Re: 32-bit powerpc kernel builds (head -r356187): old ld (works) vs. devel/binutils@powerpc based (fails to boot): DYNAMIC vs. EXEC_P (powerpc64 lld too)
Mark Millard via freebsd-toolchain <[email protected]>
| Newsgroups | gmane.os.freebsd.devel.toolchain,gmane.os.freebsd.devel.ppc |
|---|---|
| Message-ID | <[email protected]> |
On 2019-Dec-31, at 16:44, Mark Millard <marklmi at yahoo.com> wrote: > On 2019-Dec-31, at 14:52, Mark Millard <marklmi at yahoo.com> wrote: > >> My attempt to buildkernel via devel/binutils@powerpc >> produces a kernel that gets a very early crash. >> >> Looking at the normal and alternate kernels a little >> shows. . . >> >> >> >> Old ld (and such): >> >> /boot/kernel/kernel: file format elf32-powerpc-freebsd >> /boot/kernel/kernel >> architecture: powerpc:common, flags 0x00000150: >> HAS_SYMS, DYNAMIC, D_PAGED >> start address 0x001001e0 >> . . . >> 00e7a034 l O *ABS* 00000000 .hidden _DYNAMIC >> >> Produced via (from kernel.full.meta): >> >> CMD @ld -m elf32ppc_fbsd -Bdynamic -T /usr/src/sys/conf/ldscript.powerpc --secure-plt -pie --no-warn-mismatch --warn-common --export-dynamic --dynamic-linker /red/herring -X -o kernel.full locore.o . . . >> >> >> devel/binutils@powerpc based: >> >> /boot/kerbad/kernel: file format elf32-powerpc-freebsd >> /boot/kerbad/kernel >> architecture: powerpc:common, flags 0x00000112: >> EXEC_P, HAS_SYMS, D_PAGED >> start address 0x00100200 >> >> 00e7a034 l O .dynamic 00000000 _DYNAMIC >> >> Produced via (from kernel.full.meta): >> >> CMD @/usr/local/powerpc-unknown-freebsd13.0/bin/ld -m elf32ppc_fbsd -Bdynamic -T /usr/src/sys/conf/ldscript.powerpc --secure-plt --build-id=sha1 -pie --no-warn-mismatch --warn-common --export-dynamic >> --dynamic-linker /red/herring -X -o kernel.full locore.o . . . > I see the same sort of thing for powerpc64 kernels, based on system lld and such vs. based on devel/binutils@powerpc64 . (The old 2 socket/2-cores-per PowerMac is rebuilding the 435 ports on head -r356187 .) > _GLOBAL_OFFSET_TABLE_ has a similar status. > > In fact, there is a big difference in the two > context's ABS lists: devel/binutils@powerpc > produces a very short list: > > # readelf -a /boot/kerbad/kernel | grep "\<ABS\>" | more > 2: 0000000000000070 0 NOTYPE GLOBAL DEFAULT ABS dlmisssize > 569: 0000000000100100 0 NOTYPE GLOBAL DEFAULT ABS kernbase > 5103: 0000000000000020 0 NOTYPE GLOBAL DEFAULT ABS testppc64size > 8156: 0000000000000018 0 NOTYPE GLOBAL DEFAULT ABS restorebridgesize > 9078: 00000000000000b0 0 NOTYPE GLOBAL DEFAULT ABS imisssize > 12351: 00000000000000f0 0 NOTYPE GLOBAL DEFAULT ABS dsmisssize > 25923: 0000000000000070 0 NOTYPE GLOBAL DEFAULT ABS dlmisssize > 26490: 0000000000100100 0 NOTYPE GLOBAL DEFAULT ABS kernbase > 31024: 0000000000000020 0 NOTYPE GLOBAL DEFAULT ABS testppc64size > 34077: 0000000000000018 0 NOTYPE GLOBAL DEFAULT ABS restorebridgesize > 34999: 00000000000000b0 0 NOTYPE GLOBAL DEFAULT ABS imisssize > 38272: 00000000000000f0 0 NOTYPE GLOBAL DEFAULT ABS dsmisssize > > but the old ld produces a much longer list: > > # readelf -a /boot/kernel/kernel | grep "\<ABS\>" | more > 2: 0000000000000070 0 NOTYPE GLOBAL DEFAULT ABS dlmisssize > 212: 0000000000e793dc 0 NOTYPE GLOBAL DEFAULT ABS __start_set_gfb_set > 462: 0000000000e793c8 0 NOTYPE GLOBAL DEFAULT ABS __start_set_mmu_set > 569: 0000000000100100 0 NOTYPE GLOBAL DEFAULT ABS kernbase > 1334: 0000000000dd5728 0 NOTYPE GLOBAL DEFAULT ABS __start_set_sdt_probes_set > 1395: 0000000000e5e608 0 NOTYPE GLOBAL DEFAULT ABS __start_set_vnet > 1765: 0000000001183648 0 NOTYPE GLOBAL DEFAULT ABS end > 1798: 0000000000dd36d0 0 NOTYPE GLOBAL DEFAULT ABS __stop_set_sysinit_set > 1857: 0000000000dd4e34 0 NOTYPE GLOBAL DEFAULT ABS __stop_set_modmetadata_set . . . (detail dropped) . . . > 35963: 0000000000dd5674 0 NOTYPE GLOBAL DEFAULT ABS __stop_set_kbddriver_set > 35987: 0000000000dd5728 0 NOTYPE GLOBAL DEFAULT ABS __stop_set_sdt_providers_set > 36015: 0000000001183648 0 NOTYPE GLOBAL DEFAULT ABS _end > 36625: 0000000000e793d4 0 NOTYPE GLOBAL DEFAULT ABS __start_set_videodriver_set > 36828: 0000000000dd7954 0 NOTYPE GLOBAL DEFAULT ABS __start_set_scterm_set > 37429: 0000000000e9786c 0 NOTYPE GLOBAL DEFAULT ABS _edata > 37854: 0000000000dd7988 0 NOTYPE GLOBAL DEFAULT ABS __start_set_vt_drv_set > 38049: 0000000000dc2240 0 NOTYPE GLOBAL DEFAULT ABS __stop_set_sysctl_set > 38270: 00000000000000f0 0 NOTYPE GLOBAL DEFAULT ABS dsmisssize > > So what do the __start_set_* and __stop_set_* symbols > show up as in /boot/kerbad/kernel generally? PROTECTED > visibility as a GLOBAL as it turns out: > > # readelf -a /boot/kerbad/kernel | egrep "__(start|stop)_set_" | more > 00e83680 002c0701 R_PPC_ADDR32 00000000 __stop_set_uart_fdt_class_set + 0 > 00e83684 0021a801 R_PPC_ADDR32 00000000 __start_set_uart_fdt_class_set + 0 > 212: 0000000000e793dc 0 NOTYPE GLOBAL PROTECTED 48 __start_set_gfb_set > 462: 0000000000e793c8 0 NOTYPE GLOBAL PROTECTED 45 __start_set_mmu_set . . . (detail dropped) . . . > 35681: 0000000000dd565c 0 NOTYPE GLOBAL PROTECTED 29 __start_set_kdb_dbbe_set > 35695: 0000000000dd5668 0 NOTYPE GLOBAL PROTECTED 30 __start_set_kbddriver_set > 35965: 0000000000dd5674 0 NOTYPE GLOBAL PROTECTED 30 __stop_set_kbddriver_set > 35989: 0000000000dd5728 0 NOTYPE GLOBAL PROTECTED 33 __stop_set_sdt_providers_set > 36627: 0000000000e793d4 0 NOTYPE GLOBAL PROTECTED 46 __start_set_videodriver_set > 36830: 0000000000dd7954 0 NOTYPE GLOBAL PROTECTED 36 __start_set_scterm_set > 37192: 0000000000000000 0 NOTYPE WEAK DEFAULT UND __stop_set_uart_fdt_class_set > 37856: 0000000000dd7988 0 NOTYPE GLOBAL PROTECTED 40 __start_set_vt_drv_set > 38051: 0000000000dd2468 0 NOTYPE GLOBAL PROTECTED 22 __stop_set_sysctl_set > > > > (Note: the "bad" in "/boot/kerbad/" is just a > reference to the boot failure, a relationship > of the link results and code handling the loaded > kernel code. Which style of linker output for the > kernel is "more correct" for the linker options > supplied is not part of my claims. I am noting > the mismatches, however.) > === Mark Millard marklmi at yahoo.com ( dsl-only.net went away in early 2018-Mar)