I've reached an impass, all mine to be sure, but I hope you'll help me explore a hypothesis for a way forward.
I got up to the point of building Firefox-31.4esr, kicked it off on the Conroe box I've been building on, it ground-on for over an hour then died, I suppose over a dependency. <grump> If it's gonna die, I want it to die sooner! I've got an i7 box useful as a "compile engine", that would compile FF faster, or fail faster. Eventually I'm going to want to migrate it to other boxes anyway. So I updated my usual migrate/install script, migrated what I'd built so far to the i7 and fixed the oversight glitches that always arise. It's running now with my initial non-SMP, "any likely hardware", kernels 3.5.2 & 3.9.11. (3.10.6x-lts caused unbearable console noise over the USB thumb drives you may recall.) Now it's time to rebuild the kernel, with SMP & SMT for all four hyperthreaded cores. It keeps failing to compile, log file to follow. (One possible issue is 3.9.11 only goes as modern as Core2 & newer Xeon processors, which MAY not support i7's, but I laid down the 3.5.2 source and the compile results from LFS-7.2 Chapter 8, and that does compile with SMP. Another slight possibility is this may not be a retail CPU, though I doubt it's an engineering sample. Someone who works at Intel gave me the board. :) I've not had trouble like this compiling the kernel before. I've spent a couple days now making sure my migration script has made a clean copy. I'm really pretty sure what's working on the Conroe SHOULD also work on the i7. It does compile on the Conroe. I haven't tried "importing" a SMP version of 3.9.11 compiled on the Conroe. That may fail if there's something different about the i7/X58 APIC. So, should I try patching the kernel up beyond 3.9 & 3.10, looking for a "sweet spot" that will work (IIRC Ken suggested 3.14 might be one), or is that a waste of time? I'm already beyond the LFS-7.2 recommendation. Here's the main bits of the compile console log file, note two errors: HOSTLD scripts/kconfig/conf scripts/kconfig/conf --silentoldconfig Kconfig # # configuration written to .config # make[1]: Nothing to be done for `all'. HOSTCC arch/x86/tools/relocs CHK include/generated/uapi/linux/version.h CHK include/generated/utsrelease.h UPD include/generated/utsrelease.h CC kernel/bounds.s GEN include/generated/bounds.h CC arch/x86/kernel/asm-offsets.s GEN include/generated/asm-offsets.h CALL scripts/checksyscalls.sh <stdin>:1223:2: warning: #warning syscall finit_module not implemented [-Wcpp] CC scripts/mod/empty.o HOSTCC scripts/mod/mk_elfconfig MKELF scripts/mod/elfconfig.h ... CC arch/x86/kernel/acpi/boot.o CC arch/x86/kernel/acpi/cstate.o LD arch/x86/kernel/acpi/built-in.o CC arch/x86/kernel/apic/apic.o arch/x86/kernel/apic/apic.c: In function 'lapic_next_deadline': arch/x86/kernel/apic/apic.c:475:2: error: 'MSR_IA32_TSC_DEADLINE' undeclared (f$ arch/x86/kernel/apic/apic.c:475:2: note: each undeclared identifier is reported$ make[3]: *** [arch/x86/kernel/apic/apic.o] Error 1 make[2]: *** [arch/x86/kernel/apic] Error 2 make[1]: *** [arch/x86/kernel] Error 2 make: *** [arch/x86] Error 2 -- Paul Rogers [email protected] Rogers' Second Law: "Everything you do communicates." (I do not personally endorse any additions after this line. TANSTAAFL :-) -- http://www.fastmail.com - Same, same, but different... -- http://lists.linuxfromscratch.org/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
