pkg/lwip: use ztimer_msec instead of xtimer#17115
Merged
fjmolinas merged 2 commits intoRIOT-OS:masterfrom Mar 4, 2022
Merged
Conversation
fjmolinas
added
CI: ready for build
github-actions
bot
added
Area: network
22 tasks
miri64
reviewed
Nov 2, 2021
Contributor
Author
|
May I squash? |
Member
|
Yes, please go ahead. |
Member
|
It appears you need to address this unrelated issue, pointed out by |
fjmolinas
requested review from
MichelRottleuthner,
aabadie,
leandrolanzieri and
smlng
as code owners
fjmolinas
removed
the
CI: ready for build
Contributor
Author
Maybe a void cast would be enough though...
Went with an alternative, let me know if it's OK with you or you prefer the ones you suggested. |
Member
I would prefer either one of the suggested or the void cast you proposed. The initialization you use in your alternative introduces additional code (via an implicit |
fjmolinas
commented
Nov 4, 2021
tests/lwip/ip.c
Outdated
| sizeof(addrstr))); | ||
| break; | ||
| } | ||
| printf("unspecified source\n"); |
Contributor
Author
There was a problem hiding this comment.
I used c-conditionals and it seems to work, should I change de message here to something like inet unsupported?
Contributor
There was a problem hiding this comment.
since it continues the "Received IP data from" line, maybe say "unsupported protocol" ?
Member
There was a problem hiding this comment.
[…] maybe say "unsupported protocol" ?
Maybe, to immediately give a hint what is missing "unsupported protocol IPv4"/"unsupported protocol IPv6"?
Contributor
|
I am not familiar with the timer implementations, but the changes look reasonable to me |
Contributor
Author
|
May I squash? |
fjmolinas
force-pushed
the
pr_lwip_use_ztimer
branch
from
515573a to
a33271d
Compare
fjmolinas
added
the
CI: ready for build
maribu
added
CI: ready for build
Contributor
Author
|
Ci failures are mostly unrelated. The
|
fjmolinas
removed
the
CI: run tests
Contributor
Author
|
Hmm those |
Contributor
Author
|
@miri64 how can we get this unblocked? It seems from you testing that this is failing independent of this PR |
Member
Where is the block? The only NACK I see was addressed AFAIK. I do not have much time for review, so please, if a reviewer can be found, go ahead. |
kfessel
reviewed
Mar 1, 2022
| printf("Success: send %u byte over TCP to server\n", (unsigned)data_len); | ||
| } | ||
| xtimer_usleep(delay); | ||
| ztimer_sleep(ZTIMER_USEC, delay); |
Contributor
There was a problem hiding this comment.
something like
d5dee30
is missing here but this might be an oportunity for another PR
Suggested change
| ztimer_sleep(ZTIMER_USEC, delay); | |
| if( i+1 < num) ztimer_sleep(ZTIMER_USEC, delay); |
| (unsigned)data_len, addr_str); | ||
| } | ||
| xtimer_usleep(delay); | ||
| ztimer_sleep(ZTIMER_USEC, delay); |
Contributor
There was a problem hiding this comment.
something like
d5dee30
is missing here but this might be an oportunity for another PR
Suggested change
| ztimer_sleep(ZTIMER_USEC, delay); | |
| if( i+1 < num) ztimer_sleep(ZTIMER_USEC, delay); |
Contributor
Author
There was a problem hiding this comment.
True but IMO not worth it, this is just for the test, the actual module is only using msec
tests/lwip/ip.c
Outdated
| ipv6_addr_to_str(addrstr, | ||
| (ipv6_addr_t *)&src.addr.ipv6, | ||
| sizeof(addrstr))); | ||
| if (IS_USED(MODULE_LWIP_IPV6)) { |
Contributor
There was a problem hiding this comment.
this does not do as intended:
tests/lwip$ LWIP_IPV4=1 make
tests/lwip/ip.c: In function ‘_ip_recv’:
tests/lwip/ip.c:74:74: error: ‘union <anonymous>’ has no member named ‘ipv6’; did you mean ‘ipv4’?
74 | (ipv6_addr_t *)&src.addr.ipv6,
| ^~~~
| ipv4
Contributor
Author
Hmm strange that this was not caught before... |
fjmolinas
force-pushed
the
pr_lwip_use_ztimer
branch
from
55770f9 to
19a28ad
Compare
Contributor
Author
|
Thanks for testing @kfessel, I pushed and squashed the fixup => new build triggered (good idea anyway since this PR was old) |
Contributor
|
btw.: |
kfessel
approved these changes
Mar 2, 2022
Contributor
Author
|
Well seems like another xtimer call was added in the meantime... good we ran the ci again |
Contributor
Author
|
I'll have to rebase to fix this |
fjmolinas
force-pushed
the
pr_lwip_use_ztimer
branch
2 times, most recently
from
fe1fefd to
409d5df
Compare
Contributor
Author
|
@kfessel seems like re-triggering the python job is not enough... maybe I need to rebase, can I skip the ci build? |
Contributor
Author
|
For reference last build was yesterday https://ci.riot-os.org/RIOT-OS/RIOT/17115/409d5df061a95787f4136461d74c40c9f4c33cce/output.html: |
fjmolinas
force-pushed
the
pr_lwip_use_ztimer
branch
from
409d5df to
8516e25
Compare
Contributor
Author
|
I rebased and now the static job failed, I let you device if we let the ci run again @kfessel |
kfessel
added
the
CI: skip compile test
Contributor
|
There is no reason to compile again #17115 (comment) murdock says it is green and i trust that this is just a rebase |
Member
|
Did you test IPv4 and dual stack as well? ;-) |
Contributor
|
the provided lwip/test/run.py is not able to test IPv4, but since there are only timer changes left keeping every thing else the same i assume they work the same as before before ACK I redid the build test: #17115 (comment) |
Contributor
Author
|
Re-run of native tests as @miri64 did in #17145 I ran for all combinations, all still passing for test in tests/lwip_sock_{ip,tcp,udp}/; do
for ipv6 in 0 1; do
for ipv4 in 0 1; do
if [ $((ipv4 + ipv6)) -eq 0 ]; then
continue
fi
LWIP_IPV4=${ipv4} LWIP_IPV6=${ipv6} make -C ${test} flash -j test || break
done
done
doneDetails |
Member
You have fingers to type ;-). Since you need DHCP for it, most likely, it is hard to automate. |
Contributor
For the moment i think we can trust the lwip_socket_.. test done by @fjmolinas and keep the ACK I tested the aspect that changed in the in the lwip test and it failed that -> that specific aspect change was undone -> it succeeded, the rest are 1 for 1 timer changes. most of our current ztimer transitions ignore #17614, -and ai was a bit hassitant to ack any ztimer transition, but i think that is easier to account for that after a solution like #17607 is implemented and easier from ztimer -> ztimer taking the now issue into account that going the full path at once. (any move toward ztimer (especialy 32 or msec) is a huge improvement) |
Contributor
Author
|
GO! |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
7 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Contribution description
This package changes
lwipto useztimer_msecinstead ofxtimer.Testing procedure
Run all
tests/lwip*teststest/lwip_sock_iptest/lwip_sock_udptest/lwip_sock_tcpRelated Issues
Ticks item off #13667