Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.debian.bugs.dist > #1281894 > unrolled thread

Bug#1127767: transition: Qt6 6.10.2

Back to article view | Back to linux.debian.bugs.dist

Contents

  Bug#1127767: transition: Qt6 6.10.2 Patrick Franz <deltaone@debian.org> - 2026-02-12 20:00 +0100
    Bug#1127767: transition: Qt6 6.10.2 Sebastian Ramacher <sramacher@debian.org> - 2026-02-13 11:00 +0100
      Bug#1127767: transition: Qt6 6.10.2 Patrick Franz <deltaone@debian.org> - 2026-02-21 12:40 +0100
        Bug#1127767: transition: Qt6 6.10.2 Sebastian Ramacher <sramacher@debian.org> - 2026-02-28 22:20 +0100
          Bug#1127767: transition: Qt6 6.10.2 Sebastian Ramacher <sramacher@debian.org> - 2026-03-01 19:40 +0100
            Bug#1127767: transition: Qt6 6.10.2 Sebastian Ramacher <sramacher@debian.org> - 2026-03-18 23:40 +0100
              Bug#1127767: transition: Qt6 6.10.2 Paul Gevers <elbrus@debian.org> - 2026-03-23 20:50 +0100
                Bug#1127767: transition: Qt6 6.10.2 Sebastian Ramacher <sramacher@debian.org> - 2026-03-23 22:40 +0100

#1281894 — Bug#1127767: transition: Qt6 6.10.2

Package: release.debian.org
Severity: normal
X-Debbugs-Cc: deltaone@debian.org
User: release.debian.org@packages.debian.org
Usertags: transition

Hi Release Team,

we would like to request a transition for Qt 6.10.2 which is currently
available in experimental.
We have the usual ABI changes for the private headers and one binary
package has been removed (qt6-wayland-dev-tools, the contents have been
moved into qt6-base-dev-tools, so the fix will be easy).
Most reverse dependencies of qt6-wayland-dev-tools are maintained in
our team.

I have not done any rebuilds yet, but if you create the transition page,
it'll me which packages need to be rebuild and crucially in which order.

Once I've done those rebuilds, I'll get back to you.


Ben file:
---
title = "Qt 6.10.2";
is_affected = .depends ~ /qt6-.*-private-abi \(= 6\.10\.2\)/ | .depends ~ /qt6-.*-private-abi \(= 6\.9\.2\)/;
is_good = .depends ~ /qt6-.*-private-abi \(= 6\.10\.2\)/;
is_bad = .depends ~ /qt6-.*-private-abi \(= 6\.9\.2\)/;
---

Thank you very much.


-- 
Med vänliga hälsningar

Patrick Franz

[toc] | [next] | [standalone]

#1281965

Control: tags -1 moreinfo

On 2026-02-12 19:50:47 +0100, Patrick Franz wrote:
> Package: release.debian.org
> Severity: normal
> X-Debbugs-Cc: deltaone@debian.org
> User: release.debian.org@packages.debian.org
> Usertags: transition
> 
> Hi Release Team,
> 
> we would like to request a transition for Qt 6.10.2 which is currently
> available in experimental.
> We have the usual ABI changes for the private headers and one binary
> package has been removed (qt6-wayland-dev-tools, the contents have been
> moved into qt6-base-dev-tools, so the fix will be easy).
> Most reverse dependencies of qt6-wayland-dev-tools are maintained in
> our team.
> 
> I have not done any rebuilds yet, but if you create the transition page,
> it'll me which packages need to be rebuild and crucially in which order.

The transition page should appear with the next run of ben.

> Once I've done those rebuilds, I'll get back to you.

Adding the moreinfo tag for now. Please remove it once the rebuilds have
been done.

Cheers
-- 
Sebastian Ramacher

[toc] | [prev] | [next] | [standalone]

#1282985

Control: tags -1 -moreinfo

Hi Sebastian,

Am Freitag, 13. Februar 2026, 10:51:04 Mitteleuropäische Normalzeit 
schrieb Sebastian Ramacher:
[...]
> > I have not done any rebuilds yet, but if you create the transition
> > page, it'll me which packages need to be rebuild and crucially in
> > which order.
> The transition page should appear with the next run of ben.
> 
> > Once I've done those rebuilds, I'll get back to you.
> 
> Adding the moreinfo tag for now. Please remove it once the rebuilds
> have been done.

I've done all rebuilds now and it looks like this:

Many packages actually failed to build and we've managed to fix some of 
those already.

Almost 10 packages still FTBFS due to changes in Qt's build system (if 
you link against private targets, you now have to find them explicitly in 
CMake). I filed bugs against of all them but it may be time to raise the 
severity to serious. Most of these bugs should be relatively easy to fix 
and for a number of those, there is an upstream fix available.

