Skip to contentSkip to conversation

is bzImage container large enough?

DISCUSSION Samium Gromoff 5 participants 6 messages Sep 1, 2001

lore
Share & tools
lore
#0 lore raw

If one wanting to turn on virtually every kernel CONFIG_* option
in order to check if the kernel compiles and then report possible gcc errors to lkml, will the resulting kernel fit the bzImage format? Will the bootup be possible?

thanks in advance...

cheers,
Sam

#1 lore raw
Quoted 3 lines · Samium Gromoff

If one wanting to turn on virtually every kernel CONFIG_* option
in order to check if the kernel compiles and then report possible gcc errors to lkml, will the resulting kernel fit the bzImage format?

No, it is far too big.

BTW, if you want to test compiles against various combinations of config, there are kbuild patches that add make allyes, make allno, make allmod and make randconfig. Included in separate mail.

#2 lore raw
Quoted 1 line · Keith Owens

allmod and make randconfig. Included in separate mail.

randconfig?

This scares me. :->

Btw, are such things independent of CML being used? Was wondering if a "randconfig" might use CML2 to validate the configuration (as far as CML2 knows about it). _Might_ be useful to enhance CML2 to track down incompatible configurations.

mrc

Signature · 3 lines
     Mike Castle      [email protected]      www.netcom.com/~dalgoda/
    We are all of us living in the shadow of Manhattan.  -- Watchmen
fatal ("You are in a maze of twisty compiler features, all different"); -- gcc
#3 lore raw
Quoted 6 lines · Mike Castle
Quoted 1 line · Keith Owens

allmod and make randconfig. Included in separate mail.

randconfig?

This scares me. :->

The config is random but valid, it passes the CML1 validation checks. randconfig is useful for finding errors in the CML1 checks, it also finds errors in code which assume that a feature is always present.

Quoted 4 lines

Btw, are such things independent of CML being used? Was wondering if a "randconfig" might use CML2 to validate the configuration (as far as CML2 knows about it). _Might_ be useful to enhance CML2 to track down incompatible configurations.

At the moment, randconfig, allyes, allno, allmod only work for CML1. Doing randconfig in CML2 is awkward, it may be added later.

#4 lore raw
In reply to Keith Owens
Quoted 7 lines · Keith Owens
Quoted 3 lines · Samium Gromoff

If one wanting to turn on virtually every kernel CONFIG_* option
in order to check if the kernel compiles and then report possible gcc errors to lkml, will the resulting kernel fit the bzImage format?

No, it is far too big.

We just "fixed" this sort of problem for ppc64 the other day. Course we don't use bzImages but rather zImages but none the less it was still anoying since we can boot zImages over the network and
of course that makes it quite reasonable to blow past the 1.44 floppy limitation. The bug we fixed was that an uncompressed kernel could only be up to 4 meg in size, if it was larger, at uncompression time you'd just lose everything past the 4M mark. Todd Inglett raised the limit to 8 meg for us, and that's a mighty large penguin....

Regards,

Tom

--
Tom Gall - PPC64 Code Monkey "Where's the ka-boom? There was
Linux Technology Center supposed to be an earth
(w) [email protected] shattering ka-boom!"
(w) 507-253-4558 -- Marvin Martian
(h) [email protected]
http://www.ibm.com/linux/ltc/projects/ppc

#5 lore raw
In reply to Keith Owens
Quoted 6 lines

The bug we fixed was that an uncompressed
kernel could only be up to 4 meg in size, if it was larger, at uncompression time you'd just lose everything past the 4M mark. Todd Inglett raised the limit to 8 meg for us, and that's a mighty large penguin....

Computing the size at link time is neater and this is what he actually does.

Anton