted 15 lines
doc/usage/environment.rst fdt_high · line 209
Quoted 12 lines
doc/usage/environment.rst +3
-1@@ -204,7 +204,9 @@ fdt_high to work it must reside in writable memory, have sufficient padding on the end of it for u-boot to add the information it needs into it, and the memory - must be accessible by the kernel. + must be accessible by the kernel. This usage is strongly discouraged + however as it also stops U-Boot from ensuring the device tree starting + address is properly aligned and a misaligned tree will cause OS failures.
fdt_addr_ris typically 4 byte aligned. Is there a bug in some part of the code?
It must be 8 byte aligned. And we don't know the alignment inside of FIT images. Do not disable device tree relocation except under the most calculated of circumstances.
Quoted 4 lines
Quoted 4 lines
doc/usage/environment.rst +0
-0fdtcontroladdr if set this is the address of the control flattened@@ -240,14 +242,21 @@ initrd_high memory. In this case U-Boot will NOT COPY the
Quoted 8 lines
doc/usage/environment.rst initrd_high · line 246
Quoted 5 lines
doc/usage/environment.rst +2
-1ramdisk at all. This may be useful to reduce the boot time on your system, but requires that this - feature is supported by your Linux kernel. + feature is supported by your Linux kernel. This usage however requires + that the user ensure that there will be no overlap with other parts of theIt is not the user but the developer who define $kernel_addr_r, $initr_addr_r, and $fdt_addr_r.
No, it can be either. See stackoverflow and assorted blog posts. Person who picked up an SBC and just wants the thing to work will find various posts saying to do ... something ... here.
Quoted 3 lines
Relocating initrd does not stop the user from loading the kernel to
initrd_high. The user is always responsible for avoiding overlap. In this regard there is nothing special about initrd_high=~0UL.
And we're responsible for providing reasonable defaults. Do not disable initrd relocation by default as it gains little and can break the OS.
Quoted 3 lines
doc/usage/environment.rst initrd_high · line 247
Quoted 1 line
doc/usage/environment.rst +1
-0+ iamge such as the Linux kernel BSS. It should not be enabled by default%s/iamge/image
Please fix when applying.
Quoted 5 lines
doc/usage/environment.rst initrd_high · line 248
Quoted 1 line
doc/usage/environment.rst +1
-0+ and only done as part of optimizing a deployment.Relocating initrd and fdt is using boot time. On many boards it is simply superfluous. On these boards I cannot see any reason not to disable it by default.
The way to avoid "superfluous" time here is to set reasonable default locations, not disabling relocation.
Please note this is what the policy is, not something new.
Signature · 1 line
Tom
PATCH 3/3 doc: environment: Further expand on Image locations and provide example
In reply to Tom RiniStart by elaborating on what some of our constraints tend to be with image location values, and document where these external constraints come from. Provide a new subsection, an example based on the TI ARMv7 OMAP2PLUS families of chips, that gives sample values and explains why we use these particular values. This is based on what is in include/configs/ti_armv7_common.h as of fb3ad9bd923d ("TI: Add, use a DEFAULT_LINUX_BOOT_ENV environment string") as this contains just the values referenced in this document now and not some of the further additions that are less generic.
--- doc/usage/environment.rst | 39 +++++++++++++++++++++++++++++++++++++++ 1 file changed, 39 insertions(+)
doc/usage/environment.rst +39 -0
@@ -404,6 +404,42 @@ device tree blob fdtfile fdt_addr_r fdt_addr ramdisk ramdiskfile ramdisk_addr_r ramdisk_addr ================= ============== ================ ============== +When setting the RAM addresses for `kernel_addr_r`, `fdt_addr_r` and +`ramdisk_addr_r` there are several constraints to keep in mind. When booting +Linux, the `Booting ARM Linux`_ and `Booting AArch64 Linux`_ documents lay out +the requirements for booting all ARM platforms, including both alignment and +where within memory various things must be. These guidelines tend to also be +correct for other OSes and unless specifically contradicted by documentation +specific to another architecture, are good rules to follow for other +architectures as well. + +Example Image locations +^^^^^^^^^^^^^^^^^^^^^^^ + +If we take the Texas Instruments OMAP2PLUS family of ARMv7 processors as an +example for the above listed variables, we would do:: + + loadaddr=0x82000000 + kernel_addr_r=${loadaddr} + fdt_addr_r=0x88000000 + ramdisk_addr_r=0x88080000 + bootm_size=0x10000000 + +To explain this, we start by noting that DRAM starts at 0x80000000. A 32MiB +buffer from the start of memory as our default load address, and so where the +kernel would also be loaded to. This will hopefully allow for us to have the +whole of the compressed kernel image exist in memory above where the whole of +the decompressed kernel image will be, and allow for a quicker boot. Next, we +say that the device tree will be placed at 128MiB offset from the start of +memory. This is suggested by the kernel documment as it is exceedingly +unlikely to be overwritten by the kernel itself given other architectural +constraints. We then allow for the device tree to be up to 512KiB in size +before placing the ramdisk in memory. We then say that everything should be +within the first 256MiB of memory so that U-Boot can relocate things as needed +to ensure proper alignment. We pick 256MiB as our value here because we know +there are very few platforms on in this family with less memory. It could be +as high as 768MiB and still ensure that everything would be visible to the +kernel, but again we go with what we assume is the safest assumption. Automatically updated variables -------------------------------
@@ -472,3 +508,6 @@ Implementation -------------- See :doc:`../develop/environment` for internal development details. + +.. _`Booting ARM Linux`: https://www.kernel.org/doc/html/latest/arm/booting.html +.. _`Booting AArch64 Linux`: https://www.kernel.org/doc/html/latest/arm64/booting.html
Signature · 1 line
2.25.1
Re: [PATCH 3/3] doc: environment: Further expand on Image locations and provide example
Quoted 11 lines · Tom Rini
Start by elaborating on what some of our constraints tend to be with image location values, and document where these external constraints come from. Provide a new subsection, an example based on the TI ARMv7 OMAP2PLUS families of chips, that gives sample values and explains why we use these particular values. This is based on what is in
include/configs/ti_armv7_common.has offb3ad9bd923d("TI: Add, use aDEFAULT_LINUX_BOOT_ENVenvironment string") as this contains just the values referenced in this document now and not some of the further additions that are less generic.