A further 5-6 packages FTBFS due to a packaging change in Qt 6.10. All 
but one of those are maintained in our team, so we can take care of them 
ourselves. The fix is absolutely trivial but cannot be done until Qt 6.10 
arrives in unstable.

2 packages or so fail for different reasons, but bugs have been filed as 
well.


-- 
Med vänliga hälsningar

Patrick Franz

[toc] | [prev] | [next] | [standalone]

#1284042

Control: block -1 by 1125701

On 2026-02-21 12:35:53 +0100, Patrick Franz wrote:
> Control: tags -1 -moreinfo
> 
> Hi Sebastian,
> 
> Am Freitag, 13. Februar 2026, 10:51:04 Mitteleuropäische Normalzeit 
> schrieb Sebastian Ramacher:
> [...]
> > > I have not done any rebuilds yet, but if you create the transition
> > > page, it'll me which packages need to be rebuild and crucially in
> > > which order.
> > The transition page should appear with the next run of ben.
> > 
> > > Once I've done those rebuilds, I'll get back to you.
> > 
> > Adding the moreinfo tag for now. Please remove it once the rebuilds
> > have been done.
> 
> I've done all rebuilds now and it looks like this:
> 
> Many packages actually failed to build and we've managed to fix some of 
> those already.
> 
> Almost 10 packages still FTBFS due to changes in Qt's build system (if 
> you link against private targets, you now have to find them explicitly in 
> CMake). I filed bugs against of all them but it may be time to raise the 
> severity to serious. Most of these bugs should be relatively easy to fix 
> and for a number of those, there is an upstream fix available.
> 
> A further 5-6 packages FTBFS due to a packaging change in Qt 6.10. All 
> but one of those are maintained in our team, so we can take care of them 
> ourselves. The fix is absolutely trivial but cannot be done until Qt 6.10 
> arrives in unstable.
> 
> 2 packages or so fail for different reasons, but bugs have been filed as 
> well.

As this is a lock-step migration, we will need the onnxruntime
transition to finish first.

Cheers
-- 
Sebastian Ramacher

[toc] | [prev] | [next] | [standalone]

#1284162

Control: block -1 by 1129225

On 2026-02-28 22:12:19 +0100, Sebastian Ramacher wrote:
> Control: block -1 by 1125701
> 
> On 2026-02-21 12:35:53 +0100, Patrick Franz wrote:
> > Control: tags -1 -moreinfo
> > 
> > Hi Sebastian,
> > 
> > Am Freitag, 13. Februar 2026, 10:51:04 Mitteleuropäische Normalzeit 
> > schrieb Sebastian Ramacher:
> > [...]
> > > > I have not done any rebuilds yet, but if you create the transition
> > > > page, it'll me which packages need to be rebuild and crucially in
> > > > which order.
> > > The transition page should appear with the next run of ben.
> > > 
> > > > Once I've done those rebuilds, I'll get back to you.
> > > 
> > > Adding the moreinfo tag for now. Please remove it once the rebuilds
> > > have been done.
> > 
> > I've done all rebuilds now and it looks like this:
> > 
> > Many packages actually failed to build and we've managed to fix some of 
> > those already.
> > 
> > Almost 10 packages still FTBFS due to changes in Qt's build system (if 
> > you link against private targets, you now have to find them explicitly in 
> > CMake). I filed bugs against of all them but it may be time to raise the 
> > severity to serious. Most of these bugs should be relatively easy to fix 
> > and for a number of those, there is an upstream fix available.
> > 
> > A further 5-6 packages FTBFS due to a packaging change in Qt 6.10. All 
> > but one of those are maintained in our team, so we can take care of them 
> > ourselves. The fix is absolutely trivial but cannot be done until Qt 6.10 
> > arrives in unstable.
> > 
> > 2 packages or so fail for different reasons, but bugs have been filed as 
> > well.
> 
> As this is a lock-step migration, we will need the onnxruntime
> transition to finish first.

kaidan also will need to be handled. Adding also as a blocking bug.

Cheers
-- 
Sebastian Ramacher

[toc] | [prev] | [next] | [standalone]

