On 2/3/21 12:41 PM, Andrii Nakryiko wrote:
> On Wed, Feb 3, 2021 at 12:36 PM Randy Dunlap <rdunlap@xxxxxxxxxxxxx> wrote:
>>
>> On 2/3/21 12:33 PM, Andrii Nakryiko wrote:
>>> On Wed, Feb 3, 2021 at 12:15 PM Randy Dunlap <rdunlap@xxxxxxxxxxxxx> wrote:
>>>>
>>>> On 2/3/21 12:12 PM, Andrii Nakryiko wrote:
>>>>> On Wed, Feb 3, 2021 at 12:09 PM Randy Dunlap <rdunlap@xxxxxxxxxxxxx> wrote:
>>>>>>
>>>>>> On 2/3/21 11:39 AM, Andrii Nakryiko wrote:
>>>>>>> On Wed, Feb 3, 2021 at 9:22 AM Randy Dunlap <rdunlap@xxxxxxxxxxxxx> wrote:
>>>>>>>>
>>>>>>>> On 2/3/21 2:57 AM, Toke Høiland-Jørgensen wrote:
>>>>>>>>> Randy Dunlap <rdunlap@xxxxxxxxxxxxx> writes:
>>>>>>>>>
>>>>>>>>>> Hi,
>>>>>>>>>>
>>>>>>>>>> I see this sometimes when building a kernel: (on x86_64,
>>>>>>>>>> with today's linux-next 20210202):
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> CONFIG_CGROUP_BPF=y
>>>>>>>>>> CONFIG_BPF=y
>>>>>>>>>> CONFIG_BPF_SYSCALL=y
>>>>>>>>>> CONFIG_ARCH_WANT_DEFAULT_BPF_JIT=y
>>>>>>>>>> CONFIG_BPF_PRELOAD=y
>>>>>>>>>> CONFIG_BPF_PRELOAD_UMD=m
>>>>>>>>>> CONFIG_HAVE_EBPF_JIT=y
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> Auto-detecting system features:
>>>>>>>>>> ... libelf: [ [31mOFF[m ]
>>>>>>>>>> ... zlib: [ [31mOFF[m ]
>>>>>>>>>> ... bpf: [ [31mOFF[m ]
>>>>>>>>>>
>>>>>>>>>> No libelf found
>>>>>>>>>> make[5]: [Makefile:287: elfdep] Error 1 (ignored)
>>>>>>>>>> No zlib found
>>>>>>>>>> make[5]: [Makefile:290: zdep] Error 1 (ignored)
>>>>>>>>>> BPF API too old
>>>>>>>>>> make[5]: [Makefile:293: bpfdep] Error 1 (ignored)
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> but pkg-config tells me:
>>>>>>>>>>
>>>>>>>>>> $ pkg-config --modversion libelf
>>>>>>>>>> 0.168
>>>>>>>>>> $ pkg-config --libs libelf
>>>>>>>>>> -lelf
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> Any ideas?
>>>>>>>>>
>>>>>>>>> This usually happens because there's a stale cache of the feature
>>>>>>>>> detection tests lying around somewhere. Look for a 'feature' directory
>>>>>>>>> in whatever subdir you got that error. Just removing the feature
>>>>>>>>> directory usually fixes this; I've fixed a couple of places where this
>>>>>>>>> is not picked up by 'make clean' (see, e.g., 9d9aae53b96d ("bpf/preload:
>>>>>>>>> Make sure Makefile cleans up after itself, and add .gitignore")) but I
>>>>>>>>> wouldn't be surprised if there are still some that are broken.
>>>>>>>>
>>>>>>>> Hi,
>>>>>>>>
>>>>>>>> Thanks for replying.
>>>>>>>>
>>>>>>>> I removed the feature subdir and still got this build error, so I
>>>>>>>> removed everything in BUILDDIR/kernel/bpf/preload and rebuilt --
>>>>>>>> and still got the same libelf build error.
>>>>>>>
>>>>>>> I hate the complexity of feature detection framework to the point that
>>>>>>> I'm willing to rip it out from libbpf's Makefile completely. I just
>>>>>>> spent an hour trying to understand what's going on in a very similar
>>>>>>> situation. Extremely frustrating.
>>>>>>>
>>>>>>> In your case, it might be feature detection triggered from
>>>>>>> resolve_btfids, so try removing
>>>>>>> $(OUTPUT)/tools/bpf/resolve_btfids/{feature/,FEATURE-DUMP.libbpf}.
>>>>>>>
>>>>>>> It seems like we don't do proper cleanup in resolve_btfids (it should
>>>>>>> probably call libbpf's clean as well). And it's beyond me why `make -C
>>>>>>> tools/build/feature clean` doesn't clean up FEATURE-DUMP.<use-case>
>>>>>>> file as well.
>>>>>>
>>>>>>
>>>>>> I don't think it's related to improper cleanup or old files/dirs
>>>>>> laying around. I say that because I did a full build in a new output dir.
>>>>>> and it still failed in the same way.
>>>>>
>>>>> If you cd tools/lib/bpf and run make there, does it detect those libraries?
>>>>
>>>> Yes:
>>>>
>>>> Auto-detecting system features:
>>>> ... libelf: [ on ]
>>>> ... zlib: [ on ]
>>>> ... bpf: [ on ]
>>>>
>>>>
>>>
>>> Sounds exactly like my case. I removed
>>> $(O)/tools/bpf/resolve_btfids/{feature/,FEATURE-DUMP.libbpf} and it
>>> started working.
>>
>> I already tried that with no success.
>>
>> I suppose that it could be related to how I do builds:
>>
>> make ARCH=x86_64 O=subdir -j4 all
>>
>> so subdir is a relative path, not an absolute path.
>
> so can you confirm this by specifying the absolute path to subdir?
Yes, absolute output path works for libelf, zlib, and bpf:
Auto-detecting system features:
... libelf: [ [32mon[m ]
... zlib: [ [32mon[m ]
... bpf: [ [32mon[m ]
Are {feature,FEATURE-DUMP} created in multiple places?
I don't see them in $(OUTDIR)/tools -- there is no /bpf/ subdir there
at all.
I do see them in $(OUTDIR)/kernel/bpf/preload/ -- are those different
feature files?
thanks.
--
~Randy
