[ipxe-devel] Regression with commit fc87adb46c1395b09302085e9d15fcd8ab3c31fe - [efi] Expose downloaded images via EFI_SIMPLE_FILE_SYSTEM_PROTOCOL on IMAGE_EFI

[Joshua C.]

Actually I'm built the rom image and wanted to test this before flasing it.
As I said my platform supports efi and I'd like to have the efi support so
that i can load and boot an efi image when stil in POST. This should be
possible since I'm starting the shell once ipxe has initilized (I've done
this in the past: redirect the shell to an efi image). That's why I'd like
to have the ef support build in in the rom.


2013/4/6 Michael Brown <mbrown at fensystems.co.uk>

> On 06/04/13 09:52, Joshua C. wrote:
>
>> Commit fc87adb46c1395b09302085e9d15fc**d8ab3c31fe ([efi] Expose
>> downloaded
>> images via EFI_SIMPLE_FILE_SYSTEM_**PROTOCOL) introduced a regression
>> when
>> compiling ipxe with IMAGE_EFI support. In my case initialise(); finishes
>> fine but startup(); cannot finish and my mashine restarts all the time.
>> After reverting this commit (and all other commits that introduced
>> changes to those files) everything works fine again. I can circumvent
>> this with disabling the IMAGE_EFI support. Then everything works fine
>> again.
>>
>> Testing was done with 10ec8168.lkrn on a p67 chipset (ami bios with efi
>> support). Can someone take a look at this?
>>
>
> Do not enable IMAGE_EFI when building a non-EFI version of iPXE. There's
> no way to load an EFI binary on a non-EFI platform anyway, so it makes no
> sense to ask for IMAGE_EFI support.
>
> IMAGE_EFI is enabled by default when building EFI versions of iPXE. There
> should never be a need to explicitly enable it.
>
> Michael
>



-- 
--joshua
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ipxe.org/pipermail/ipxe-devel/attachments/20130406/cde41f61/attachment.htm>