[U-Boot] [PATCH 1/2] efi_loader: rework fdt handling in distro boot script
Alexander Graf
agraf at suse.de
Thu Oct 18 07:31:57 UTC 2018
On 18.10.18 04:07, AKASHI Takahiro wrote:
> On Tue, Oct 16, 2018 at 03:15:13PM +0200, Alexander Graf wrote:
>>
>>
>> On 12.10.18 07:07, AKASHI Takahiro wrote:
>>> The current scenario for default UEFI booting, scan_dev_for_efi, has
>>> several issues:
>>> * invoke 'bootmgr' only if BOOTEFI_NAME binary does exit even though
>>> 'bootmgr' can and should work independently whether or not the binary
>>> exist,
>>> * always assume that a 'fdtfile' variable is defined,
>>> * redundantly check for 'fdt_addr_r' in boot_efi_binary
>>>
>>> In this patch, all the issues above are sorted out.
>>>
>>> Signed-off-by: AKASHI Takahiro <takahiro.akashi at linaro.org>
>>> ---
>>> include/config_distro_bootcmd.h | 43 ++++++++++++++++-----------------
>>> 1 file changed, 21 insertions(+), 22 deletions(-)
>>>
>>> diff --git a/include/config_distro_bootcmd.h b/include/config_distro_bootcmd.h
>>> index 373fee78a999..76e12b7bf4ee 100644
>>> --- a/include/config_distro_bootcmd.h
>>> +++ b/include/config_distro_bootcmd.h
>>> @@ -124,42 +124,41 @@
>>>
>>> #define BOOTENV_SHARED_EFI \
>>> "boot_efi_binary=" \
>>> - "if fdt addr ${fdt_addr_r}; then " \
>>> - "bootefi bootmgr ${fdt_addr_r};" \
>>> - "else " \
>>> - "bootefi bootmgr ${fdtcontroladdr};" \
>>> - "fi;" \
>>> "load ${devtype} ${devnum}:${distro_bootpart} " \
>>> "${kernel_addr_r} efi/boot/"BOOTEFI_NAME"; " \
>>> - "if fdt addr ${fdt_addr_r}; then " \
>>> - "bootefi ${kernel_addr_r} ${fdt_addr_r};" \
>>> - "else " \
>>> - "bootefi ${kernel_addr_r} ${fdtcontroladdr};" \
>>> - "fi\0" \
>>> + "bootefi ${kernel_addr_r} ${efi_fdt_addr};\0" \
>>> \
>>> "load_efi_dtb=" \
>>> "load ${devtype} ${devnum}:${distro_bootpart} " \
>>> - "${fdt_addr_r} ${prefix}${efi_fdtfile}\0" \
>>> + "${fdt_addr_r} ${prefix}${efi_fdtfile};" \
>>> + "if fdt addr ${fdt_addr_r}; then " \
>>> + "setenv efi_fdt_addr ${fdt_addr_r}; " \
>>> + "else; " \
>>> + "setenv efi_fdt_addr ${fdtcontroladdr}; " \
>>> + "fi;\0" \
>>> \
>>> - "efi_dtb_prefixes=/ /dtb/ /dtb/current/\0" \
>>> - "scan_dev_for_efi=" \
>>> + "set_efi_fdt_addr=" \
>>> "setenv efi_fdtfile ${fdtfile}; " \
>>> BOOTENV_EFI_SET_FDTFILE_FALLBACK \
>>> - "for prefix in ${efi_dtb_prefixes}; do " \
>>
>> I fail to see where the prefix logic went? Without that, our distro's
>> dtb loading will break.
>
> Oops, I missed it.
>
>>
>>> - "if test -e ${devtype} " \
>>> - "${devnum}:${distro_bootpart} " \
>>> - "${prefix}${efi_fdtfile}; then " \
>>> - "run load_efi_dtb; " \
>>> - "fi;" \
>>> - "done;" \
>>> + "if test x${efi_fdtfile} != x -a x{$fdt_addr_r} != x ; then " \
>>> + "run load_efi_dtb; " \
>>> + "else; " \
>>> + "setenv efi_fdt_addr ${fdtcontroladdr}; " \
>>> + "fi; " \
>>
>> Just unconditionally set efi_fdt_addr=$fdtcontrolladdr before the check
>> (and invocation of load_efi_dtb).
