| Message ID | 20240904111338.4095848-2-matttbe@kernel.org (mailing list archive) |
|---|---|
| State | Superseded, archived |
| Delegated to: | Matthieu Baerts |
| Headers | show |
| Series | [6.1.y] selftests: mptcp: join: validate event numbers | expand |
On Wed, Sep 04, 2024 at 01:13:39PM +0200, Matthieu Baerts (NGI0) wrote: > commit 20ccc7c5f7a3aa48092441a4b182f9f40418392e upstream. > This did not apply either. I think I've gone through all of the 6.1 patches now. If I've missed anything, please let me know. thanks, greg k-h
Hi Greg, On 04/09/2024 16:38, Greg KH wrote: > On Wed, Sep 04, 2024 at 01:13:39PM +0200, Matthieu Baerts (NGI0) wrote: >> commit 20ccc7c5f7a3aa48092441a4b182f9f40418392e upstream. >> > > This did not apply either. > > I think I've gone through all of the 6.1 patches now. If I've missed > anything, please let me know. It looks like there are some conflicts with the patches Sasha recently added: queue-6.1/selftests-mptcp-add-explicit-test-case-for-remove-re.patch queue-6.1/selftests-mptcp-join-check-re-adding-init-endp-with-.patch queue-6.1/selftests-mptcp-join-check-re-using-id-of-unused-add.patch From commit 0d8d8d5bcef1 ("Fixes for 6.1") from the stable-queue tree: https://git.kernel.org/pub/scm/linux/kernel/git/stable/stable-queue.git/commit/?id=0d8d8d5bcef1 I have also added these patches -- we can see patches with almost the same name -- but I adapted them to the v6.1 kernel: it was possible to apply them without conflicts, but they were causing issues because they were calling functions that are not available in v6.1, or taking different parameters. Do you mind removing the ones from Sasha please? I hope that will not cause any issues. After that, the two patches you had errors with should apply without conflicts: - selftests: mptcp: join: validate event numbers - selftests: mptcp: join: check re-re-adding ID 0 signal Cheers, Matt
On Wed, Sep 04, 2024 at 05:20:59PM +0200, Matthieu Baerts wrote: > Hi Greg, > > On 04/09/2024 16:38, Greg KH wrote: > > On Wed, Sep 04, 2024 at 01:13:39PM +0200, Matthieu Baerts (NGI0) wrote: > >> commit 20ccc7c5f7a3aa48092441a4b182f9f40418392e upstream. > >> > > > > This did not apply either. > > > > I think I've gone through all of the 6.1 patches now. If I've missed > > anything, please let me know. > It looks like there are some conflicts with the patches Sasha recently > added: > > queue-6.1/selftests-mptcp-add-explicit-test-case-for-remove-re.patch > queue-6.1/selftests-mptcp-join-check-re-adding-init-endp-with-.patch > queue-6.1/selftests-mptcp-join-check-re-using-id-of-unused-add.patch > > >From commit 0d8d8d5bcef1 ("Fixes for 6.1") from the stable-queue tree: > > > https://git.kernel.org/pub/scm/linux/kernel/git/stable/stable-queue.git/commit/?id=0d8d8d5bcef1 > > I have also added these patches -- we can see patches with almost the > same name -- but I adapted them to the v6.1 kernel: it was possible to > apply them without conflicts, but they were causing issues because they > were calling functions that are not available in v6.1, or taking > different parameters. > > Do you mind removing the ones from Sasha please? I hope that will not > cause any issues. After that, the two patches you had errors with should > apply without conflicts: Ok, I've now dropped them, that actually fixes an error I was seeing where we had duplicated patches in the tree. > - selftests: mptcp: join: validate event numbers > - selftests: mptcp: join: check re-re-adding ID 0 signal I'll go add these now, thanks! And thanks for all the backports, much appreciated. greg k-h