Quoted 4 lines
Signed-off-by Tom Rini <[email protected]>--- doc/usage/environment.rst | 39 +++++++++++++++++++++++++++++++++++++++ 1 file changed, 39 insertions(+)
Quoted 21 lines
doc/usage/environment.rst +12
-0@@ -404,6 +404,42 @@ device tree blob fdtfile fdt_addr_r fdt_addr ramdisk ramdiskfile ramdisk_addr_r ramdisk_addr ================= ============== ================ ============== +When setting the RAM addresses for `kernel_addr_r`, `fdt_addr_r` and +`ramdisk_addr_r` there are several constraints to keep in mind. When booting +Linux, the `Booting ARM Linux`_ and `Booting AArch64 Linux`_ documents lay out +the requirements for booting all ARM platforms, including both alignment and +where within memory various things must be. These guidelines tend to also be +correct for other OSes and unless specifically contradicted by documentation +specific to another architecture, are good rules to follow for other +architectures as well. + +Example Image locations +^^^^^^^^^^^^^^^^^^^^^^^ +
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 428
Quoted 10 lines
doc/usage/environment.rst +10
-0+If we take the Texas Instruments OMAP2PLUS family of ARMv7 processors as an +example for the above listed variables, we would do:: + + loadaddr=0x82000000 + kernel_addr_r=${loadaddr} + fdt_addr_r=0x88000000 + ramdisk_addr_r=0x88080000 + bootm_size=0x10000000 + +To explain this, we start by noting that DRAM starts at 0x80000000. A 32MiB
Should it say 'We use a 32MiB' ?
Quoted 27 lines
doc/usage/environment.rst +17
-1+buffer from the start of memory as our default load address, and so where the +kernel would also be loaded to. This will hopefully allow for us to have the +whole of the compressed kernel image exist in memory above where the whole of +the decompressed kernel image will be, and allow for a quicker boot. Next, we +say that the device tree will be placed at 128MiB offset from the start of +memory. This is suggested by the kernel documment as it is exceedingly +unlikely to be overwritten by the kernel itself given other architectural +constraints. We then allow for the device tree to be up to 512KiB in size +before placing the ramdisk in memory. We then say that everything should be +within the first 256MiB of memory so that U-Boot can relocate things as needed +to ensure proper alignment. We pick 256MiB as our value here because we know +there are very few platforms on in this family with less memory. It could be +as high as 768MiB and still ensure that everything would be visible to the +kernel, but again we go with what we assume is the safest assumption. Automatically updated variables -------------------------------@@ -472,3 +508,6 @@ Implementation -------------- See :doc:`../develop/environment` for internal development details. + +.. _`Booting ARM Linux`: https://www.kernel.org/doc/html/latest/arm/booting.html +.. _`Booting AArch64 Linux`: https://www.kernel.org/doc/html/latest/arm64/booting.html --2.25.1
Re: [PATCH 3/3] doc: environment: Further expand on Image locations and provide example
Quoted 12 lines · Simon Glass
Quoted 11 lines · Tom Rini
Start by elaborating on what some of our constraints tend to be with image location values, and document where these external constraints come from. Provide a new subsection, an example based on the TI ARMv7 OMAP2PLUS families of chips, that gives sample values and explains why we use these particular values. This is based on what is in
include/configs/ti_armv7_common.has offb3ad9bd923d("TI: Add, use aDEFAULT_LINUX_BOOT_ENVenvironment string") as this contains just the values referenced in this document now and not some of the further additions that are less generic.
Quoted 6 lines
Quoted 4 lines
Signed-off-by Tom Rini <[email protected]>--- doc/usage/environment.rst | 39 +++++++++++++++++++++++++++++++++++++++ 1 file changed, 39 insertions(+)Reviewed-by Simon Glass <[email protected]>
Below you want a change?
Quoted 10 lines
Quoted 10 lines
doc/usage/environment.rst +0
-0@@ -404,6 +404,42 @@ device tree blob fdtfile fdt_addr_r fdt_addr ramdisk ramdiskfile ramdisk_addr_r ramdisk_addr ================= ============== ================ ==============
Quoted 6 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 412
Quoted 6 lines
doc/usage/environment.rst +6
-0+When setting the RAM addresses for `kernel_addr_r`, `fdt_addr_r` and +`ramdisk_addr_r` there are several constraints to keep in mind. When booting +Linux, the `Booting ARM Linux`_ and `Booting AArch64 Linux`_ documents lay out +the requirements for booting all ARM platforms, including both alignment and +where within memory various things must be. These guidelines tend to also be +correct for other OSes and unless specifically contradicted by documentation
What makes you think that BSD or Haiku have the same constraints as Linux?
Quoted 2 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 414
Quoted 2 lines
doc/usage/environment.rst +2
-0+specific to another architecture, are good rules to follow for other +architectures as well.
No. RISC-V does not have the same requirements as ARM. E.g. the initrd can be located anywhere in memory.
Quoted 3 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 417
Quoted 3 lines
doc/usage/environment.rst +3
-0+ +Example Image locations +^^^^^^^^^^^^^^^^^^^^^^^
You seem not to refer to a file 'Image'.
%s/Image/image/
Quoted 3 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 420
Quoted 3 lines
doc/usage/environment.rst +3
-0+ +If we take the Texas Instruments OMAP2PLUS family of ARMv7 processors as an +example for the above listed variables, we would do::
%s/we would do/we chose/ ?
Quoted 10 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 428
Quoted 8 lines
doc/usage/environment.rst +8
-0+ + loadaddr=0x82000000 + kernel_addr_r=${loadaddr} + fdt_addr_r=0x88000000 + ramdisk_addr_r=0x88080000 + bootm_size=0x10000000 + +To explain this, we start by noting that DRAM starts at 0x80000000. A 32MiBShould it say 'We use a 32MiB' ?
Please, mention that MAX_UNCOMP_KERNEL_SIZE = 32 MiB is ARMv7 specific.
Quoted 4 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 430
Quoted 4 lines
doc/usage/environment.rst +2
-0+buffer from the start of memory as our default load address, and so where the +kernel would also be loaded to. This will hopefully allow for us to have the
%s/allow for us/allow/
Quoted 2 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 432
Quoted 2 lines
doc/usage/environment.rst +2
-0+whole of the compressed kernel image exist in memory above where the whole of +the decompressed kernel image will be, and allow for a quicker boot. Next, we
Please, mention that decompressor code otherwise will have to relocate the compressed kernel.
Quoted 1 line
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 433
Quoted 1 line
doc/usage/environment.rst +1
-0+say that the device tree will be placed at 128MiB offset from the start of
Please, mention that initrd must be
* within 512 MiB (0x20000000) of the memory start on arm (which restricts initrd_high)
* in a a 1 GB aligned region of size '1UL << (VA_BITS_MIN - 1)' that includes the kernel on arm64
On RISC-V such a limitation does not exist.
Quoted 1 line
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 434
Quoted 1 line
doc/usage/environment.rst +1
-0+memory. This is suggested by the kernel documment as it is exceedingly
%s/documment/document
Quoted 5 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 439
Quoted 5 lines
doc/usage/environment.rst +5
-0+unlikely to be overwritten by the kernel itself given other architectural +constraints. We then allow for the device tree to be up to 512KiB in size +before placing the ramdisk in memory. We then say that everything should be +within the first 256MiB of memory so that U-Boot can relocate things as needed +to ensure proper alignment. We pick 256MiB as our value here because we know
If the load address ranges occupy the first 256 MiB, where will be the space for relocation on a 256 MiB board? Without mentioning initrd_high this paragraph is incomplete.
Quoted 3 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 442
Quoted 3 lines
doc/usage/environment.rst +3
-0+there are very few platforms on in this family with less memory. It could be +as high as 768MiB and still ensure that everything would be visible to the +kernel, but again we go with what we assume is the safest assumption.