>
> OK.
>
>> That way you don't need to check for
>> the failure case in load_efi_dtb again.
>
> Well, I think that we need a check for validity with "fdt addr" command
> just in case that a fdt file exist but its content be corrupted.
>
> My modified version looks like:
> ===8<===
> #define BOOTENV_SHARED_EFI \
> "boot_efi_binary=" \
> "load ${devtype} ${devnum}:${distro_bootpart} " \
> "${kernel_addr_r} efi/boot/"BOOTEFI_NAME"; " \
> "bootefi ${kernel_addr_r} ${efi_fdt_addr};\0" \
> \
> "load_efi_dtb=" \
> "efi_dtb_prefixes=/ /dtb/ /dtb/current/" \
You want to keep those in a global variable so they can get adapted in
the environment easily. Also I think the way you wrote it right now
fails to execute anyway, as it's missing a semicolon :).
> "for prefix in ${efi_dtb_prefixes}; do " \
> "load ${devtype} ${devnum}:${distro_bootpart} " \
> "${fdt_addr_r} ${prefix}${efi_fdtfile};" \
Will this fail silently or throw error messaged on the screen? If there
are error messages, you need to keep the test -e around :). We don't
want to confuse users with bogus error messages.
> "if fdt addr ${fdt_addr_r}; then " \
> "setenv efi_fdt_addr ${fdt_addr_r}; " \
> "fi; " \
> "done;\0" \
> \
> "set_efi_fdt_addr=" \
> "efi_fdtfile=${fdtfile}; " \
> BOOTENV_EFI_SET_FDTFILE_FALLBACK \
> "setenv efi_fdt_addr ${fdtcontroladdr}; " \
> "if test x${efi_fdtfile} != x -a x{$fdt_addr_r} != x ; then " \
> "run load_efi_dtb; " \
> "fi;\0" \
> "setenv efi_fdtfile\0" \
> \
> "scan_dev_for_efi=" \
> "run set_efi_fdt_addr; " \
> "bootefi bootmgr ${efi_fdt_addr};" \
So we're running the bootmgr over and over again for every target
device? I don't think that's very useful. It should only run once.
I'm personally also perfectly ok with not supporting dynamic DT loading
with bootmgr. The whole dynamic DT loading logic is only there for
boards where we don't have a stable DT yet. On those we usually don't
have bootmgr support either, because we're lacking persistent variable
storage ;).
Alex
> "if test -e ${devtype} ${devnum}:${distro_bootpart} " \
> "efi/boot/"BOOTEFI_NAME"; then " \
> "echo Found EFI removable media binary " \
> "efi/boot/"BOOTEFI_NAME"; " \
> "run boot_efi_binary; " \
> "echo EFI LOAD FAILED: continuing...; " \
> "fi; " \
> "setenv efi_fdt_addr\0"
> ===>8===
> (not tested yet)
>
> -Takahiro Akashi
>
>> Alex
>>
>>> + "setenv efi_fdtfile\0" \
>>> + \
>>> + "efi_dtb_prefixes=/ /dtb/ /dtb/current/\0" \
>>> + "scan_dev_for_efi=" \
>>> + "run set_efi_fdt_addr; " \
>>> + "bootefi bootmgr ${efi_fdt_addr};" \
>>> "if test -e ${devtype} ${devnum}:${distro_bootpart} " \
>>> "efi/boot/"BOOTEFI_NAME"; then " \
>>> "echo Found EFI removable media binary " \
>>> "efi/boot/"BOOTEFI_NAME"; " \
>>> "run boot_efi_binary; " \
>>> "echo EFI LOAD FAILED: continuing...; " \
>>> - "fi; " \
>>> - "setenv efi_fdtfile\0"
>>> + "fi; " \
>>> + "setenv efi_fdt_addr\0"
>>> #define SCAN_DEV_FOR_EFI "run scan_dev_for_efi;"
>>> #else
>>> #define BOOTENV_SHARED_EFI
>>>
More information about the U-Boot
mailing list