| 1 |
On Tue, Jan 23, 2018 at 10:03 AM, Corbin Bird <corbinbird@×××××××.net> wrote: |
| 2 |
> |
| 3 |
> |
| 4 |
> On 01/23/2018 01:42 AM, Alexander Kapshuk wrote: |
| 5 |
>> On Tue, Jan 23, 2018 at 4:34 AM, Corbin Bird <corbinbird@×××××××.net> wrote: |
| 6 |
>>> On 01/22/2018 11:56 AM, Rich Freeman wrote: |
| 7 |
>>>> On Mon, Jan 22, 2018 at 12:11 PM, Alexander Kapshuk |
| 8 |
>>>> <alexander.kapshuk@×××××.com> wrote: |
| 9 |
>>>>> On Mon, Jan 22, 2018 at 5:51 PM, Corbin Bird <corbinbird@×××××××.net> wrote: |
| 10 |
>>>>>> Anyone else getting this error? ( kernel 4.9.77 ) |
| 11 |
>>>>>> |
| 12 |
>>>>>>> CC fs/ext4/mballoc.o |
| 13 |
>>>>>>> CC fs/ext4/block_validity.o |
| 14 |
>>>>>>> CC fs/ext4/move_extent.o |
| 15 |
>>>>>>> CC fs/ext4/mmp.o |
| 16 |
>>>>>>> CC fs/ext4/indirect.o |
| 17 |
>>>>>>> fs/ext4/indirect.o: warning: objtool: ext4_clear_blocks()+0x242: can't |
| 18 |
>>>>>>> find jump dest instruction at .text+0x56c |
| 19 |
>>>>>>> CC fs/ext4/extents_status.o |
| 20 |
>>>>>>> /bin/sh: line 1: 31735 Segmentation fault ./tools/objtool/objtool |
| 21 |
>>>>>>> check "fs/ext4/extents_status.o" |
| 22 |
>>>>>>> make[2]: *** [scripts/Makefile.build:294: fs/ext4/extents_status.o] |
| 23 |
>>>>>>> Error 139 |
| 24 |
>>>>>>> make[1]: *** [scripts/Makefile.build:544: fs/ext4] Error 2 |
| 25 |
>>>>>>> make: *** [Makefile:995: fs] Error 2 |
| 26 |
>>>>>> Corbin |
| 27 |
>>>>>> |
| 28 |
>>>>>> |
| 29 |
>>>>> Based on the output you supplied, objtool isn't happy with |
| 30 |
>>>>> fs/ext4/indirect.o and fs/ext4/extents_status.o. |
| 31 |
>>>>> Did you do 'make clean' prior to building the kernel? |
| 32 |
>>>>> If you did, it might be worth doing a 'make mrproper' before |
| 33 |
>>>>> rebuilding the kernel. |
| 34 |
>>>>> Don't forget to back up your .config before running mrproper. |
| 35 |
>>>>> |
| 36 |
>>>> While this is going to cause a bit more building, I personally tend to |
| 37 |
>>>> redirect kernel build output. I add O=/var/tmp/linux to all my make |
| 38 |
>>>> commands, which leaves /usr/src untouched and builds faster besides |
| 39 |
>>>> (tmpfs). |
| 40 |
>>>> |
| 41 |
>>>> The main downsides to this are: |
| 42 |
>>>> 1. Make can't re-use objects from previous builds, so rebuilds will |
| 43 |
>>>> go slower. Though, IMO this is worthwhile if re-using those objects |
| 44 |
>>>> causes issues (which I assume is due to imperfect makefiles). |
| 45 |
>>>> 2. When building packages that install kernel modules the prepared |
| 46 |
>>>> sources won't exist. This requires going back and re-preparing them |
| 47 |
>>>> (make O=/var/tmp/linux modules_prepare). For packages like zfs-kmod |
| 48 |
>>>> that use Module.symvers this is even more painful as you can only get |
| 49 |
>>>> that by rebuilding the whole thing. |
| 50 |
>>>> |
| 51 |
>>> Tried both approaches ... the results : |
| 52 |
>>>> make distclean |
| 53 |
>>>> make mrproper |
| 54 |
>>>> --> copy over .config |
| 55 |
>>>> make modules_prepare |
| 56 |
>>>> make && make modules_install |
| 57 |
>>>> |
| 58 |
>>>> CC fs/ext4/block_validity.o |
| 59 |
>>>> CC fs/ext4/move_extent.o |
| 60 |
>>>> CC fs/ext4/mmp.o |
| 61 |
>>>> CC fs/ext4/indirect.o |
| 62 |
>>>> fs/ext4/indirect.o: warning: objtool: ext4_clear_blocks()+0x242: can't |
| 63 |
>>>> find jump dest instruction at .text+0x56c |
| 64 |
>>>> CC fs/ext4/extents_status.o |
| 65 |
>>>> /bin/sh: line 1: 24095 Segmentation fault ./tools/objtool/objtool |
| 66 |
>>>> check "fs/ext4/extents_status.o" |
| 67 |
>>>> make[2]: *** [scripts/Makefile.build:294: fs/ext4/extents_status.o] |
| 68 |
>>>> Error 139 |
| 69 |
>>>> make[1]: *** [scripts/Makefile.build:544: fs/ext4] Error 2 |
| 70 |
