On Tue, Jun 20, 2023 at 9:43 AM Alex Kiernan <alex.kier...@gmail.com> wrote: > > On Tue, Jun 20, 2023 at 5:38 PM Beniamin Sandu <beniaminsa...@gmail.com> > wrote: > > > > On Tue, Jun 20, 2023 at 4:12 PM Beniamin Sandu <beniaminsa...@gmail.com> > > wrote: > > > > > > On Tue, Jun 20, 2023 at 3:42 PM Alex Kiernan <alex.kier...@gmail.com> > > > wrote: > > > > > > > > On Tue, Jun 20, 2023 at 1:03 PM Alex Kiernan via > > > > lists.openembedded.org <alex.kiernan=gmail....@lists.openembedded.org> > > > > wrote: > > > > > > > > > > On Tue, Jun 20, 2023 at 11:53 AM Beniamin Sandu > > > > > <beniaminsa...@gmail.com> wrote: > > > > > > > > > > > > On Tue, Jun 20, 2023 at 12:56 PM Alex Kiernan > > > > > > <alex.kier...@gmail.com> wrote: > > > > > > > > > > > > > > On Mon, Jun 19, 2023 at 8:38 PM Beniamin Sandu > > > > > > > <beniaminsa...@gmail.com> wrote: > > > > > > > > > > > > > > > > I am trying to add a v3.x mbedtls recipe to master branch, and > > > > > > > > without > > > > > > > > > > > > > > Sorry, was reading the thread out of order and missed that you > > > > > > > were > > > > > > > trying to add an Mbed TLS 3 recipe! > > > > > > > > > > > > > > > this patch, mdns does not build. I could not find an obvious > > > > > > > > way to > > > > > > > > submit upstream patches to mdns from a quick search, otherwise > > > > > > > > I would > > > > > > > > gladly do it. > > > > > > > > > > > > > > > > > > > > > > Several folk have tried sending them to d...@apple.com (seemingly > > > > > > > without success), there's also a couple of attempted PRs on github > > > > > > > (again seemingly going nowhere). Winding back about 20 years, I > > > > > > > tried > > > > > > > to get a bunch of portability fixes into the Quicktime Streaming > > > > > > > Server without success - the model appears to very much be open > > > > > > > source, not open contribution. > > > > > > > > > > > > > > > While I am with you on trying to reduce technical debt within > > > > > > > > yocto > > > > > > > > layers, this one is very small and rather harmless, so I hope > > > > > > > > we can > > > > > > > > make a sacrifice for the greater good. :) > > > > > > > > > > > > > > > > > > > > > > I'm guessing we'll need to scope an Mbed TLS 3 recipe as being > > > > > > > explicitly mbedtls3_...? So sticking with 2.x doesn't feel like a > > > > > > > terrible option, but I agree its not ideal. > > > > > > > > > > > > This also sounds like a fairly decent option, having a separate > > > > > > mbedtls3 package might make it easier in general, although I am not > > > > > > aware of how used the current v2 version actually is. > > > > > > > > > > > > > > > > Looks like we're out of date (there's a 2.28.3), but it's still > > > > > supported for now... "Mbed TLS 2.28 is a long-time support branch. It > > > > > will be supported with bug-fixes and security fixes until end of > > > > > 2024." > > > > > > > > > > > > > Will send an update for 2.28.3, but constrain `devtool upgrade` to > > > > things on the 2.x branch > > > > > > > > One thought on 3.x - if mdns is the only failure maybe rename the > > > > current recipe to mbedtls2_% and add 3.x as mbedtls_% and then fix up > > > > the mdns recipe to consume mbedtls2? > > > > > > > > > > Personally I would say it is not worth to do it, just because of this > > > one package. Plus, it's building fine with both versions by including > > > this small patch, but maybe Khem Raj can chime in with an opinion > > > here.
yeah, I agree. Its better to maintain one version if we can. > > > > > > > While looking through some stuff, I discovered some variable I had no > > idea exists, called DEFAULT_PREFERENCE: > > https://docs.yoctoproject.org/ref-manual/variables.html#term-DEFAULT_PREFERENCE > > > > Seems it could be a very simple way to make mbedtls v2 be the > > "default" version unless specified by user through PREFERRED_PROVIDER? > > > > It works, but if you have two recipes with conflicting requirements > (which I do for some other things), it doesn't really help... > > Probably I'm leaning towards just upgrading to 3.x and taking the > mDNSResponder patch, I just worry about where Apple go... > > > > > > > > > > > > > > > My concern isn't really with the one liner now, its the how hard > > > > > > > does > > > > > > > it become to do the upgrades down the road - if we suddenly can't > > > > > > > upgrade because upstream's stuck on 2.x and there's something > > > > > > > incompatible with 3.x what do we do? > > > > > > > > > > > > > > > Cheers, > > > > > > > > Beni > > > > > > > > > > > > > > > > > > > > > > > > On Mon, Jun 19, 2023 at 10:14 PM Alex Kiernan > > > > > > > > <alex.kier...