On Wednesday 27 October 2021 17:27:41 Stefan Roese wrote:
On 27.10.21 16:10, Pali Rohár wrote:
<snip>
Perfect! So it really looks like that issue is in the code which resets baudrate back to the value 115200.
I have there another diff which removes usage of the stack in code which resets baudrate back to default value:
diff --git a/tools/kwboot.c b/tools/kwboot.c index b56c9a0c8104..8f0e50501398 100644 --- a/tools/kwboot.c +++ b/tools/kwboot.c @@ -1444,6 +1444,11 @@ _inject_baudrate_change_code(void *img, size_t *size, int pre, memcpy(code, kwboot_baud_code, codesz - 8); *(uint32_t *)(code + codesz - 8) = cpu_to_le32(old_baud); *(uint32_t *)(code + codesz - 4) = cpu_to_le32(new_baud);
- if (!pre) {
Ou, there is a mistake, it should be "if (pre) {"
*(uint32_t *)code = cpu_to_le32(0xe1a00000); /* arm nop */*(uint32_t *)(code + codesz - 4*7) = cpu_to_le32(0xe12fff1e); /* bx lr */- } } static int
Could you try to apply this change instead of my previous one?
Tested with this corrected patch:
97 % [......................................................................] 99 % [................... ] Done Finishing transfer Waiting 2s for baudrate change magic
Changing baudrate back to 115200 Bd
[Type Ctrl-\ + c to quit]
U-Boot 2021.10-00916-gc6142e537e88-dirty (Oct 27 2021 - 17:23:24 +0200)
SoC: MV78260-B0 at 1333 MHz I2C: ready DRAM: 2 GiB (667 MHz, 64-bit, ECC not enabled) Loading Environment from SPIFlash... SF: Detected m25p128 with page size 256 Bytes, erase size 256 KiB, total 16 MiB OK Model: Marvell Armada XP theadorable ...
Perfect. This works on Armada XP. :)
Nice! Therefore I will prepare proper patch for removing stack usage.
Still I see no speed change. But this is a different story...
Thanks, Stefan