Hi Heinrich,
On Thu, 30 Dec 2021 at 23:18, Heinrich Schuchardt xypron.glpk@gmx.de wrote:
On 12/29/21 19:57, Simon Glass wrote:
Add info about how to select vidconsole or serial.
Also set up a demo boot command.
Signed-off-by: Simon Glass sjg@chromium.org
(no changes since v2)
Changes in v2:
Add a better boot command too
include/configs/efi-x86_app.h | 25 +++++++++++++++++++++++++ 1 file changed, 25 insertions(+)
diff --git a/include/configs/efi-x86_app.h b/include/configs/efi-x86_app.h index 6061a6db0a4..33afb7ca0f9 100644 --- a/include/configs/efi-x86_app.h +++ b/include/configs/efi-x86_app.h @@ -10,8 +10,33 @@
#undef CONFIG_TPM_TIS_BASE_ADDRESS
+/*
- Select the output device: Put an 'x' prefix before one of these to disable it
- */
+/*
- Video output - can normally continue after exit_boot_services has been
- called, since output to the display does not require EFI services at that
- point. U-Boot sets up the console memory and does its own drawing.
- */ #define CONFIG_STD_DEVICES_SETTINGS "stdin=serial\0" \ "stdout=vidconsole\0" \ "stderr=vidconsole\0"
+/*
- Serial output with no console. Run qemu with:
- -display none -serial mon:stdio
- This will hang or fail to output on the console after exit_boot_services is
- called.
In lib/efi/efi_stub.c, efi_main() you set use_uart=true. Couldn't you do a similar thing in the app?
Well use_uart is a bit of a hack - just using a hard-coded x86 UART. See putc() in efi_stub.c - it is just for debugging. There seem to be importants periods in EFI where debug output is not supported.
Instead of hard coding a specific UART it would be preferable both for the EFI stub as well as for the app to use the debug console defined by CONFIG_DEBUG_UART.
Well the app uses the con_out thing, so doesn't use a specific UART.
The stub uses the same thing while EFI is available, but uses the hack when it is not. The console used by DEBUG_UART should be the same driver as the 'serial' one selected here, so I think it is already what you want.
I suspect we can make the serial and video enable/disable at runtime but that seems like a separate problem.
Regards, Simon