A section per architecture clearly naming the architecture specific restrictions of Linux would be much more helpful.
Best regards
Heinrich
Quoted 13 lines
Quoted 13 lines
doc/usage/environment.rst +3
-1Automatically updated variables -------------------------------@@ -472,3 +508,6 @@ Implementation -------------- See :doc:`../develop/environment` for internal development details. + +.. _`Booting ARM Linux`: https://www.kernel.org/doc/html/latest/arm/booting.html +.. _`Booting AArch64 Linux`: https://www.kernel.org/doc/html/latest/arm64/booting.html --2.25.1
Re: [PATCH 3/3] doc: environment: Further expand on Image locations and provide example
Quoted 13 lines · Heinrich Schuchardt
Quoted 12 lines · Simon Glass
Quoted 11 lines · Tom Rini
Start by elaborating on what some of our constraints tend to be with image location values, and document where these external constraints come from. Provide a new subsection, an example based on the TI ARMv7 OMAP2PLUS families of chips, that gives sample values and explains why we use these particular values. This is based on what is in
include/configs/ti_armv7_common.has offb3ad9bd923d("TI: Add, use aDEFAULT_LINUX_BOOT_ENVenvironment string") as this contains just the values referenced in this document now and not some of the further additions that are less generic.
Quoted 8 lines
Quoted 6 lines
Quoted 4 lines
Signed-off-by Tom Rini <[email protected]>--- doc/usage/environment.rst | 39 +++++++++++++++++++++++++++++++++++++++ 1 file changed, 39 insertions(+)Reviewed-by Simon Glass <[email protected]>Below you want a change?
Yes, often Simon does that (and it's fine) to both offer a tag but if another iteration is needed to make a minor adjustment to some wording or another, or to make when applying. Which is fine with me.
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
doc/usage/environment.rst +0
-0@@ -404,6 +404,42 @@ device tree blob fdtfile fdt_addr_r fdt_addr ramdisk ramdiskfile ramdisk_addr_r ramdisk_addr ================= ============== ================ ==============
Quoted 8 lines
Quoted 6 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 412
Quoted 6 lines
doc/usage/environment.rst +6
-0+When setting the RAM addresses for `kernel_addr_r`, `fdt_addr_r` and +`ramdisk_addr_r` there are several constraints to keep in mind. When booting +Linux, the `Booting ARM Linux`_ and `Booting AArch64 Linux`_ documents lay out +the requirements for booting all ARM platforms, including both alignment and +where within memory various things must be. These guidelines tend to also be +correct for other OSes and unless specifically contradicted by documentationWhat makes you think that BSD or Haiku have the same constraints as Linux?
Because of what I said, and experience? Now, one may be a subset of another, but that still means it will work for both. This is intended to be general best practices. If you follow this then it's likely anything else will work too. The danger comes from trying to optimize the sizes to be as small as possible, rather than as large/flexible as will likely work anywhere. I will try and expand on that idea in the next iteration.
Quoted 5 lines
Quoted 2 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 414
Quoted 2 lines
doc/usage/environment.rst +2
-0+specific to another architecture, are good rules to follow for other +architectures as well.No. RISC-V does not have the same requirements as ARM. E.g. the initrd can be located anywhere in memory.
Please point to documentation that confirms that, and some otherwise bad examples that actually work.
Quoted 7 lines
Quoted 3 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 417
Quoted 3 lines
doc/usage/environment.rst +3
-0+ +Example Image locations +^^^^^^^^^^^^^^^^^^^^^^^You seem not to refer to a file 'Image'.
%s/Image/image/
OK.
Quoted 5 lines
Quoted 3 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 420
Quoted 3 lines
doc/usage/environment.rst +3
-0+ +If we take the Texas Instruments OMAP2PLUS family of ARMv7 processors as an +example for the above listed variables, we would do::%s/we would do/we chose/ ?
Either? I don't see it mattering either way.
Quoted 11 lines
Quoted 10 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 428
Quoted 8 lines
doc/usage/environment.rst +8
-0+ + loadaddr=0x82000000 + kernel_addr_r=${loadaddr} + fdt_addr_r=0x88000000 + ramdisk_addr_r=0x88080000 + bootm_size=0x10000000 + +To explain this, we start by noting that DRAM starts at 0x80000000. A 32MiBShould it say 'We use a 32MiB' ?
Quoted 1 line
Please, mention that
MAX_UNCOMP_KERNEL_SIZE= 32 MiB is ARMv7 specific.
Sorry? As I understood it last, the maximum size was something like 96MiB before you have to employ some funky tricks.
Quoted 7 lines
Quoted 2 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 430
Quoted 2 lines
doc/usage/environment.rst +2
-0+buffer from the start of memory as our default load address, and so where the +kernel would also be loaded to. This will hopefully allow for us to have the%s/allow for us/allow/
Quoted 2 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 432
Quoted 2 lines
doc/usage/environment.rst +2
-0+whole of the compressed kernel image exist in memory above where the whole of +the decompressed kernel image will be, and allow for a quicker boot. Next, we
We use 32MiB for the reason I said here. Which is only a slight rewording of the arm32 Linux booting document, and the section starts out by saying this is an example for ARMv7 platforms.
Quoted 2 lines
Please, mention that decompressor code otherwise will have to relocate the compressed kernel.
I'm not sure. Perhaps it would be good to also link to some of the articles expanding upon how Linux on ARM32 boots, as part of more general documentation, rather than a specific example here.
Quoted 8 lines
Quoted 1 line
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 433
Quoted 1 line
doc/usage/environment.rst +1
-0+say that the device tree will be placed at 128MiB offset from the start ofPlease, mention that initrd must be
* within 512 MiB (
0x20000000) of the memory start on arm (which restrictsinitrd_high)
* in a a 1 GB aligned region of size '1UL << (VA_BITS_MIN- 1)' that includes the kernel on arm64
No, because this is not intended to list every constraint on ARM32 (nor arm64, which would benefit from an example that's not TI, as TI arm64 platforms share the same base address for memory).
Quoted 1 line
On RISC-V such a limitation does not exist.
I do wish that RISC-V had a similar document to ARM/ARM64, in Linux.
Quoted 4 lines
Quoted 1 line
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 434
Quoted 1 line
doc/usage/environment.rst +1
-0+memory. This is suggested by the kernel documment as it is exceedingly%s/documment/document
Quoted 9 lines
Quoted 5 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 439
Quoted 5 lines
doc/usage/environment.rst +5
-0+unlikely to be overwritten by the kernel itself given other architectural +constraints. We then allow for the device tree to be up to 512KiB in size +before placing the ramdisk in memory. We then say that everything should be +within the first 256MiB of memory so that U-Boot can relocate things as needed +to ensure proper alignment. We pick 256MiB as our value here because we knowIf the load address ranges occupy the first 256 MiB, where will be the space for relocation on a 256 MiB board? Without mentioning
initrd_highthis paragraph is incomplete.
You don't need initrd_high / fdt_high when bootm_size is used. Maybe that needs to be clearer in this paragraph and also the rest of the documentation.
