[PATCH-for-5.2 v4 00/17] ci: Move various jobs from Travis to GitLab CI

[PATCH-for-5.2 v4 00/17] ci: Move various jobs from Travis to GitLab CI

[Philippe Mathieu-Daudé]

We have various job covered on Travis-CI, but recommend the
community to use GitLab, so build failures are missed.

We need help to move the jobs to GitLab (we will keep the s390x
and ppc64 containerized jobs on Travis as there is no similar
offer on GitLab). Start moving jobs.

Resulting pipelines (96 jobs in 71 minutes and 4 seconds)
  https://gitlab.com/philmd/qemu/-/pipelines/213351318
  https://gitlab.com/philmd/qemu/-/pipelines/213378500

Patches 1-3 are included because without them some jobs are
failing; but they will be merged via different trees.

Since v3:
- moving various other jobs
- make some jobs optional

Since v2:
- Run m68k qtests (Thomas)
- Reworded hw/9pfs description (Greg)
- Added tags

Since v1:
- Fix Xen+9pfs config (Paolo)
- Run AVR qtests (Thomas)

Based-on: <20201104115706.3101190-1-philmd@redhat.com>

Philippe Mathieu-Daudé (16):
  configure: Check vhost-user is available to build
    vhost-user-blk-server
  hw/i386/acpi-build: Fix maybe-uninitialized error when ACPI hotplug
    off
  MAINTAINERS: Add gitlab-pipeline-status script to GitLab CI section
  gitlab-ci: Split CONFIGURE_ARGS one argument per line for
    build-disabled
  gitlab-ci: Update 'build-disabled' to cover all configurable options
  gitlab-ci: Move job testing --without-default-devices across to gitlab
  gitlab-ci: Move linux-user debug-tcg test across to gitlab
  gitlab-ci: Move gprof/gcov test across to gitlab
  gitlab-ci: Move user-static test across to gitlab
  gitlab-ci: Move tools-only and softfloat tests across to gitlab
  gitlab-ci: Move coroutine tests across to gitlab
  tests/docker: Install lttng-ust-devel package in Fedora image
  gitlab-ci: Move trace backend tests across to gitlab
  gitlab-ci: Add test for Xen (on CentOS 7)
  gitlab-ci: Make some ARM 32-bit jobs optional
  gitlab-ci: Make MIPS user-mode 32-bit little-endian job optional

Stefan Hajnoczi (1):
  meson: always include contrib/libvhost-user

 configure                                  |   3 +
 meson.build                                |   6 +-
 hw/i386/acpi-build.c                       |  45 ++--
 .gitlab-ci.d/crossbuilds.yml               |   9 +
 .gitlab-ci.yml                             | 228 +++++++++++++++++++--
 .travis.yml                                |  78 -------
 MAINTAINERS                                |   3 +-
 scripts/{travis => ci}/coverage-summary.sh |   2 +-
 tests/docker/dockerfiles/fedora.docker     |   1 +
 9 files changed, 249 insertions(+), 126 deletions(-)
 rename scripts/{travis => ci}/coverage-summary.sh (92%)

-- 
2.26.2

[PATCH-for-5.2 v4 01/17] configure: Check vhost-user is available to build vhost-user-blk-server

[Philippe Mathieu-Daudé]

Check vhost-user is available to build vhost-user-blk-server to fix:

 $ ../configure \
      --disable-vhost-user --enable-vhost-user-blk-server && \
   make qemu-nbd
 ...
 [505/505] Linking target qemu-nbd
 FAILED: qemu-nbd
 cc  -o qemu-nbd qemu-nbd.p/qemu-nbd.c.o -Wl,--as-needed -Wl,--no-undefined -pie -Wl,--whole-archive libblockdev.fa libblock.fa libcrypto.fa libauthz.fa libqom.fa libio.fa -Wl,--no-whole-archive -Wl,--warn-common -Wl,-z,relro -Wl,-z,now -m64 -fstack-protector-strong -Wl,--start-group libqemuutil.a libblockdev.fa libblock.fa libcrypto.fa libauthz.fa libqom.fa libio.fa @block.syms -lgio-2.0 -lgobject-2.0 -lglib-2.0 -lgio-2.0 -lgobject-2.0 -lglib-2.0 -pthread -lgnutls -lutil -lm -lgthread-2.0 -lglib-2.0 -lbz2 -lgthread-2.0 -lglib-2.0 -lssh -lrbd -lrados -lcurl -lxml2 -lzstd -lacl -lgfapi -lglusterfs -lgfrpc -lgfxdr -luuid -laio /usr/lib64/libz.so -L/usr/lib64/iscsi -liscsi -lnettle -lgnutls -lpam -Wl,--end-group
 /usr/bin/ld: libblockdev.fa(block_export_vhost-user-blk-server.c.o): in function `vu_blk_process_vq':
 block/export/vhost-user-blk-server.c:203: undefined reference to `vu_get_queue'
 /usr/bin/ld: block/export/vhost-user-blk-server.c:208: undefined reference to `vu_queue_pop'
 /usr/bin/ld: libblockdev.fa(block_export_vhost-user-blk-server.c.o): in function `vu_blk_queue_set_started':
 block/export/vhost-user-blk-server.c:228: undefined reference to `vu_get_queue'
 /usr/bin/ld: libblockdev.fa(block_export_vhost-user-blk-server.c.o): in function `vu_blk_req_complete':
 block/export/vhost-user-blk-server.c:55: undefined reference to `vu_queue_push'
 /usr/bin/ld: block/export/vhost-user-blk-server.c:56: undefined reference to `vu_queue_notify'
 /usr/bin/ld: libblockdev.fa(block_export_vhost-user-blk-server.c.o): in function `vu_blk_queue_set_started':
 block/export/vhost-user-blk-server.c:229: undefined reference to `vu_set_queue_handler'
 /usr/bin/ld: libqemuutil.a(util_vhost-user-server.c.o): in function `vu_client_trip':
 util/vhost-user-server.c:176: undefined reference to `vu_dispatch'
 /usr/bin/ld: util/vhost-user-server.c:180: undefined reference to `vu_deinit'
 /usr/bin/ld: libqemuutil.a(util_vhost-user-server.c.o): in function `vu_accept':
 util/vhost-user-server.c:291: undefined reference to `vu_init'
 collect2: error: ld returned 1 exit status
 ninja: build stopped: subcommand failed.
 make: *** [Makefile:171: run-ninja] Error 1

Fixes: bc15e44cb21 ("configure: introduce --enable-vhost-user-blk-server")
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
---
 configure | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/configure b/configure
index 805f7791503..5079ebb416a 100755
--- a/configure
+++ b/configure
@@ -2390,6 +2390,9 @@ fi
 
 # libvhost-user is Linux-only
 test "$vhost_user_blk_server" = "" && vhost_user_blk_server=$linux
+if test "$vhost_user_blk_server" = "yes" && test "$vhost_user" = "no"; then
+  error_exit "--enable-vhost-user-blk-server requires --enable-vhost-user"
+fi
 if test "$vhost_user_blk_server" = "yes" && test "$linux" = "no"; then
   error_exit "--enable-vhost-user-blk-server is only available on Linux"
 fi
-- 
2.26.2

[PATCH-for-5.2 v4 02/17] meson: always include contrib/libvhost-user

[Philippe Mathieu-Daudé]

From: Stefan Hajnoczi <stefanha@redhat.com>

libvhost-user is needed when CONFIG_LINUX is set. The CONFIG_VHOST_USER
check in meson.build is incorrect.

In fact, no explicit check is needed since this dependency is not built
by default. If something declares a dependency on libvhost-user then it
will be built, otherwise it won't be built (i.e. on non-Linux hosts).

This fixes ./configure --disable-vhost-user && make.

Fixes: bc15e44cb2191bbb2318878acdf5038134e56394 ("configure: introduce --enable-vhost-user-blk-server")
Reported-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reported-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-Id: <20201106210340.698771-1-stefanha@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
---
 meson.build | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/meson.build b/meson.build
index f5175010dfa..b4736203219 100644
--- a/meson.build
+++ b/meson.build
@@ -1450,11 +1450,7 @@
   'util',
 ]
 
-vhost_user = not_found
-if 'CONFIG_VHOST_USER' in config_host
-  subdir('contrib/libvhost-user')
-endif
-
+subdir('contrib/libvhost-user')
 subdir('qapi')
 subdir('qobject')
 subdir('stubs')
-- 
2.26.2

[PATCH-for-5.2 v4 03/17] hw/i386/acpi-build: Fix maybe-uninitialized error when ACPI hotplug off

[Philippe Mathieu-Daudé]

GCC 9.3.0 thinks that 'method' can be left uninitialized. This code
is already in the "if (bsel || pcihp_bridge_en)" block statement,
but it isn't smart enough to figure it out.

Restrict the code to be used only in the "if (bsel || pcihp_bridge_en)"
block statement to fix (on Ubuntu):

  ../hw/i386/acpi-build.c: In function 'build_append_pci_bus_devices':
  ../hw/i386/acpi-build.c:496:9: error: 'method' may be used uninitialized
  in this function [-Werror=maybe-uninitialized]
    496 |         aml_append(parent_scope, method);
        |         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  cc1: all warnings being treated as errors

Fixes: df4008c9c59 ("piix4: don't reserve hw resources when hotplug is off globally")
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
---
 hw/i386/acpi-build.c | 45 +++++++++++++++++++++-----------------------
 1 file changed, 21 insertions(+), 24 deletions(-)

diff --git a/hw/i386/acpi-build.c b/hw/i386/acpi-build.c
index 4f66642d887..1f5c2112452 100644
--- a/hw/i386/acpi-build.c
+++ b/hw/i386/acpi-build.c
@@ -465,34 +465,31 @@ static void build_append_pci_bus_devices(Aml *parent_scope, PCIBus *bus,
      */
     if (bsel || pcihp_bridge_en) {
         method = aml_method("PCNT", 0, AML_NOTSERIALIZED);
-    }
-    /* If bus supports hotplug select it and notify about local events */
-    if (bsel) {
-        uint64_t bsel_val = qnum_get_uint(qobject_to(QNum, bsel));
 
-        aml_append(method, aml_store(aml_int(bsel_val), aml_name("BNUM")));
-        aml_append(method,
-            aml_call2("DVNT", aml_name("PCIU"), aml_int(1) /* Device Check */)
-        );
-        aml_append(method,
-            aml_call2("DVNT", aml_name("PCID"), aml_int(3)/* Eject Request */)
-        );
-    }
+        /* If bus supports hotplug select it and notify about local events */
+        if (bsel) {
+            uint64_t bsel_val = qnum_get_uint(qobject_to(QNum, bsel));
 
-    /* Notify about child bus events in any case */
-    if (pcihp_bridge_en) {
-        QLIST_FOREACH(sec, &bus->child, sibling) {
-            int32_t devfn = sec->parent_dev->devfn;
-
-            if (pci_bus_is_root(sec) || pci_bus_is_express(sec)) {
-                continue;
-            }
-
-            aml_append(method, aml_name("^S%.02X.PCNT", devfn));
+            aml_append(method, aml_store(aml_int(bsel_val), aml_name("BNUM")));
+            aml_append(method, aml_call2("DVNT", aml_name("PCIU"),
+                                         aml_int(1))); /* Device Check */
+            aml_append(method, aml_call2("DVNT", aml_name("PCID"),
+                                         aml_int(3))); /* Eject Request */
+        }
+
+        /* Notify about child bus events in any case */
+        if (pcihp_bridge_en) {
+            QLIST_FOREACH(sec, &bus->child, sibling) {
+                int32_t devfn = sec->parent_dev->devfn;
+
+                if (pci_bus_is_root(sec) || pci_bus_is_express(sec)) {
+                    continue;
+                }
+
+                aml_append(method, aml_name("^S%.02X.PCNT", devfn));
+            }
         }
-    }
 
-    if (bsel || pcihp_bridge_en) {
         aml_append(parent_scope, method);
     }
     qobject_unref(bsel);
-- 
2.26.2

[PATCH-for-6.0 v4 04/17] MAINTAINERS: Add gitlab-pipeline-status script to GitLab CI section

[Philippe Mathieu-Daudé]

Do not let the gitlab-pipeline-status script unmaintained,
add it to the 'GitLab Continuous Integration' section.

Fixes: c02b2eac55e ("GitLab Gating CI: introduce pipeline-status contrib script")
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
---
Cc: Cleber Rosa <crosa@redhat.com>
---
 MAINTAINERS | 1 +
 1 file changed, 1 insertion(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index 63223e1183d..817c7c93f91 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -3146,6 +3146,7 @@ S: Maintained
 F: .gitlab-ci.yml
 F: .gitlab-ci.d/crossbuilds.yml
 F: .gitlab-ci.d/*py
+F: scripts/ci/gitlab-pipeline-status
 
 Guest Test Compilation Support
 M: Alex Bennée <alex.bennee@linaro.org>
-- 
2.26.2

[PATCH-for-6.0 v4 05/17] gitlab-ci: Split CONFIGURE_ARGS one argument per line for build-disabled

[Philippe Mathieu-Daudé]

We will keep adding/removing options to our 'configure' script,
so for easier maintainability it makes sense to have CONFIGURE_ARGS
declared as one option per line. This way we can review diff easily
(or rebase/cherry-pick).

No logical change.

Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
---
Cc: Daniel P . Berrange <berrange@redhat.com>
---
 .gitlab-ci.yml | 80 +++++++++++++++++++++++++++++++++++++++-----------
 1 file changed, 63 insertions(+), 17 deletions(-)

diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
index 3b15ae5c302..9f090cca5e3 100644
--- a/.gitlab-ci.yml
+++ b/.gitlab-ci.yml
@@ -203,23 +203,69 @@ build-disabled:
   <<: *native_build_job_definition
   variables:
     IMAGE: fedora
-    CONFIGURE_ARGS: --disable-attr --disable-avx2 --disable-bochs
-      --disable-brlapi --disable-bzip2 --disable-cap-ng --disable-capstone
-      --disable-cloop --disable-coroutine-pool --disable-curl --disable-curses
-      --disable-dmg --disable-docs --disable-glusterfs --disable-gnutls
-      --disable-gtk --disable-guest-agent --disable-iconv --disable-kvm
-      --disable-libiscsi --disable-libpmem --disable-libssh --disable-libusb
-      --disable-libxml2 --disable-linux-aio --disable-live-block-migration
-      --disable-lzo --disable-malloc-trim --disable-mpath --disable-nettle
-      --disable-numa --disable-parallels --disable-pie --disable-qcow1
-      --disable-qed --disable-qom-cast-debug --disable-rbd --disable-rdma
-      --disable-replication --disable-sdl --disable-seccomp --disable-sheepdog
-      --disable-slirp --disable-smartcard --disable-snappy --disable-spice
-      --disable-strip --disable-tpm --disable-usb-redir --disable-vdi
-      --disable-vhost-crypto --disable-vhost-net --disable-vhost-scsi
-      --disable-vhost-user --disable-vhost-vdpa --disable-vhost-vsock
-      --disable-virglrenderer --disable-vnc --disable-vte --disable-vvfat
-      --disable-xen --disable-zstd
+    CONFIGURE_ARGS:
+      --disable-attr
+      --disable-avx2
+      --disable-bochs
+      --disable-brlapi
+      --disable-bzip2
+      --disable-cap-ng
+      --disable-capstone
+      --disable-cloop
+      --disable-coroutine-pool
+      --disable-curl
+      --disable-curses
+      --disable-dmg
+      --disable-docs
+      --disable-glusterfs
+      --disable-gnutls
+      --disable-gtk
+      --disable-guest-agent
+      --disable-iconv
+      --disable-kvm
+      --disable-libiscsi
+      --disable-libpmem
+      --disable-libssh
+      --disable-libusb
+      --disable-libxml2
+      --disable-linux-aio
+      --disable-live-block-migration
+      --disable-lzo
+      --disable-malloc-trim
+      --disable-mpath
+      --disable-nettle
+      --disable-numa
+      --disable-parallels
+      --disable-pie
+      --disable-qcow1
+      --disable-qed
+      --disable-qom-cast-debug
+      --disable-rbd
+      --disable-rdma
+      --disable-replication
+      --disable-sdl
+      --disable-seccomp
+      --disable-sheepdog
+      --disable-slirp
+      --disable-smartcard
+      --disable-snappy
+      --disable-spice
+      --disable-strip
+      --disable-tpm
+      --disable-usb-redir
+      --disable-vdi
+      --disable-vhost-crypto
+      --disable-vhost-net
+      --disable-vhost-scsi
+      --disable-vhost-user
+      --disable-vhost-vdpa
+      --disable-vhost-vsock
+      --disable-virglrenderer
+      --disable-vnc
+      --disable-vte
+      --disable-vvfat
+      --disable-xen
+      --disable-zstd
     TARGETS: arm-softmmu i386-softmmu ppc64-softmmu mips64-softmmu
       s390x-softmmu i386-linux-user
     MAKE_CHECK_ARGS: check-qtest SPEED=slow
-- 
2.26.2

[PATCH-for-6.0 v4 06/17] gitlab-ci: Update 'build-disabled' to cover all configurable options

[Philippe Mathieu-Daudé]

Add these missing options to the 'build-disabled' job:
  --disable-auth-pam             (commit 8953caf3cd38)
  --disable-gcrypt               (commit 91bfcdb01d48)
  --disable-keyring              (commit 54e7aac05624)
  --disable-libudev              (commit 5c53015a480b)
  --disable-opengl               (commit da076ffed6b9)
  --disable-sparse               (commit 03b4fe7de226)

Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
---
 .gitlab-ci.yml | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
index 9f090cca5e3..b98800462ed 100644
--- a/.gitlab-ci.yml
+++ b/.gitlab-ci.yml
@@ -205,6 +205,7 @@ build-disabled:
     IMAGE: fedora
     CONFIGURE_ARGS:
       --disable-attr
+      --disable-auth-pam
       --disable-avx2
       --disable-bochs
       --disable-brlapi
@@ -217,15 +218,18 @@ build-disabled:
       --disable-curses
       --disable-dmg
       --disable-docs
+      --disable-gcrypt
       --disable-glusterfs
       --disable-gnutls
       --disable-gtk
       --disable-guest-agent
       --disable-iconv
+      --disable-keyring
       --disable-kvm
       --disable-libiscsi
       --disable-libpmem
       --disable-libssh
+      --disable-libudev
       --disable-libusb
       --disable-libxml2
       --disable-linux-aio
@@ -235,6 +239,7 @@ build-disabled:
       --disable-mpath
       --disable-nettle
       --disable-numa
+      --disable-opengl
       --disable-parallels
       --disable-pie
       --disable-qcow1
@@ -249,6 +254,7 @@ build-disabled:
       --disable-slirp
       --disable-smartcard
       --disable-snappy
+      --disable-sparse
       --disable-spice
       --disable-strip
       --disable-tpm
@@ -258,6 +264,7 @@ build-disabled:
       --disable-vhost-net
       --disable-vhost-scsi
       --disable-vhost-user
+      --disable-vhost-user-blk-server
       --disable-vhost-vdpa
       --disable-vhost-vsock
       --disable-virglrenderer
-- 
2.26.2

[PATCH-for-6.0 v4 07/17] gitlab-ci: Move job testing --without-default-devices across to gitlab

[Philippe Mathieu-Daudé]

Similarly to commit 8cdb2cef3f1, move the job testing the
'--without-default-devices' configure option to GitLab.

Since building all softmmu targets takes too long, split
the job in 2.

As smoke test, run the qtests on the AVR and m68k targets.

Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
---
Cc: Paolo Bonzini <pbonzini@redhat.com>

v3 had:
Reviewed-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Wainer dos Santos Moschetta <wainersm@redhat.com>
---
 .gitlab-ci.yml | 46 ++++++++++++++++++++++++++++++++++++++++++++++
 .travis.yml    |  8 --------
 2 files changed, 46 insertions(+), 8 deletions(-)

diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
index b98800462ed..3fc3d0568c6 100644
--- a/.gitlab-ci.yml
+++ b/.gitlab-ci.yml
@@ -315,6 +315,52 @@ build-user-plugins:
     MAKE_CHECK_ARGS: check-tcg
   timeout: 1h 30m
 
+build-system-ubuntu-without-default-devices 1/2:
+  <<: *native_build_job_definition
+  variables:
+    IMAGE: ubuntu2004
+    CONFIGURE_ARGS: --without-default-devices --disable-tools --disable-docs
+    TARGETS:
+      aarch64-softmmu
+      alpha-softmmu
+      arm-softmmu
+      avr-softmmu
+      cris-softmmu
+      hppa-softmmu
+      i386-softmmu
+      m68k-softmmu
+      microblazeel-softmmu
+      microblaze-softmmu
+      mips64el-softmmu
+      mips64-softmmu
+      mipsel-softmmu
+      mips-softmmu
+      moxie-softmmu
+    MAKE_CHECK_ARGS: check-qtest-avr check-qtest-m68k
+
+build-system-ubuntu-without-default-devices 2/2:
+  <<: *native_build_job_definition
+  variables:
+    IMAGE: ubuntu2004
+    CONFIGURE_ARGS: --without-default-devices --disable-tools --disable-docs
+    TARGETS:
+      nios2-softmmu
+      or1k-softmmu
+      ppc64-softmmu
+      ppc-softmmu
+      riscv32-softmmu
+      riscv64-softmmu
+      rx-softmmu
+      s390x-softmmu
+      sh4eb-softmmu
+      sh4-softmmu
+      sparc64-softmmu
+      sparc-softmmu
+      tricore-softmmu
+      x86_64-softmmu
+      xtensaeb-softmmu
+      xtensa-softmmu
+
 build-clang:
   <<: *native_build_job_definition
   variables:
diff --git a/.travis.yml b/.travis.yml
index a3d78171cab..15d92291358 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -224,14 +224,6 @@ jobs:
         - ${SRC_DIR}/scripts/travis/coverage-summary.sh
 
 
-    # We manually include builds which we disable "make check" for
-    - name: "GCC without-default-devices (softmmu)"
-      env:
-        - CONFIG="--without-default-devices --disable-user"
-        - CACHE_NAME="${TRAVIS_BRANCH}-linux-gcc-default"
-        - TEST_CMD=""
-
-
     # We don't need to exercise every backend with every front-end
     - name: "GCC trace log,simple,syslog (user)"
       env:
-- 
2.26.2

[PATCH-for-6.0 v4 08/17] gitlab-ci: Move linux-user debug-tcg test across to gitlab

[Philippe Mathieu-Daudé]

Similarly to commit 8cdb2cef3f1, move the linux-user (debug-tcg)
test to GitLab.

Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
---
Cc: Laurent Vivier <laurent@vivier.eu>
---
 .gitlab-ci.yml | 7 +++++++
 .travis.yml    | 9 ---------
 2 files changed, 7 insertions(+), 9 deletions(-)

diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
index 3fc3d0568c6..80082a602b8 100644
--- a/.gitlab-ci.yml
+++ b/.gitlab-ci.yml
@@ -304,6 +304,13 @@ build-user:
     CONFIGURE_ARGS: --disable-tools --disable-system
     MAKE_CHECK_ARGS: check-tcg
 
+build-user-debug:
+  <<: *native_build_job_definition
+  variables:
+    IMAGE: debian-all-test-cross
+    CONFIGURE_ARGS: --disable-tools --disable-system --enable-debug-tcg
+    MAKE_CHECK_ARGS: check-tcg
+
 # Run check-tcg against linux-user (with plugins)
 # we skip sparc64-linux-user until it has been fixed somewhat
 # we skip cris-linux-user as it doesn't use the common run loop
diff --git a/.travis.yml b/.travis.yml
index 15d92291358..bee6197290d 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -293,15 +293,6 @@ jobs:
         - ${SRC_DIR}/configure ${CONFIG} --extra-cflags="-g3 -O0 -fsanitize=thread" || { cat config.log meson-logs/meson-log.txt && exit 1; }
 
 
-    # Run check-tcg against linux-user
-    - name: "GCC check-tcg (user)"
-      env:
-        - CONFIG="--disable-system --enable-debug-tcg"
-        - TEST_BUILD_CMD="make build-tcg"
-        - TEST_CMD="make check-tcg"
-        - CACHE_NAME="${TRAVIS_BRANCH}-linux-gcc-debug-tcg"
-
-
     # Run check-tcg against softmmu targets
     - name: "GCC check-tcg (some-softmmu)"
       env:
-- 
2.26.2

[PATCH-for-6.0 v4 09/17] gitlab-ci: Move gprof/gcov test across to gitlab

[Philippe Mathieu-Daudé]

Similarly to commit 8cdb2cef3f1, move the gprof/gcov test to GitLab.

The coverage-summary.sh script is not Travis-CI specific, make it
generic.

Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
---
 .gitlab-ci.yml                             | 12 ++++++++++++
 .travis.yml                                | 14 --------------
 MAINTAINERS                                |  2 +-
 scripts/{travis => ci}/coverage-summary.sh |  2 +-
 4 files changed, 14 insertions(+), 16 deletions(-)
 rename scripts/{travis => ci}/coverage-summary.sh (92%)

diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
index 80082a602b8..d1583cee5db 100644
--- a/.gitlab-ci.yml
+++ b/.gitlab-ci.yml
@@ -403,6 +403,18 @@ check-deprecated:
     MAKE_CHECK_ARGS: check-tcg
   allow_failure: true
 
+# gprof/gcov are GCC features
+build-gprof-gcov:
+  <<: *native_build_job_definition
+  variables:
+    IMAGE: ubuntu2004
+    CONFIGURE_ARGS: --enable-gprof --enable-gcov
+    MAKE_CHECK_ARGS: build-tcg
+    TARGETS: aarch64-softmmu mips64-softmmu ppc64-softmmu
+             riscv64-softmmu s390x-softmmu x86_64-softmmu
+  after_script:
+    - ${CI_PROJECT_DIR}/scripts/ci/coverage-summary.sh
+
 build-oss-fuzz:
   <<: *native_build_job_definition
   variables:
diff --git a/.travis.yml b/.travis.yml
index bee6197290d..6087e2909cd 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -210,20 +210,6 @@ jobs:
       compiler: clang
 
 
-    # gprof/gcov are GCC features
-    - name: "GCC gprof/gcov"
-      dist: bionic
-      addons:
-        apt:
-          packages:
-            - ninja-build
-      env:
-        - CONFIG="--enable-gprof --enable-gcov --disable-libssh
-                  --target-list=${MAIN_SOFTMMU_TARGETS}"
-      after_success:
-        - ${SRC_DIR}/scripts/travis/coverage-summary.sh
-
-
     # We don't need to exercise every backend with every front-end
     - name: "GCC trace log,simple,syslog (user)"
       env:
diff --git a/MAINTAINERS b/MAINTAINERS
index 817c7c93f91..3d8c586d444 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -3121,7 +3121,7 @@ R: Philippe Mathieu-Daudé <philmd@redhat.com>
 S: Maintained
 F: .github/lockdown.yml
 F: .travis.yml
-F: scripts/travis/
+F: scripts/ci/
 F: .shippable.yml
 F: tests/docker/
 F: tests/vm/
diff --git a/scripts/travis/coverage-summary.sh b/scripts/ci/coverage-summary.sh
similarity index 92%
rename from scripts/travis/coverage-summary.sh
rename to scripts/ci/coverage-summary.sh
index d7086cf9ca6..8d9fb4de401 100755
--- a/scripts/travis/coverage-summary.sh
+++ b/scripts/ci/coverage-summary.sh
@@ -3,7 +3,7 @@
 # Author: Alex Bennée <alex.bennee@linaro.org>
 #
 # Summerise the state of code coverage with gcovr and tweak the output
-# to be more sane on Travis hosts. As we expect to be executed on a
+# to be more sane on CI runner. As we expect to be executed on a
 # throw away CI instance we do spam temp files all over the shop. You
 # most likely don't want to execute this script but just call gcovr
 # directly. See also "make coverage-report"
-- 
2.26.2

[PATCH-for-6.0 v4 10/17] gitlab-ci: Move user-static test across to gitlab

[Philippe Mathieu-Daudé]

Similarly to commit 8cdb2cef3f1, move the user-static test to GitLab.

Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
---
Cc: Laurent Vivier <laurent@vivier.eu>
---
 .gitlab-ci.yml | 7 +++++++
 .travis.yml    | 5 -----
 2 files changed, 7 insertions(+), 5 deletions(-)

diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
index d1583cee5db..a91870cf85e 100644
--- a/.gitlab-ci.yml
+++ b/.gitlab-ci.yml
@@ -311,6 +311,13 @@ build-user-debug:
     CONFIGURE_ARGS: --disable-tools --disable-system --enable-debug-tcg
     MAKE_CHECK_ARGS: check-tcg
 
+build-user-static:
+  <<: *native_build_job_definition
+  variables:
+    IMAGE: debian-all-test-cross
+    CONFIGURE_ARGS: --disable-tools --disable-system --static
+    MAKE_CHECK_ARGS: check-tcg
+
 # Run check-tcg against linux-user (with plugins)
 # we skip sparc64-linux-user until it has been fixed somewhat
 # we skip cris-linux-user as it doesn't use the common run loop
diff --git a/.travis.yml b/.travis.yml
index 6087e2909cd..5fb69c5fb02 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -119,11 +119,6 @@ after_script:
 
 jobs:
   include:
-    - name: "GCC static (user)"
-      env:
-        - CONFIG="--disable-system --static"
-        - CACHE_NAME="${TRAVIS_BRANCH}-linux-gcc-default"
-
     # Just build tools and run minimal unit and softfloat checks
     - name: "GCC check-unit and check-softfloat"
       env:
-- 
2.26.2

[PATCH-for-6.0 v4 11/17] gitlab-ci: Move tools-only and softfloat tests across to gitlab

[Philippe Mathieu-Daudé]

Similarly to commit 8cdb2cef3f1, move the tools-only and softfloat
test to GitLab (splitting it in 2 different jobs).

Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
---
Cc: Alex Bennée <alex.bennee@linaro.org>
Cc: Emilio G. Cota <cota@braap.org>
---
 .gitlab-ci.yml | 14 ++++++++++++++
 .travis.yml    |  9 ---------
 2 files changed, 14 insertions(+), 9 deletions(-)

diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
index a91870cf85e..a799803afb9 100644
--- a/.gitlab-ci.yml
+++ b/.gitlab-ci.yml
@@ -297,6 +297,20 @@ build-tcg-disabled:
             208 209 215 216 218 222 227 234 246 247 248 250 254 255 257 258
             260 261 262 263 264 270 272 273 277 279
 
+build-tools:
+  <<: *native_build_job_definition
+  variables:
+    IMAGE: debian-all-test-cross
+    CONFIGURE_ARGS: --enable-tools --disable-system --disable-user
+
+# run minimal unit and softfloat checks
+build-softfloat:
+  <<: *native_build_job_definition
+  variables:
+    IMAGE: debian-all-test-cross
+    CONFIGURE_ARGS: --disable-tools --disable-system --disable-user
+    MAKE_CHECK_ARGS: check-unit check-softfloat
+
 build-user:
   <<: *native_build_job_definition
   variables:
diff --git a/.travis.yml b/.travis.yml
index 5fb69c5fb02..15550fb023f 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -119,15 +119,6 @@ after_script:
 
 jobs:
   include:
-    # Just build tools and run minimal unit and softfloat checks
-    - name: "GCC check-unit and check-softfloat"
-      env:
-        - BASE_CONFIG="--enable-tools"
-        - CONFIG="--disable-user --disable-system"
-        - TEST_CMD="make check-unit check-softfloat -j${JOBS}"
-        - CACHE_NAME="${TRAVIS_BRANCH}-linux-gcc-default"
-
-
     # --enable-debug implies --enable-debug-tcg, also runs quite a bit slower
     - name: "GCC debug (main-softmmu)"
       env:
-- 
2.26.2

[PATCH-for-6.0 v4 12/17] gitlab-ci: Move coroutine tests across to gitlab

[Philippe Mathieu-Daudé]

Similarly to commit 8cdb2cef3f1, move the coroutine tests to GitLab.

Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
---
Cc: Stefan Hajnoczi <stefanha@redhat.com>
Cc: Kevin Wolf <kwolf@redhat.com>
---
 .gitlab-ci.yml | 16 ++++++++++++++++
 .travis.yml    | 14 --------------
 2 files changed, 16 insertions(+), 14 deletions(-)

diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
index a799803afb9..6552a832939 100644
--- a/.gitlab-ci.yml
+++ b/.gitlab-ci.yml
@@ -475,6 +475,22 @@ build-tci:
     - QTEST_QEMU_BINARY="./qemu-system-x86_64" ./tests/qtest/pxe-test
     - QTEST_QEMU_BINARY="./qemu-system-s390x" ./tests/qtest/pxe-test -m slow
 
+# Alternate coroutines implementations are only really of interest to KVM users
+# However we can't test against KVM on Travis so we can only run unit tests
+build-coroutine-ucontext:
+  <<: *native_build_job_definition
+  variables:
+    IMAGE: ubuntu2004
+    CONFIGURE_ARGS: --with-coroutine=ucontext --disable-tcg
+    MAKE_CHECK_ARGS: check-unit
+
+build-coroutine-sigaltstack:
+  <<: *native_build_job_definition
+  variables:
+    IMAGE: ubuntu2004
+    CONFIGURE_ARGS: --with-coroutine=sigaltstack --disable-tcg
+    MAKE_CHECK_ARGS: check-unit
+
 # Most jobs test latest gcrypt or nettle builds
 #
 # These jobs test old gcrypt and nettle from RHEL7
diff --git a/.travis.yml b/.travis.yml
index 15550fb023f..8ef31f8d8b6 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -139,20 +139,6 @@ jobs:
         - CACHE_NAME="${TRAVIS_BRANCH}-linux-gcc-default"
 
 
-    # Alternate coroutines implementations are only really of interest to KVM users
-    # However we can't test against KVM on Travis so we can only run unit tests
-    - name: "check-unit coroutine=ucontext"
-      env:
-        - CONFIG="--with-coroutine=ucontext --disable-tcg"
-        - TEST_CMD="make check-unit -j${JOBS} V=1"
-
-
-    - name: "check-unit coroutine=sigaltstack"
-      env:
-        - CONFIG="--with-coroutine=sigaltstack --disable-tcg"
-        - TEST_CMD="make check-unit -j${JOBS} V=1"
-
-
     # Check we can build docs and tools (out of tree)
     - name: "tools and docs (bionic)"
       dist: bionic
-- 
2.26.2

[PATCH-for-6.0 v4 13/17] tests/docker: Install lttng-ust-devel package in Fedora image

[Philippe Mathieu-Daudé]

Install the lttng-ust-devel package to be able to
build QEMU using the User-Space Tracer trace backend
(configure --enable-trace-backends=ust).

Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
---
Cc: Stefan Hajnoczi <stefanha@redhat.com>
Cc: Daniel P . Berrange <berrange@redhat.com>
---
 tests/docker/dockerfiles/fedora.docker | 1 +
 1 file changed, 1 insertion(+)

diff --git a/tests/docker/dockerfiles/fedora.docker b/tests/docker/dockerfiles/fedora.docker
index 0b5053f2d09..ac90f187882 100644
--- a/tests/docker/dockerfiles/fedora.docker
+++ b/tests/docker/dockerfiles/fedora.docker
@@ -44,6 +44,7 @@ ENV PACKAGES \
     libxml2-devel \
     libzstd-devel \
     llvm \
+    lttng-ust-devel \
     lzo-devel \
     make \
     mingw32-bzip2 \
-- 
2.26.2

[PATCH-for-6.0 v4 14/17] gitlab-ci: Move trace backend tests across to gitlab

[Philippe Mathieu-Daudé]

Similarly to commit 8cdb2cef3f1, move the trace backend
tests to GitLab.

Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
---
Cc: Stefan Hajnoczi <stefanha@redhat.com>
---
 .gitlab-ci.yml | 18 ++++++++++++++++++
 .travis.yml    | 19 -------------------
 2 files changed, 18 insertions(+), 19 deletions(-)

diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
index 6552a832939..2f0da7b3dc1 100644
--- a/.gitlab-ci.yml
+++ b/.gitlab-ci.yml
@@ -557,6 +557,24 @@ check-crypto-only-gnutls:
     IMAGE: centos7
     MAKE_CHECK_ARGS: check
 
+# We don't need to exercise every backend with every front-end
+build-trace-multi-user:
+  <<: *native_build_job_definition
+  variables:
+    IMAGE: ubuntu2004
+    CONFIGURE_ARGS: --enable-trace-backends=log,simple,syslog --disable-system
+
+build-trace-ftrace-system:
+  <<: *native_build_job_definition
+  variables:
+    IMAGE: ubuntu2004
+    CONFIGURE_ARGS: --enable-trace-backends=ftrace --target-list=aarch64-softmmu
+
+build-trace-ust-system:
+  <<: *native_build_job_definition
+  variables:
+    IMAGE: fedora
+    CONFIGURE_ARGS: --enable-trace-backends=ust --target-list=x86_64-softmmu --disable-tcg
 
 check-patch:
   stage: build
diff --git a/.travis.yml b/.travis.yml
index 8ef31f8d8b6..ff5d5ead579 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -182,25 +182,6 @@ jobs:
       compiler: clang
 
 
-    # We don't need to exercise every backend with every front-end
-    - name: "GCC trace log,simple,syslog (user)"
-      env:
-        - CONFIG="--enable-trace-backends=log,simple,syslog --disable-system"
-        - TEST_CMD=""
-
-
-    - name: "GCC trace ftrace (x86_64-softmmu)"
-      env:
-        - CONFIG="--enable-trace-backends=ftrace --target-list=x86_64-softmmu"
-        - TEST_CMD=""
-
-
-    - name: "GCC trace ust (x86_64-softmmu)"
-      env:
-        - CONFIG="--enable-trace-backends=ust --target-list=x86_64-softmmu"
-        - TEST_CMD=""
-
-
     # Using newer GCC with sanitizers
     - name: "GCC9 with sanitizers (softmmu)"
       dist: bionic
-- 
2.26.2

[PATCH-for-6.0 v4 15/17] gitlab-ci: Add test for Xen (on CentOS 7)

[Philippe Mathieu-Daudé]

Xen packages are available in CentOS 7, but have been
removed from CentOS 8. Use the CentOS 7 container.

Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
---
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Stefano Stabellini <sstabellini@kernel.org>
Cc: Anthony Perard <anthony.perard@citrix.com>
Cc: Paul Durrant <paul@xen.org>
Cc: xen-devel@lists.xenproject.org
---
 .gitlab-ci.yml | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)

diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
index 2f0da7b3dc1..8e15266c277 100644
--- a/.gitlab-ci.yml
+++ b/.gitlab-ci.yml
@@ -557,6 +557,27 @@ check-crypto-only-gnutls:
     IMAGE: centos7
     MAKE_CHECK_ARGS: check
 
+build-xen-centos:
+  <<: *native_build_job_definition
+  variables:
+    IMAGE: centos7
+    TARGETS: i386-softmmu x86_64-softmmu
+    CONFIGURE_ARGS: --enable-xen
+    MAKE_CHECK_ARGS: check-build
+  artifacts:
+    paths:
+      - build
+
+check-xen-centos:
+  <<: *native_test_job_definition
+  needs:
+    - job: build-xen-centos
+      artifacts: true
+  variables:
+    IMAGE: centos7
+    MAKE_CHECK_ARGS: check
+
+
 # We don't need to exercise every backend with every front-end
 build-trace-multi-user:
   <<: *native_build_job_definition
-- 
2.26.2

[PATCH-for-6.0 v4 15/17] gitlab-ci: Add test for Xen (on CentOS 7)

[Philippe Mathieu-Daudé]

Xen packages are available in CentOS 7, but have been
removed from CentOS 8. Use the CentOS 7 container.

Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
---
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Stefano Stabellini <sstabellini@kernel.org>
Cc: Anthony Perard <anthony.perard@citrix.com>
Cc: Paul Durrant <paul@xen.org>
Cc: xen-devel@lists.xenproject.org
---
 .gitlab-ci.yml | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)

diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
index 2f0da7b3dc1..8e15266c277 100644
--- a/.gitlab-ci.yml
+++ b/.gitlab-ci.yml
@@ -557,6 +557,27 @@ check-crypto-only-gnutls:
     IMAGE: centos7
     MAKE_CHECK_ARGS: check
 
+build-xen-centos:
+  <<: *native_build_job_definition
+  variables:
+    IMAGE: centos7
+    TARGETS: i386-softmmu x86_64-softmmu
+    CONFIGURE_ARGS: --enable-xen
+    MAKE_CHECK_ARGS: check-build
+  artifacts:
+    paths:
+      - build
+
+check-xen-centos:
+  <<: *native_test_job_definition
+  needs:
+    - job: build-xen-centos
+      artifacts: true
+  variables:
+    IMAGE: centos7
+    MAKE_CHECK_ARGS: check
+
+
 # We don't need to exercise every backend with every front-end
 build-trace-multi-user:
   <<: *native_build_job_definition
-- 
2.26.2

[PATCH-for-6.0 v4 16/17] gitlab-ci: Make some ARM 32-bit jobs optional

[Philippe Mathieu-Daudé]

The armel/armhf targets have a lot in common. Instead of
testing both user/system emulation on each target, test
each one once. Mark the other combination optional, so
user can still run the tests manually.

We now test system-mode emulation on armel target, and
user-mode emulation to armhf target:

- armel-system          Yes
- armel-user            Optional
- armhf-system          Optional
- armhf-user            Yes

Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
---
Cc: qemu-arm@nongnu.org
Cc: Laurent Vivier <laurent@vivier.eu>
---
 .gitlab-ci.d/crossbuilds.yml | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/.gitlab-ci.d/crossbuilds.yml b/.gitlab-ci.d/crossbuilds.yml
index 03ebfabb3fa..e4526e98232 100644
--- a/.gitlab-ci.d/crossbuilds.yml
+++ b/.gitlab-ci.d/crossbuilds.yml
@@ -32,11 +32,17 @@ cross-armel-user:
   <<: *cross_user_build_job_definition
   variables:
     IMAGE: debian-armel-cross
+  rules:
+    - when: manual
+      allow_failure: true
 
 cross-armhf-system:
   <<: *cross_system_build_job_definition
   variables:
     IMAGE: debian-armhf-cross
+  rules:
+    - when: manual
+      allow_failure: true
 
 cross-armhf-user:
   <<: *cross_user_build_job_definition
-- 
2.26.2

[PATCH-for-6.0 v4 17/17] gitlab-ci: Make MIPS user-mode 32-bit little-endian job optional

[Philippe Mathieu-Daudé]

Keep the mips user-mode emulation by default on 32-bit big endian
and 64-bit little endian targets, but make it optional on 32-bit
little endian. User can still run the job manually.

Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
---
Cc: Laurent Vivier <laurent@vivier.eu>
---
 .gitlab-ci.d/crossbuilds.yml | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/.gitlab-ci.d/crossbuilds.yml b/.gitlab-ci.d/crossbuilds.yml
index e4526e98232..5f611b5ff42 100644
--- a/.gitlab-ci.d/crossbuilds.yml
+++ b/.gitlab-ci.d/crossbuilds.yml
@@ -78,6 +78,9 @@ cross-mipsel-user:
   <<: *cross_user_build_job_definition
   variables:
     IMAGE: debian-mipsel-cross
+  rules:
+    - when: manual
+      allow_failure: true
 
 cross-mips64el-system:
   <<: *cross_system_build_job_definition
-- 
2.26.2

Re: [PATCH-for-6.0 v4 16/17] gitlab-ci: Make some ARM 32-bit jobs optional

[Philippe Mathieu-Daudé]

On 11/8/20 9:45 PM, Philippe Mathieu-Daudé wrote:
> The armel/armhf targets have a lot in common. Instead of
> testing both user/system emulation on each target, test
> each one once. Mark the other combination optional, so
> user can still run the tests manually.
> 
> We now test system-mode emulation on armel target, and
> user-mode emulation to armhf target:
> 
> - armel-system          Yes
> - armel-user            Optional
> - armhf-system          Optional
> - armhf-user            Yes
> 
> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
> ---
> Cc: qemu-arm@nongnu.org
> Cc: Laurent Vivier <laurent@vivier.eu>
> ---
>  .gitlab-ci.d/crossbuilds.yml | 6 ++++++
>  1 file changed, 6 insertions(+)
> 
> diff --git a/.gitlab-ci.d/crossbuilds.yml b/.gitlab-ci.d/crossbuilds.yml
> index 03ebfabb3fa..e4526e98232 100644
> --- a/.gitlab-ci.d/crossbuilds.yml
> +++ b/.gitlab-ci.d/crossbuilds.yml
> @@ -32,11 +32,17 @@ cross-armel-user:
>    <<: *cross_user_build_job_definition
>    variables:
>      IMAGE: debian-armel-cross
> +  rules:

As I only want to restrict this to push events, I'll add:

       - if: '$CI_PIPELINE_SOURCE != "push"'
         when: always

So all other events (merge request, pipeline triggered via WebUI)
the job is automatically executed.

> +    - when: manual
> +      allow_failure: true
>  
>  cross-armhf-system:
>    <<: *cross_system_build_job_definition
>    variables:
>      IMAGE: debian-armhf-cross
> +  rules:
> +    - when: manual
> +      allow_failure: true

Ditto (and in the MIPS job in next patch).

>  
>  cross-armhf-user:
>    <<: *cross_user_build_job_definition
> 

Re: [PATCH-for-5.2 v4 01/17] configure: Check vhost-user is available to build vhost-user-blk-server

[Thomas Huth]

On 08/11/2020 21.45, Philippe Mathieu-Daudé wrote:
> Check vhost-user is available to build vhost-user-blk-server to fix:
> 
>  $ ../configure \
>       --disable-vhost-user --enable-vhost-user-blk-server && \
>    make qemu-nbd
>  ...
>  [505/505] Linking target qemu-nbd
>  FAILED: qemu-nbd
>  cc  -o qemu-nbd qemu-nbd.p/qemu-nbd.c.o -Wl,--as-needed -Wl,--no-undefined -pie -Wl,--whole-archive libblockdev.fa libblock.fa libcrypto.fa libauthz.fa libqom.fa libio.fa -Wl,--no-whole-archive -Wl,--warn-common -Wl,-z,relro -Wl,-z,now -m64 -fstack-protector-strong -Wl,--start-group libqemuutil.a libblockdev.fa libblock.fa libcrypto.fa libauthz.fa libqom.fa libio.fa @block.syms -lgio-2.0 -lgobject-2.0 -lglib-2.0 -lgio-2.0 -lgobject-2.0 -lglib-2.0 -pthread -lgnutls -lutil -lm -lgthread-2.0 -lglib-2.0 -lbz2 -lgthread-2.0 -lglib-2.0 -lssh -lrbd -lrados -lcurl -lxml2 -lzstd -lacl -lgfapi -lglusterfs -lgfrpc -lgfxdr -luuid -laio /usr/lib64/libz.so -L/usr/lib64/iscsi -liscsi -lnettle -lgnutls -lpam -Wl,--end-group
>  /usr/bin/ld: libblockdev.fa(block_export_vhost-user-blk-server.c.o): in function `vu_blk_process_vq':
>  block/export/vhost-user-blk-server.c:203: undefined reference to `vu_get_queue'
>  /usr/bin/ld: block/export/vhost-user-blk-server.c:208: undefined reference to `vu_queue_pop'
>  /usr/bin/ld: libblockdev.fa(block_export_vhost-user-blk-server.c.o): in function `vu_blk_queue_set_started':
>  block/export/vhost-user-blk-server.c:228: undefined reference to `vu_get_queue'
>  /usr/bin/ld: libblockdev.fa(block_export_vhost-user-blk-server.c.o): in function `vu_blk_req_complete':
>  block/export/vhost-user-blk-server.c:55: undefined reference to `vu_queue_push'
>  /usr/bin/ld: block/export/vhost-user-blk-server.c:56: undefined reference to `vu_queue_notify'
>  /usr/bin/ld: libblockdev.fa(block_export_vhost-user-blk-server.c.o): in function `vu_blk_queue_set_started':
>  block/export/vhost-user-blk-server.c:229: undefined reference to `vu_set_queue_handler'
>  /usr/bin/ld: libqemuutil.a(util_vhost-user-server.c.o): in function `vu_client_trip':
>  util/vhost-user-server.c:176: undefined reference to `vu_dispatch'
>  /usr/bin/ld: util/vhost-user-server.c:180: undefined reference to `vu_deinit'
>  /usr/bin/ld: libqemuutil.a(util_vhost-user-server.c.o): in function `vu_accept':
>  util/vhost-user-server.c:291: undefined reference to `vu_init'
>  collect2: error: ld returned 1 exit status
>  ninja: build stopped: subcommand failed.
>  make: *** [Makefile:171: run-ninja] Error 1
> 
> Fixes: bc15e44cb21 ("configure: introduce --enable-vhost-user-blk-server")
> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
> ---
>  configure | 3 +++
>  1 file changed, 3 insertions(+)
> 
> diff --git a/configure b/configure
> index 805f7791503..5079ebb416a 100755
> --- a/configure
> +++ b/configure
> @@ -2390,6 +2390,9 @@ fi
>  
>  # libvhost-user is Linux-only
>  test "$vhost_user_blk_server" = "" && vhost_user_blk_server=$linux
> +if test "$vhost_user_blk_server" = "yes" && test "$vhost_user" = "no"; then
> +  error_exit "--enable-vhost-user-blk-server requires --enable-vhost-user"
> +fi

This patch looks incomplete to me. If I now do:

$ .../configure --target-list=x86_64-softmmu --disable-vhost-user

I get:

ERROR: --enable-vhost-user-blk-server requires --enable-vhost-user

... but I did not say --enable-vhost-user-blk-server ! So if
--disable-vhost-user has been given, it should also automatically disable
all the other vhost-user switches, I think.

 Thomas

Re: [PATCH-for-6.0 v4 12/17] gitlab-ci: Move coroutine tests across to gitlab

[Stefan Hajnoczi]

[-- Attachment #1: Type: text/plain, Size: 964 bytes --]

On Sun, Nov 08, 2020 at 09:45:30PM +0100, Philippe Mathieu-Daudé wrote:
> Similarly to commit 8cdb2cef3f1, move the coroutine tests to GitLab.
> 
> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
> ---
> Cc: Stefan Hajnoczi <stefanha@redhat.com>
> Cc: Kevin Wolf <kwolf@redhat.com>
> ---
>  .gitlab-ci.yml | 16 ++++++++++++++++
>  .travis.yml    | 14 --------------
>  2 files changed, 16 insertions(+), 14 deletions(-)
> 
> diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
> index a799803afb9..6552a832939 100644
> --- a/.gitlab-ci.yml
> +++ b/.gitlab-ci.yml
> @@ -475,6 +475,22 @@ build-tci:
>      - QTEST_QEMU_BINARY="./qemu-system-x86_64" ./tests/qtest/pxe-test
>      - QTEST_QEMU_BINARY="./qemu-system-s390x" ./tests/qtest/pxe-test -m slow
>  
> +# Alternate coroutines implementations are only really of interest to KVM users
> +# However we can't test against KVM on Travis so we can only run unit tests

s/Travis/GitLab CI/

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: [PATCH-for-6.0 v4 13/17] tests/docker: Install lttng-ust-devel package in Fedora image

[Stefan Hajnoczi]

[-- Attachment #1: Type: text/plain, Size: 537 bytes --]

On Sun, Nov 08, 2020 at 09:45:31PM +0100, Philippe Mathieu-Daudé wrote:
> Install the lttng-ust-devel package to be able to
> build QEMU using the User-Space Tracer trace backend
> (configure --enable-trace-backends=ust).
> 
> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
> ---
> Cc: Stefan Hajnoczi <stefanha@redhat.com>
> Cc: Daniel P . Berrange <berrange@redhat.com>
> ---
>  tests/docker/dockerfiles/fedora.docker | 1 +
>  1 file changed, 1 insertion(+)

Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: [PATCH-for-6.0 v4 14/17] gitlab-ci: Move trace backend tests across to gitlab

[Stefan Hajnoczi]

[-- Attachment #1: Type: text/plain, Size: 472 bytes --]

On Sun, Nov 08, 2020 at 09:45:32PM +0100, Philippe Mathieu-Daudé wrote:
> Similarly to commit 8cdb2cef3f1, move the trace backend
> tests to GitLab.
> 
> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
> ---
> Cc: Stefan Hajnoczi <stefanha@redhat.com>
> ---
>  .gitlab-ci.yml | 18 ++++++++++++++++++
>  .travis.yml    | 19 -------------------
>  2 files changed, 18 insertions(+), 19 deletions(-)

Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: [PATCH-for-6.0 v4 04/17] MAINTAINERS: Add gitlab-pipeline-status script to GitLab CI section

[Thomas Huth]

On 08/11/2020 21.45, Philippe Mathieu-Daudé wrote:
> Do not let the gitlab-pipeline-status script unmaintained,
> add it to the 'GitLab Continuous Integration' section.
> 
> Fixes: c02b2eac55e ("GitLab Gating CI: introduce pipeline-status contrib script")
> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
> ---
> Cc: Cleber Rosa <crosa@redhat.com>
> ---
>  MAINTAINERS | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/MAINTAINERS b/MAINTAINERS
> index 63223e1183d..817c7c93f91 100644
> --- a/MAINTAINERS
> +++ b/MAINTAINERS
> @@ -3146,6 +3146,7 @@ S: Maintained
>  F: .gitlab-ci.yml
>  F: .gitlab-ci.d/crossbuilds.yml
>  F: .gitlab-ci.d/*py
> +F: scripts/ci/gitlab-pipeline-status

Reviewed-by: Thomas Huth <thuth@redhat.com>

Re: [PATCH-for-6.0 v4 06/17] gitlab-ci: Update 'build-disabled' to cover all configurable options

[Thomas Huth]

On 08/11/2020 21.45, Philippe Mathieu-Daudé wrote:
> Add these missing options to the 'build-disabled' job:
>   --disable-auth-pam             (commit 8953caf3cd38)
>   --disable-gcrypt               (commit 91bfcdb01d48)
>   --disable-keyring              (commit 54e7aac05624)
>   --disable-libudev              (commit 5c53015a480b)
>   --disable-opengl               (commit da076ffed6b9)
>   --disable-sparse               (commit 03b4fe7de226)
> 
> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
> ---
>  .gitlab-ci.yml | 7 +++++++
>  1 file changed, 7 insertions(+)
[...]
> @@ -258,6 +264,7 @@ build-disabled:
>        --disable-vhost-net
>        --disable-vhost-scsi
>        --disable-vhost-user
> +      --disable-vhost-user-blk-server
>        --disable-vhost-vdpa
>        --disable-vhost-vsock
>        --disable-virglrenderer

vhost-user-blk-server should get automatically disabled by
--disable-vhost-user ... so maybe drop this hunk, so that we're able to
detect problems with this logic in the future, too? (as we've just seen with
the current build failures in the CI)
... if you insist on keeping this hunk, I think you should at least mention
it in the patch description (since you've mentioned all the other options there)

 Thomas

Re: [PATCH-for-6.0 v4 07/17] gitlab-ci: Move job testing --without-default-devices across to gitlab

[Thomas Huth]

On 08/11/2020 21.45, Philippe Mathieu-Daudé wrote:
> Similarly to commit 8cdb2cef3f1, move the job testing the
> '--without-default-devices' configure option to GitLab.
> 
> Since building all softmmu targets takes too long, split
> the job in 2.
> 
> As smoke test, run the qtests on the AVR and m68k targets.
> 
> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
> ---
> Cc: Paolo Bonzini <pbonzini@redhat.com>
> 
> v3 had:
> Reviewed-by: Thomas Huth <thuth@redhat.com>
> Reviewed-by: Wainer dos Santos Moschetta <wainersm@redhat.com>
> ---
>  .gitlab-ci.yml | 46 ++++++++++++++++++++++++++++++++++++++++++++++
>  .travis.yml    |  8 --------
>  2 files changed, 46 insertions(+), 8 deletions(-)
> 
> diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
> index b98800462ed..3fc3d0568c6 100644
> --- a/.gitlab-ci.yml
> +++ b/.gitlab-ci.yml
> @@ -315,6 +315,52 @@ build-user-plugins:
>      MAKE_CHECK_ARGS: check-tcg
>    timeout: 1h 30m
>  
> +build-system-ubuntu-without-default-devices 1/2:
> +  <<: *native_build_job_definition
> +  variables:
> +    IMAGE: ubuntu2004
> +    CONFIGURE_ARGS: --without-default-devices --disable-tools --disable-docs
> +    TARGETS:
> +      aarch64-softmmu
> +      alpha-softmmu
> +      arm-softmmu
> +      avr-softmmu
> +      cris-softmmu
> +      hppa-softmmu
> +      i386-softmmu
> +      m68k-softmmu
> +      microblazeel-softmmu
> +      microblaze-softmmu
> +      mips64el-softmmu
> +      mips64-softmmu
> +      mipsel-softmmu
> +      mips-softmmu
> +      moxie-softmmu
> +    MAKE_CHECK_ARGS: check-qtest-avr check-qtest-m68k
> +
> +build-system-ubuntu-without-default-devices 2/2:
> +  <<: *native_build_job_definition
> +  variables:
> +    IMAGE: ubuntu2004
> +    CONFIGURE_ARGS: --without-default-devices --disable-tools --disable-docs
> +    TARGETS:
> +      nios2-softmmu
> +      or1k-softmmu
> +      ppc64-softmmu
> +      ppc-softmmu
> +      riscv32-softmmu
> +      riscv64-softmmu
> +      rx-softmmu
> +      s390x-softmmu
> +      sh4eb-softmmu
> +      sh4-softmmu
> +      sparc64-softmmu
> +      sparc-softmmu
> +      tricore-softmmu
> +      x86_64-softmmu
> +      xtensaeb-softmmu
> +      xtensa-softmmu

Could you please add a MAKE_CHECK_ARGS here, too? check-qtest-rx or
check-qtest-sh4 sound like good candidates...

And could you please also use Fedora and/or CentOS for these tests now?
We're using --without-default-devices in downstream RHEL, so testing with
that option in Fedora and CentOS would be helpful for us.

 Thanks,
  Thomas

Re: [PATCH-for-6.0 v4 07/17] gitlab-ci: Move job testing --without-default-devices across to gitlab

[Philippe Mathieu-Daudé]

On 11/9/20 11:20 AM, Thomas Huth wrote:
> On 08/11/2020 21.45, Philippe Mathieu-Daudé wrote:
>> Similarly to commit 8cdb2cef3f1, move the job testing the
>> '--without-default-devices' configure option to GitLab.
>>
>> Since building all softmmu targets takes too long, split
>> the job in 2.
>>
>> As smoke test, run the qtests on the AVR and m68k targets.
>>
>> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
>> ---
>> Cc: Paolo Bonzini <pbonzini@redhat.com>
>>
>> v3 had:
>> Reviewed-by: Thomas Huth <thuth@redhat.com>
>> Reviewed-by: Wainer dos Santos Moschetta <wainersm@redhat.com>
>> ---
>>  .gitlab-ci.yml | 46 ++++++++++++++++++++++++++++++++++++++++++++++
>>  .travis.yml    |  8 --------
>>  2 files changed, 46 insertions(+), 8 deletions(-)
>>
>> diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
>> index b98800462ed..3fc3d0568c6 100644
>> --- a/.gitlab-ci.yml
>> +++ b/.gitlab-ci.yml
>> @@ -315,6 +315,52 @@ build-user-plugins:
>>      MAKE_CHECK_ARGS: check-tcg
>>    timeout: 1h 30m
>>  
>> +build-system-ubuntu-without-default-devices 1/2:
>> +  <<: *native_build_job_definition
>> +  variables:
>> +    IMAGE: ubuntu2004
>> +    CONFIGURE_ARGS: --without-default-devices --disable-tools --disable-docs
>> +    TARGETS:
>> +      aarch64-softmmu
>> +      alpha-softmmu
>> +      arm-softmmu
>> +      avr-softmmu
>> +      cris-softmmu
>> +      hppa-softmmu
>> +      i386-softmmu
>> +      m68k-softmmu
>> +      microblazeel-softmmu
>> +      microblaze-softmmu
>> +      mips64el-softmmu
>> +      mips64-softmmu
>> +      mipsel-softmmu
>> +      mips-softmmu
>> +      moxie-softmmu
>> +    MAKE_CHECK_ARGS: check-qtest-avr check-qtest-m68k
>> +
>> +build-system-ubuntu-without-default-devices 2/2:
>> +  <<: *native_build_job_definition
>> +  variables:
>> +    IMAGE: ubuntu2004
>> +    CONFIGURE_ARGS: --without-default-devices --disable-tools --disable-docs
>> +    TARGETS:
>> +      nios2-softmmu
>> +      or1k-softmmu
>> +      ppc64-softmmu
>> +      ppc-softmmu
>> +      riscv32-softmmu
>> +      riscv64-softmmu
>> +      rx-softmmu
>> +      s390x-softmmu
>> +      sh4eb-softmmu
>> +      sh4-softmmu
>> +      sparc64-softmmu
>> +      sparc-softmmu
>> +      tricore-softmmu
>> +      x86_64-softmmu
>> +      xtensaeb-softmmu
>> +      xtensa-softmmu
> 
> Could you please add a MAKE_CHECK_ARGS here, too? check-qtest-rx or
> check-qtest-sh4 sound like good candidates...

OK.

> And could you please also use Fedora and/or CentOS for these tests now?
> We're using --without-default-devices in downstream RHEL, so testing with
> that option in Fedora and CentOS would be helpful for us.

I wanted to do that, but expected someone to object as "we are not
testing the same than before". I'll change one (maybe CentOS 8, which
is less tested than Fedora).

> 
>  Thanks,
>   Thomas
> 

Re: [PATCH-for-6.0 v4 07/17] gitlab-ci: Move job testing --without-default-devices across to gitlab

[Thomas Huth]

On 09/11/2020 12.04, Philippe Mathieu-Daudé wrote:
> On 11/9/20 11:20 AM, Thomas Huth wrote:
>> On 08/11/2020 21.45, Philippe Mathieu-Daudé wrote:
>>> Similarly to commit 8cdb2cef3f1, move the job testing the
>>> '--without-default-devices' configure option to GitLab.
>>>
>>> Since building all softmmu targets takes too long, split
>>> the job in 2.
>>>
>>> As smoke test, run the qtests on the AVR and m68k targets.
>>>
>>> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
>>> ---
>>> Cc: Paolo Bonzini <pbonzini@redhat.com>
>>>
>>> v3 had:
>>> Reviewed-by: Thomas Huth <thuth@redhat.com>
>>> Reviewed-by: Wainer dos Santos Moschetta <wainersm@redhat.com>
>>> ---
>>>  .gitlab-ci.yml | 46 ++++++++++++++++++++++++++++++++++++++++++++++
>>>  .travis.yml    |  8 --------
>>>  2 files changed, 46 insertions(+), 8 deletions(-)
>>>
>>> diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
>>> index b98800462ed..3fc3d0568c6 100644
>>> --- a/.gitlab-ci.yml
>>> +++ b/.gitlab-ci.yml
>>> @@ -315,6 +315,52 @@ build-user-plugins:
>>>      MAKE_CHECK_ARGS: check-tcg
>>>    timeout: 1h 30m
>>>  
>>> +build-system-ubuntu-without-default-devices 1/2:
>>> +  <<: *native_build_job_definition
>>> +  variables:
>>> +    IMAGE: ubuntu2004
>>> +    CONFIGURE_ARGS: --without-default-devices --disable-tools --disable-docs
>>> +    TARGETS:
>>> +      aarch64-softmmu
>>> +      alpha-softmmu
>>> +      arm-softmmu
>>> +      avr-softmmu
>>> +      cris-softmmu
>>> +      hppa-softmmu
>>> +      i386-softmmu
>>> +      m68k-softmmu
>>> +      microblazeel-softmmu
>>> +      microblaze-softmmu
>>> +      mips64el-softmmu
>>> +      mips64-softmmu
>>> +      mipsel-softmmu
>>> +      mips-softmmu
>>> +      moxie-softmmu
>>> +    MAKE_CHECK_ARGS: check-qtest-avr check-qtest-m68k
>>> +
>>> +build-system-ubuntu-without-default-devices 2/2:
>>> +  <<: *native_build_job_definition
>>> +  variables:
>>> +    IMAGE: ubuntu2004
>>> +    CONFIGURE_ARGS: --without-default-devices --disable-tools --disable-docs
>>> +    TARGETS:
>>> +      nios2-softmmu
>>> +      or1k-softmmu
>>> +      ppc64-softmmu
>>> +      ppc-softmmu
>>> +      riscv32-softmmu
>>> +      riscv64-softmmu
>>> +      rx-softmmu
>>> +      s390x-softmmu
>>> +      sh4eb-softmmu
>>> +      sh4-softmmu
>>> +      sparc64-softmmu
>>> +      sparc-softmmu
>>> +      tricore-softmmu
>>> +      x86_64-softmmu
>>> +      xtensaeb-softmmu
>>> +      xtensa-softmmu
>>
>> Could you please add a MAKE_CHECK_ARGS here, too? check-qtest-rx or
>> check-qtest-sh4 sound like good candidates...
> 
> OK.
> 
>> And could you please also use Fedora and/or CentOS for these tests now?
>> We're using --without-default-devices in downstream RHEL, so testing with
>> that option in Fedora and CentOS would be helpful for us.
> 
> I wanted to do that, but expected someone to object as "we are not
> testing the same than before". I'll change one (maybe CentOS 8, which
> is less tested than Fedora).

Well, we simply used Ubuntu on Travis since there was no other option. Now
we are free to choose. Thus I'd prefer if you could use one Fedora and one
CentOS job here now if possible. We've already got some other Ubuntu jobs in
the gitlab-CI, so I doubt that anybody will complain.

 Thomas

Re: [PATCH-for-6.0 v4 07/17] gitlab-ci: Move job testing --without-default-devices across to gitlab

[Philippe Mathieu-Daudé]

On 11/9/20 12:18 PM, Thomas Huth wrote:
> On 09/11/2020 12.04, Philippe Mathieu-Daudé wrote:
>> On 11/9/20 11:20 AM, Thomas Huth wrote:
>>> On 08/11/2020 21.45, Philippe Mathieu-Daudé wrote:
>>>> Similarly to commit 8cdb2cef3f1, move the job testing the
>>>> '--without-default-devices' configure option to GitLab.
>>>>
>>>> Since building all softmmu targets takes too long, split
>>>> the job in 2.
>>>>
>>>> As smoke test, run the qtests on the AVR and m68k targets.
>>>>
>>>> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
>>>> ---
>>>> Cc: Paolo Bonzini <pbonzini@redhat.com>
>>>>
>>>> v3 had:
>>>> Reviewed-by: Thomas Huth <thuth@redhat.com>
>>>> Reviewed-by: Wainer dos Santos Moschetta <wainersm@redhat.com>
>>>> ---
>>>>  .gitlab-ci.yml | 46 ++++++++++++++++++++++++++++++++++++++++++++++
>>>>  .travis.yml    |  8 --------
>>>>  2 files changed, 46 insertions(+), 8 deletions(-)
>>>>
>>>> diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
>>>> index b98800462ed..3fc3d0568c6 100644
>>>> --- a/.gitlab-ci.yml
>>>> +++ b/.gitlab-ci.yml
>>>> @@ -315,6 +315,52 @@ build-user-plugins:
>>>>      MAKE_CHECK_ARGS: check-tcg
>>>>    timeout: 1h 30m
>>>>  
>>>> +build-system-ubuntu-without-default-devices 1/2:
>>>> +  <<: *native_build_job_definition
>>>> +  variables:
>>>> +    IMAGE: ubuntu2004
>>>> +    CONFIGURE_ARGS: --without-default-devices --disable-tools --disable-docs
>>>> +    TARGETS:
>>>> +      aarch64-softmmu
>>>> +      alpha-softmmu
>>>> +      arm-softmmu
>>>> +      avr-softmmu
>>>> +      cris-softmmu
>>>> +      hppa-softmmu
>>>> +      i386-softmmu
>>>> +      m68k-softmmu
>>>> +      microblazeel-softmmu
>>>> +      microblaze-softmmu
>>>> +      mips64el-softmmu
>>>> +      mips64-softmmu
>>>> +      mipsel-softmmu
>>>> +      mips-softmmu
>>>> +      moxie-softmmu
>>>> +    MAKE_CHECK_ARGS: check-qtest-avr check-qtest-m68k
>>>> +
>>>> +build-system-ubuntu-without-default-devices 2/2:
>>>> +  <<: *native_build_job_definition
>>>> +  variables:
>>>> +    IMAGE: ubuntu2004
>>>> +    CONFIGURE_ARGS: --without-default-devices --disable-tools --disable-docs
>>>> +    TARGETS:
>>>> +      nios2-softmmu
>>>> +      or1k-softmmu
>>>> +      ppc64-softmmu
>>>> +      ppc-softmmu
>>>> +      riscv32-softmmu
>>>> +      riscv64-softmmu
>>>> +      rx-softmmu
>>>> +      s390x-softmmu
>>>> +      sh4eb-softmmu
>>>> +      sh4-softmmu
>>>> +      sparc64-softmmu
>>>> +      sparc-softmmu
>>>> +      tricore-softmmu
>>>> +      x86_64-softmmu
>>>> +      xtensaeb-softmmu
>>>> +      xtensa-softmmu
>>>
>>> Could you please add a MAKE_CHECK_ARGS here, too? check-qtest-rx or
>>> check-qtest-sh4 sound like good candidates...
>>
>> OK.
>>
>>> And could you please also use Fedora and/or CentOS for these tests now?
>>> We're using --without-default-devices in downstream RHEL, so testing with
>>> that option in Fedora and CentOS would be helpful for us.
>>
>> I wanted to do that, but expected someone to object as "we are not
>> testing the same than before". I'll change one (maybe CentOS 8, which
>> is less tested than Fedora).
> 
> Well, we simply used Ubuntu on Travis since there was no other option. Now
> we are free to choose. Thus I'd prefer if you could use one Fedora and one
> CentOS job here now if possible. We've already got some other Ubuntu jobs in
> the gitlab-CI, so I doubt that anybody will complain.

Also we still use Ubuntu on Travis-CI.

> 
>  Thomas
> 

Re: [PATCH-for-6.0 v4 04/17] MAINTAINERS: Add gitlab-pipeline-status script to GitLab CI section

[Wainer dos Santos Moschetta]

On 11/8/20 6:45 PM, Philippe Mathieu-Daudé wrote:
> Do not let the gitlab-pipeline-status script unmaintained,
> add it to the 'GitLab Continuous Integration' section.
>
> Fixes: c02b2eac55e ("GitLab Gating CI: introduce pipeline-status contrib script")
> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
> ---
> Cc: Cleber Rosa <crosa@redhat.com>
> ---
>   MAINTAINERS | 1 +
>   1 file changed, 1 insertion(+)


Reviewed-by: Wainer dos Santos Moschetta <wainersm@redhat.com>


>
> diff --git a/MAINTAINERS b/MAINTAINERS
> index 63223e1183d..817c7c93f91 100644
> --- a/MAINTAINERS
> +++ b/MAINTAINERS
> @@ -3146,6 +3146,7 @@ S: Maintained
>   F: .gitlab-ci.yml
>   F: .gitlab-ci.d/crossbuilds.yml
>   F: .gitlab-ci.d/*py
> +F: scripts/ci/gitlab-pipeline-status
>   
>   Guest Test Compilation Support
>   M: Alex Bennée <alex.bennee@linaro.org>

Re: [PATCH-for-6.0 v4 05/17] gitlab-ci: Split CONFIGURE_ARGS one argument per line for build-disabled

[Wainer dos Santos Moschetta]

On 11/8/20 6:45 PM, Philippe Mathieu-Daudé wrote:
> We will keep adding/removing options to our 'configure' script,
> so for easier maintainability it makes sense to have CONFIGURE_ARGS
> declared as one option per line. This way we can review diff easily
> (or rebase/cherry-pick).
>
> No logical change.
>
> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
> ---
> Cc: Daniel P . Berrange <berrange@redhat.com>
> ---
>   .gitlab-ci.yml | 80 +++++++++++++++++++++++++++++++++++++++-----------
>   1 file changed, 63 insertions(+), 17 deletions(-)
>
> diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
> index 3b15ae5c302..9f090cca5e3 100644
> --- a/.gitlab-ci.yml
> +++ b/.gitlab-ci.yml
> @@ -203,23 +203,69 @@ build-disabled:
>     <<: *native_build_job_definition
>     variables:
>       IMAGE: fedora
> -    CONFIGURE_ARGS: --disable-attr --disable-avx2 --disable-bochs
> -      --disable-brlapi --disable-bzip2 --disable-cap-ng --disable-capstone
> -      --disable-cloop --disable-coroutine-pool --disable-curl --disable-curses
> -      --disable-dmg --disable-docs --disable-glusterfs --disable-gnutls
> -      --disable-gtk --disable-guest-agent --disable-iconv --disable-kvm
> -      --disable-libiscsi --disable-libpmem --disable-libssh --disable-libusb
> -      --disable-libxml2 --disable-linux-aio --disable-live-block-migration
> -      --disable-lzo --disable-malloc-trim --disable-mpath --disable-nettle
> -      --disable-numa --disable-parallels --disable-pie --disable-qcow1
> -      --disable-qed --disable-qom-cast-debug --disable-rbd --disable-rdma
> -      --disable-replication --disable-sdl --disable-seccomp --disable-sheepdog
> -      --disable-slirp --disable-smartcard --disable-snappy --disable-spice
> -      --disable-strip --disable-tpm --disable-usb-redir --disable-vdi
> -      --disable-vhost-crypto --disable-vhost-net --disable-vhost-scsi
> -      --disable-vhost-user --disable-vhost-vdpa --disable-vhost-vsock
> -      --disable-virglrenderer --disable-vnc --disable-vte --disable-vvfat
> -      --disable-xen --disable-zstd
> +    CONFIGURE_ARGS:

I suggest to use '>-' like 'CONFIGURE_ARGS: >-'. It's up to you.

Reviewed-by: Wainer dos Santos Moschetta <wainersm@redhat.com>

> +      --disable-attr
> +      --disable-avx2
> +      --disable-bochs
> +      --disable-brlapi
> +      --disable-bzip2
> +      --disable-cap-ng
> +      --disable-capstone
> +      --disable-cloop
> +      --disable-coroutine-pool
> +      --disable-curl
> +      --disable-curses
> +      --disable-dmg
> +      --disable-docs
> +      --disable-glusterfs
> +      --disable-gnutls
> +      --disable-gtk
> +      --disable-guest-agent
> +      --disable-iconv
> +      --disable-kvm
> +      --disable-libiscsi
> +      --disable-libpmem
> +      --disable-libssh
> +      --disable-libusb
> +      --disable-libxml2
> +      --disable-linux-aio
> +      --disable-live-block-migration
> +      --disable-lzo
> +      --disable-malloc-trim
> +      --disable-mpath
> +      --disable-nettle
> +      --disable-numa
> +      --disable-parallels
> +      --disable-pie
> +      --disable-qcow1
> +      --disable-qed
> +      --disable-qom-cast-debug
> +      --disable-rbd
> +      --disable-rdma
> +      --disable-replication
> +      --disable-sdl
> +      --disable-seccomp
> +      --disable-sheepdog
> +      --disable-slirp
> +      --disable-smartcard
> +      --disable-snappy
> +      --disable-spice
> +      --disable-strip
> +      --disable-tpm
> +      --disable-usb-redir
> +      --disable-vdi
> +      --disable-vhost-crypto
> +      --disable-vhost-net
> +      --disable-vhost-scsi
> +      --disable-vhost-user
> +      --disable-vhost-vdpa
> +      --disable-vhost-vsock
> +      --disable-virglrenderer
> +      --disable-vnc
> +      --disable-vte
> +      --disable-vvfat
> +      --disable-xen
> +      --disable-zstd
>       TARGETS: arm-softmmu i386-softmmu ppc64-softmmu mips64-softmmu
>         s390x-softmmu i386-linux-user
>       MAKE_CHECK_ARGS: check-qtest SPEED=slow

Re: [PATCH-for-6.0 v4 07/17] gitlab-ci: Move job testing --without-default-devices across to gitlab

[Wainer dos Santos Moschetta]

On 11/8/20 6:45 PM, Philippe Mathieu-Daudé wrote:
> Similarly to commit 8cdb2cef3f1, move the job testing the
> '--without-default-devices' configure option to GitLab.
>
> Since building all softmmu targets takes too long, split
> the job in 2.
>
> As smoke test, run the qtests on the AVR and m68k targets.
>
> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
> ---
> Cc: Paolo Bonzini <pbonzini@redhat.com>
>
> v3 had:
> Reviewed-by: Thomas Huth <thuth@redhat.com>
> Reviewed-by: Wainer dos Santos Moschetta <wainersm@redhat.com>
> ---
>   .gitlab-ci.yml | 46 ++++++++++++++++++++++++++++++++++++++++++++++
>   .travis.yml    |  8 --------
>   2 files changed, 46 insertions(+), 8 deletions(-)
>
> diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
> index b98800462ed..3fc3d0568c6 100644
> --- a/.gitlab-ci.yml
> +++ b/.gitlab-ci.yml
> @@ -315,6 +315,52 @@ build-user-plugins:
>       MAKE_CHECK_ARGS: check-tcg
>     timeout: 1h 30m
>   
> +build-system-ubuntu-without-default-devices 1/2:

Nit: It seems a non-usual key naming. Maybe 
"build-system-ubuntu-without-default-devices_1_2"  instead.

- Wainer

> +  <<: *native_build_job_definition
> +  variables:
> +    IMAGE: ubuntu2004
> +    CONFIGURE_ARGS: --without-default-devices --disable-tools --disable-docs
> +    TARGETS:
> +      aarch64-softmmu
> +      alpha-softmmu
> +      arm-softmmu
> +      avr-softmmu
> +      cris-softmmu
> +      hppa-softmmu
> +      i386-softmmu
> +      m68k-softmmu
> +      microblazeel-softmmu
> +      microblaze-softmmu
> +      mips64el-softmmu
> +      mips64-softmmu
> +      mipsel-softmmu
> +      mips-softmmu
> +      moxie-softmmu
> +    MAKE_CHECK_ARGS: check-qtest-avr check-qtest-m68k
> +
> +build-system-ubuntu-without-default-devices 2/2:
> +  <<: *native_build_job_definition
> +  variables:
> +    IMAGE: ubuntu2004
> +    CONFIGURE_ARGS: --without-default-devices --disable-tools --disable-docs
> +    TARGETS:
> +      nios2-softmmu
> +      or1k-softmmu
> +      ppc64-softmmu
> +      ppc-softmmu
> +      riscv32-softmmu
> +      riscv64-softmmu
> +      rx-softmmu
> +      s390x-softmmu
> +      sh4eb-softmmu
> +      sh4-softmmu
> +      sparc64-softmmu
> +      sparc-softmmu
> +      tricore-softmmu
> +      x86_64-softmmu
> +      xtensaeb-softmmu
> +      xtensa-softmmu
> +
>   build-clang:
>     <<: *native_build_job_definition
>     variables:
> diff --git a/.travis.yml b/.travis.yml
> index a3d78171cab..15d92291358 100644
> --- a/.travis.yml
> +++ b/.travis.yml
> @@ -224,14 +224,6 @@ jobs:
>           - ${SRC_DIR}/scripts/travis/coverage-summary.sh
>   
>   
> -    # We manually include builds which we disable "make check" for
> -    - name: "GCC without-default-devices (softmmu)"
> -      env:
> -        - CONFIG="--without-default-devices --disable-user"
> -        - CACHE_NAME="${TRAVIS_BRANCH}-linux-gcc-default"
> -        - TEST_CMD=""
> -
> -
>       # We don't need to exercise every backend with every front-end
>       - name: "GCC trace log,simple,syslog (user)"
>         env:

Re: [PATCH-for-6.0 v4 07/17] gitlab-ci: Move job testing --without-default-devices across to gitlab

[Philippe Mathieu-Daudé]

On 11/10/20 2:44 PM, Wainer dos Santos Moschetta wrote:
> 
> On 11/8/20 6:45 PM, Philippe Mathieu-Daudé wrote:
>> Similarly to commit 8cdb2cef3f1, move the job testing the
>> '--without-default-devices' configure option to GitLab.
>>
>> Since building all softmmu targets takes too long, split
>> the job in 2.
>>
>> As smoke test, run the qtests on the AVR and m68k targets.
>>
>> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
>> ---
>> Cc: Paolo Bonzini <pbonzini@redhat.com>
>>
>> v3 had:
>> Reviewed-by: Thomas Huth <thuth@redhat.com>
>> Reviewed-by: Wainer dos Santos Moschetta <wainersm@redhat.com>
>> ---
>>   .gitlab-ci.yml | 46 ++++++++++++++++++++++++++++++++++++++++++++++
>>   .travis.yml    |  8 --------
>>   2 files changed, 46 insertions(+), 8 deletions(-)
>>
>> diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
>> index b98800462ed..3fc3d0568c6 100644
>> --- a/.gitlab-ci.yml
>> +++ b/.gitlab-ci.yml
>> @@ -315,6 +315,52 @@ build-user-plugins:
>>       MAKE_CHECK_ARGS: check-tcg
>>     timeout: 1h 30m
>>   +build-system-ubuntu-without-default-devices 1/2:
> 
> Nit: It seems a non-usual key naming. Maybe
> "build-system-ubuntu-without-default-devices_1_2"  instead.

This is the GitLab convention:
https://docs.gitlab.com/ee/ci/pipelines/#group-jobs-in-a-pipeline

 This regular expression evaluates the job
 names: \d+[\s:\/\\]+\d+\s*.

I'll add a comment about it, but I don't think we should each
time we use the convention.

Thanks,

Phil.

Re: [PATCH-for-6.0 v4 08/17] gitlab-ci: Move linux-user debug-tcg test across to gitlab

[Wainer dos Santos Moschetta]

On 11/8/20 6:45 PM, Philippe Mathieu-Daudé wrote:
> Similarly to commit 8cdb2cef3f1, move the linux-user (debug-tcg)
> test to GitLab.
>
> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
> ---
> Cc: Laurent Vivier <laurent@vivier.eu>
> ---
>   .gitlab-ci.yml | 7 +++++++
>   .travis.yml    | 9 ---------
>   2 files changed, 7 insertions(+), 9 deletions(-)
>
> diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
> index 3fc3d0568c6..80082a602b8 100644
> --- a/.gitlab-ci.yml
> +++ b/.gitlab-ci.yml
> @@ -304,6 +304,13 @@ build-user:
>       CONFIGURE_ARGS: --disable-tools --disable-system
>       MAKE_CHECK_ARGS: check-tcg
>   
> +build-user-debug:
> +  <<: *native_build_job_definition
> +  variables:
> +    IMAGE: debian-all-test-cross
> +    CONFIGURE_ARGS: --disable-tools --disable-system --enable-debug-tcg
> +    MAKE_CHECK_ARGS: check-tcg
> +

It would be useful a variable like 'MAKE_BUILD_ARGS' so that we could 
overwrite the default make target (as in the diff below). So that here 
it would eval to  `make -j"$JOBS" build-tcg` as is done in Travis CI.

On the absence of such as variable,

Reviewed-by: Wainer dos Santos Moschetta <wainersm@redhat.com>

---

diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
index 3b15ae5c30..27517afd22 100644
--- a/.gitlab-ci.yml
+++ b/.gitlab-ci.yml
@@ -34,7 +34,7 @@ include:
        else
          ../configure --enable-werror $CONFIGURE_ARGS ;
        fi || { cat config.log meson-logs/meson-log.txt && exit 1; }
-    - make -j"$JOBS"
+    - make -j"$JOBS" "$MAKE_BUILD_ARGS"
      - if test -n "$MAKE_CHECK_ARGS";
        then
          make -j"$JOBS" $MAKE_CHECK_ARGS ;

>   # Run check-tcg against linux-user (with plugins)
>   # we skip sparc64-linux-user until it has been fixed somewhat
>   # we skip cris-linux-user as it doesn't use the common run loop
> diff --git a/.travis.yml b/.travis.yml
> index 15d92291358..bee6197290d 100644
> --- a/.travis.yml
> +++ b/.travis.yml
> @@ -293,15 +293,6 @@ jobs:
>           - ${SRC_DIR}/configure ${CONFIG} --extra-cflags="-g3 -O0 -fsanitize=thread" || { cat config.log meson-logs/meson-log.txt && exit 1; }
>   
>   
> -    # Run check-tcg against linux-user
> -    - name: "GCC check-tcg (user)"
> -      env:
> -        - CONFIG="--disable-system --enable-debug-tcg"
> -        - TEST_BUILD_CMD="make build-tcg"
> -        - TEST_CMD="make check-tcg"
> -        - CACHE_NAME="${TRAVIS_BRANCH}-linux-gcc-debug-tcg"
> -
> -
>       # Run check-tcg against softmmu targets
>       - name: "GCC check-tcg (some-softmmu)"
>         env:

Re: [PATCH-for-6.0 v4 09/17] gitlab-ci: Move gprof/gcov test across to gitlab

[Wainer dos Santos Moschetta]

On 11/8/20 6:45 PM, Philippe Mathieu-Daudé wrote:
> Similarly to commit 8cdb2cef3f1, move the gprof/gcov test to GitLab.
>
> The coverage-summary.sh script is not Travis-CI specific, make it
> generic.
>
> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
> ---
>   .gitlab-ci.yml                             | 12 ++++++++++++
>   .travis.yml                                | 14 --------------
>   MAINTAINERS                                |  2 +-
>   scripts/{travis => ci}/coverage-summary.sh |  2 +-
>   4 files changed, 14 insertions(+), 16 deletions(-)
>   rename scripts/{travis => ci}/coverage-summary.sh (92%)
>
> diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
> index 80082a602b8..d1583cee5db 100644
> --- a/.gitlab-ci.yml
> +++ b/.gitlab-ci.yml
> @@ -403,6 +403,18 @@ check-deprecated:
>       MAKE_CHECK_ARGS: check-tcg
>     allow_failure: true
>   
> +# gprof/gcov are GCC features
> +build-gprof-gcov:
> +  <<: *native_build_job_definition
> +  variables:
> +    IMAGE: ubuntu2004
> +    CONFIGURE_ARGS: --enable-gprof --enable-gcov
> +    MAKE_CHECK_ARGS: build-tcg
> +    TARGETS: aarch64-softmmu mips64-softmmu ppc64-softmmu
> +             riscv64-softmmu s390x-softmmu x86_64-softmmu
> +  after_script:
> +    - ${CI_PROJECT_DIR}/scripts/ci/coverage-summary.sh
> +

The script uses gcovr and it is not installed in the Ubuntu 20.04 
container image:

docker run registry.gitlab.com/qemu-project/qemu/qemu/ubuntu2004 gcovr
Emulate Docker CLI using podman. Create /etc/containers/nodocker to 
quiet msg.
Error: executable file not found in $PATH: No such file or directory: 
OCI runtime command not found error

- Wainer

>   build-oss-fuzz:
>     <<: *native_build_job_definition
>     variables:
> diff --git a/.travis.yml b/.travis.yml
> index bee6197290d..6087e2909cd 100644
> --- a/.travis.yml
> +++ b/.travis.yml
> @@ -210,20 +210,6 @@ jobs:
>         compiler: clang
>   
>   
> -    # gprof/gcov are GCC features
> -    - name: "GCC gprof/gcov"
> -      dist: bionic
> -      addons:
> -        apt:
> -          packages:
> -            - ninja-build
> -      env:
> -        - CONFIG="--enable-gprof --enable-gcov --disable-libssh
> -                  --target-list=${MAIN_SOFTMMU_TARGETS}"
> -      after_success:
> -        - ${SRC_DIR}/scripts/travis/coverage-summary.sh
> -
> -
>       # We don't need to exercise every backend with every front-end
>       - name: "GCC trace log,simple,syslog (user)"
>         env:
> diff --git a/MAINTAINERS b/MAINTAINERS
> index 817c7c93f91..3d8c586d444 100644
> --- a/MAINTAINERS
> +++ b/MAINTAINERS
> @@ -3121,7 +3121,7 @@ R: Philippe Mathieu-Daudé <philmd@redhat.com>
>   S: Maintained
>   F: .github/lockdown.yml
>   F: .travis.yml
> -F: scripts/travis/
> +F: scripts/ci/
>   F: .shippable.yml
>   F: tests/docker/
>   F: tests/vm/
> diff --git a/scripts/travis/coverage-summary.sh b/scripts/ci/coverage-summary.sh
> similarity index 92%
> rename from scripts/travis/coverage-summary.sh
> rename to scripts/ci/coverage-summary.sh
> index d7086cf9ca6..8d9fb4de401 100755
> --- a/scripts/travis/coverage-summary.sh
> +++ b/scripts/ci/coverage-summary.sh
> @@ -3,7 +3,7 @@
>   # Author: Alex Bennée <alex.bennee@linaro.org>
>   #
>   # Summerise the state of code coverage with gcovr and tweak the output
> -# to be more sane on Travis hosts. As we expect to be executed on a
> +# to be more sane on CI runner. As we expect to be executed on a
>   # throw away CI instance we do spam temp files all over the shop. You
>   # most likely don't want to execute this script but just call gcovr
>   # directly. See also "make coverage-report"

Re: [PATCH-for-6.0 v4 10/17] gitlab-ci: Move user-static test across to gitlab

[Wainer dos Santos Moschetta]

On 11/8/20 6:45 PM, Philippe Mathieu-Daudé wrote:
> Similarly to commit 8cdb2cef3f1, move the user-static test to GitLab.
>
> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
> ---
> Cc: Laurent Vivier <laurent@vivier.eu>
> ---
>   .gitlab-ci.yml | 7 +++++++
>   .travis.yml    | 5 -----
>   2 files changed, 7 insertions(+), 5 deletions(-)

LGTM.

Reviewed-by: Wainer dos Santos Moschetta <wainersm@redhat.com>

>
> diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
> index d1583cee5db..a91870cf85e 100644
> --- a/.gitlab-ci.yml
> +++ b/.gitlab-ci.yml
> @@ -311,6 +311,13 @@ build-user-debug:
>       CONFIGURE_ARGS: --disable-tools --disable-system --enable-debug-tcg
>       MAKE_CHECK_ARGS: check-tcg
>   
> +build-user-static:
> +  <<: *native_build_job_definition
> +  variables:
> +    IMAGE: debian-all-test-cross
> +    CONFIGURE_ARGS: --disable-tools --disable-system --static
> +    MAKE_CHECK_ARGS: check-tcg
> +
>   # Run check-tcg against linux-user (with plugins)
>   # we skip sparc64-linux-user until it has been fixed somewhat
>   # we skip cris-linux-user as it doesn't use the common run loop
> diff --git a/.travis.yml b/.travis.yml
> index 6087e2909cd..5fb69c5fb02 100644
> --- a/.travis.yml
> +++ b/.travis.yml
> @@ -119,11 +119,6 @@ after_script:
>   
>   jobs:
>     include:
> -    - name: "GCC static (user)"
> -      env:
> -        - CONFIG="--disable-system --static"
> -        - CACHE_NAME="${TRAVIS_BRANCH}-linux-gcc-default"
> -
>       # Just build tools and run minimal unit and softfloat checks
>       - name: "GCC check-unit and check-softfloat"
>         env:

Re: [PATCH-for-5.2 v4 00/17] ci: Move various jobs from Travis to GitLab CI

[Alex Bennée]

Philippe Mathieu-Daudé <philmd@redhat.com> writes:

> We have various job covered on Travis-CI, but recommend the
> community to use GitLab, so build failures are missed.
>
> We need help to move the jobs to GitLab (we will keep the s390x
> and ppc64 containerized jobs on Travis as there is no similar
> offer on GitLab). Start moving jobs.
>
> Resulting pipelines (96 jobs in 71 minutes and 4 seconds)
>   https://gitlab.com/philmd/qemu/-/pipelines/213351318
>   https://gitlab.com/philmd/qemu/-/pipelines/213378500
>
> Patches 1-3 are included because without them some jobs are
> failing; but they will be merged via different trees.

Which trees? I'm currently holding 2/3 in my tree as I kick the tyres
although I dropped 1 following breakage and discussion on IRC.

-- 
Alex Bennée

Re: [PATCH-for-5.2 v4 00/17] ci: Move various jobs from Travis to GitLab CI

[Philippe Mathieu-Daudé]

On 11/10/20 4:55 PM, Alex Bennée wrote:
> 
> Philippe Mathieu-Daudé <philmd@redhat.com> writes:
> 
>> We have various job covered on Travis-CI, but recommend the
>> community to use GitLab, so build failures are missed.
>>
>> We need help to move the jobs to GitLab (we will keep the s390x
>> and ppc64 containerized jobs on Travis as there is no similar
>> offer on GitLab). Start moving jobs.
>>
>> Resulting pipelines (96 jobs in 71 minutes and 4 seconds)
>>   https://gitlab.com/philmd/qemu/-/pipelines/213351318
>>   https://gitlab.com/philmd/qemu/-/pipelines/213378500
>>
>> Patches 1-3 are included because without them some jobs are
>> failing; but they will be merged via different trees.
> 
> Which trees? I'm currently holding 2/3 in my tree as I kick the tyres
> although I dropped 1 following breakage and discussion on IRC.

Well I'm not sure... The point of this series is not to fix GitLab
(as it will keep failing) but to move jobs from Travis to it. I
simply wanted to show a green GitLab.

I suppose they belong to Michael Tsirkin (vhost-user, ACPI).

Regards,

Phil.

Re: [PATCH-for-6.0 v4 12/17] gitlab-ci: Move coroutine tests across to gitlab

[Wainer dos Santos Moschetta]

On 11/8/20 6:45 PM, Philippe Mathieu-Daudé wrote:
> Similarly to commit 8cdb2cef3f1, move the coroutine tests to GitLab.
>
> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
> ---
> Cc: Stefan Hajnoczi <stefanha@redhat.com>
> Cc: Kevin Wolf <kwolf@redhat.com>
> ---
>   .gitlab-ci.yml | 16 ++++++++++++++++
>   .travis.yml    | 14 --------------
>   2 files changed, 16 insertions(+), 14 deletions(-)
>
> diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
> index a799803afb9..6552a832939 100644
> --- a/.gitlab-ci.yml
> +++ b/.gitlab-ci.yml
> @@ -475,6 +475,22 @@ build-tci:
>       - QTEST_QEMU_BINARY="./qemu-system-x86_64" ./tests/qtest/pxe-test
>       - QTEST_QEMU_BINARY="./qemu-system-s390x" ./tests/qtest/pxe-test -m slow
>   
> +# Alternate coroutines implementations are only really of interest to KVM users
> +# However we can't test against KVM on Travis so we can only run unit tests


s/Travis/Gitlab

Otherwise,

Reviewed-by: Wainer dos Santos Moschetta <wainersm@redhat.com>


> +build-coroutine-ucontext:
> +  <<: *native_build_job_definition
> +  variables:
> +    IMAGE: ubuntu2004
> +    CONFIGURE_ARGS: --with-coroutine=ucontext --disable-tcg
> +    MAKE_CHECK_ARGS: check-unit
> +
> +build-coroutine-sigaltstack:
> +  <<: *native_build_job_definition
> +  variables:
> +    IMAGE: ubuntu2004
> +    CONFIGURE_ARGS: --with-coroutine=sigaltstack --disable-tcg
> +    MAKE_CHECK_ARGS: check-unit
> +
>   # Most jobs test latest gcrypt or nettle builds
>   #
>   # These jobs test old gcrypt and nettle from RHEL7
> diff --git a/.travis.yml b/.travis.yml
> index 15550fb023f..8ef31f8d8b6 100644
> --- a/.travis.yml
> +++ b/.travis.yml
> @@ -139,20 +139,6 @@ jobs:
>           - CACHE_NAME="${TRAVIS_BRANCH}-linux-gcc-default"
>   
>   
> -    # Alternate coroutines implementations are only really of interest to KVM users
> -    # However we can't test against KVM on Travis so we can only run unit tests
> -    - name: "check-unit coroutine=ucontext"
> -      env:
> -        - CONFIG="--with-coroutine=ucontext --disable-tcg"
> -        - TEST_CMD="make check-unit -j${JOBS} V=1"
> -
> -
> -    - name: "check-unit coroutine=sigaltstack"
> -      env:
> -        - CONFIG="--with-coroutine=sigaltstack --disable-tcg"
> -        - TEST_CMD="make check-unit -j${JOBS} V=1"
> -
> -
>       # Check we can build docs and tools (out of tree)
>       - name: "tools and docs (bionic)"
>         dist: bionic

Re: [PATCH-for-6.0 v4 13/17] tests/docker: Install lttng-ust-devel package in Fedora image

[Wainer dos Santos Moschetta]

On 11/8/20 6:45 PM, Philippe Mathieu-Daudé wrote:
> Install the lttng-ust-devel package to be able to
> build QEMU using the User-Space Tracer trace backend
> (configure --enable-trace-backends=ust).
>
> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
> ---
> Cc: Stefan Hajnoczi <stefanha@redhat.com>
> Cc: Daniel P . Berrange <berrange@redhat.com>
> ---
>   tests/docker/dockerfiles/fedora.docker | 1 +
>   1 file changed, 1 insertion(+)


Reviewed-by: Wainer dos Santos Moschetta <wainersm@redhat.com>


>
> diff --git a/tests/docker/dockerfiles/fedora.docker b/tests/docker/dockerfiles/fedora.docker
> index 0b5053f2d09..ac90f187882 100644
> --- a/tests/docker/dockerfiles/fedora.docker
> +++ b/tests/docker/dockerfiles/fedora.docker
> @@ -44,6 +44,7 @@ ENV PACKAGES \
>       libxml2-devel \
>       libzstd-devel \
>       llvm \
> +    lttng-ust-devel \
>       lzo-devel \
>       make \
>       mingw32-bzip2 \

Re: [PATCH-for-6.0 v4 14/17] gitlab-ci: Move trace backend tests across to gitlab

[Wainer dos Santos Moschetta]

On 11/8/20 6:45 PM, Philippe Mathieu-Daudé wrote:
> Similarly to commit 8cdb2cef3f1, move the trace backend
> tests to GitLab.
>
> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
> ---
> Cc: Stefan Hajnoczi <stefanha@redhat.com>
> ---
>   .gitlab-ci.yml | 18 ++++++++++++++++++
>   .travis.yml    | 19 -------------------
>   2 files changed, 18 insertions(+), 19 deletions(-)
>
> diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
> index 6552a832939..2f0da7b3dc1 100644
> --- a/.gitlab-ci.yml
> +++ b/.gitlab-ci.yml
> @@ -557,6 +557,24 @@ check-crypto-only-gnutls:
>       IMAGE: centos7
>       MAKE_CHECK_ARGS: check
>   
> +# We don't need to exercise every backend with every front-end
> +build-trace-multi-user:
> +  <<: *native_build_job_definition
> +  variables:
> +    IMAGE: ubuntu2004

Doesn't it need the lttng-ust-dev package in Ubuntu's image likewise you 
did for Fedora (patch 13)?

> +    CONFIGURE_ARGS: --enable-trace-backends=log,simple,syslog --disable-system
> +
> +build-trace-ftrace-system:
> +  <<: *native_build_job_definition
> +  variables:
> +    IMAGE: ubuntu2004
> +    CONFIGURE_ARGS: --enable-trace-backends=ftrace --target-list=aarch64-softmmu
On Travis it builds the x86_64 softmmu target. Changed to aarch64 to 
increase coverage?
> +
> +build-trace-ust-system:
> +  <<: *native_build_job_definition
> +  variables:
> +    IMAGE: fedora

Similar question here, increasing coverage by using Fedora?

- Wainer

> +    CONFIGURE_ARGS: --enable-trace-backends=ust --target-list=x86_64-softmmu --disable-tcg
>   
>   check-patch:
>     stage: build
> diff --git a/.travis.yml b/.travis.yml
> index 8ef31f8d8b6..ff5d5ead579 100644
> --- a/.travis.yml
> +++ b/.travis.yml
> @@ -182,25 +182,6 @@ jobs:
>         compiler: clang
>   
>   
> -    # We don't need to exercise every backend with every front-end
> -    - name: "GCC trace log,simple,syslog (user)"
> -      env:
> -        - CONFIG="--enable-trace-backends=log,simple,syslog --disable-system"
> -        - TEST_CMD=""
> -
> -
> -    - name: "GCC trace ftrace (x86_64-softmmu)"
> -      env:
> -        - CONFIG="--enable-trace-backends=ftrace --target-list=x86_64-softmmu"
> -        - TEST_CMD=""
> -
> -
> -    - name: "GCC trace ust (x86_64-softmmu)"
> -      env:
> -        - CONFIG="--enable-trace-backends=ust --target-list=x86_64-softmmu"
> -        - TEST_CMD=""
> -
> -
>       # Using newer GCC with sanitizers
>       - name: "GCC9 with sanitizers (softmmu)"
>         dist: bionic

Re: [PATCH-for-6.0 v4 08/17] gitlab-ci: Move linux-user debug-tcg test across to gitlab

[Alex Bennée]

Philippe Mathieu-Daudé <philmd@redhat.com> writes:

> Similarly to commit 8cdb2cef3f1, move the linux-user (debug-tcg)
> test to GitLab.
>
> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
> ---
> Cc: Laurent Vivier <laurent@vivier.eu>
> ---
>  .gitlab-ci.yml | 7 +++++++
>  .travis.yml    | 9 ---------
>  2 files changed, 7 insertions(+), 9 deletions(-)
>
> diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
> index 3fc3d0568c6..80082a602b8 100644
> --- a/.gitlab-ci.yml
> +++ b/.gitlab-ci.yml
> @@ -304,6 +304,13 @@ build-user:
>      CONFIGURE_ARGS: --disable-tools --disable-system
>      MAKE_CHECK_ARGS: check-tcg
>  
> +build-user-debug:
> +  <<: *native_build_job_definition
> +  variables:
> +    IMAGE: debian-all-test-cross
> +    CONFIGURE_ARGS: --disable-tools --disable-system --enable-debug-tcg
> +    MAKE_CHECK_ARGS: check-tcg
> +
>  # Run check-tcg against linux-user (with plugins)
>  # we skip sparc64-linux-user until it has been fixed somewhat
>  # we skip cris-linux-user as it doesn't use the common run loop
> diff --git a/.travis.yml b/.travis.yml
> index 15d92291358..bee6197290d 100644
> --- a/.travis.yml
> +++ b/.travis.yml
> @@ -293,15 +293,6 @@ jobs:
>          - ${SRC_DIR}/configure ${CONFIG} --extra-cflags="-g3 -O0 -fsanitize=thread" || { cat config.log meson-logs/meson-log.txt && exit 1; }
>  
>  
> -    # Run check-tcg against linux-user
> -    - name: "GCC check-tcg (user)"
> -      env:
> -        - CONFIG="--disable-system --enable-debug-tcg"
> -        - TEST_BUILD_CMD="make build-tcg"
> -        - TEST_CMD="make check-tcg"
> -        - CACHE_NAME="${TRAVIS_BRANCH}-linux-gcc-debug-tcg"
> -
> -
>      # Run check-tcg against softmmu targets
>      - name: "GCC check-tcg (some-softmmu)"
>        env:

I just realised I replicated this is a slightly different way - by
dropping --debug-tcg and moving the rest in one commit. I skipped over
the for 6.0 stuff when looking over your series but it's certainly worth
moving the check-tcg ones now given the stability issues.

-- 
Alex Bennée

Re: [PATCH-for-6.0 v4 14/17] gitlab-ci: Move trace backend tests across to gitlab

[Philippe Mathieu-Daudé]

On 11/10/20 7:12 PM, Wainer dos Santos Moschetta wrote:
> 
> On 11/8/20 6:45 PM, Philippe Mathieu-Daudé wrote:
>> Similarly to commit 8cdb2cef3f1, move the trace backend
>> tests to GitLab.
>>
>> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
>> ---
>> Cc: Stefan Hajnoczi <stefanha@redhat.com>
>> ---
>>   .gitlab-ci.yml | 18 ++++++++++++++++++
>>   .travis.yml    | 19 -------------------
>>   2 files changed, 18 insertions(+), 19 deletions(-)
>>
>> diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
>> index 6552a832939..2f0da7b3dc1 100644
>> --- a/.gitlab-ci.yml
>> +++ b/.gitlab-ci.yml
>> @@ -557,6 +557,24 @@ check-crypto-only-gnutls:
>>       IMAGE: centos7
>>       MAKE_CHECK_ARGS: check
>>   +# We don't need to exercise every backend with every front-end
>> +build-trace-multi-user:
>> +  <<: *native_build_job_definition
>> +  variables:
>> +    IMAGE: ubuntu2004
> 
> Doesn't it need the lttng-ust-dev package in Ubuntu's image likewise you
> did for Fedora (patch 13)?

No, because ...

>> +    CONFIGURE_ARGS: --enable-trace-backends=log,simple,syslog
>> --disable-system

... we don't build the User-Space Tracer there.

>> +
>> +build-trace-ftrace-system:
>> +  <<: *native_build_job_definition
>> +  variables:
>> +    IMAGE: ubuntu2004
>> +    CONFIGURE_ARGS: --enable-trace-backends=ftrace
>> --target-list=aarch64-softmmu
> On Travis it builds the x86_64 softmmu target. Changed to aarch64 to
> increase coverage?

Maybe because the next job already test this target? I'll change.

>> +
>> +build-trace-ust-system:
>> +  <<: *native_build_job_definition
>> +  variables:
>> +    IMAGE: fedora
> 
> Similar question here, increasing coverage by using Fedora?

No, because User-Space Tracer is a feature Red Hat cares
about, and plan to maintain, so it makes sense to test it
on Fedora. I'll comment this change in the description.

> 
> - Wainer
> 
>> +    CONFIGURE_ARGS: --enable-trace-backends=ust
>> --target-list=x86_64-softmmu --disable-tcg
>>     check-patch:
>>     stage: build
>> diff --git a/.travis.yml b/.travis.yml
>> index 8ef31f8d8b6..ff5d5ead579 100644
>> --- a/.travis.yml
>> +++ b/.travis.yml
>> @@ -182,25 +182,6 @@ jobs:
>>         compiler: clang
>>     -    # We don't need to exercise every backend with every front-end
>> -    - name: "GCC trace log,simple,syslog (user)"
>> -      env:
>> -        - CONFIG="--enable-trace-backends=log,simple,syslog
>> --disable-system"
>> -        - TEST_CMD=""
>> -
>> -
>> -    - name: "GCC trace ftrace (x86_64-softmmu)"
>> -      env:
>> -        - CONFIG="--enable-trace-backends=ftrace
>> --target-list=x86_64-softmmu"
>> -        - TEST_CMD=""
>> -
>> -
>> -    - name: "GCC trace ust (x86_64-softmmu)"
>> -      env:
>> -        - CONFIG="--enable-trace-backends=ust
>> --target-list=x86_64-softmmu"
>> -        - TEST_CMD=""
>> -
>> -
>>       # Using newer GCC with sanitizers
>>       - name: "GCC9 with sanitizers (softmmu)"
>>         dist: bionic
> 

Re: [PATCH-for-6.0 v4 14/17] gitlab-ci: Move trace backend tests across to gitlab

[Philippe Mathieu-Daudé]

On 11/11/20 9:42 AM, Philippe Mathieu-Daudé wrote:
> On 11/10/20 7:12 PM, Wainer dos Santos Moschetta wrote:
>>
>> On 11/8/20 6:45 PM, Philippe Mathieu-Daudé wrote:
>>> Similarly to commit 8cdb2cef3f1, move the trace backend
>>> tests to GitLab.
>>>
>>> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
>>> ---
>>> Cc: Stefan Hajnoczi <stefanha@redhat.com>
>>> ---
>>>   .gitlab-ci.yml | 18 ++++++++++++++++++
>>>   .travis.yml    | 19 -------------------
>>>   2 files changed, 18 insertions(+), 19 deletions(-)
>>>
>>> diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
>>> index 6552a832939..2f0da7b3dc1 100644
>>> --- a/.gitlab-ci.yml
>>> +++ b/.gitlab-ci.yml
>>> @@ -557,6 +557,24 @@ check-crypto-only-gnutls:
>>>       IMAGE: centos7
>>>       MAKE_CHECK_ARGS: check
>>>   +# We don't need to exercise every backend with every front-end
>>> +build-trace-multi-user:
>>> +  <<: *native_build_job_definition
>>> +  variables:
>>> +    IMAGE: ubuntu2004
>>
>> Doesn't it need the lttng-ust-dev package in Ubuntu's image likewise you
>> did for Fedora (patch 13)?
> 
> No, because ...
> 
>>> +    CONFIGURE_ARGS: --enable-trace-backends=log,simple,syslog
>>> --disable-system
> 
> ... we don't build the User-Space Tracer there.

And User-Space Tracer is covered on Ubuntu by the s390x
jobs on Travis. I'll add a comment so this is clear.

Thanks.

Re: [PATCH-for-6.0 v4 14/17] gitlab-ci: Move trace backend tests across to gitlab

[Philippe Mathieu-Daudé]

On 11/11/20 9:42 AM, Philippe Mathieu-Daudé wrote:
> On 11/10/20 7:12 PM, Wainer dos Santos Moschetta wrote:
>>
>> On 11/8/20 6:45 PM, Philippe Mathieu-Daudé wrote:
>>> Similarly to commit 8cdb2cef3f1, move the trace backend
>>> tests to GitLab.
>>>
>>> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
>>> ---
>>> Cc: Stefan Hajnoczi <stefanha@redhat.com>
>>> ---
>>>   .gitlab-ci.yml | 18 ++++++++++++++++++
>>>   .travis.yml    | 19 -------------------
>>>   2 files changed, 18 insertions(+), 19 deletions(-)
>>>
>>> diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
>>> index 6552a832939..2f0da7b3dc1 100644
>>> --- a/.gitlab-ci.yml
>>> +++ b/.gitlab-ci.yml
>>> @@ -557,6 +557,24 @@ check-crypto-only-gnutls:
>>>       IMAGE: centos7
>>>       MAKE_CHECK_ARGS: check
>>>   +# We don't need to exercise every backend with every front-end
>>> +build-trace-multi-user:
>>> +  <<: *native_build_job_definition
>>> +  variables:
>>> +    IMAGE: ubuntu2004
>>
>> Doesn't it need the lttng-ust-dev package in Ubuntu's image likewise you
>> did for Fedora (patch 13)?
> 
> No, because ...
> 
>>> +    CONFIGURE_ARGS: --enable-trace-backends=log,simple,syslog
>>> --disable-system
> 
> ... we don't build the User-Space Tracer there.
> 
>>> +
>>> +build-trace-ftrace-system:
>>> +  <<: *native_build_job_definition
>>> +  variables:
>>> +    IMAGE: ubuntu2004
>>> +    CONFIGURE_ARGS: --enable-trace-backends=ftrace
>>> --target-list=aarch64-softmmu
>> On Travis it builds the x86_64 softmmu target. Changed to aarch64 to
>> increase coverage?
> 
> Maybe because the next job already test this target? I'll change.
> 
>>> +
>>> +build-trace-ust-system:
>>> +  <<: *native_build_job_definition
>>> +  variables:
>>> +    IMAGE: fedora
>>
>> Similar question here, increasing coverage by using Fedora?
> 
> No, because User-Space Tracer is a feature Red Hat cares
> about, and plan to maintain, so it makes sense to test it
> on Fedora. I'll comment this change in the description.

Bah, this is a different change in itself, out of the scope of
this series. I'll keep Ubuntu image. Whoever is interested to
test this on Fedora can add a proper job. Thanks.

Re: [PATCH-for-6.0 v4 15/17] gitlab-ci: Add test for Xen (on CentOS 7)

[Anthony PERARD via]

On Sun, Nov 08, 2020 at 09:45:33PM +0100, Philippe Mathieu-Daudé wrote:
> Xen packages are available in CentOS 7, but have been
> removed from CentOS 8. Use the CentOS 7 container.

Technically Xen has never been in CentOS 8, I'm working on it, slowly.

> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
> ---
> Cc: Eduardo Habkost <ehabkost@redhat.com>
> Cc: Stefano Stabellini <sstabellini@kernel.org>
> Cc: Anthony Perard <anthony.perard@citrix.com>
> Cc: Paul Durrant <paul@xen.org>
> Cc: xen-devel@lists.xenproject.org
> ---
>  .gitlab-ci.yml | 21 +++++++++++++++++++++
>  1 file changed, 21 insertions(+)
> 
> diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
> index 2f0da7b3dc1..8e15266c277 100644
> --- a/.gitlab-ci.yml
> +++ b/.gitlab-ci.yml
> @@ -557,6 +557,27 @@ check-crypto-only-gnutls:
>      IMAGE: centos7
>      MAKE_CHECK_ARGS: check
>  
> +build-xen-centos:
> +  <<: *native_build_job_definition
> +  variables:
> +    IMAGE: centos7
> +    TARGETS: i386-softmmu x86_64-softmmu
> +    CONFIGURE_ARGS: --enable-xen
> +    MAKE_CHECK_ARGS: check-build
> +  artifacts:
> +    paths:
> +      - build
> +
> +check-xen-centos:
> +  <<: *native_test_job_definition
> +  needs:
> +    - job: build-xen-centos
> +      artifacts: true
> +  variables:
> +    IMAGE: centos7
> +    MAKE_CHECK_ARGS: check

Is `make check` going to do something useful with the Xen support? Or is
it going to need more work in order to test the Xen support of QEMU?
(Like starting an actual Xen guest.)

Cheers,

-- 
Anthony PERARD

Re: [PATCH-for-6.0 v4 15/17] gitlab-ci: Add test for Xen (on CentOS 7)

[Anthony PERARD]

On Sun, Nov 08, 2020 at 09:45:33PM +0100, Philippe Mathieu-Daudé wrote:
> Xen packages are available in CentOS 7, but have been
> removed from CentOS 8. Use the CentOS 7 container.

Technically Xen has never been in CentOS 8, I'm working on it, slowly.

> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
> ---
> Cc: Eduardo Habkost <ehabkost@redhat.com>
> Cc: Stefano Stabellini <sstabellini@kernel.org>
> Cc: Anthony Perard <anthony.perard@citrix.com>
> Cc: Paul Durrant <paul@xen.org>
> Cc: xen-devel@lists.xenproject.org
> ---
>  .gitlab-ci.yml | 21 +++++++++++++++++++++
>  1 file changed, 21 insertions(+)
> 
> diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
> index 2f0da7b3dc1..8e15266c277 100644
> --- a/.gitlab-ci.yml
> +++ b/.gitlab-ci.yml
> @@ -557,6 +557,27 @@ check-crypto-only-gnutls:
>      IMAGE: centos7
>      MAKE_CHECK_ARGS: check
>  
> +build-xen-centos:
> +  <<: *native_build_job_definition
> +  variables:
> +    IMAGE: centos7
> +    TARGETS: i386-softmmu x86_64-softmmu
> +    CONFIGURE_ARGS: --enable-xen
> +    MAKE_CHECK_ARGS: check-build
> +  artifacts:
> +    paths:
> +      - build
> +
> +check-xen-centos:
> +  <<: *native_test_job_definition
> +  needs:
> +    - job: build-xen-centos
> +      artifacts: true
> +  variables:
> +    IMAGE: centos7
> +    MAKE_CHECK_ARGS: check

Is `make check` going to do something useful with the Xen support? Or is
it going to need more work in order to test the Xen support of QEMU?
(Like starting an actual Xen guest.)

Cheers,

-- 
Anthony PERARD

Re: [PATCH-for-6.0 v4 15/17] gitlab-ci: Add test for Xen (on CentOS 7)

[Eduardo Habkost]

On Thu, Nov 26, 2020 at 05:38:24PM +0000, Anthony PERARD wrote:
> On Sun, Nov 08, 2020 at 09:45:33PM +0100, Philippe Mathieu-Daudé wrote:
> > Xen packages are available in CentOS 7, but have been
> > removed from CentOS 8. Use the CentOS 7 container.
> 
> Technically Xen has never been in CentOS 8, I'm working on it, slowly.
> 
> > Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
> > ---
> > Cc: Eduardo Habkost <ehabkost@redhat.com>
> > Cc: Stefano Stabellini <sstabellini@kernel.org>
> > Cc: Anthony Perard <anthony.perard@citrix.com>
> > Cc: Paul Durrant <paul@xen.org>
> > Cc: xen-devel@lists.xenproject.org
> > ---
> >  .gitlab-ci.yml | 21 +++++++++++++++++++++
> >  1 file changed, 21 insertions(+)
> > 
> > diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
> > index 2f0da7b3dc1..8e15266c277 100644
> > --- a/.gitlab-ci.yml
> > +++ b/.gitlab-ci.yml
> > @@ -557,6 +557,27 @@ check-crypto-only-gnutls:
> >      IMAGE: centos7
> >      MAKE_CHECK_ARGS: check
> >  
> > +build-xen-centos:
> > +  <<: *native_build_job_definition
> > +  variables:
> > +    IMAGE: centos7
> > +    TARGETS: i386-softmmu x86_64-softmmu
> > +    CONFIGURE_ARGS: --enable-xen
> > +    MAKE_CHECK_ARGS: check-build
> > +  artifacts:
> > +    paths:
> > +      - build
> > +
> > +check-xen-centos:
> > +  <<: *native_test_job_definition
> > +  needs:
> > +    - job: build-xen-centos
> > +      artifacts: true
> > +  variables:
> > +    IMAGE: centos7
> > +    MAKE_CHECK_ARGS: check
> 
> Is `make check` going to do something useful with the Xen support? Or is
> it going to need more work in order to test the Xen support of QEMU?
> (Like starting an actual Xen guest.)

I don't think it will test Xen support, but we still want to at
least check if --enable-xen doesn't break anything else.

Is there any public CI system anywhere where Xen support is
tested today?

-- 
Eduardo

Re: [PATCH-for-6.0 v4 15/17] gitlab-ci: Add test for Xen (on CentOS 7)

[Eduardo Habkost]

On Thu, Nov 26, 2020 at 05:38:24PM +0000, Anthony PERARD wrote:
> On Sun, Nov 08, 2020 at 09:45:33PM +0100, Philippe Mathieu-Daudé wrote:
> > Xen packages are available in CentOS 7, but have been
> > removed from CentOS 8. Use the CentOS 7 container.
> 
> Technically Xen has never been in CentOS 8, I'm working on it, slowly.
> 
> > Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
> > ---
> > Cc: Eduardo Habkost <ehabkost@redhat.com>
> > Cc: Stefano Stabellini <sstabellini@kernel.org>
> > Cc: Anthony Perard <anthony.perard@citrix.com>
> > Cc: Paul Durrant <paul@xen.org>
> > Cc: xen-devel@lists.xenproject.org
> > ---
> >  .gitlab-ci.yml | 21 +++++++++++++++++++++
> >  1 file changed, 21 insertions(+)
> > 
> > diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
> > index 2f0da7b3dc1..8e15266c277 100644
> > --- a/.gitlab-ci.yml
> > +++ b/.gitlab-ci.yml
> > @@ -557,6 +557,27 @@ check-crypto-only-gnutls:
> >      IMAGE: centos7
> >      MAKE_CHECK_ARGS: check
> >  
> > +build-xen-centos:
> > +  <<: *native_build_job_definition
> > +  variables:
> > +    IMAGE: centos7
> > +    TARGETS: i386-softmmu x86_64-softmmu
> > +    CONFIGURE_ARGS: --enable-xen
> > +    MAKE_CHECK_ARGS: check-build
> > +  artifacts:
> > +    paths:
> > +      - build
> > +
> > +check-xen-centos:
> > +  <<: *native_test_job_definition
> > +  needs:
> > +    - job: build-xen-centos
> > +      artifacts: true
> > +  variables:
> > +    IMAGE: centos7
> > +    MAKE_CHECK_ARGS: check
> 
> Is `make check` going to do something useful with the Xen support? Or is
> it going to need more work in order to test the Xen support of QEMU?
> (Like starting an actual Xen guest.)

I don't think it will test Xen support, but we still want to at
least check if --enable-xen doesn't break anything else.

Is there any public CI system anywhere where Xen support is
tested today?

-- 
Eduardo

Re: [PATCH-for-6.0 v4 15/17] gitlab-ci: Add test for Xen (on CentOS 7)

[Anthony PERARD via]

On Thu, Nov 26, 2020 at 12:45:59PM -0500, Eduardo Habkost wrote:
> On Thu, Nov 26, 2020 at 05:38:24PM +0000, Anthony PERARD wrote:
> > Is `make check` going to do something useful with the Xen support? Or is
> > it going to need more work in order to test the Xen support of QEMU?
> > (Like starting an actual Xen guest.)
> 
> I don't think it will test Xen support, but we still want to at
> least check if --enable-xen doesn't break anything else.

That sound good.

> Is there any public CI system anywhere where Xen support is
> tested today?

Yes, we have osstest which regularly test Xen with QEMU from upstream.
Result are sent to xen-devel. But that might not be very useful for
qemu-devel.

We also have a GitLab CI which does some Xen tests, but I don't think
QEMU is tested there.
https://xenbits.xen.org/gitweb/?p=xen.git;a=blob;f=automation/gitlab-ci/test.yaml;hb=HEAD
https://gitlab.com/xen-project/xen/

Cheers,

-- 
Anthony PERARD

Re: [PATCH-for-6.0 v4 15/17] gitlab-ci: Add test for Xen (on CentOS 7)

[Anthony PERARD]

On Thu, Nov 26, 2020 at 12:45:59PM -0500, Eduardo Habkost wrote:
> On Thu, Nov 26, 2020 at 05:38:24PM +0000, Anthony PERARD wrote:
> > Is `make check` going to do something useful with the Xen support? Or is
> > it going to need more work in order to test the Xen support of QEMU?
> > (Like starting an actual Xen guest.)
> 
> I don't think it will test Xen support, but we still want to at
> least check if --enable-xen doesn't break anything else.

That sound good.

> Is there any public CI system anywhere where Xen support is
> tested today?

Yes, we have osstest which regularly test Xen with QEMU from upstream.
Result are sent to xen-devel. But that might not be very useful for
qemu-devel.

We also have a GitLab CI which does some Xen tests, but I don't think
QEMU is tested there.
https://xenbits.xen.org/gitweb/?p=xen.git;a=blob;f=automation/gitlab-ci/test.yaml;hb=HEAD
https://gitlab.com/xen-project/xen/

Cheers,

-- 
Anthony PERARD

Re: [PATCH-for-6.0 v4 15/17] gitlab-ci: Add test for Xen (on CentOS 7)

[Stefano Stabellini]

On Fri, 27 Nov 2020, Anthony PERARD wrote:
> On Thu, Nov 26, 2020 at 12:45:59PM -0500, Eduardo Habkost wrote:
> > On Thu, Nov 26, 2020 at 05:38:24PM +0000, Anthony PERARD wrote:
> > > Is `make check` going to do something useful with the Xen support? Or is
> > > it going to need more work in order to test the Xen support of QEMU?
> > > (Like starting an actual Xen guest.)
> > 
> > I don't think it will test Xen support, but we still want to at
> > least check if --enable-xen doesn't break anything else.
> 
> That sound good.
> 
> > Is there any public CI system anywhere where Xen support is
> > tested today?
> 
> Yes, we have osstest which regularly test Xen with QEMU from upstream.
> Result are sent to xen-devel. But that might not be very useful for
> qemu-devel.
> 
> We also have a GitLab CI which does some Xen tests, but I don't think
> QEMU is tested there.
> https://xenbits.xen.org/gitweb/?p=xen.git;a=blob;f=automation/gitlab-ci/test.yaml;hb=HEAD
> https://gitlab.com/xen-project/xen/

QEMU (the version of QEMU picked by the Xen tools) is built but not used
in the Xen Project CI-loop yet.

I am extending the CI-loop with more tests [1], and I would like to have at
least one QEMU test at some point soon. Probably something based on Xen 9pfs.

[1] https://marc.info/?l=xen-devel&m=160627845825763 

Re: [PATCH-for-6.0 v4 15/17] gitlab-ci: Add test for Xen (on CentOS 7)

[Stefano Stabellini]

On Fri, 27 Nov 2020, Anthony PERARD wrote:
> On Thu, Nov 26, 2020 at 12:45:59PM -0500, Eduardo Habkost wrote:
> > On Thu, Nov 26, 2020 at 05:38:24PM +0000, Anthony PERARD wrote:
> > > Is `make check` going to do something useful with the Xen support? Or is
> > > it going to need more work in order to test the Xen support of QEMU?
> > > (Like starting an actual Xen guest.)
> > 
> > I don't think it will test Xen support, but we still want to at
> > least check if --enable-xen doesn't break anything else.
> 
> That sound good.
> 
> > Is there any public CI system anywhere where Xen support is
> > tested today?
> 
> Yes, we have osstest which regularly test Xen with QEMU from upstream.
> Result are sent to xen-devel. But that might not be very useful for
> qemu-devel.
> 
> We also have a GitLab CI which does some Xen tests, but I don't think
> QEMU is tested there.
> https://xenbits.xen.org/gitweb/?p=xen.git;a=blob;f=automation/gitlab-ci/test.yaml;hb=HEAD
> https://gitlab.com/xen-project/xen/

QEMU (the version of QEMU picked by the Xen tools) is built but not used
in the Xen Project CI-loop yet.

I am extending the CI-loop with more tests [1], and I would like to have at
least one QEMU test at some point soon. Probably something based on Xen 9pfs.

[1] https://marc.info/?l=xen-devel&m=160627845825763 

Re: [PATCH-for-5.2 v4 00/17] ci: Move various jobs from Travis to GitLab CI

[Thomas Huth]

On 08/11/2020 21.45, Philippe Mathieu-Daudé wrote:
> We have various job covered on Travis-CI, but recommend the
> community to use GitLab, so build failures are missed.
> 
> We need help to move the jobs to GitLab (we will keep the s390x
> and ppc64 containerized jobs on Travis as there is no similar
> offer on GitLab). Start moving jobs.

Now that the hard freeze is over, I've put some of your patches (that looked
ready to go) in my current pull request. Once that has been merged, could
you please respin your series with the remaining patches? That would be
great, since Travis is likely unusable for every average developer by the
end of the year...

 Thomas