Hello,

On 19.06.2013 17:35, Daniel Waddington wrote:

Hi, I recently changed to using the Fiasco.OC source from Genode. I have just noticed problems with the bootstrap not outputting to the console on certain machines. The h/w is checked as ok, and if we replace bootstrap with r56 version from TU Dresden then everything is ok.

Are there some Genode-specific changes to bootstrap on r46?

Do you maybe referring to r47 ? For this branch a commit was added to read out the Bios Data Area and looking the I/O ports of the first serial device [0] reported by the Bios.

If you boot with 'bender' before bootstrap (which is the case for the pulsar configs generated by the Genode run scripts) then in the BDA 'bender' will write the i/o ports of the first PCI device implementing serial output support.

If you just use 'bootstrap' with the above commit the values of the Bios of the PC will be taken if there are any. If the values are bogus for some reason, then it could explain your observation. For such machines you may want to use the "-comport" option of bootstrap and use some default values like 0x3f8.

Cheers,

Alex.

[0] https://github.com/ssumpf/foc/commit/203cd7607436a847facdc952ab999d6e63c9ce1...

P.S. I didn't notice the problem until I did a full rebuild so I'm not sure when the issue crept in.

Daniel

---

This SF.net email is sponsored by Windows:

Build for Windows Store.

http://p.sf.net/sfu/windows-dev2dev _______________________________________________ Genode-main mailing list Genode-main@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/genode-main