Quoted 6 lines
Quoted 3 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 442
Quoted 3 lines
doc/usage/environment.rst +3
-0+there are very few platforms on in this family with less memory. It could be +as high as 768MiB and still ensure that everything would be visible to the +kernel, but again we go with what we assume is the safest assumption.A section per architecture clearly naming the architecture specific restrictions of Linux would be much more helpful.
I did ask for further explained examples as it would indeed be good to have at least one per architecture.
Signature · 1 line
Tom
Re: [PATCH 3/3] doc: environment: Further expand on Image locations and provide example
Quoted 52 lines · Tom Rini
Quoted 13 lines · Heinrich Schuchardt
Quoted 12 lines · Simon Glass
Quoted 11 lines · Tom Rini
Start by elaborating on what some of our constraints tend to be with image location values, and document where these external constraints come from. Provide a new subsection, an example based on the TI ARMv7 OMAP2PLUS families of chips, that gives sample values and explains why we use these particular values. This is based on what is in
include/configs/ti_armv7_common.has offb3ad9bd923d("TI: Add, use aDEFAULT_LINUX_BOOT_ENVenvironment string") as this contains just the values referenced in this document now and not some of the further additions that are less generic.Quoted 8 lines
Quoted 6 lines
Quoted 4 lines
Signed-off-by Tom Rini <[email protected]>--- doc/usage/environment.rst | 39 +++++++++++++++++++++++++++++++++++++++ 1 file changed, 39 insertions(+)Reviewed-by Simon Glass <[email protected]>Below you want a change?
Yes, often Simon does that (and it's fine) to both offer a tag but if another iteration is needed to make a minor adjustment to some wording or another, or to make when applying. Which is fine with me.
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
doc/usage/environment.rst +0
-0@@ -404,6 +404,42 @@ device tree blob fdtfile fdt_addr_r fdt_addr ramdisk ramdiskfile ramdisk_addr_r ramdisk_addr ================= ============== ================ ==============Quoted 8 lines
Quoted 6 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 412
Quoted 6 lines
doc/usage/environment.rst +6
-0+When setting the RAM addresses for `kernel_addr_r`, `fdt_addr_r` and +`ramdisk_addr_r` there are several constraints to keep in mind. When booting +Linux, the `Booting ARM Linux`_ and `Booting AArch64 Linux`_ documents lay out +the requirements for booting all ARM platforms, including both alignment and +where within memory various things must be. These guidelines tend to also be +correct for other OSes and unless specifically contradicted by documentationWhat makes you think that BSD or Haiku have the same constraints as Linux?
Because of what I said, and experience? Now, one may be a subset of another, but that still means it will work for both. This is intended to be general best practices. If you follow this then it's likely anything else will work too. The danger comes from trying to optimize the sizes to be as small as possible, rather than as large/flexible as will likely work anywhere. I will try and expand on that idea in the next iteration.
Quoted 8 lines
Quoted 5 lines
Quoted 2 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 414
Quoted 2 lines
doc/usage/environment.rst +2
-0+specific to another architecture, are good rules to follow for other +architectures as well.No. RISC-V does not have the same requirements as ARM. E.g. the initrd can be located anywhere in memory.
Please point to documentation that confirms that, and some otherwise bad examples that actually work.
[PATCH 1/1] RISC-V: load initrd wherever it fits into memory
https://lore.kernel.org/all/[email protected]/
Please, have a look at efi_get_max_initrd_addr() in these files:
arch/arm/include/asm/efi.h:73arch/arm64/include/asm/efi.h:77arch/loongarch/include/asm/efi.h:36arch/riscv/include/asm/efi.h:31
MAX_UNCOMP_KERNEL_SIZE = 32 MiB is only enforced in drivers/firmware/efi/libstub/arm32-stub.c.
Quoted 30 lines
Quoted 8 lines
Quoted 4 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 417
Quoted 4 lines
doc/usage/environment.rst +3
-0+ +Example Image locations +^^^^^^^^^^^^^^^^^^^^^^^You seem not to refer to a file 'Image'.
%s/Image/image/
OK.
Quoted 5 lines
Quoted 3 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 420
Quoted 3 lines
doc/usage/environment.rst +3
-0+ +If we take the Texas Instruments OMAP2PLUS family of ARMv7 processors as an +example for the above listed variables, we would do::%s/we would do/we chose/ ?
Either? I don't see it mattering either way.
Quoted 11 lines
Quoted 10 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 428
Quoted 8 lines
doc/usage/environment.rst +8
-0+ + loadaddr=0x82000000 + kernel_addr_r=${loadaddr} + fdt_addr_r=0x88000000 + ramdisk_addr_r=0x88080000 + bootm_size=0x10000000 + +To explain this, we start by noting that DRAM starts at 0x80000000. A 32MiBShould it say 'We use a 32MiB' ?
Quoted 4 lines
Quoted 1 line
Please, mention that
MAX_UNCOMP_KERNEL_SIZE= 32 MiB is ARMv7 specific.Sorry? As I understood it last, the maximum size was something like 96MiB before you have to employ some funky tricks.
Look at the use of MAX_UNCOMP_KERNEL_SIZE in handle_kernel_image() of the EFI stub (drivers/firmware/efi/libstub/arm32-stub.c).
Quoted 6 lines
Quoted 6 lines
Quoted 3 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 430
Quoted 3 lines
doc/usage/environment.rst +2
-0+buffer from the start of memory as our default load address, and so where the +kernel would also be loaded to. This will hopefully allow for us to have the%s/allow for us/allow/
Quoted 6 lines
Quoted 2 lines
Quoted 2 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 432
Quoted 2 lines
doc/usage/environment.rst +2
-0+whole of the compressed kernel image exist in memory above where the whole of +the decompressed kernel image will be, and allow for a quicker boot. Next, weWe use 32MiB for the reason I said here. Which is only a slight rewording of the arm32 Linux booting document, and the section starts out by saying this is an example for ARMv7 platforms.
You ask all other architectures to follow this example?
Quoted 7 lines
Quoted 3 lines
Please, mention that decompressor code otherwise will have to relocate the compressed kernel.
I'm not sure. Perhaps it would be good to also link to some of the articles expanding upon how Linux on ARM32 boots, as part of more general documentation, rather than a specific example here.
Just look at the comment above the definition of MAX_UNCOMP_KERNEL_SIZE in arch/arm/include/asm/efi.h.
Quoted 5 lines
Quoted 5 lines
Quoted 2 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 433
Quoted 2 lines
doc/usage/environment.rst +1
-0+say that the device tree will be placed at 128MiB offset from the start ofPlease, mention that initrd must be
Quoted 8 lines
Quoted 4 lines
* within 512 MiB (
0x20000000) of the memory start on arm (which restrictsinitrd_high)
* in a a 1 GB aligned region of size '1UL << (VA_BITS_MIN- 1)' that includes the kernel on arm64No, because this is not intended to list every constraint on ARM32 (nor arm64, which would benefit from an example that's not TI, as TI arm64 platforms share the same base address for memory).