On Thu, Sep 05, 2024 at 11:33:46AM +0200, Greg KH wrote: > On Wed, Sep 04, 2024 at 05:20:59PM +0200, Matthieu Baerts wrote: > > Hi Greg, > > > > On 04/09/2024 16:38, Greg KH wrote: > > > On Wed, Sep 04, 2024 at 01:13:39PM +0200, Matthieu Baerts (NGI0) wrote: > > >> commit 20ccc7c5f7a3aa48092441a4b182f9f40418392e upstream. > > >> > > > > > > This did not apply either. > > > > > > I think I've gone through all of the 6.1 patches now. If I've missed > > > anything, please let me know. > > It looks like there are some conflicts with the patches Sasha recently > > added: > > > > queue-6.1/selftests-mptcp-add-explicit-test-case-for-remove-re.patch > > queue-6.1/selftests-mptcp-join-check-re-adding-init-endp-with-.patch > > queue-6.1/selftests-mptcp-join-check-re-using-id-of-unused-add.patch > > > > >From commit 0d8d8d5bcef1 ("Fixes for 6.1") from the stable-queue tree: > > > > > > https://git.kernel.org/pub/scm/linux/kernel/git/stable/stable-queue.git/commit/?id=0d8d8d5bcef1 > > > > I have also added these patches -- we can see patches with almost the > > same name -- but I adapted them to the v6.1 kernel: it was possible to > > apply them without conflicts, but they were causing issues because they > > were calling functions that are not available in v6.1, or taking > > different parameters. > > > > Do you mind removing the ones from Sasha please? I hope that will not > > cause any issues. After that, the two patches you had errors with should > > apply without conflicts: > > Ok, I've now dropped them, that actually fixes an error I was seeing > where we had duplicated patches in the tree. > > > - selftests: mptcp: join: validate event numbers > > - selftests: mptcp: join: check re-re-adding ID 0 signal > > I'll go add these now, thanks! I just tried, and they still fail to apply. How about we wait for this next 6.1.y release to happen and then you rebase and see what I messed up and send me the remaining ones as this is getting confusing on my end... thanks, greg k-h
On 05/09/2024 11:36, Greg KH wrote: > On Thu, Sep 05, 2024 at 11:33:46AM +0200, Greg KH wrote: >> On Wed, Sep 04, 2024 at 05:20:59PM +0200, Matthieu Baerts wrote: >>> Hi Greg, >>> >>> On 04/09/2024 16:38, Greg KH wrote: >>>> On Wed, Sep 04, 2024 at 01:13:39PM +0200, Matthieu Baerts (NGI0) wrote: >>>>> commit 20ccc7c5f7a3aa48092441a4b182f9f40418392e upstream. >>>>> >>>> >>>> This did not apply either. >>>> >>>> I think I've gone through all of the 6.1 patches now. If I've missed >>>> anything, please let me know. >>> It looks like there are some conflicts with the patches Sasha recently >>> added: >>> >>> queue-6.1/selftests-mptcp-add-explicit-test-case-for-remove-re.patch >>> queue-6.1/selftests-mptcp-join-check-re-adding-init-endp-with-.patch >>> queue-6.1/selftests-mptcp-join-check-re-using-id-of-unused-add.patch >>> >>> >From commit 0d8d8d5bcef1 ("Fixes for 6.1") from the stable-queue tree: >>> >>> >>> https://git.kernel.org/pub/scm/linux/kernel/git/stable/stable-queue.git/commit/?id=0d8d8d5bcef1 >>> >>> I have also added these patches -- we can see patches with almost the >>> same name -- but I adapted them to the v6.1 kernel: it was possible to >>> apply them without conflicts, but they were causing issues because they >>> were calling functions that are not available in v6.1, or taking >>> different parameters. >>> >>> Do you mind removing the ones from Sasha please? I hope that will not >>> cause any issues. After that, the two patches you had errors with should >>> apply without conflicts: >> >> Ok, I've now dropped them, that actually fixes an error I was seeing >> where we had duplicated patches in the tree. >> >>> - selftests: mptcp: join: validate event numbers >>> - selftests: mptcp: join: check re-re-adding ID 0 signal >> >> I'll go add these now, thanks! > > I just tried, and they still fail to apply. How about we wait for this > next 6.1.y release to happen and then you rebase and see what I messed > up and send me the remaining ones as this is getting confusing on my > end... Thank you for having tried! Sure, no problem, I can wait. I just hope the previous patches I sent have been applied properly. I mean: I don't know how quilt handled the duplicated patches with different content. If you prefer, we can also drop all the MPTCP related patches -- or only the ones related to the selftests -- and I send the whole series I have on my side: it is still on top of the v6.1.108-rc1, so without the patches Sasha backported yesterday. Cheers, Matt
On Thu, Sep 05, 2024 at 11:42:21AM +0200, Matthieu Baerts wrote: > On 05/09/2024 11:36, Greg KH wrote: > > On Thu, Sep 05, 2024 at 11:33:46AM +0200, Greg KH wrote: > >> On Wed, Sep 04, 2024 at 05:20:59PM +0200, Matthieu Baerts wrote: > >>> Hi Greg, > >>> > >>> On 04/09/2024 16:38, Greg KH wrote: > >>>> On Wed, Sep 04, 2024 at 01:13:39PM +0200, Matthieu Baerts (NGI0) wrote: > >>>>> commit 20ccc7c5f7a3aa48092441a4b182f9f40418392e upstream. > >>>>> > >>>> > >>>> This did not apply either. > >>>> > >>>> I think I've gone through all of the 6.1 patches now. If I've missed > >>>> anything, please let me know. > >>> It looks like there are some conflicts with the patches Sasha recently > >>> added: > >>> > >>> queue-6.1/selftests-mptcp-add-explicit-test-case-for-remove-re.patch > >>> queue-6.1/selftests-mptcp-join-check-re-adding-init-endp-with-.patch > >>> queue-6.1/selftests-mptcp-join-check-re-using-id-of-unused-add.patch > >>> > >>> >From commit 0d8d8d5bcef1 ("Fixes for 6.1") from the stable-queue tree: > >>> > >>> > >>> https://git.kernel.org/pub/scm/linux/kernel/git/stable/stable-queue.git/commit/?id=0d8d8d5bcef1 > >>> > >>> I have also added these patches -- we can see patches with almost the > >>> same name -- but I adapted them to the v6.1 kernel: it was possible to > >>> apply them without conflicts, but they were causing issues because they > >>> were calling functions that are not available in v6.1, or taking > >>> different parameters. > >>> > >>> Do you mind removing the ones from Sasha please? I hope that will not > >>> cause any issues. After that, the two patches you had errors with should > >>> apply without conflicts: > >> > >> Ok, I've now dropped them, that actually fixes an error I was seeing > >> where we had duplicated patches in the tree. > >> > >>> - selftests: mptcp: join: validate event numbers > >>> - selftests: mptcp: join: check re-re-adding ID 0 signal > >> > >> I'll go add these now, thanks! > > > > I just tried, and they still fail to apply. How about we wait for this > > next 6.1.y release to happen and then you rebase and see what I messed > > up and send me the remaining ones as this is getting confusing on my > > end... > > Thank you for having tried! > > Sure, no problem, I can wait. > > I just hope the previous patches I sent have been applied properly. I > mean: I don't know how quilt handled the duplicated patches with > different content. > > If you prefer, we can also drop all the MPTCP related patches -- or only > the ones related to the selftests -- and I send the whole series I have > on my side: it is still on top of the v6.1.108-rc1, so without the > patches Sasha backported yesterday. Let's see how this -rc release looks, if it's broken or messed up, I can easily drop patches as needed. thanks, greg k-h
