On Thu, 26 Jan 2023 10:01:44 +0000 Andrei Gherzan wrote:
On 23/01/25 11:08PM, Jakub Kicinski wrote:
On Wed, 25 Jan 2023 21:13:49 +0000 Andrei Gherzan wrote:
The udpgro_frglist.sh uses nat6to4.o which is tested for existence in bpf/nat6to4.o (relative to the script). This is where the object is compiled. Even so, the script attempts to use it as part of tc with a different path (../bpf/nat6to4.o). As a consequence, this fails the script:
Is this a recent regression? Can you add a Fixes tag?
This issue seems to be included from the beginning (edae34a3ed92). I can't say why this was not seen before upstream but on our side, this test was disabled internally due to lack of CC support in BPF programs. This was fixed in the meanwhile in 837a3d66d698 (selftests: net: Add cross-compilation support for BPF programs) and we found this issue while trying to reenable the test.
So if you think that is reasonable, I could add a Fixes tag for the initial script commit edae34a3ed92 (selftests net: add UDP GRO fraglist + bpf self-tests) and push a v3.
We have queued commit 3c107f36db06 ("selftests/net: mv bpf/nat6to4.c to net folder") in net-next, I think that should fix it, too?
What tree did you base this patch on? Doesn't seem to apply
The patches were done on top of git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git, the master branch - 948ef7bb70c4 (Merge tag 'modules-6.2-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/mcgrof/linux). There is another merge that happened in the meanwhile but the rebase works without issues. I can send a rebased v3 if needed.
Could you try linux-next or net-next ?