You ask above to follow the example of ARMv7 on all architectures. Hence it is necessary to point out the differences.
Best regards
Heinrich
Quoted 33 lines
Quoted 2 lines
On RISC-V such a limitation does not exist.
I do wish that RISC-V had a similar document to ARM/ARM64, in Linux.
Quoted 4 lines
Quoted 1 line
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 434
Quoted 1 line
doc/usage/environment.rst +1
-0+memory. This is suggested by the kernel documment as it is exceedingly%s/documment/document
Quoted 9 lines
Quoted 5 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 439
Quoted 5 lines
doc/usage/environment.rst +5
-0+unlikely to be overwritten by the kernel itself given other architectural +constraints. We then allow for the device tree to be up to 512KiB in size +before placing the ramdisk in memory. We then say that everything should be +within the first 256MiB of memory so that U-Boot can relocate things as needed +to ensure proper alignment. We pick 256MiB as our value here because we knowIf the load address ranges occupy the first 256 MiB, where will be the space for relocation on a 256 MiB board? Without mentioning
initrd_highthis paragraph is incomplete.You don't need
initrd_high/fdt_highwhenbootm_sizeis used. Maybe that needs to be clearer in this paragraph and also the rest of the documentation.Quoted 6 lines
Quoted 3 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 442
Quoted 3 lines
doc/usage/environment.rst +3
-0+there are very few platforms on in this family with less memory. It could be +as high as 768MiB and still ensure that everything would be visible to the +kernel, but again we go with what we assume is the safest assumption.A section per architecture clearly naming the architecture specific restrictions of Linux would be much more helpful.
I did ask for further explained examples as it would indeed be good to have at least one per architecture.
Re: [PATCH 3/3] doc: environment: Further expand on Image locations and provide example
Quoted 59 lines · Heinrich Schuchardt
Quoted 58 lines · Tom Rini
Quoted 13 lines · Heinrich Schuchardt
Quoted 12 lines · Simon Glass
Quoted 11 lines · Tom Rini
Start by elaborating on what some of our constraints tend to be with image location values, and document where these external constraints come from. Provide a new subsection, an example based on the TI ARMv7 OMAP2PLUS families of chips, that gives sample values and explains why we use these particular values. This is based on what is in
include/configs/ti_armv7_common.has offb3ad9bd923d("TI: Add, use aDEFAULT_LINUX_BOOT_ENVenvironment string") as this contains just the values referenced in this document now and not some of the further additions that are less generic.Quoted 8 lines
Quoted 6 lines
Quoted 4 lines
Signed-off-by Tom Rini <[email protected]>--- doc/usage/environment.rst | 39 +++++++++++++++++++++++++++++++++++++++ 1 file changed, 39 insertions(+)Reviewed-by Simon Glass <[email protected]>Below you want a change?
Yes, often Simon does that (and it's fine) to both offer a tag but if another iteration is needed to make a minor adjustment to some wording or another, or to make when applying. Which is fine with me.
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
doc/usage/environment.rst +0
-0@@ -404,6 +404,42 @@ device tree blob fdtfile fdt_addr_r fdt_addr ramdisk ramdiskfile ramdisk_addr_r ramdisk_addr ================= ============== ================ ==============Quoted 8 lines
Quoted 6 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 412
Quoted 6 lines
doc/usage/environment.rst +6
-0+When setting the RAM addresses for `kernel_addr_r`, `fdt_addr_r` and +`ramdisk_addr_r` there are several constraints to keep in mind. When booting +Linux, the `Booting ARM Linux`_ and `Booting AArch64 Linux`_ documents lay out +the requirements for booting all ARM platforms, including both alignment and +where within memory various things must be. These guidelines tend to also be +correct for other OSes and unless specifically contradicted by documentationWhat makes you think that BSD or Haiku have the same constraints as Linux?
Because of what I said, and experience? Now, one may be a subset of another, but that still means it will work for both. This is intended to be general best practices. If you follow this then it's likely anything else will work too. The danger comes from trying to optimize the sizes to be as small as possible, rather than as large/flexible as will likely work anywhere. I will try and expand on that idea in the next iteration.
Quoted 5 lines
Quoted 2 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 414
Quoted 2 lines
doc/usage/environment.rst +2
-0+specific to another architecture, are good rules to follow for other +architectures as well.No. RISC-V does not have the same requirements as ARM. E.g. the initrd can be located anywhere in memory.
Quoted 5 lines
Quoted 2 lines
Please point to documentation that confirms that, and some otherwise bad examples that actually work.
[PATCH 1/1] RISC-V: load initrd wherever it fits into memory
https://lore.kernel.org/all/[email protected]/
Looks like someone ran in to the first common case of "oops, overwrote the ramdisk with the kernel bss" or something along those lines.
Which is why I'm asking for more architectures to add good examples of where to load each payload in to memory, with explanations of why and how big of a gap to have. I _think_ in Linux RISC-V (and hopefully for 32bit and 64bit) used the arm64 Image format and so BSS size is available in the header and so we can safely check for that overlap and relocate rather than fail to boot. Checking for, and avoiding to start with, these types of problems is why I want to add the examples.
Quoted 9 lines
Please, have a look at
efi_get_max_initrd_addr()in these files:
arch/arm/include/asm/efi.h:73arch/arm64/include/asm/efi.h:77arch/loongarch/include/asm/efi.h:36arch/riscv/include/asm/efi.h:31
MAX_UNCOMP_KERNEL_SIZE= 32 MiB is only enforced indrivers/firmware/efi/libstub/arm32-stub.c.
This isn't an EFI thing however. The max uncompressed Linux kernel image for arm32 is something along the lines of 96MiB I recall rmk telling me when I asked about it at the time. The base+32MiB in the example here is for optimal (but not REQUIRED) decompressor location.
Quoted 42 lines
Quoted 28 lines
Quoted 6 lines
Quoted 2 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 417
Quoted 2 lines
doc/usage/environment.rst +2
-0+Example Image locations +^^^^^^^^^^^^^^^^^^^^^^^You seem not to refer to a file 'Image'.
%s/Image/image/
OK.
Quoted 5 lines
Quoted 3 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 420
Quoted 3 lines
doc/usage/environment.rst +3
-0+ +If we take the Texas Instruments OMAP2PLUS family of ARMv7 processors as an +example for the above listed variables, we would do::%s/we would do/we chose/ ?
Either? I don't see it mattering either way.
Quoted 11 lines
Quoted 10 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 428
Quoted 8 lines
doc/usage/environment.rst +8
-0+ + loadaddr=0x82000000 + kernel_addr_r=${loadaddr} + fdt_addr_r=0x88000000 + ramdisk_addr_r=0x88080000 + bootm_size=0x10000000 + +To explain this, we start by noting that DRAM starts at 0x80000000. A 32MiBShould it say 'We use a 32MiB' ?
Quoted 4 lines
Quoted 1 line
Please, mention that
MAX_UNCOMP_KERNEL_SIZE= 32 MiB is ARMv7 specific.Sorry? As I understood it last, the maximum size was something like 96MiB before you have to employ some funky tricks.