#1286522

Control: tags -1 = confirmed

On 2026-03-01 19:38:05 +0100, Sebastian Ramacher wrote:
> Control: block -1 by 1129225
> 
> On 2026-02-28 22:12:19 +0100, Sebastian Ramacher wrote:
> > Control: block -1 by 1125701
> > 
> > On 2026-02-21 12:35:53 +0100, Patrick Franz wrote:
> > > Control: tags -1 -moreinfo
> > > 
> > > Hi Sebastian,
> > > 
> > > Am Freitag, 13. Februar 2026, 10:51:04 Mitteleuropäische Normalzeit 
> > > schrieb Sebastian Ramacher:
> > > [...]
> > > > > I have not done any rebuilds yet, but if you create the transition
> > > > > page, it'll me which packages need to be rebuild and crucially in
> > > > > which order.
> > > > The transition page should appear with the next run of ben.
> > > > 
> > > > > Once I've done those rebuilds, I'll get back to you.
> > > > 
> > > > Adding the moreinfo tag for now. Please remove it once the rebuilds
> > > > have been done.
> > > 
> > > I've done all rebuilds now and it looks like this:
> > > 
> > > Many packages actually failed to build and we've managed to fix some of 
> > > those already.
> > > 
> > > Almost 10 packages still FTBFS due to changes in Qt's build system (if 
> > > you link against private targets, you now have to find them explicitly in 
> > > CMake). I filed bugs against of all them but it may be time to raise the 
> > > severity to serious. Most of these bugs should be relatively easy to fix 
> > > and for a number of those, there is an upstream fix available.
> > > 
> > > A further 5-6 packages FTBFS due to a packaging change in Qt 6.10. All 
> > > but one of those are maintained in our team, so we can take care of them 
> > > ourselves. The fix is absolutely trivial but cannot be done until Qt 6.10 
> > > arrives in unstable.
> > > 
> > > 2 packages or so fail for different reasons, but bugs have been filed as 
> > > well.
> > 
> > As this is a lock-step migration, we will need the onnxruntime
> > transition to finish first.
> 
> kaidan also will need to be handled. Adding also as a blocking bug.

Go ahead

Cheers
-- 
Sebastian Ramacher

[toc] | [prev] | [next] | [standalone]

#1287115

[Multipart message — attachments visible in raw view] — view raw

Hi

On 3/18/26 23:32, Sebastian Ramacher wrote:
> Go ahead

I'm slightly worried about the dolfin autopkgtest after the binNMU [1], 
as I think it's trying to check for its abi (but I could be wrong):

https://ci.debian.net/packages/d/dolphin/unstable/amd64/69713934/

102s dh_acc: error: abi-compliance-checker -q -l libdolphinvcs-dev -v1 
4:25.12.1-1 -dump debian/libdolphinvcs-dev.acc -dump-path 
debian/libdolphinvcs-dev/usr/lib/x86_64-linux-gnu/dh-acc/libdolphinvcs-dev_4:25.12.1-1.abi.tar.gz 
returned exit code 6

Does that worry anyone?

Paul

[1] also visible in the excuses [2] for the binNMU itself
[2] https://release.debian.org/britney/update_excuses.html

[toc] | [prev] | [next] | [standalone]

#1287126

On 2026-03-23 20:44:51 +0100, Paul Gevers wrote:
> Hi
> 
> On 3/18/26 23:32, Sebastian Ramacher wrote:
> > Go ahead
> 
> I'm slightly worried about the dolfin autopkgtest after the binNMU [1], as I
> think it's trying to check for its abi (but I could be wrong):
> 
> https://ci.debian.net/packages/d/dolphin/unstable/amd64/69713934/
> 
> 102s dh_acc: error: abi-compliance-checker -q -l libdolphinvcs-dev -v1
> 4:25.12.1-1 -dump debian/libdolphinvcs-dev.acc -dump-path debian/libdolphinvcs-dev/usr/lib/x86_64-linux-gnu/dh-acc/libdolphinvcs-dev_4:25.12.1-1.abi.tar.gz
> returned exit code 6
> 
> Does that worry anyone?

Not really. It's abi-checker failing:

The GCC parameters:
  gcc -fdump-lang-raw -fkeep-inline-functions -c -x c++ -fpermissive -w  -fPIC -I/usr/lib/x86_64-linux-gnu/qt5/mkspecs/common/posix/ "/tmp/_LCZ3A3J1M/dump1.h"  -I/usr/include/x86_64-linux-gnu/qt6 -I/usr/include/x86_64-linux-gnu/qt6/QtGui -I/usr/include/x86_64-linux-gnu/qt6/QtCore

/tmp/_LCZ3A3J1M/dump1.h:9:13: error: expected identifier before 'const'
    9 |   namespace const{typedef int tmp_add_type_1;}
      |             ^~~~~
/tmp/_LCZ3A3J1M/dump1.h:9:18: error: expected unqualified-id before '{' token
    9 |   namespace const{typedef int tmp_add_type_1;}
      |                  ^
/tmp/_LCZ3A3J1M/dump1.h:10:10: error: 'tmp_add_type_1' in namespace '::' does not name a type
   10 |   const::tmp_add_type_1 tmp_add_func_1(){return 0;};
      |          ^~~~~~~~~~~~~~

Also note the qt5 paths in the command line. This will need a fix in
dolphin.

Cheers

> Paul
> 
> [1] also visible in the excuses [2] for the binNMU itself
> [2] https://release.debian.org/britney/update_excuses.html




-- 
Sebastian Ramacher

[toc] | [prev] | [standalone]

Back to top | Article view | linux.debian.bugs.dist

csiph-web