In message 1058653888.774.80.camel@localhost you wrote:
Nothing wrong, or am I missing something?
My guess is that your tftp server correctly appends \0 to the error message and atftpd does not. Strange. Having a look at atftpd source code I suspect that its use of strncpy for the error message is the cause, but I think u-boot should not rely on the trailing \0, because the length of the error message is known anyway. I'll report when I have been able to test it.
The TFTP protocol explicitely requires the error string to be '\0' terminated:
Quote from RFC 1350 "THE TFTP PROTOCOL (REVISION 2)":
... 2 bytes 2 bytes string 1 byte ----------------------------------------- | Opcode | ErrorCode | ErrMsg | 0 | -----------------------------------------
Figure 5-4: ERROR packet
An ERROR packet (opcode 5) takes the form depicted in Figure 5-4. An ERROR packet can be the acknowledgment of any other type of packet. The error code is an integer indicating the nature of the error. A table of values and meanings is given in the appendix. (Note that several error codes have been added to this version of this document.) The error message is intended for human consumption, and should be in netascii. Like all other strings, it is terminated with a zero byte. ...
IMHO the "it is terminated with a zero byte" is mandatory, so no change to U-Boot sems to be necessary.
Best regards,
Wolfgang Denk