Look at the use of
MAX_UNCOMP_KERNEL_SIZEinhandle_kernel_image()of the EFI stub (drivers/firmware/efi/libstub/arm32-stub.c).Quoted 6 lines
Quoted 6 lines
Quoted 3 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 430
Quoted 3 lines
doc/usage/environment.rst +2
-0+buffer from the start of memory as our default load address, and so where the +kernel would also be loaded to. This will hopefully allow for us to have the%s/allow for us/allow/
Quoted 8 lines
Quoted 6 lines
Quoted 2 lines
Quoted 2 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 432
Quoted 2 lines
doc/usage/environment.rst +2
-0+whole of the compressed kernel image exist in memory above where the whole of +the decompressed kernel image will be, and allow for a quicker boot. Next, weWe use 32MiB for the reason I said here. Which is only a slight rewording of the arm32 Linux booting document, and the section starts out by saying this is an example for ARMv7 platforms.
You ask all other architectures to follow this example?
I could have sworn that somewhere within the comments of this series I asked for more examples to be added, yes. And I know I intended to (since we _need_ them, and I think I've expressed me desire to have them before) and I am asking now.
Quoted 9 lines
Quoted 6 lines
Quoted 2 lines
Please, mention that decompressor code otherwise will have to relocate the compressed kernel.
I'm not sure. Perhaps it would be good to also link to some of the articles expanding upon how Linux on ARM32 boots, as part of more general documentation, rather than a specific example here.
Just look at the comment above the definition of
MAX_UNCOMP_KERNEL_SIZEinarch/arm/include/asm/efi.h.
Keep in mind this is only vaguely EFI-related. Given how long ago edkII for beagleboard was done, it doesn't quite predate EFI on ARM, but this example has been in long use for the common non-EFI case on 32bit ARM.
Quoted 9 lines
Quoted 9 lines
Quoted 8 lines
Quoted 1 line
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 433
Quoted 1 line
doc/usage/environment.rst +1
-0+say that the device tree will be placed at 128MiB offset from the start ofPlease, mention that initrd must be
* within 512 MiB (
0x20000000) of the memory start on arm (which restrictsinitrd_high)
* in a a 1 GB aligned region of size '1UL << (VA_BITS_MIN- 1)' that includes the kernel on arm64
Quoted 6 lines
Quoted 3 lines
No, because this is not intended to list every constraint on ARM32 (nor arm64, which would benefit from an example that's not TI, as TI arm64 platforms share the same base address for memory).
You ask above to follow the example of ARMv7 on all architectures. Hence it is necessary to point out the differences.
No, I'm asking for more examples to be added for each architecture.
Signature · 1 line
Tom
Re: [PATCH 3/3] doc: environment: Further expand on Image locations and provide example
Quoted 60 lines · Tom Rini
Quoted 59 lines · Heinrich Schuchardt
Quoted 58 lines · Tom Rini
Quoted 13 lines · Heinrich Schuchardt
Quoted 12 lines · Simon Glass
Quoted 11 lines · Tom Rini
Start by elaborating on what some of our constraints tend to be with image location values, and document where these external constraints come from. Provide a new subsection, an example based on the TI ARMv7 OMAP2PLUS families of chips, that gives sample values and explains why we use these particular values. This is based on what is in
include/configs/ti_armv7_common.has offb3ad9bd923d("TI: Add, use aDEFAULT_LINUX_BOOT_ENVenvironment string") as this contains just the values referenced in this document now and not some of the further additions that are less generic.Quoted 8 lines
Quoted 6 lines
Quoted 4 lines
Signed-off-by Tom Rini <[email protected]>--- doc/usage/environment.rst | 39 +++++++++++++++++++++++++++++++++++++++ 1 file changed, 39 insertions(+)Reviewed-by Simon Glass <[email protected]>Below you want a change?
Yes, often Simon does that (and it's fine) to both offer a tag but if another iteration is needed to make a minor adjustment to some wording or another, or to make when applying. Which is fine with me.
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
doc/usage/environment.rst +0
-0@@ -404,6 +404,42 @@ device tree blob fdtfile fdt_addr_r fdt_addr ramdisk ramdiskfile ramdisk_addr_r ramdisk_addr ================= ============== ================ ==============Quoted 8 lines
Quoted 6 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 412
Quoted 6 lines
doc/usage/environment.rst +6
-0+When setting the RAM addresses for `kernel_addr_r`, `fdt_addr_r` and +`ramdisk_addr_r` there are several constraints to keep in mind. When booting +Linux, the `Booting ARM Linux`_ and `Booting AArch64 Linux`_ documents lay out +the requirements for booting all ARM platforms, including both alignment and +where within memory various things must be. These guidelines tend to also be +correct for other OSes and unless specifically contradicted by documentationWhat makes you think that BSD or Haiku have the same constraints as Linux?
Because of what I said, and experience? Now, one may be a subset of another, but that still means it will work for both. This is intended to be general best practices. If you follow this then it's likely anything else will work too. The danger comes from trying to optimize the sizes to be as small as possible, rather than as large/flexible as will likely work anywhere. I will try and expand on that idea in the next iteration.
Quoted 5 lines
Quoted 2 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 414
Quoted 2 lines
doc/usage/environment.rst +2
-0+specific to another architecture, are good rules to follow for other +architectures as well.No. RISC-V does not have the same requirements as ARM. E.g. the initrd can be located anywhere in memory.
Quoted 8 lines
Quoted 5 lines
Quoted 2 lines
Please point to documentation that confirms that, and some otherwise bad examples that actually work.
[PATCH 1/1] RISC-V: load initrd wherever it fits into memory
https://lore.kernel.org/all/[email protected]/Looks like someone ran in to the first common case of "oops, overwrote the ramdisk with the kernel bss" or something along those lines.
Not at all. The ramdisk was relocated by U-Boot unnecessarily to above 256 MiB after start of RAM and the EFI stub before the patch did not accept this address for no reason.
Quoted 8 lines
Which is why I'm asking for more architectures to add good examples of where to load each payload in to memory, with explanations of why and how big of a gap to have. I _think_ in Linux RISC-V (and hopefully for 32bit and 64bit) used the arm64 Image format and so BSS size is available in the header and so we can safely check for that overlap and relocate rather than fail to boot. Checking for, and avoiding to start with, these types of problems is why I want to add the examples.
I am not aware of any restrictions for the placement of kernel, initrd, fdt for RISC-V. Therefore there is no need to relocate anything after loading (without overlap).
The bootefi command will never relocate a kernel or an fdt on any architecture. You just pass the original load addresses.
Quoted 8 lines
Quoted 8 lines
Please, have a look at
efi_get_max_initrd_addr()in these files:
arch/arm/include/asm/efi.h:73arch/arm64/include/asm/efi.h:77arch/loongarch/include/asm/efi.h:36arch/riscv/include/asm/efi.h:31
Quoted 7 lines
Quoted 2 lines
MAX_UNCOMP_KERNEL_SIZE= 32 MiB is only enforced indrivers/firmware/efi/libstub/arm32-stub.c.This isn't an EFI thing however. The max uncompressed Linux kernel image for arm32 is something along the lines of 96MiB I recall rmk telling me when I asked about it at the time. The base+32MiB in the example here is for optimal (but not REQUIRED) decompressor location.