>>>> make: *** [Makefile:995: fs] Error 2 |
| 71 |
>>>> |
| 72 |
>>>> |
| 73 |
>>> Try 2 : |
| 74 |
>>>> make distclean |
| 75 |
>>>> make mrproper |
| 76 |
>>>> --> copy over .config |
| 77 |
>>>> make O=/var/tmp/linux modules_prepare |
| 78 |
>>>> make O=/var/tmp/linux && make O=/var/tmp/linux modules_install |
| 79 |
>>>> |
| 80 |
>>>> CC fs/ext4/move_extent.o |
| 81 |
>>>> CC fs/ext4/mmp.o |
| 82 |
>>>> CC fs/ext4/indirect.o |
| 83 |
>>>> fs/ext4/indirect.o: warning: objtool: ext4_clear_blocks()+0x242: can't |
| 84 |
>>>> find jump dest instruction at .text+0x56c |
| 85 |
>>>> CC fs/ext4/extents_status.o |
| 86 |
>>>> /bin/sh: line 1: 21028 Segmentation fault ./tools/objtool/objtool |
| 87 |
>>>> check "fs/ext4/extents_status.o" |
| 88 |
>>>> make[3]: *** [/usr/src/linux-4.9.77-gentoo/scripts/Makefile.build:294: |
| 89 |
>>>> fs/ext4/extents_status.o] Error 139 |
| 90 |
>>>> make[2]: *** [/usr/src/linux-4.9.77-gentoo/scripts/Makefile.build:544: |
| 91 |
>>>> fs/ext4] Error 2 |
| 92 |
>>>> make[1]: *** [/usr/src/linux-4.9.77-gentoo/Makefile:995: fs] Error 2 |
| 93 |
>>>> make[1]: Leaving directory '/var/tmp/linux' |
| 94 |
>>>> make: *** [Makefile:150: sub-make] Error 2 |
| 95 |
>>> Try 2 gave an error warning about needing an 'make mrproper' ... the error : |
| 96 |
>>>> CHK include/config/kernel.release |
| 97 |
>>>> UPD include/config/kernel.release |
| 98 |
>>>> Using /usr/src/linux-4.9.77-gentoo as source for kernel |
| 99 |
>>>> /usr/src/linux-4.9.77-gentoo is not clean, please run 'make mrproper' |
| 100 |
>>>> in the '/usr/src/linux-4.9.77-gentoo' directory. |
| 101 |
>>>> make[1]: *** [/usr/src/linux-4.9.77-gentoo/Makefile:1021: prepare3] |
| 102 |
>>>> Error 1 |
| 103 |
>>>> make[1]: Leaving directory '/var/tmp/linux' |
| 104 |
>>>> make: *** [Makefile:150: sub-make] Error 2 |
| 105 |
>>> So ... I followed directions : |
| 106 |
>>>> Darkstar /usr/src/linux # cd /usr/src/linux-4.9.77-gentoo |
| 107 |
>>>> Darkstar /usr/src/linux-4.9.77-gentoo # make mrproper |
| 108 |
>>>> CLEAN scripts/basic |
| 109 |
>>>> CLEAN scripts/kconfig |
| 110 |
>>>> CLEAN .config |
| 111 |
>>> Still get the same error ... |
| 112 |
>>> |
| 113 |
>>> Corbin |
| 114 |
>>> |
| 115 |
>> Is this a distribution supplied kernel or vanilla or one with your own |
| 116 |
>> customisations/modifications? |
| 117 |
>> |
| 118 |
> |
| 119 |
> Using 'sys-kernel/gentoo-sources-4.9.77' with USE flags : modules |
| 120 |
> symlink experimental. |
| 121 |
> Both gcc && binutils have been upgraded to the latest stable / no signs |
| 122 |
> of them giving trouble. |
| 123 |
> |
| 124 |
> Beginning to think that something corrupted during the emerge process ... |
| 125 |
> |
| 126 |
> Corbin |
| 127 |
> |
| 128 |
> |
| 129 |
|
| 130 |
The 4.9.77 announcement upstream did introduce a number of objtool |
| 131 |
changes. See below. |
| 132 |
https://lkml.org/lkml/2018/1/17/198 |
| 133 |
Josh Poimboeuf (4): |
| 134 |
objtool, modules: Discard objtool annotation sections for modules |
| 135 |
objtool: Detect jumps to retpoline thunks |
| 136 |
objtool: Allow alternatives to be ignored |
| 137 |
objtool: Fix retpoline support for pre-ORC objtool |
| 138 |
|
| 139 |
I can't seem to remember seeing reports of the issue you're having on |
| 140 |
the LMKL, since 4.9.77 was announced. |
| 141 |
As you said, try re-emerging the gentoo kernel, make mrproper, no need |
| 142 |
for distclean, which does run mrproper as part of its operation, and |
| 143 |
see if that helps. |
| 144 |
Alternatively, you could try building the upstream version of 4.9.77, |
| 145 |
and see if that works. Let me know if you need instructions on how to |
| 146 |
proceed with that. |
Archives