On Fri, 1 Oct 2021 12:09:04 +0200 Pali Rohár pali@kernel.org wrote:
Hello!
On Friday 01 October 2021 11:58:14 Stefan Roese wrote:
I checked by quickly starting a terminal app and it does not seem that the CPU did go through a reboot. Nothing on the UART at this stage.
BUT: The tests I did in this mail were done with your 39 patches applied on top of "master-kwboot" [1]. Here booting with 115k works as explained below.
Now I applied the same 39 patches on top of "next" [2]. And here not even booting with 115k works on my AXP target. It always hangs at this stage:
So default baudrate 115200 works fine on master+39patches and does not work on next+39patches? Could you check if it works on next without applying any patch?
This seems as if the ACK was not sent by the BootROM.
Stefan, could you apply the patch below, recompile and then send the output again?
diff --git a/tools/kwboot.c b/tools/kwboot.c index 6a1a030308..d433ec6d9c 100644 --- a/tools/kwboot.c +++ b/tools/kwboot.c @@ -865,6 +865,7 @@ kwboot_xm_recv_reply(int fd, char *c, int allow_non_xm, int *non_xm_print, while (1) { rc = kwboot_tty_recv(fd, c, 1, timeout); if (rc) { + printf("rc=%i errno=%i allow_non_xm=%i *non_xm_print=%i\n", rc, errno, allow_non_xm, allow_non_xm ? *non_xm_print : -1); if (errno != ETIMEDOUT) return rc; else if (allow_non_xm && *non_xm_print)