The decompressor is what follows the EFI stub in the image?
Quoted 64 lines
Quoted 43 lines
Quoted 29 lines
Quoted 7 lines
Quoted 3 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 417
Quoted 3 lines
doc/usage/environment.rst +2
-0+Example Image locations +^^^^^^^^^^^^^^^^^^^^^^^You seem not to refer to a file 'Image'.
%s/Image/image/
OK.
Quoted 5 lines
Quoted 3 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 420
Quoted 3 lines
doc/usage/environment.rst +3
-0+ +If we take the Texas Instruments OMAP2PLUS family of ARMv7 processors as an +example for the above listed variables, we would do::%s/we would do/we chose/ ?
Either? I don't see it mattering either way.
Quoted 11 lines
Quoted 10 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 428
Quoted 8 lines
doc/usage/environment.rst +8
-0+ + loadaddr=0x82000000 + kernel_addr_r=${loadaddr} + fdt_addr_r=0x88000000 + ramdisk_addr_r=0x88080000 + bootm_size=0x10000000 + +To explain this, we start by noting that DRAM starts at 0x80000000. A 32MiBShould it say 'We use a 32MiB' ?
Quoted 4 lines
Quoted 1 line
Please, mention that
MAX_UNCOMP_KERNEL_SIZE= 32 MiB is ARMv7 specific.Sorry? As I understood it last, the maximum size was something like 96MiB before you have to employ some funky tricks.
Look at the use of
MAX_UNCOMP_KERNEL_SIZEinhandle_kernel_image()of the EFI stub (drivers/firmware/efi/libstub/arm32-stub.c).Quoted 6 lines
Quoted 6 lines
Quoted 3 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 430
Quoted 3 lines
doc/usage/environment.rst +2
-0+buffer from the start of memory as our default load address, and so where the +kernel would also be loaded to. This will hopefully allow for us to have the%s/allow for us/allow/
Quoted 8 lines
Quoted 6 lines
Quoted 2 lines
Quoted 2 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 432
Quoted 2 lines
doc/usage/environment.rst +2
-0+whole of the compressed kernel image exist in memory above where the whole of +the decompressed kernel image will be, and allow for a quicker boot. Next, weWe use 32MiB for the reason I said here. Which is only a slight rewording of the arm32 Linux booting document, and the section starts out by saying this is an example for ARMv7 platforms.
You ask all other architectures to follow this example?
I could have sworn that somewhere within the comments of this series I asked for more examples to be added, yes. And I know I intended to (since we _need_ them, and I think I've expressed me desire to have them before) and I am asking now.
Quoted 7 lines
Quoted 6 lines
Quoted 2 lines
Please, mention that decompressor code otherwise will have to relocate the compressed kernel.
I'm not sure. Perhaps it would be good to also link to some of the articles expanding upon how Linux on ARM32 boots, as part of more general documentation, rather than a specific example here.
Quoted 6 lines
Quoted 2 lines
Just look at the comment above the definition of
MAX_UNCOMP_KERNEL_SIZEinarch/arm/include/asm/efi.h.Keep in mind this is only vaguely EFI-related. Given how long ago edkII for beagleboard was done, it doesn't quite predate EFI on ARM, but this example has been in long use for the common non-EFI case on 32bit ARM.
The EFI stub is using the value due to (assumed) restrictions in the decompressor and main kernel.
Best regards
Heinrich
Quoted 19 lines
Quoted 10 lines
Quoted 10 lines
Quoted 9 lines
Quoted 2 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 433
Quoted 2 lines
doc/usage/environment.rst +1
-0+say that the device tree will be placed at 128MiB offset from the start ofPlease, mention that initrd must be
* within 512 MiB (
0x20000000) of the memory start on arm (which restrictsinitrd_high)
* in a a 1 GB aligned region of size '1UL << (VA_BITS_MIN- 1)' that includes the kernel on arm64Quoted 6 lines
Quoted 3 lines
No, because this is not intended to list every constraint on ARM32 (nor arm64, which would benefit from an example that's not TI, as TI arm64 platforms share the same base address for memory).
You ask above to follow the example of ARMv7 on all architectures. Hence it is necessary to point out the differences.
No, I'm asking for more examples to be added for each architecture.
Re: [PATCH 3/3] doc: environment: Further expand on Image locations and provide example
Quoted 67 lines · Heinrich Schuchardt
Quoted 66 lines · Tom Rini
Quoted 64 lines · Heinrich Schuchardt
Quoted 52 lines · Tom Rini
Quoted 13 lines · Heinrich Schuchardt
Quoted 12 lines · Simon Glass
Quoted 11 lines · Tom Rini
Start by elaborating on what some of our constraints tend to be with image location values, and document where these external constraints come from. Provide a new subsection, an example based on the TI ARMv7 OMAP2PLUS families of chips, that gives sample values and explains why we use these particular values. This is based on what is in
include/configs/ti_armv7_common.has offb3ad9bd923d("TI: Add, use aDEFAULT_LINUX_BOOT_ENVenvironment string") as this contains just the values referenced in this document now and not some of the further additions that are less generic.Quoted 8 lines
Quoted 6 lines
Quoted 4 lines
Signed-off-by Tom Rini <[email protected]>--- doc/usage/environment.rst | 39 +++++++++++++++++++++++++++++++++++++++ 1 file changed, 39 insertions(+)Reviewed-by Simon Glass <[email protected]>Below you want a change?
Yes, often Simon does that (and it's fine) to both offer a tag but if another iteration is needed to make a minor adjustment to some wording or another, or to make when applying. Which is fine with me.
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
doc/usage/environment.rst +0
-0@@ -404,6 +404,42 @@ device tree blob fdtfile fdt_addr_r fdt_addr ramdisk ramdiskfile ramdisk_addr_r ramdisk_addr ================= ============== ================ ==============Quoted 8 lines
Quoted 6 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 412
Quoted 6 lines
doc/usage/environment.rst +6
-0+When setting the RAM addresses for `kernel_addr_r`, `fdt_addr_r` and +`ramdisk_addr_r` there are several constraints to keep in mind. When booting +Linux, the `Booting ARM Linux`_ and `Booting AArch64 Linux`_ documents lay out +the requirements for booting all ARM platforms, including both alignment and +where within memory various things must be. These guidelines tend to also be +correct for other OSes and unless specifically contradicted by documentationWhat makes you think that BSD or Haiku have the same constraints as Linux?
Because of what I said, and experience? Now, one may be a subset of another, but that still means it will work for both. This is intended to be general best practices. If you follow this then it's likely anything else will work too. The danger comes from trying to optimize the sizes to be as small as possible, rather than as large/flexible as will likely work anywhere. I will try and expand on that idea in the next iteration.
Quoted 8 lines
Quoted 5 lines
Quoted 2 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 414
Quoted 2 lines
doc/usage/environment.rst +2
-0+specific to another architecture, are good rules to follow for other +architectures as well.No. RISC-V does not have the same requirements as ARM. E.g. the initrd can be located anywhere in memory.
