[commit] r5900 - trunk/src/lib

[repository service]

Author: myles
Date: Fri Oct  1 17:23:41 2010
New Revision: 5900
URL: https://tracker.coreboot.org/trac/coreboot/changeset/5900

Log:
Fix make warnings:

Makefile:261: warning: overriding commands for target `coreboot-builds/a-trend_atc-6220/lib/lzma.ramstage.o'
Makefile:261: warning: ignoring old commands for target `coreboot-builds/a-trend_atc-6220/lib/lzma.ramstage.o'

lzma.c is already included unconditionally in the same file.

Signed-off-by: Myles Watson 
Acked-by: Myles Watson

Modified:
   trunk/src/lib/Makefile.inc

Modified: trunk/src/lib/Makefile.inc
==============================================================================
--- trunk/src/lib/Makefile.inc	Fri Oct  1 17:16:20 2010	(r5899)
+++ trunk/src/lib/Makefile.inc	Fri Oct  1 17:23:41 2010	(r5900)
@@ -28,8 +28,6 @@

ramstage-$(CONFIG_USBDEBUG) += usbdebug.c

-ramstage-$(CONFIG_COMPRESSED_PAYLOAD_LZMA) += lzma.c
-
 ramstage-$(CONFIG_BOOTSPLASH) += jpeg.c

smm-y += memcpy.c

[Peter Stuge]

repository service wrote:
> lzma.c is already included unconditionally in the same file.

..
> -ramstage-$(CONFIG_COMPRESSED_PAYLOAD_LZMA) += lzma.c

Why include lzma.c unconditionally if the payload will not be
compressed?

//Peter

[Myles Watson]

> repository service wrote:
> > lzma.c is already included unconditionally in the same file.
>
> ..
> > -ramstage-$(CONFIG_COMPRESSED_PAYLOAD_LZMA) += lzma.c
>
> Why include lzma.c unconditionally if the payload will not be
> compressed?

CONFIG_COMPRESSED_PAYLOAD_LZMA != CONFIG_COMPRESSED_PAYLOAD_LZMA_SUPPORTED
This config option is for adding the payload, not for all payloads.

Right now we don't have a good way to advertise lzma support in a coreboot
image.

If build coreboot with no payload (CONFIG_COMPRESSED_PAYLOAD_LZMA = n), then
add a payload with cbfstool, how does cbfstool know that you can't handle a
compressed payload?

In selfboot.c, the code assumes that lzma support is always present for that
reason.

I don't know how much difference it makes.  You'd have to modify a few
places in the code to be able to build an image without lzma support.

Thanks,
Myles

[Peter Stuge]

Myles Watson wrote:
> > > lzma.c is already included unconditionally in the same file.
> >
> > Why include lzma.c unconditionally if the payload will not be
> > compressed?

..
> If build coreboot with no payload (CONFIG_COMPRESSED_PAYLOAD_LZMA = n),
> then add a payload with cbfstool, how does cbfstool know that you can't
> handle a compressed payload?

Good point. Thanks!

//Peter

[Myles Watson]

> Myles Watson wrote:
> > > > lzma.c is already included unconditionally in the same file.
> > >
> > > Why include lzma.c unconditionally if the payload will not be
> > > compressed?
>
> ..
> > If build coreboot with no payload (CONFIG_COMPRESSED_PAYLOAD_LZMA = n),
> > then add a payload with cbfstool, how does cbfstool know that you can't
> > handle a compressed payload?
>
> Good point. Thanks!

You're welcome.  It was a good question.  Sorry about the grammar of the
response :)  "If you build coreboot ..."

Thanks,
Myles