Skip to contentSkip to conversation

board: samsung: e850-96: Fix MMC devices in EFI

PATCH In review Sam Protsenko 2 participants 7 messages Jun 20, 2025

lore
Share & tools
lore
#0 lore raw

cover board: samsung: e850-96: Fix MMC devices in EFI

Avoid using MMC in board_init() to make sure disk objects are created in EFI subsystem later.

If it's still possible, please take this series for 2025.07, as it makes EFI functional on E850-96 board.

Sam Protsenko (2):
configs: e850-96: Enable BOARD_LATE_INIT
board: samsung: e850-96: Load LDFW in board_late_init()

 board/samsung/e850-96/e850-96.c | 9 +++++++++
 configs/e850-96_defconfig       | 1 +
 2 files changed, 10 insertions(+)
Signature · 1 line
2.39.5
#1 lore raw

PATCH 1/2 configs: e850-96: Enable BOARD_LATE_INIT

LDFW firmware should be loaded in board_late_init() when MMC driver is already probed and initialized. Doing so in board_init() causes MMC driver to be probed earlier, which leads to EFI disk objects not being created. Enable CONFIG_BOARD_LATE_INIT to make that possible.

---
 configs/e850-96_defconfig | 1 +
 1 file changed, 1 insertion(+)
configs/e850-96_defconfig +1 -0
index 5facd418260c..e4c6aa28e0b7 100644
@@ -19,6 +19,7 @@ CONFIG_ANDROID_BOOT_IMAGE=y
 CONFIG_BOOTSTD_FULL=y
 CONFIG_DEFAULT_FDT_FILE="exynos850-e850-96.dtb"
 # CONFIG_DISPLAY_CPUINFO is not set
+CONFIG_BOARD_LATE_INIT=y
 CONFIG_CMD_BOOTEFI_SELFTEST=y
 CONFIG_CMD_ABOOTIMG=y
 CONFIG_CMD_NVEDIT_EFI=y
Signature · 1 line
2.39.5
#2 lore raw

PATCH 2/2 board: samsung: e850-96: Load LDFW in board_late_init()

In reply to Sam Protsenko

As stated in 5e847f7729b3 ("efi_loader: call efi_init_early() earlier"):

efi_init_early() creates an event hook for block device probing. It has to be called before any block device is probed.

Indeed, efi_bl_init() registers EVT_DM_POST_PROBE event, which calls efi_disk_probe() whenever any block device is probed. And to make that hook work, the initialization of all block devices was put after efi_init_early() in initcall_run_r():

    INITCALL(efi_init_early);
    INITCALL(initr_nand);
    INITCALL(initr_onenand);
    INITCALL(initr_mmc);

Because LDFW firmware is being read from MMC, attempt to load LDFW in board_init() causes MMC driver to be probed. And because board_init() is executed before efi_init_early(), the hook mentioned above won't work for MMC devices anymore. So EFI disk objects won't be created, which in turn makes the EFI subsystem non-functional, showing next symptoms:
- 'efidebug dh' output is empty
- attempt to add boot devices in 'eficonfig' shows this message:
"No block device found!"
- 'bootefi selftest $fdtcontroladdr' shows this warning:
"Cannot persist EFI variables without system partition"
- booting GRUB with 'bootefi' runs minimal GRUB shell which doesn't see any block devices as well, probably because EFI vars weren't passed

Load LDFW in board_late_init() instead, as it's called after efi_init_early(). This fixes the described problem and makes it possible to run EFI apps like GRUB correctly, add entries in 'eficonfig', and makes 'efivar --list' command in Linux rootfs actually show EFI variables.

The only user of LDFW at the moment is the TRNG driver, and it's probed later, only when it's going to be used (e.g. on "rng" command). So it's fine to load LDFW in board_late_init(). Now the corresponding call order will look like this:

    efi_init_early()
    initr_mmc()
      mmc_probe()
        EVT_DM_POST_PROBE -> efi_disk_probe()
    board_late_init()
      load_ldfw() -> fs_read(), blk_dread()
    exynos_trng_probe()
Fixes ccfd8de541a8 ("board: samsung: e850-96: Report LDFW loading failures")
Fixes f04e58cc9788 ("board: samsung: e850-96: Load LDFW firmware on board init")
---
 board/samsung/e850-96/e850-96.c | 9 +++++++++
 1 file changed, 9 insertions(+)
board/samsung/e850-96/e850-96.c +9 -0
index 0bef68d2fb20..3bbd95201b5f 100644
@@ -18,9 +18,18 @@ int dram_init_banksize(void)
 }
 
 int board_init(void)
+{
+	return 0;
+}
+
+int board_late_init(void)
 {
 	int err;
 
+	/*
+	 * Do this in board_late_init() to make sure MMC is not probed before
+	 * efi_init_early().
+	 */
 	err = load_ldfw();
 	if (err)
 		printf("ERROR: LDFW loading failed (%d)\n", err);
Signature · 1 line
2.39.5
#3 lore raw
In reply to Sam Protsenko
Quoted 6 lines · Sam Protsenko

Avoid using MMC in board_init() to make sure disk objects are created in EFI subsystem later.

If it's still possible, please take this series for 2025.07, as it makes EFI functional on E850-96 board.

Quoted 7 lines

Sam Protsenko (2):
configs: e850-96: Enable BOARD_LATE_INIT
board: samsung: e850-96: Load LDFW in board_late_init()

 board/samsung/e850-96/e850-96.c | 9 +++++++++
 configs/e850-96_defconfig       | 1 +
 2 files changed, 10 insertions(+)

These should be one patch and also I think you want to disable CONFIG_BOARD_INIT ?

Signature · 1 line
Tom
#4 lore raw
Quoted 13 lines · Tom Rini
Quoted 10 lines · Sam Protsenko

Avoid using MMC in board_init() to make sure disk objects are created in EFI subsystem later.

If it's still possible, please take this series for 2025.07, as it makes EFI functional on E850-96 board.

Sam Protsenko (2):
configs: e850-96: Enable BOARD_LATE_INIT
board: samsung: e850-96: Load LDFW in board_late_init()

Quoted 7 lines
Quoted 3 lines
 board/samsung/e850-96/e850-96.c | 9 +++++++++
 configs/e850-96_defconfig       | 1 +
 2 files changed, 10 insertions(+)

These should be one patch and also I think you want to disable CONFIG_BOARD_INIT ?

Thanks for the review! I'll send v2 in an hour. But I wasn't able to find CONFIG_BOARD_INIT option:

    $ find . -name Kconfig* -exec grep -Hn "\bBOARD_INIT" {} \;

initcall_run_r() seems to always call board_init() for ARM and other architectures:

    #if CONFIG_IS_ENABLED(ARM) || CONFIG_IS_ENABLED(RISCV) || \
        CONFIG_IS_ENABLED(SANDBOX)
        INITCALL(board_init);    /* Setup chipselects */
    #endif

Interesting thought though. I did a brief analysis, and it looks like there are 79 boards which have empty board_init() because it's not possible to disable it at the moment:

    $ grep -Pzrl "\bboard_init\(void\)\n{\n.*return 0;\n}\n" board/ | wc -l
    79

Do you think U-Boot could benefit from having such an option?

Thanks!

Quoted 2 lines

--
Tom

#5 lore raw
Quoted 35 lines · Sam Protsenko
Quoted 13 lines · Tom Rini
Quoted 10 lines · Sam Protsenko

Avoid using MMC in board_init() to make sure disk objects are created in EFI subsystem later.

If it's still possible, please take this series for 2025.07, as it makes EFI functional on E850-96 board.

Sam Protsenko (2):
configs: e850-96: Enable BOARD_LATE_INIT
board: samsung: e850-96: Load LDFW in board_late_init()

Quoted 7 lines
Quoted 3 lines
 board/samsung/e850-96/e850-96.c | 9 +++++++++
 configs/e850-96_defconfig       | 1 +
 2 files changed, 10 insertions(+)

These should be one patch and also I think you want to disable CONFIG_BOARD_INIT ?

Thanks for the review! I'll send v2 in an hour. But I wasn't able to find CONFIG_BOARD_INIT option:

    $ find . -name Kconfig* -exec grep -Hn "\bBOARD_INIT" {} \;

initcall_run_r() seems to always call board_init() for ARM and other architectures:

    #if CONFIG_IS_ENABLED(ARM) || CONFIG_IS_ENABLED(RISCV) || \
        CONFIG_IS_ENABLED(SANDBOX)
        INITCALL(board_init);    /* Setup chipselects */
    #endif
Quoted 8 lines

Interesting thought though. I did a brief analysis, and it looks like there are 79 boards which have empty board_init() because it's not possible to disable it at the moment:

    $ grep -Pzrl "\bboard_init\(void\)\n{\n.*return 0;\n}\n" board/ | wc -l
    79

Do you think U-Boot could benefit from having such an option?

Yes, it looks like we need to clean the code up to match what we have in Kconfig now, thanks!

Signature · 1 line
Tom
#6 lore raw
Quoted 41 lines · Tom Rini
Quoted 39 lines · Sam Protsenko
Quoted 13 lines · Tom Rini
Quoted 10 lines · Sam Protsenko

Avoid using MMC in board_init() to make sure disk objects are created in EFI subsystem later.

If it's still possible, please take this series for 2025.07, as it makes EFI functional on E850-96 board.

Sam Protsenko (2):
configs: e850-96: Enable BOARD_LATE_INIT
board: samsung: e850-96: Load LDFW in board_late_init()

Quoted 7 lines
Quoted 3 lines
 board/samsung/e850-96/e850-96.c | 9 +++++++++
 configs/e850-96_defconfig       | 1 +
 2 files changed, 10 insertions(+)

These should be one patch and also I think you want to disable CONFIG_BOARD_INIT ?

Thanks for the review! I'll send v2 in an hour. But I wasn't able to find CONFIG_BOARD_INIT option:

    $ find . -name Kconfig* -exec grep -Hn "\bBOARD_INIT" {} \;

initcall_run_r() seems to always call board_init() for ARM and other architectures:

    #if CONFIG_IS_ENABLED(ARM) || CONFIG_IS_ENABLED(RISCV) || \
        CONFIG_IS_ENABLED(SANDBOX)
        INITCALL(board_init);    /* Setup chipselects */
    #endif

Interesting thought though. I did a brief analysis, and it looks like there are 79 boards which have empty board_init() because it's not possible to disable it at the moment:

Quoted 8 lines
Quoted 4 lines
    $ grep -Pzrl "\bboard_init\(void\)\n{\n.*return 0;\n}\n" board/ | wc -l
    79

Do you think U-Boot could benefit from having such an option?

Yes, it looks like we need to clean the code up to match what we have in Kconfig now, thanks!

On it. Would appreciate it if you can take v2 as is for v2025.07, it fixes EFI on E850-96.

Quoted 2 lines

--
Tom