* multiboot code asking for stubs-32.h under x86-64? @ 2007-10-04 6:28 Julio Meca Hansen 2007-10-04 20:53 ` Robert Millan 0 siblings, 1 reply; 6+ messages in thread From: Julio Meca Hansen @ 2007-10-04 6:28 UTC (permalink / raw) To: grub-devel Hi, I'm trying to compile grub2 from a cvs tarball but while it's compiling the code it stops at the multiboot subdirectory producing an error refered to not finding stubs-32.h, but I'm not compiling grub2 for i386, not even x86-64 in biarch mode, but x86-64 in pure mode, without *ANY* kind of 32bit library The steps I used to try to compile grub2 are the following: cat > config.cache < "EOF" grub_cv_check_edata_symbol=yes grub_cv_check_end_symbol=yes grub_cv_check_uscore_edata_symbol=yes grub_cv_check_uscore_end_symbol=yes grub_cv_check_uscore_start_symbol=yes grub_cv_check_uscore_uscore_bss_start_symbol=yes grub_cv_i386_check_nested_functions=no EOF ./configure --prefix=/usr --cache-file=config.cache make make install How can I cleanly compile grub2 without this error? grub-1.95 compiles just fine, but not the cvs version, so something must have changed since that... and 1.95 doesn't serve well my purpose as, if I recall correctly, you mentioned something in the previous month threads about 1.95 not supporting the journal on an ext3 partition? or does it work? Julio ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: multiboot code asking for stubs-32.h under x86-64? 2007-10-04 6:28 multiboot code asking for stubs-32.h under x86-64? Julio Meca Hansen @ 2007-10-04 20:53 ` Robert Millan 2007-10-04 21:50 ` Julio Meca Hansen 0 siblings, 1 reply; 6+ messages in thread From: Robert Millan @ 2007-10-04 20:53 UTC (permalink / raw) To: The development of GRUB 2 On Thu, Oct 04, 2007 at 08:28:51AM +0200, Julio Meca Hansen wrote: > > How can I cleanly compile grub2 without this error? grub-1.95 compiles > just fine, but not the cvs version, so something must have changed since > that... If the bug is not obvious, sounds like a regression test [1] would be helpful. [1] e.g. see http://www.winehq.org/site/docs/winedev-guide/x1344 -- Robert Millan <GPLv2> I know my rights; I want my phone call! <DRM> What use is a phone call, if you are unable to speak? (as seen on /.) ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: multiboot code asking for stubs-32.h under x86-64? 2007-10-04 20:53 ` Robert Millan @ 2007-10-04 21:50 ` Julio Meca Hansen 2007-10-05 11:59 ` Robert Millan 0 siblings, 1 reply; 6+ messages in thread From: Julio Meca Hansen @ 2007-10-04 21:50 UTC (permalink / raw) To: The development of GRUB 2 hmm... a regression test? but... the only thing that's not behaving properly in my almost-fully-compiled system is the search for stubs-32.h, I suppose the correct behaviour would be to search just for stubs.h or in the excessive case, for stubs-64.h, but as I have a pure64 system there's not any presence of 32bit libraries... anyway... if grub-1.95 can support journaled ext3 filesystems, it would suffice to me, but I'm not sure and I was quite afraid of messing something... to be honest, I'm not much of an experienced programmer in linux and performing a regression tests looks like a bit difficult, and you're refering to wine instructions, but my system, to be clear... just needs grub2 in order to obtain the 'stage 1' system status, that's... just pure console linux, no aditional libraries, not any kind of X server, no desktop, and so on... so I need this step successfully fulfilled (and to continue being honest, I thought maybe you as developers would have an idea of what could have been changed :S) Julio On Thu, 2007-10-04 at 22:53 +0200, Robert Millan wrote: > On Thu, Oct 04, 2007 at 08:28:51AM +0200, Julio Meca Hansen wrote: > > > > How can I cleanly compile grub2 without this error? grub-1.95 compiles > > just fine, but not the cvs version, so something must have changed since > > that... > > If the bug is not obvious, sounds like a regression test [1] would be helpful. > > [1] e.g. see http://www.winehq.org/site/docs/winedev-guide/x1344 > ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: multiboot code asking for stubs-32.h under x86-64? 2007-10-04 21:50 ` Julio Meca Hansen @ 2007-10-05 11:59 ` Robert Millan 2007-10-05 13:15 ` Julio Meca Hansen 0 siblings, 1 reply; 6+ messages in thread From: Robert Millan @ 2007-10-05 11:59 UTC (permalink / raw) To: The development of GRUB 2 On Thu, Oct 04, 2007 at 11:50:07PM +0200, Julio Meca Hansen wrote: > hmm... a regression test? but... the only thing that's not behaving > properly in my almost-fully-compiled system is the search for > stubs-32.h, I suppose the correct behaviour would be to search just for > stubs.h or in the excessive case, for stubs-64.h, but as I have a pure64 > system there's not any presence of 32bit libraries... > > anyway... if grub-1.95 can support journaled ext3 filesystems, it would > suffice to me, but I'm not sure and I was quite afraid of messing > something... > > to be honest, I'm not much of an experienced programmer in linux and > performing a regression tests looks like a bit difficult, and you're > refering to wine instructions, but my system, to be clear... just needs > grub2 in order to obtain the 'stage 1' system status, that's... > > just pure console linux, no aditional libraries, not any kind of X > server, no desktop, and so on... so I need this step successfully > fulfilled Why are you compiling GRUB in 64-bit mode then? It's not supported yet. Doesn't the build system automaticaly ressort to 32-bit build for you? -- Robert Millan <GPLv2> I know my rights; I want my phone call! <DRM> What use is a phone call, if you are unable to speak? (as seen on /.) ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: multiboot code asking for stubs-32.h under x86-64? 2007-10-05 11:59 ` Robert Millan @ 2007-10-05 13:15 ` Julio Meca Hansen 2007-10-05 20:45 ` Robert Millan 0 siblings, 1 reply; 6+ messages in thread From: Julio Meca Hansen @ 2007-10-05 13:15 UTC (permalink / raw) To: The development of GRUB 2 On Fri, 2007-10-05 at 13:59 +0200, Robert Millan wrote: > On Thu, Oct 04, 2007 at 11:50:07PM +0200, Julio Meca Hansen wrote: > > hmm... a regression test? but... the only thing that's not behaving > > properly in my almost-fully-compiled system is the search for > > stubs-32.h, I suppose the correct behaviour would be to search just for > > stubs.h or in the excessive case, for stubs-64.h, but as I have a pure64 > > system there's not any presence of 32bit libraries... > > > > anyway... if grub-1.95 can support journaled ext3 filesystems, it would > > suffice to me, but I'm not sure and I was quite afraid of messing > > something... > > > > to be honest, I'm not much of an experienced programmer in linux and > > performing a regression tests looks like a bit difficult, and you're > > refering to wine instructions, but my system, to be clear... just needs > > grub2 in order to obtain the 'stage 1' system status, that's... > > > > just pure console linux, no aditional libraries, not any kind of X > > server, no desktop, and so on... so I need this step successfully > > fulfilled > > Why are you compiling GRUB in 64-bit mode then? It's not supported yet. > > Doesn't the build system automaticaly ressort to 32-bit build for you? > You leave me quite 'impressed' in the desillusional way haha but it's ok, but try to understand my point of view for a moment... I have a 64bit capable machine so I would like to have a 64bit system, the previous time I built my 64bit system, used grub-0.97 because I actually had a 32bit installation in another partition, so it did the trick for me, but if I'm going for the *absolute* way I can't have a 32bit bootloader but a 64bit one, that's the reason I want to compile grub2 for 64bit mode and you have to take into account my toolchain is non-multilib (at least that's what suggests the installation method), so I cannot produce any kind of 32bit code, but from the console output is like I *almost* managed to get it done, just this stubs-32.h non-existent reference blew my attempt... And in the installation instructions I posted... they belong to the CLFS Trac, and if the author (Joe Ciccone) stands grub2 can be compiled with the given options, at least we can confirm that grub2 actually *compiles* in pure 64bit mode, but not cvs grub2, that's what I asked what could I do with this error, just that... :( Isn't really any solution to this? Julio ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: multiboot code asking for stubs-32.h under x86-64? 2007-10-05 13:15 ` Julio Meca Hansen @ 2007-10-05 20:45 ` Robert Millan 0 siblings, 0 replies; 6+ messages in thread From: Robert Millan @ 2007-10-05 20:45 UTC (permalink / raw) To: The development of GRUB 2 On Fri, Oct 05, 2007 at 03:15:13PM +0200, Julio Meca Hansen wrote: > > > > Why are you compiling GRUB in 64-bit mode then? It's not supported yet. > > > > Doesn't the build system automaticaly ressort to 32-bit build for you? > > > > You leave me quite 'impressed' in the desillusional way haha but it's > ok, but try to understand my point of view for a moment... > > I have a 64bit capable machine so I would like to have a 64bit system, > the previous time I built my 64bit system, used grub-0.97 because I > actually had a 32bit installation in another partition, so it did the > trick for me, but if I'm going for the *absolute* way I can't have a > 32bit bootloader but a 64bit one, that's the reason I want to compile > grub2 for 64bit mode and you have to take into account my toolchain is > non-multilib (at least that's what suggests the installation method), so > I cannot produce any kind of 32bit code, but from the console output is > like I *almost* managed to get it done, just this stubs-32.h > non-existent reference blew my attempt... > > And in the installation instructions I posted... they belong to the CLFS > Trac, and if the author (Joe Ciccone) stands grub2 can be compiled with > the given options, at least we can confirm that grub2 actually > *compiles* in pure 64bit mode, but not cvs grub2, that's what I asked > what could I do with this error, just that... :( Most of GRUB 2 compiles (and works) in 64-bit mode. See grub-emu for example. > Isn't really any solution to this? As I said, if you can't find the reason for this problem, maybe a regression test can help. There can be other ways to find it, of course, but that's an easy one if you don't know where to look. -- Robert Millan <GPLv2> I know my rights; I want my phone call! <DRM> What use is a phone call, if you are unable to speak? (as seen on /.) ^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2007-10-05 20:45 UTC | newest] Thread overview: 6+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2007-10-04 6:28 multiboot code asking for stubs-32.h under x86-64? Julio Meca Hansen 2007-10-04 20:53 ` Robert Millan 2007-10-04 21:50 ` Julio Meca Hansen 2007-10-05 11:59 ` Robert Millan 2007-10-05 13:15 ` Julio Meca Hansen 2007-10-05 20:45 ` Robert Millan
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.