On 05/09/2024 12:10, Greg KH wrote: > On Thu, Sep 05, 2024 at 11:42:21AM +0200, Matthieu Baerts wrote: >> On 05/09/2024 11:36, Greg KH wrote: >>> On Thu, Sep 05, 2024 at 11:33:46AM +0200, Greg KH wrote: >>>> On Wed, Sep 04, 2024 at 05:20:59PM +0200, Matthieu Baerts wrote: >>>>> Hi Greg, >>>>> >>>>> On 04/09/2024 16:38, Greg KH wrote: >>>>>> On Wed, Sep 04, 2024 at 01:13:39PM +0200, Matthieu Baerts (NGI0) wrote: >>>>>>> commit 20ccc7c5f7a3aa48092441a4b182f9f40418392e upstream. >>>>>>> >>>>>> >>>>>> This did not apply either. >>>>>> >>>>>> I think I've gone through all of the 6.1 patches now. If I've missed >>>>>> anything, please let me know. >>>>> It looks like there are some conflicts with the patches Sasha recently >>>>> added: >>>>> >>>>> queue-6.1/selftests-mptcp-add-explicit-test-case-for-remove-re.patch >>>>> queue-6.1/selftests-mptcp-join-check-re-adding-init-endp-with-.patch >>>>> queue-6.1/selftests-mptcp-join-check-re-using-id-of-unused-add.patch >>>>> >>>>> >From commit 0d8d8d5bcef1 ("Fixes for 6.1") from the stable-queue tree: >>>>> >>>>> >>>>> https://git.kernel.org/pub/scm/linux/kernel/git/stable/stable-queue.git/commit/?id=0d8d8d5bcef1 >>>>> >>>>> I have also added these patches -- we can see patches with almost the >>>>> same name -- but I adapted them to the v6.1 kernel: it was possible to >>>>> apply them without conflicts, but they were causing issues because they >>>>> were calling functions that are not available in v6.1, or taking >>>>> different parameters. >>>>> >>>>> Do you mind removing the ones from Sasha please? I hope that will not >>>>> cause any issues. After that, the two patches you had errors with should >>>>> apply without conflicts: >>>> >>>> Ok, I've now dropped them, that actually fixes an error I was seeing >>>> where we had duplicated patches in the tree. >>>> >>>>> - selftests: mptcp: join: validate event numbers >>>>> - selftests: mptcp: join: check re-re-adding ID 0 signal >>>> >>>> I'll go add these now, thanks! >>> >>> I just tried, and they still fail to apply. How about we wait for this >>> next 6.1.y release to happen and then you rebase and see what I messed >>> up and send me the remaining ones as this is getting confusing on my >>> end... >> >> Thank you for having tried! >> >> Sure, no problem, I can wait. >> >> I just hope the previous patches I sent have been applied properly. I >> mean: I don't know how quilt handled the duplicated patches with >> different content. >> >> If you prefer, we can also drop all the MPTCP related patches -- or only >> the ones related to the selftests -- and I send the whole series I have >> on my side: it is still on top of the v6.1.108-rc1, so without the >> patches Sasha backported yesterday. > > Let's see how this -rc release looks, if it's broken or messed up, I can > easily drop patches as needed. Sounds good to me. I will check this RC before looking at the remaining failed backports. Thank you for your support! Cheers, Matt
