Hi,
When the last output of a make does not exist (in this case the linked
menu editor) in some configurations it will re-make the intermediates
and in this case I believe it's needed.
Keep in mind that make menuconfig is not ./configure that tests for
dependencies.
Since a make is usually I/O bound, I would assume it will be much
slower on a VM of any type.
David Kohen

On Tue, Mar 8, 2011 at 5:33 AM, Rob Landley <[email protected]> wrote:
> On 03/07/2011 08:50 PM, Denys Vlasenko wrote:
>> On Tuesday 08 March 2011 03:18, Rob Landley wrote:
>>> $ time make menuconfig
>>> ....
>>> real 1m6.573s
>>> user 0m56.292s
>>> sys  0m9.462s
>>>
>>> Now in that time it eventually figures out that ncurses isn't installed,
>>> but sits there fore over a minute before attempting to build the bits
>>> that need ncurses, which then time out.
>>
>> This does not happen to me. Can you determine where exactly it waits,
>> and for what?
>>
>
> It's gen_build_files.
>
> I'm running it under qemu, so I'm not getting my full laptop speed.
> Possibly this is just considered an acceptable speed on modern hardware.
>  Except that it's not just doing it the first time, it's doing it EVERY
> time I run menuconfig.  The generated files are already generated, and
> none of the *.c or *.h files changed, but it's doing it again every time.
>
> (The ncurses bits didn't time out, they died because I didn't have
> ncurses installed.  Fixed it.  Sorry, my cold is flaring back up and I'm
> getting all unfocused...)
>
> Rob
> _______________________________________________
> busybox mailing list
> [email protected]
> http://lists.busybox.net/mailman/listinfo/busybox
>
_______________________________________________
busybox mailing list
[email protected]
http://lists.busybox.net/mailman/listinfo/busybox

Reply via email to