Please point to documentation that confirms that, and some otherwise bad examples that actually work.
[PATCH 1/1] RISC-V: load initrd wherever it fits into memory
https://lore.kernel.org/all/[email protected]/
Quoted 6 lines
Quoted 2 lines
Looks like someone ran in to the first common case of "oops, overwrote the ramdisk with the kernel bss" or something along those lines.
Not at all. The ramdisk was relocated by U-Boot unnecessarily to above 256 MiB after start of RAM and the EFI stub before the patch did not accept this address for no reason.
Then bootm_size / bootm_low should have been set appropriately in U-Boot.
Quoted 8 lines
Quoted 7 lines
Which is why I'm asking for more architectures to add good examples of where to load each payload in to memory, with explanations of why and how big of a gap to have. I _think_ in Linux RISC-V (and hopefully for 32bit and 64bit) used the arm64 Image format and so BSS size is available in the header and so we can safely check for that overlap and relocate rather than fail to boot. Checking for, and avoiding to start with, these types of problems is why I want to add the examples.
Quoted 6 lines
I am not aware of any restrictions for the placement of kernel, initrd, fdt for RISC-V. Therefore there is no need to relocate anything after loading (without overlap).
The bootefi command will never relocate a kernel or an fdt on any architecture. You just pass the original load addresses.
Which is all the more reason we need to ensure that the fdt and initrd are placed in appropriately aligned and non-overlapping locations. Because if you put the initrd too close to where the kernel ends up, the BSS will eat your ramdisk before it can go "oh, there's a ramdisk there". That's a practical, not architectural, problem. Same with the device tree.
Quoted 10 lines
Quoted 10 lines
Quoted 9 lines
Please, have a look at
efi_get_max_initrd_addr()in these files:
arch/arm/include/asm/efi.h:73arch/arm64/include/asm/efi.h:77arch/loongarch/include/asm/efi.h:36arch/riscv/include/asm/efi.h:31
MAX_UNCOMP_KERNEL_SIZE= 32 MiB is only enforced indrivers/firmware/efi/libstub/arm32-stub.c.
Quoted 6 lines
Quoted 4 lines
This isn't an EFI thing however. The max uncompressed Linux kernel image for arm32 is something along the lines of 96MiB I recall rmk telling me when I asked about it at the time. The base+32MiB in the example here is for optimal (but not REQUIRED) decompressor location.
The decompressor is what follows the EFI stub in the image?
I'm not talking about EFI at all in this example, as it's 32bit ARM, not 64bit ARM or RISC-V.
Quoted 66 lines
Quoted 66 lines
Quoted 42 lines
Quoted 28 lines
Quoted 6 lines
Quoted 2 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 416
Quoted 2 lines
doc/usage/environment.rst +2
-0+Example Image locations +^^^^^^^^^^^^^^^^^^^^^^^You seem not to refer to a file 'Image'.
%s/Image/image/
OK.
Quoted 5 lines
Quoted 3 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 419
Quoted 3 lines
doc/usage/environment.rst +3
-0+ +If we take the Texas Instruments OMAP2PLUS family of ARMv7 processors as an +example for the above listed variables, we would do::%s/we would do/we chose/ ?
Either? I don't see it mattering either way.
Quoted 11 lines
Quoted 10 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 427
Quoted 8 lines
doc/usage/environment.rst +8
-0+ + loadaddr=0x82000000 + kernel_addr_r=${loadaddr} + fdt_addr_r=0x88000000 + ramdisk_addr_r=0x88080000 + bootm_size=0x10000000 + +To explain this, we start by noting that DRAM starts at 0x80000000. A 32MiBShould it say 'We use a 32MiB' ?
Quoted 4 lines
Quoted 1 line
Please, mention that
MAX_UNCOMP_KERNEL_SIZE= 32 MiB is ARMv7 specific.Sorry? As I understood it last, the maximum size was something like 96MiB before you have to employ some funky tricks.
Look at the use of
MAX_UNCOMP_KERNEL_SIZEinhandle_kernel_image()of the EFI stub (drivers/firmware/efi/libstub/arm32-stub.c).Quoted 6 lines
Quoted 6 lines
Quoted 3 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 429
Quoted 3 lines
doc/usage/environment.rst +2
-0+buffer from the start of memory as our default load address, and so where the +kernel would also be loaded to. This will hopefully allow for us to have the%s/allow for us/allow/
Quoted 8 lines
Quoted 6 lines
Quoted 2 lines
Quoted 2 lines
doc/usage/environment.rst device tree blob fdtfile fdt_addr_r fdt_addr · line 431
Quoted 2 lines
doc/usage/environment.rst +2
-0+whole of the compressed kernel image exist in memory above where the whole of +the decompressed kernel image will be, and allow for a quicker boot. Next, weWe use 32MiB for the reason I said here. Which is only a slight rewording of the arm32 Linux booting document, and the section starts out by saying this is an example for ARMv7 platforms.
You ask all other architectures to follow this example?
I could have sworn that somewhere within the comments of this series I asked for more examples to be added, yes. And I know I intended to (since we _need_ them, and I think I've expressed me desire to have them before) and I am asking now.
Quoted 9 lines
Quoted 6 lines
Quoted 2 lines
Please, mention that decompressor code otherwise will have to relocate the compressed kernel.
I'm not sure. Perhaps it would be good to also link to some of the articles expanding upon how Linux on ARM32 boots, as part of more general documentation, rather than a specific example here.
Just look at the comment above the definition of
MAX_UNCOMP_KERNEL_SIZEinarch/arm/include/asm/efi.h.
Quoted 6 lines
Quoted 3 lines
Keep in mind this is only vaguely EFI-related. Given how long ago edkII for beagleboard was done, it doesn't quite predate EFI on ARM, but this example has been in long use for the common non-EFI case on 32bit ARM.
The EFI stub is using the value due to (assumed) restrictions in the decompressor and main kernel.
OK, but EFI is largely irrelevant to the 32bit ARM case, which is what this example is for.
Signature · 1 line
Tom
Quoted 7 lines · Tom Rini
This variable is never set nor explained why it would be set, drop it.
Signed-off-by Tom Rini <[email protected]>--- doc/usage/environment.rst | 1 - 1 file changed, 1 deletion(-)
Quoted 3 lines · Tom Rini
This variable is never set nor explained why it would be set, drop it.
Signed-off-by Tom Rini <[email protected]>
Quoted 16 lines
--- doc/usage/environment.rst | 1 - 1 file changed, 1 deletion(-)doc/usage/environment.rst +0
-1@@ -390,7 +390,6 @@ in U-Boot code. ================= ============== ================ ============== Image File Name RAM Address Flash Location ================= ============== ================ ============== -u-boot u-boot u-boot_addr_r u-boot_addr Linux kernel bootfile kernel_addr_r kernel_addr device tree blob fdtfile fdt_addr_r fdt_addr ramdisk ramdiskfile ramdisk_addr_r ramdisk_addr