diff --git a/tools/testing/selftests/net/mptcp/mptcp_join.sh b/tools/testing/selftests/net/mptcp/mptcp_join.sh index 79e9cac89cd2..79a39e30842b 100755 --- a/tools/testing/selftests/net/mptcp/mptcp_join.sh +++ b/tools/testing/selftests/net/mptcp/mptcp_join.sh @@ -394,6 +394,25 @@ reset_with_fail() fi } +start_events() +{ + evts_ns1=$(mktemp) + evts_ns2=$(mktemp) + :> "$evts_ns1" + :> "$evts_ns2" + ip netns exec "${ns1}" ./pm_nl_ctl events >> "$evts_ns1" 2>&1 & + evts_ns1_pid=$! + ip netns exec "${ns2}" ./pm_nl_ctl events >> "$evts_ns2" 2>&1 & + evts_ns2_pid=$! +} + +reset_with_events() +{ + reset "${1}" || return 1 + + start_events +} + reset_with_tcp_filter() { reset "${1}" || return 1 @@ -596,6 +615,14 @@ kill_tests_wait() wait } +kill_events_pids() +{ + kill_wait $evts_ns1_pid + evts_ns1_pid=0 + kill_wait $evts_ns2_pid + evts_ns2_pid=0 +} + pm_nl_set_limits() { local ns=$1 @@ -3146,6 +3173,32 @@ fail_tests() fi } +# $1: ns ; $2: event type ; $3: count +chk_evt_nr() +{ + local ns=${1} + local evt_name="${2}" + local exp="${3}" + + local evts="${evts_ns1}" + local evt="${!evt_name}" + local count + + evt_name="${evt_name:16}" # without MPTCP_LIB_EVENT_ + [ "${ns}" == "ns2" ] && evts="${evts_ns2}" + + printf "%-${nr_blank}s %s" " " "event ${ns} ${evt_name} (${exp})" + + count=$(grep -cw "type:${evt}" "${evts}") + if [ "${count}" != "${exp}" ]; then + echo "[fail] got $count events, expected $exp" + fail_test + dump_stats + else + echo "[ ok ]" + fi +} + userspace_tests() { # userspace pm type prevents add_addr @@ -3268,11 +3321,13 @@ endpoint_tests() if reset_with_tcp_filter "delete and re-add" ns2 10.0.3.2 REJECT OUTPUT && mptcp_lib_kallsyms_has "subflow_rebuild_header$"; then + start_events pm_nl_set_limits $ns1 0 3 pm_nl_set_limits $ns2 0 3 pm_nl_add_endpoint $ns2 10.0.1.2 id 1 dev ns2eth1 flags subflow pm_nl_add_endpoint $ns2 10.0.2.2 id 2 dev ns2eth2 flags subflow run_tests $ns1 $ns2 10.0.1.1 4 0 0 speed_5 2>/dev/null & + local tests_pid=$! wait_mpj $ns2 pm_nl_del_endpoint $ns2 2 10.0.2.2 @@ -3304,14 +3359,30 @@ endpoint_tests() chk_subflow_nr "" "after re-add id 0 ($i)" 3 done + kill_wait "${tests_pid}" + kill_events_pids kill_tests_wait + chk_evt_nr ns1 MPTCP_LIB_EVENT_CREATED 1 + chk_evt_nr ns1 MPTCP_LIB_EVENT_ESTABLISHED 1 + chk_evt_nr ns1 MPTCP_LIB_EVENT_ANNOUNCED 0 + chk_evt_nr ns1 MPTCP_LIB_EVENT_REMOVED 4 + chk_evt_nr ns1 MPTCP_LIB_EVENT_SUB_ESTABLISHED 6 + chk_evt_nr ns1 MPTCP_LIB_EVENT_SUB_CLOSED 4 + + chk_evt_nr ns2 MPTCP_LIB_EVENT_CREATED 1 + chk_evt_nr ns2 MPTCP_LIB_EVENT_ESTABLISHED 1 + chk_evt_nr ns2 MPTCP_LIB_EVENT_ANNOUNCED 0 + chk_evt_nr ns2 MPTCP_LIB_EVENT_REMOVED 0 + chk_evt_nr ns2 MPTCP_LIB_EVENT_SUB_ESTABLISHED 6 + chk_evt_nr ns2 MPTCP_LIB_EVENT_SUB_CLOSED 5 # one has been closed before estab + chk_join_nr 6 6 6 chk_rm_nr 4 4 fi # remove and re-add - if reset "delete re-add signal" && + if reset_with_events "delete re-add signal" && mptcp_lib_kallsyms_has "subflow_rebuild_header$"; then pm_nl_set_limits $ns1 0 3 pm_nl_set_limits $ns2 3 3 @@ -3342,8 +3413,25 @@ endpoint_tests() pm_nl_add_endpoint $ns1 10.0.1.1 id 99 flags signal wait_mpj $ns2 chk_subflow_nr "" "after re-add" 3 + + kill_wait "${tests_pid}" + kill_events_pids kill_tests_wait + chk_evt_nr ns1 MPTCP_LIB_EVENT_CREATED 1 + chk_evt_nr ns1 MPTCP_LIB_EVENT_ESTABLISHED 1 + chk_evt_nr ns1 MPTCP_LIB_EVENT_ANNOUNCED 0 + chk_evt_nr ns1 MPTCP_LIB_EVENT_REMOVED 0 + chk_evt_nr ns1 MPTCP_LIB_EVENT_SUB_ESTABLISHED 4 + chk_evt_nr ns1 MPTCP_LIB_EVENT_SUB_CLOSED 2 + + chk_evt_nr ns2 MPTCP_LIB_EVENT_CREATED 1 + chk_evt_nr ns2 MPTCP_LIB_EVENT_ESTABLISHED 1 + chk_evt_nr ns2 MPTCP_LIB_EVENT_ANNOUNCED 5 + chk_evt_nr ns2 MPTCP_LIB_EVENT_REMOVED 3 + chk_evt_nr ns2 MPTCP_LIB_EVENT_SUB_ESTABLISHED 4 + chk_evt_nr ns2 MPTCP_LIB_EVENT_SUB_CLOSED 2 + chk_join_nr 4 4 4 chk_add_nr 5 5 chk_rm_nr 3 2 invert diff --git a/tools/testing/selftests/net/mptcp/mptcp_lib.sh b/tools/testing/selftests/net/mptcp/mptcp_lib.sh index f32045b23b89..4b1bef34d6d8 100644 --- a/tools/testing/selftests/net/mptcp/mptcp_lib.sh +++ b/tools/testing/selftests/net/mptcp/mptcp_lib.sh @@ -4,6 +4,21 @@ readonly KSFT_FAIL=1 readonly KSFT_SKIP=4 +# These variables are used in some selftests, read-only +declare -rx MPTCP_LIB_EVENT_CREATED=1 # MPTCP_EVENT_CREATED +declare -rx MPTCP_LIB_EVENT_ESTABLISHED=2 # MPTCP_EVENT_ESTABLISHED +declare -rx MPTCP_LIB_EVENT_CLOSED=3 # MPTCP_EVENT_CLOSED +declare -rx MPTCP_LIB_EVENT_ANNOUNCED=6 # MPTCP_EVENT_ANNOUNCED +declare -rx MPTCP_LIB_EVENT_REMOVED=7 # MPTCP_EVENT_REMOVED +declare -rx MPTCP_LIB_EVENT_SUB_ESTABLISHED=10 # MPTCP_EVENT_SUB_ESTABLISHED +declare -rx MPTCP_LIB_EVENT_SUB_CLOSED=11 # MPTCP_EVENT_SUB_CLOSED +declare -rx MPTCP_LIB_EVENT_SUB_PRIORITY=13 # MPTCP_EVENT_SUB_PRIORITY +declare -rx MPTCP_LIB_EVENT_LISTENER_CREATED=15 # MPTCP_EVENT_LISTENER_CREATED +declare -rx MPTCP_LIB_EVENT_LISTENER_CLOSED=16 # MPTCP_EVENT_LISTENER_CLOSED + +declare -rx MPTCP_LIB_AF_INET=2 +declare -rx MPTCP_LIB_AF_INET6=10 + # SELFTESTS_MPTCP_LIB_EXPECT_ALL_FEATURES env var can be set when validating all # features using the last version of the kernel and the selftests to make sure # a test is not being skipped by mistake.
commit 20ccc7c5f7a3aa48092441a4b182f9f40418392e upstream. This test extends "delete and re-add" and "delete re-add signal" to validate the previous commit: the number of MPTCP events are checked to make sure there are no duplicated or unexpected ones. A new helper has been introduced to easily check these events. The missing events have been added to the lib. The 'Fixes' tag here below is the same as the one from the previous commit: this patch here is not fixing anything wrong in the selftests, but it validates the previous fix for an issue introduced by this commit ID. Fixes: b911c97c7dc7 ("mptcp: add netlink event support") Cc: stable@vger.kernel.org Reviewed-by: Mat Martineau <martineau@kernel.org> Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org> Signed-off-by: Paolo Abeni <pabeni@redhat.com> [ Conflicts in mptcp_join.sh and mptcp_lib.sh, due to commit 38f027fca1b7 ("selftests: mptcp: dump userspace addrs list") -- linked to a new feature, not backportable to stable -- and commit 23a0485d1c04 ("selftests: mptcp: declare event macros in mptcp_lib") -- depending on the previous one -- not in this version. The conflicts in mptcp_join.sh were in the context, because a new helper had to be added after others that are not in this version. The conflicts in mptcp_lib.sh were due to the fact the other MPTCP_LIB_EVENT_* constants were not present. They have all been added in this version to ease future backports if any. In this version, it was also needed to import reset_with_events and kill_events_pids from the newer version, and adapt chk_evt_nr to how the results are printed in this version, plus remove the LISTENER events checks because the linked feature is not available in this kernel version. ] Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org> --- .../testing/selftests/net/mptcp/mptcp_join.sh | 90 ++++++++++++++++++- .../testing/selftests/net/mptcp/mptcp_lib.sh | 15 ++++ 2 files changed, 104 insertions(+), 1 deletion(-)