Re: do I really need "make-kpkg clean"?
On Wed, 04 May 2005 17:35:10 -0400, H S <greatexcalibur@yahoo.com>
said:
> Apparently, _Manoj Srivastava_, on 05/04/2005 03:21 PM,typed:
>> Are you sure that tine change shall be propagated to the kernel
>> that shall be built?
> I don't understant your point. If I make a change in my config file,
> whether I do *real* cleaning before compiling or not shouldn't
> matter to the issue of whether I should have a change in my
> kernel. However, if I do not do *real* cleaning, then the make
> utility should take care of what to change in the compilation and
> leave the rest alone thus saving time.
Ideally, yes. In practice, there have been glitches enough in
the kernel build systems to make that not the case, and debugging
this remotely for an novice user is not easy -- or even possible, at
times. So, for the novice user, at the expense of some time, they get
a fully deterministic build.
>>> Reberto has mentioned 'do_clean := NO' method. Have you ever used
>>> it? If not, what you do you use to avoid unnecessary
>>> recompilations during tweaking of the config file?
>> No, that won't do it. The do_clean := NO ensures that the tree is
>> not cleaned _after_ the .deb is built.
> Okay. But does that mean if I do not have that option, then the tree
> is cleaned?
Yes, after the .deb is created, the kernel source tree is
cleaned.
> I tried to recompile my kernel without "make-kpkg clean" and it took
> around only 1.5 minutes (due to a minor change in the config
> file). Had cleaning been done after the previous deb was
> created(since I do have that option), surely it should have taken
> much longer than that.
,----[ excerpt from the kernel package rules ]
| chmod -R og=rX $(IMAGE_TOP)
| chown -R root:root $(IMAGE_TOP)
| dpkg --build $(IMAGE_TOP) $(DEB_DEST)
| rm -f -r $(IMAGE_TOP)
| ifeq ($(strip $(do_clean)),YES)
| $(MAKE) $(EXTRAV_ARG) $(FLAV_ARG) $(CROSS_ARG) ARCH=$(KERNEL_ARCH) clean
| rm -f stamp-build
| endif
| echo done > stamp-image
`----
manoj
--
Many a family tree needs trimming.
Manoj Srivastava <srivasta@acm.org> <http://www.golden-gryphon.com/>
1024D/BF24424C print 4966 F272 D093 B493 410B 924B 21BA DABB BF24 424C
Reply to: