[PATCH for 4.2-rc2 v1 0/5] misc doc and testing fixes

[PATCH for 4.2-rc2 v1 0/5] misc doc and testing fixes

[Alex Bennée]

Hi,

As we approach release I'm just bundling up my random fixes for 4.2
into one tree for each release candidate. This has a minor fix for
debugging vm builds and a number of minor documentation fixes
including moving the MTTCG docs across to rst format. The final patch
is a tweak to Travis to drop the 32 bit targets from
MAIN_SOFTMMU_TARGETS. This may be too controversial although the 32
bit targets still get built and tested under GCC.

Please review.

Alex Bennée (5):
  tests/vm: make --interactive (and therefore DEBUG=1) unconditional
  docs/devel: rename plugins.rst to tcg-plugins.rst
  docs/devel: update tcg-plugins.rst with API versioning details
  docs/devel: convert multi-thread-tcg to a .rst document
  .travis.yml: drop 32 bit systems from MAIN_SOFTMMU_TARGETS

 docs/devel/index.rst                          |  3 +-
 ...ti-thread-tcg.txt => multi-thread-tcg.rst} | 28 ++++++++++++-------
 docs/devel/{plugins.rst => tcg-plugins.rst}   | 27 ++++++++++++++----
 .travis.yml                                   |  2 +-
 MAINTAINERS                                   |  1 +
 tests/vm/basevm.py                            |  2 +-
 6 files changed, 44 insertions(+), 19 deletions(-)
 rename docs/devel/{multi-thread-tcg.txt => multi-thread-tcg.rst} (96%)
 rename docs/devel/{plugins.rst => tcg-plugins.rst} (83%)

-- 
2.20.1

[PATCH v1 1/5] tests/vm: make --interactive (and therefore DEBUG=1) unconditional

[Alex Bennée]

While the concept of only dropping to ssh if a test fails is nice it
is more useful for this to be unconditional. You usually just want to
get the build up and running and then noodle around debugging or
attempting to replicate.

Cc: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>

---
v2
  - fix spelling
---
 tests/vm/basevm.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/vm/basevm.py b/tests/vm/basevm.py
index 91a9226026d..0b8c1b26576 100755
--- a/tests/vm/basevm.py
+++ b/tests/vm/basevm.py
@@ -403,7 +403,7 @@ def main(vmcls):
     exitcode = 0
     if vm.ssh(*cmd) != 0:
         exitcode = 3
-    if exitcode != 0 and args.interactive:
+    if args.interactive:
         vm.ssh()
 
     if not args.snapshot:
-- 
2.20.1

Re: [PATCH v1 1/5] tests/vm: make --interactive (and therefore DEBUG=1) unconditional

[Thomas Huth]

On 13/11/2019 12.59, Alex Bennée wrote:
> While the concept of only dropping to ssh if a test fails is nice it
> is more useful for this to be unconditional. You usually just want to
> get the build up and running and then noodle around debugging or
> attempting to replicate.
> 
> Cc: Peter Maydell <peter.maydell@linaro.org>
> Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
> 
> ---
> v2
>   - fix spelling
> ---
>  tests/vm/basevm.py | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/tests/vm/basevm.py b/tests/vm/basevm.py
> index 91a9226026d..0b8c1b26576 100755
> --- a/tests/vm/basevm.py
> +++ b/tests/vm/basevm.py
> @@ -403,7 +403,7 @@ def main(vmcls):
>      exitcode = 0
>      if vm.ssh(*cmd) != 0:
>          exitcode = 3
> -    if exitcode != 0 and args.interactive:
> +    if args.interactive:
>          vm.ssh()
>  
>      if not args.snapshot:
> 

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

[PATCH v1 2/5] docs/devel: rename plugins.rst to tcg-plugins.rst

[Alex Bennée]

This makes it a bit clearer what this is about.

Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
---
 docs/devel/index.rst                        | 2 +-
 docs/devel/{plugins.rst => tcg-plugins.rst} | 0
 MAINTAINERS                                 | 1 +
 3 files changed, 2 insertions(+), 1 deletion(-)
 rename docs/devel/{plugins.rst => tcg-plugins.rst} (100%)

diff --git a/docs/devel/index.rst b/docs/devel/index.rst
index 2ff058bae38..c86a3cdff2f 100644
--- a/docs/devel/index.rst
+++ b/docs/devel/index.rst
@@ -22,4 +22,4 @@ Contents:
    decodetree
    secure-coding-practices
    tcg
-   plugins
+   tcg-plugins
diff --git a/docs/devel/plugins.rst b/docs/devel/tcg-plugins.rst
similarity index 100%
rename from docs/devel/plugins.rst
rename to docs/devel/tcg-plugins.rst
diff --git a/MAINTAINERS b/MAINTAINERS
index ff8d0d29f4b..b160d817208 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -2369,6 +2369,7 @@ F: tcg/
 TCG Plugins
 M: Alex Bennée <alex.bennee@linaro.org>
 S: Maintained
+F: docs/devel/tcg-plugins.rst
 F: plugins/
 F: tests/plugin
 
-- 
2.20.1

Re: [PATCH v1 2/5] docs/devel: rename plugins.rst to tcg-plugins.rst

[Philippe Mathieu-Daudé]

On 11/13/19 12:59 PM, Alex Bennée wrote:
> This makes it a bit clearer what this is about.
> 
> Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
> ---
>   docs/devel/index.rst                        | 2 +-
>   docs/devel/{plugins.rst => tcg-plugins.rst} | 0
>   MAINTAINERS                                 | 1 +
>   3 files changed, 2 insertions(+), 1 deletion(-)
>   rename docs/devel/{plugins.rst => tcg-plugins.rst} (100%)
> 
> diff --git a/docs/devel/index.rst b/docs/devel/index.rst
> index 2ff058bae38..c86a3cdff2f 100644
> --- a/docs/devel/index.rst
> +++ b/docs/devel/index.rst
> @@ -22,4 +22,4 @@ Contents:
>      decodetree
>      secure-coding-practices
>      tcg
> -   plugins
> +   tcg-plugins
> diff --git a/docs/devel/plugins.rst b/docs/devel/tcg-plugins.rst
> similarity index 100%
> rename from docs/devel/plugins.rst
> rename to docs/devel/tcg-plugins.rst
> diff --git a/MAINTAINERS b/MAINTAINERS
> index ff8d0d29f4b..b160d817208 100644
> --- a/MAINTAINERS
> +++ b/MAINTAINERS
> @@ -2369,6 +2369,7 @@ F: tcg/
>   TCG Plugins
>   M: Alex Bennée <alex.bennee@linaro.org>
>   S: Maintained
> +F: docs/devel/tcg-plugins.rst
>   F: plugins/

What about (other patch):

   plugins/ -> tcg/plugins/

>   F: tests/plugin

tests/plugin/ -> tests/tcg/plugins/

Anyway for this patch:
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>

Re: [PATCH v1 2/5] docs/devel: rename plugins.rst to tcg-plugins.rst

[Paolo Bonzini]

> What about (other patch):
> 
>   plugins/ -> tcg/plugins/
> 
>>   F: tests/plugin
> 
> tests/plugin/ -> tests/tcg/plugins/

tests/tcg is for code that runs on the target in general, but perhaps
tests/tcg-plugins would make sense.

Paolo

[PATCH v1 3/5] docs/devel: update tcg-plugins.rst with API versioning details

[Alex Bennée]

While we are at it fix up the quoted code sections with the inline ::
approach.

Signed-off-by: Alex Bennée <alex.bennee@linaro.org>

---
v2
  - fix grammar
  - mention we also will fail to load outside the range
  - clean-up code sections
---
 docs/devel/tcg-plugins.rst | 27 +++++++++++++++++++++------
 1 file changed, 21 insertions(+), 6 deletions(-)

diff --git a/docs/devel/tcg-plugins.rst b/docs/devel/tcg-plugins.rst
index b18fb6729e3..718eef00f22 100644
--- a/docs/devel/tcg-plugins.rst
+++ b/docs/devel/tcg-plugins.rst
@@ -25,6 +25,23 @@ process. However the project reserves the right to change or break the
 API should it need to do so. The best way to avoid this is to submit
 your plugin upstream so they can be updated if/when the API changes.
 
+API versioning
+--------------
+
+All plugins need to declare a symbol which exports the plugin API
+version they were built against. This can be done simply by::
+
+  QEMU_PLUGIN_EXPORT int qemu_plugin_version = QEMU_PLUGIN_VERSION;
+
+The core code will refuse to load a plugin that doesn't export a
+`qemu_plugin_version` symbol or if plugin version is outside of QEMU's
+supported range of API versions.
+
+Additionally the `qemu_info_t` structure which is passed to the
+`qemu_plugin_install` method of a plugin will detail the minimum and
+current API versions supported by QEMU. The API version will be
+incremented if new APIs are added. The minimum API version will be
+incremented if existing APIs are changed or removed.
 
 Exposure of QEMU internals
 --------------------------
@@ -40,16 +57,14 @@ instructions and events are opaque to the plugins themselves.
 Usage
 =====
 
-The QEMU binary needs to be compiled for plugin support:
+The QEMU binary needs to be compiled for plugin support::
 
-::
-    configure --enable-plugins
+  configure --enable-plugins
 
 Once built a program can be run with multiple plugins loaded each with
-their own arguments:
+their own arguments::
 
-::
-    $QEMU $OTHER_QEMU_ARGS \
+  $QEMU $OTHER_QEMU_ARGS \
       -plugin tests/plugin/libhowvec.so,arg=inline,arg=hint \
       -plugin tests/plugin/libhotblocks.so
 
-- 
2.20.1

Re: FW: [PATCH v1 3/5] docs/devel: update tcg-plugins.rst with API versioning details

[Robert Foley]

On Wed, 13 Nov 2019 at 07:00, Alex Bennée <alex.bennee@linaro.org> wrote:
>
> While we are at it fix up the quoted code sections with the inline ::
> approach.
>
> Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
>
> ---
> v2
>   - fix grammar
>   - mention we also will fail to load outside the range
>   - clean-up code sections
> ---
>  docs/devel/tcg-plugins.rst | 27 +++++++++++++++++++++------
>  1 file changed, 21 insertions(+), 6 deletions(-)
>
> diff --git a/docs/devel/tcg-plugins.rst b/docs/devel/tcg-plugins.rst
> index b18fb6729e3..718eef00f22 100644
> --- a/docs/devel/tcg-plugins.rst
> +++ b/docs/devel/tcg-plugins.rst
> @@ -25,6 +25,23 @@ process. However the project reserves the right to change or break the
>  API should it need to do so. The best way to avoid this is to submit
>  your plugin upstream so they can be updated if/when the API changes.
>
> +API versioning
> +--------------
> +
> +All plugins need to declare a symbol which exports the plugin API
> +version they were built against. This can be done simply by::
> +
> +  QEMU_PLUGIN_EXPORT int qemu_plugin_version = QEMU_PLUGIN_VERSION;
> +
> +The core code will refuse to load a plugin that doesn't export a
> +`qemu_plugin_version` symbol or if plugin version is outside of QEMU's
> +supported range of API versions.
> +
> +Additionally the `qemu_info_t` structure which is passed to the
> +`qemu_plugin_install` method of a plugin will detail the minimum and
> +current API versions supported by QEMU. The API version will be
> +incremented if new APIs are added. The minimum API version will be
> +incremented if existing APIs are changed or removed.
>
>  Exposure of QEMU internals
>  --------------------------
> @@ -40,16 +57,14 @@ instructions and events are opaque to the plugins themselves.
>  Usage
>  =====
>
> -The QEMU binary needs to be compiled for plugin support:
> +The QEMU binary needs to be compiled for plugin support::
>
> -::
> -    configure --enable-plugins
> +  configure --enable-plugins
>
>  Once built a program can be run with multiple plugins loaded each with
> -their own arguments:
> +their own arguments::
>
> -::
> -    $QEMU $OTHER_QEMU_ARGS \
> +  $QEMU $OTHER_QEMU_ARGS \
>        -plugin tests/plugin/libhowvec.so,arg=inline,arg=hint \
>        -plugin tests/plugin/libhotblocks.so
>
> --
> 2.20.1
>
>
Reviewed-by: Robert Foley <robert.foley@linaro.org>

[PATCH v1 4/5] docs/devel: convert multi-thread-tcg to a .rst document

[Alex Bennée]

Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
---
 docs/devel/index.rst                          |  1 +
 ...ti-thread-tcg.txt => multi-thread-tcg.rst} | 28 ++++++++++++-------
 2 files changed, 19 insertions(+), 10 deletions(-)
 rename docs/devel/{multi-thread-tcg.txt => multi-thread-tcg.rst} (96%)

diff --git a/docs/devel/index.rst b/docs/devel/index.rst
index c86a3cdff2f..3e6624ec604 100644
--- a/docs/devel/index.rst
+++ b/docs/devel/index.rst
@@ -22,4 +22,5 @@ Contents:
    decodetree
    secure-coding-practices
    tcg
+   multi-thread-tcg
    tcg-plugins
diff --git a/docs/devel/multi-thread-tcg.txt b/docs/devel/multi-thread-tcg.rst
similarity index 96%
rename from docs/devel/multi-thread-tcg.txt
rename to docs/devel/multi-thread-tcg.rst
index 782bebc28b4..4e914bacc0c 100644
--- a/docs/devel/multi-thread-tcg.txt
+++ b/docs/devel/multi-thread-tcg.rst
@@ -1,7 +1,10 @@
-Copyright (c) 2015-2016 Linaro Ltd.
+.. Copyright (c) 2015-2016 Linaro Ltd.
+.. This work is licensed under the terms of the GNU GPL, version 2 or
+.. later. See the COPYING file in the top-level directory.
 
-This work is licensed under the terms of the GNU GPL, version 2 or
-later. See the COPYING file in the top-level directory.
+==================
+Multi-threaded TCG
+==================
 
 Introduction
 ============
@@ -40,7 +43,7 @@ Main Run Loop
 Even when there is no code being generated there are a number of
 structures associated with the hot-path through the main run-loop.
 These are associated with looking up the next translation block to
-execute. These include:
+execute. These include::
 
     tb_jmp_cache (per-vCPU, cache of recent jumps)
     tb_ctx.htable (global hash table, phys address->tb lookup)
@@ -61,7 +64,9 @@ have their block-to-block jumps patched.
 Global TCG State
 ----------------
 
-### User-mode emulation
+User-mode emulation
+~~~~~~~~~~~~~~~~~~~
+
 We need to protect the entire code generation cycle including any post
 generation patching of the translated code. This also implies a shared
 translation buffer which contains code running on all cores. Any
@@ -78,7 +83,9 @@ patching.
 
 Code generation is serialised with mmap_lock().
 
-### !User-mode emulation
+System emulation
+~~~~~~~~~~~~~~~~
+
 Each vCPU has its own TCG context and associated TCG region, thereby
 requiring no locking.
 
@@ -125,10 +132,11 @@ linked list of all Translation Blocks in that page (see page_next).
 Both the jump patching and the page cache involve linked lists that
 the invalidated TranslationBlock needs to be removed from.
 
-DESIGN REQUIREMENT: Safely handle invalidation of TBs
-                      - safely patch/revert direct jumps
-                      - remove central PageDesc lookup entries
-                      - ensure lookup caches/hashes are safely updated
+DESIGN REQUIREMENTS:
+  - Safely handle invalidation of TBs
+     - safely patch/revert direct jumps
+     - remove central PageDesc lookup entries
+     - ensure lookup caches/hashes are safely updated
 
 (Current solution)
 
-- 
2.20.1

[PATCH v1 5/5] .travis.yml: drop 32 bit systems from MAIN_SOFTMMU_TARGETS

[Alex Bennée]

The older clangs are still struggling to build and run everything
withing the 50 minute timeout so lets lighten the load a bit more. We
still have coverage for GCC and hopefully no obscure 32 bit guest only
breakages slip through the cracks.

Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
---
 .travis.yml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.travis.yml b/.travis.yml
index b9a026c8eeb..c09b6a00143 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -79,7 +79,7 @@ env:
     - BASE_CONFIG="--disable-docs --disable-tools"
     - TEST_CMD="make check V=1"
     # This is broadly a list of "mainline" softmmu targets which have support across the major distros
-    - MAIN_SOFTMMU_TARGETS="aarch64-softmmu,arm-softmmu,i386-softmmu,mips-softmmu,mips64-softmmu,ppc64-softmmu,riscv64-softmmu,s390x-softmmu,x86_64-softmmu"
+    - MAIN_SOFTMMU_TARGETS="aarch64-softmmu,mips64-softmmu,ppc64-softmmu,riscv64-softmmu,s390x-softmmu,x86_64-softmmu"
     - CCACHE_SLOPPINESS="include_file_ctime,include_file_mtime"
     - CCACHE_MAXSIZE=1G
 
-- 
2.20.1

Re: [PATCH v1 5/5] .travis.yml: drop 32 bit systems from MAIN_SOFTMMU_TARGETS

[Thomas Huth]

On 13/11/2019 12.59, Alex Bennée wrote:
> The older clangs are still struggling to build and run everything
> withing the 50 minute timeout so lets lighten the load a bit more. We
> still have coverage for GCC and hopefully no obscure 32 bit guest only
> breakages slip through the cracks.
> 
> Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
> ---
>  .travis.yml | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/.travis.yml b/.travis.yml
> index b9a026c8eeb..c09b6a00143 100644
> --- a/.travis.yml
> +++ b/.travis.yml
> @@ -79,7 +79,7 @@ env:
>      - BASE_CONFIG="--disable-docs --disable-tools"
>      - TEST_CMD="make check V=1"
>      # This is broadly a list of "mainline" softmmu targets which have support across the major distros
> -    - MAIN_SOFTMMU_TARGETS="aarch64-softmmu,arm-softmmu,i386-softmmu,mips-softmmu,mips64-softmmu,ppc64-softmmu,riscv64-softmmu,s390x-softmmu,x86_64-softmmu"
> +    - MAIN_SOFTMMU_TARGETS="aarch64-softmmu,mips64-softmmu,ppc64-softmmu,riscv64-softmmu,s390x-softmmu,x86_64-softmmu"
>      - CCACHE_SLOPPINESS="include_file_ctime,include_file_mtime"
>      - CCACHE_MAXSIZE=1G

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

Re: [PATCH v1 5/5] .travis.yml: drop 32 bit systems from MAIN_SOFTMMU_TARGETS

[Thomas Huth]

On 13/11/2019 14.30, Thomas Huth wrote:
> On 13/11/2019 12.59, Alex Bennée wrote:
>> The older clangs are still struggling to build and run everything
>> withing the 50 minute timeout so lets lighten the load a bit more. We
>> still have coverage for GCC and hopefully no obscure 32 bit guest only
>> breakages slip through the cracks.
>>
>> Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
>> ---
>>  .travis.yml | 2 +-
>>  1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/.travis.yml b/.travis.yml
>> index b9a026c8eeb..c09b6a00143 100644
>> --- a/.travis.yml
>> +++ b/.travis.yml
>> @@ -79,7 +79,7 @@ env:
>>      - BASE_CONFIG="--disable-docs --disable-tools"
>>      - TEST_CMD="make check V=1"
>>      # This is broadly a list of "mainline" softmmu targets which have support across the major distros
>> -    - MAIN_SOFTMMU_TARGETS="aarch64-softmmu,arm-softmmu,i386-softmmu,mips-softmmu,mips64-softmmu,ppc64-softmmu,riscv64-softmmu,s390x-softmmu,x86_64-softmmu"
>> +    - MAIN_SOFTMMU_TARGETS="aarch64-softmmu,mips64-softmmu,ppc64-softmmu,riscv64-softmmu,s390x-softmmu,x86_64-softmmu"
>>      - CCACHE_SLOPPINESS="include_file_ctime,include_file_mtime"
>>      - CCACHE_MAXSIZE=1G
> 
> Reviewed-by: Thomas Huth <thuth@redhat.com>

On a second glance, we also have this entry with --target-list-exclude
in our test matrix:

    - env:
        - CONFIG="--disable-user
--target-list-exclude=${MAIN_SOFTMMU_TARGETS}"
        - CACHE_NAME="${TRAVIS_BRANCH}-linux-clang-default"
      compiler: clang

So while you've speed up one target, this one might get actually slower
instead. That's a little bit unfortunate. Is there maybe a better way to
tackle this?

 Thomas

Re: [PATCH v1 5/5] .travis.yml: drop 32 bit systems from MAIN_SOFTMMU_TARGETS

[Philippe Mathieu-Daudé]

On 11/13/19 12:59 PM, Alex Bennée wrote:
> The older clangs are still struggling to build and run everything
> withing the 50 minute timeout so lets lighten the load a bit more. We
> still have coverage for GCC and hopefully no obscure 32 bit guest only
> breakages slip through the cracks.
> 
> Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
> ---
>   .travis.yml | 2 +-
>   1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/.travis.yml b/.travis.yml
> index b9a026c8eeb..c09b6a00143 100644
> --- a/.travis.yml
> +++ b/.travis.yml
> @@ -79,7 +79,7 @@ env:
>       - BASE_CONFIG="--disable-docs --disable-tools"
>       - TEST_CMD="make check V=1"
>       # This is broadly a list of "mainline" softmmu targets which have support across the major distros
> -    - MAIN_SOFTMMU_TARGETS="aarch64-softmmu,arm-softmmu,i386-softmmu,mips-softmmu,mips64-softmmu,ppc64-softmmu,riscv64-softmmu,s390x-softmmu,x86_64-softmmu"
> +    - MAIN_SOFTMMU_TARGETS="aarch64-softmmu,mips64-softmmu,ppc64-softmmu,riscv64-softmmu,s390x-softmmu,x86_64-softmmu"

Aleksandar, since you mostly test 32-bit MIPS, are you OK we keep 
mips-softmmu and drop mips64-softmmu here? Another job (acceptance-test) 
builds the mips64el-softmmu.

>       - CCACHE_SLOPPINESS="include_file_ctime,include_file_mtime"
>       - CCACHE_MAXSIZE=1G
>   
> 

Re: [EXTERNAL]Re: [PATCH v1 5/5] .travis.yml: drop 32 bit systems from MAIN_SOFTMMU_TARGETS

[Aleksandar Markovic]

> From: Philippe Mathieu-Daudé <philmd@redhat.com>
> > -    - MAIN_SOFTMMU_TARGETS="aarch64-softmmu,arm-softmmu,i386-softmmu,mips-softmmu,mips64-softmmu,ppc64-softmmu,riscv64-softmmu,s390x-softmmu,x86_64-softmmu"
> > +    - MAIN_SOFTMMU_TARGETS="aarch64-softmmu,mips64-softmmu,ppc64-softmmu,riscv64-softmmu,s390x-softmmu,x86_64-softmmu"
> 
> Aleksandar, since you mostly test 32-bit MIPS, are you OK we keep
> mips-softmmu and drop mips64-softmmu here? Another job (acceptance-test)
> builds the mips64el-softmmu.

Philippe, thanks for bringing this to my attention. Yes, 32-bit mips targets are important to us, but, what can we do, time constraints are time constraints, so I agree with Alex change, please go ahead, Alex. We can test 32-bit mips targets via other acceptance tests (those that can run longer, so-called "slow" group), and perhaps we can extend them to test more 32-bit mips systems.

Thanks to everybody,
Aleksandar

Re: [EXTERNAL]Re: [PATCH v1 5/5] .travis.yml: drop 32 bit systems from MAIN_SOFTMMU_TARGETS

[Philippe Mathieu-Daudé]

On 11/13/19 6:38 PM, Aleksandar Markovic wrote:
>> From: Philippe Mathieu-Daudé <philmd@redhat.com>
>>> -    - MAIN_SOFTMMU_TARGETS="aarch64-softmmu,arm-softmmu,i386-softmmu,mips-softmmu,mips64-softmmu,ppc64-softmmu,riscv64-softmmu,s390x-softmmu,x86_64-softmmu"
>>> +    - MAIN_SOFTMMU_TARGETS="aarch64-softmmu,mips64-softmmu,ppc64-softmmu,riscv64-softmmu,s390x-softmmu,x86_64-softmmu"
>>
>> Aleksandar, since you mostly test 32-bit MIPS, are you OK we keep
>> mips-softmmu and drop mips64-softmmu here? Another job (acceptance-test)
>> builds the mips64el-softmmu.
> 
> Philippe, thanks for bringing this to my attention. Yes, 32-bit mips targets are important to us, but, what can we do, time constraints are time constraints, so I agree with Alex change, please go ahead, Alex. We can test 32-bit mips targets via other acceptance tests (those that can run longer, so-called "slow" group), and perhaps we can extend them to test more 32-bit mips systems.

OK, let's keep mips64 as suggested Alex then.

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

Re: [EXTERNAL]Re: [PATCH v1 5/5] .travis.yml: drop 32 bit systems from MAIN_SOFTMMU_TARGETS

[Alex Bennée]

Aleksandar Markovic <amarkovic@wavecomp.com> writes:

>> From: Philippe Mathieu-Daudé <philmd@redhat.com>
>> > -    - MAIN_SOFTMMU_TARGETS="aarch64-softmmu,arm-softmmu,i386-softmmu,mips-softmmu,mips64-softmmu,ppc64-softmmu,riscv64-softmmu,s390x-softmmu,x86_64-softmmu"
>> > +    - MAIN_SOFTMMU_TARGETS="aarch64-softmmu,mips64-softmmu,ppc64-softmmu,riscv64-softmmu,s390x-softmmu,x86_64-softmmu"
>>
>> Aleksandar, since you mostly test 32-bit MIPS, are you OK we keep
>> mips-softmmu and drop mips64-softmmu here? Another job (acceptance-test)
>> builds the mips64el-softmmu.
>
> Philippe, thanks for bringing this to my attention. Yes, 32-bit mips
> targets are important to us, but, what can we do, time constraints are
> time constraints, so I agree with Alex change, please go ahead, Alex.
> We can test 32-bit mips targets via other acceptance tests (those that
> can run longer, so-called "slow" group), and perhaps we can extend
> them to test more 32-bit mips systems.

To be clear both gcc and clang have rules that test:

        - CONFIG="--disable-user --target-list-exclude=${MAIN_SOFTMMU_TARGETS}"

So the main targets which are reducing their coverage are:

        - CONFIG="--enable-debug --target-list=${MAIN_SOFTMMU_TARGETS}"

        - CONFIG="--enable-modules --target-list=${MAIN_SOFTMMU_TARGETS}"

        - CONFIG="--target-list=${MAIN_SOFTMMU_TARGETS} "
        - CACHE_NAME="${TRAVIS_BRANCH}-linux-clang-sanitize"
      compiler: clang
      before_script:
        - ./configure ${CONFIG} --extra-cflags="-fsanitize=undefined -Werror" || { cat config.log && exit 1; }

        - CONFIG="--enable-gprof --enable-gcov --disable-pie --target-list=${MAIN_SOFTMMU_TARGETS}"

and the MacOSX 9.4 build:
        # MacOSX builds
        - env:
            - CONFIG="--target-list=${MAIN_SOFTMMU_TARGETS}"
          os: osx
          osx_image: xcode9.4
          compiler: clang

The Xcode 10.3 build is already a reduced list:
        - CONFIG="--target-list=i386-softmmu,ppc-softmmu,ppc64-softmmu,m68k-softmmu,x86_64-softmmu"


>
> Thanks to everybody,
> Aleksandar


--
Alex Bennée