@gmail.com> wrote: > > > > > > > > > > > > > > > > > > Keeping up with Apple given they take nothing is frankly a > > > > > > > > > right pain, adding things which are nice to have rather than > > > > > > > > > just broken seems like we're just adding to the problem. Last > > > > > > > > > time I resynchronized this it took me days, unless there's a > > > > > > > > > really good reason to do this, or there's a way to get it > > > > > > > > > upstreamed I'd rather we didn't. > > > > > > > > > > > > > > > > > > > > > > > > > > > On Mon, 19 Jun 2023, 19:20 Beniamin Sandu, > > > > > > > > > <beniaminsa...@gmail.com> wrote: > > > > > > > > >> > > > > > > > > >> Signed-off-by: Beniamin Sandu <beniaminsa...@gmail.com> > > > > > > > > >> --- > > > > > > > > >> .../mdns/0009-remove-unneeded-headers.patch | 33 > > > > > > > > >> +++++++++++++++++++ > > > > > > > > >> .../recipes-protocols/mdns/mdns_1790.80.10.bb | 1 + > > > > > > > > >> 2 files changed, 34 insertions(+) > > > > > > > > >> create mode 100644 > > > > > > > > >> meta-networking/recipes-protocols/mdns/mdns/0009-remove-unneeded-headers.patch > > > > > > > > >> > > > > > > > > >> diff --git > > > > > > > > >> a/meta-networking/recipes-protocols/mdns/mdns/0009-remove-unneeded-headers.patch > > > > > > > > >> > > > > > > > > >> b/meta-networking/recipes-protocols/mdns/mdns/0009-remove-unneeded-headers.patch > > > > > > > > >> new file mode 100644 > > > > > > > > >> index 000000000..74d387368 > > > > > > > > >> --- /dev/null > > > > > > > > >> +++ > > > > > > > > >> b/meta-networking/recipes-protocols/mdns/mdns/0009-remove-unneeded-headers.patch > > > > > > > > >> @@ -0,0 +1,33 @@ > > > > > > > > >> +From 1d7e71e72c597ffcc19c04373a477d1fbd3ad955 Mon Sep 17 > > > > > > > > >> 00:00:00 2001 > > > > > > > > >> +From: Beniamin Sandu <beniaminsa...@gmail.com> > > > > > > > > >> +Date: Thu, 15 Jun 2023 17:02:58 +0000 > > > > > > > > >> +Subject: [PATCH] remove unneeded headers > > > > > > > > >> + > > > > > > > > >> +From a quick look, these seem to not be needed and having > > > > > > > > >> them > > > > > > > > >> +breaks the build with mbedtls 3.x. Without them it builds > > > > > > > > >> fine > > > > > > > > >> +on both 2.x and 3.x versions. > > > > > > > > >> + > > > > > > > > >> +Upstream-Status: Pending > > > > > > > > >> + > > > > > > > > >> +Signed-off-by: Beniamin Sandu <beniaminsa...@gmail.com> > > > > > > > > >> +--- > > > > > > > > >> + mDNSPosix/mbedtls.c | 2 -- > > > > > > > > >> + 1 file changed, 2 deletions(-) > > > > > > > > >> + > > > > > > > > >> +diff --git a/mDNSPosix/mbedtls.c b/mDNSPosix/mbedtls.c > > > > > > > > >> +index a73681b..ab8f8c7 100644 > > > > > > > > >> +--- a/mDNSPosix/mbedtls.c > > > > > > > > >> ++++ b/mDNSPosix/mbedtls.c > > > > > > > > >> +@@ -38,10 +38,8 @@ > > > > > > > > >> + #include <mbedtls/sha256.h> > > > > > > > > >> + #include <mbedtls/base64.h> > > > > > > > > >> + > > > > > > > > >> +-#include <mbedtls/certs.h> > > > > > > > > >> + #include <mbedtls/x509.h> > > > > > > > > >> + #include <mbedtls/ssl.h> > > > > > > > > >> +-#include <mbedtls/config.h> > > > > > > > > >> + > > > > > > > > >> + // Posix TLS server context > > > > > > > > >> + struct TLSContext_struct { > > > > > > > > >> +-- > > > > > > > > >> +2.34.1 > > > > > > > > >> diff --git > > > > > > > > >> a/meta-networking/recipes-protocols/mdns/mdns_1790.80.10.bb > > > > > > > > >> b/meta-networking/recipes-protocols/mdns/mdns_1790.80.10.bb > > > > > > > > >> index 612d343c8..46f1b70cb 100644 > > > > > > > > >> --- > > > > > > > > >> a/meta-networking/recipes-protocols/mdns/mdns_1790.80.10.bb > > > > > > > > >> +++ > > > > > > > > >> b/meta-networking/recipes-protocols/mdns/mdns_1790.80.10.bb > > > > > > > > >> @@ -21,6 +21,7 @@ SRC_URI = > > > > > > > > >> "git://github.com/apple-oss-distributions/mDNSResponder;protocol=https > > > > > > > > >> file://0006-Handle-noisy-netlink-sockets.patch \ > > > > > > > > >> > > > > > > > > >> file://0007-Mark-deleted-interfaces-as-being-changed.patch \ > > > > > > > > >> > > > > > > > > >> file://0008-Handle-errors-from-socket-calls.patch \ > > > > > > > > >> + file://0009-remove-unneeded-headers.patch \ > > > > > > > > >> file://mdns.service \ > > > > > > > > >> " > > > > > > > > >> SRCREV = "8769ab51605e465425d33d757f602ce5905ca639" > > > > > > > > >> -- > > > > > > > > >> 2.25.1 > > > > > > > > >> > > > > > > > > >> > > > > > > > > >> > > > > > > > > >> > > > > > > > > > > > > > > > > > > > > > > > > > > > > -- > > > > > > > Alex Kiernan > > > > > > > > > > > > > > > > > > > > -- > > > > > Alex Kiernan > > > > > > > > > > > > > > > > > > > > > > > > > > > -- > > > > Alex Kiernan > > > > -- > Alex Kiernan > > >
-=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#103467): https://lists.openembedded.org/g/openembedded-devel/message/103467 Mute This Topic: https://lists.openembedded.org/mt/99629056/21656 Group Owner: openembedded-devel+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
