[Qemu-devel] [PATCH for-4.0] compiler.h: Add an explicit check for the compiler version

[Qemu-devel] [PATCH for-4.0] compiler.h: Add an explicit check for the compiler version

[Thomas Huth]

The questions about our minimum compiler requirement pops up every
couple of months, and we then have to recall the details each time.
So let's document this in a proper way, by adding a comment and
check for the right compiler version to our compiler.h header.

Signed-off-by: Thomas Huth <thuth@redhat.com>
---
 include/qemu/compiler.h | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/include/qemu/compiler.h b/include/qemu/compiler.h
index ca9bc85..775446b 100644
--- a/include/qemu/compiler.h
+++ b/include/qemu/compiler.h
@@ -22,6 +22,14 @@
 # define QEMU_GNUC_PREREQ(maj, min) 0
 #endif
 
+/*
+ * We need at least GCC 4.1 for atomics support. Clang also supports these,
+ * and reports itself as GCC 4.2, so it passes this check, too.
+ */
+#if !QEMU_GNUC_PREREQ(4, 1)
+#error QEMU needs a compiler that is compatible with GCC v4.1 or newer
+#endif
+
 #define QEMU_NORETURN __attribute__ ((__noreturn__))
 
 #define QEMU_WARN_UNUSED_RESULT __attribute__((warn_unused_result))
-- 
1.8.3.1

Re: [Qemu-devel] [PATCH for-4.0] compiler.h: Add an explicit check for the compiler version

[Paolo Bonzini]

On 30/11/18 11:43, Thomas Huth wrote:
> The questions about our minimum compiler requirement pops up every
> couple of months, and we then have to recall the details each time.
> So let's document this in a proper way, by adding a comment and
> check for the right compiler version to our compiler.h header.
> 
> Signed-off-by: Thomas Huth <thuth@redhat.com>
> ---
>  include/qemu/compiler.h | 8 ++++++++
>  1 file changed, 8 insertions(+)
> 
> diff --git a/include/qemu/compiler.h b/include/qemu/compiler.h
> index ca9bc85..775446b 100644
> --- a/include/qemu/compiler.h
> +++ b/include/qemu/compiler.h
> @@ -22,6 +22,14 @@
>  # define QEMU_GNUC_PREREQ(maj, min) 0
>  #endif
>  
> +/*
> + * We need at least GCC 4.1 for atomics support. Clang also supports these,
> + * and reports itself as GCC 4.2, so it passes this check, too.
> + */
> +#if !QEMU_GNUC_PREREQ(4, 1)
> +#error QEMU needs a compiler that is compatible with GCC v4.1 or newer
> +#endif
> +
>  #define QEMU_NORETURN __attribute__ ((__noreturn__))
>  
>  #define QEMU_WARN_UNUSED_RESULT __attribute__((warn_unused_result))
> 

Queued, thanks.

Paolo

Re: [Qemu-devel] [PATCH for-4.0] compiler.h: Add an explicit check for the compiler version

[Daniel P. Berrangé]

On Fri, Nov 30, 2018 at 11:43:40AM +0100, Thomas Huth wrote:
> The questions about our minimum compiler requirement pops up every
> couple of months, and we then have to recall the details each time.
> So let's document this in a proper way, by adding a comment and
> check for the right compiler version to our compiler.h header.
> 
> Signed-off-by: Thomas Huth <thuth@redhat.com>
> ---
>  include/qemu/compiler.h | 8 ++++++++
>  1 file changed, 8 insertions(+)
> 
> diff --git a/include/qemu/compiler.h b/include/qemu/compiler.h
> index ca9bc85..775446b 100644
> --- a/include/qemu/compiler.h
> +++ b/include/qemu/compiler.h
> @@ -22,6 +22,14 @@
>  # define QEMU_GNUC_PREREQ(maj, min) 0
>  #endif
>  
> +/*
> + * We need at least GCC 4.1 for atomics support. Clang also supports these,
> + * and reports itself as GCC 4.2, so it passes this check, too.
> + */
> +#if !QEMU_GNUC_PREREQ(4, 1)
> +#error QEMU needs a compiler that is compatible with GCC v4.1 or newer
> +#endif

This encodes our current minimum which is fine as a first step.

I think we could reasonably increase our min version now that we
have declared explicitly what platforms we intend to support

      RHEL-7: 4.8.5
      Debian (Stretch): 6.3.0
      Debian (Jessie): 4.8.4
      OpenBSD (ports): 4.9.4
      FreeBSD (ports): 8.2.0
      OpenSUSE Leap 15: 7.3.1
      Ubuntu (Xenial): 5.3.1
      macOS (Homebrew): 8.2.0

Arguably we don't care about gcc version for *BSD and macOS since those
platforms normally use CLang

Anyway, with this info I think we can reasonably pick gcc 4.8.0

We would need an explicit check for clang, however, instead of
relying on it claiming gcc 4.2 support - that's a way inaccurate
claim anyway so detecting a specific clang version would be
better regardless IMHO


Regards,
Daniel
-- 
|: https://berrange.com      -o-    https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org         -o-            https://fstop138.berrange.com :|
|: https://entangle-photo.org    -o-    https://www.instagram.com/dberrange :|

Re: [Qemu-devel] [PATCH for-4.0] compiler.h: Add an explicit check for the compiler version

[Thomas Huth]

On 2018-11-30 12:15, Daniel P. Berrangé wrote:
> On Fri, Nov 30, 2018 at 11:43:40AM +0100, Thomas Huth wrote:
>> The questions about our minimum compiler requirement pops up every
>> couple of months, and we then have to recall the details each time.
>> So let's document this in a proper way, by adding a comment and
>> check for the right compiler version to our compiler.h header.
>>
>> Signed-off-by: Thomas Huth <thuth@redhat.com>
>> ---
>>  include/qemu/compiler.h | 8 ++++++++
>>  1 file changed, 8 insertions(+)
>>
>> diff --git a/include/qemu/compiler.h b/include/qemu/compiler.h
>> index ca9bc85..775446b 100644
>> --- a/include/qemu/compiler.h
>> +++ b/include/qemu/compiler.h
>> @@ -22,6 +22,14 @@
>>  # define QEMU_GNUC_PREREQ(maj, min) 0
>>  #endif
>>  
>> +/*
>> + * We need at least GCC 4.1 for atomics support. Clang also supports these,
>> + * and reports itself as GCC 4.2, so it passes this check, too.
>> + */
>> +#if !QEMU_GNUC_PREREQ(4, 1)
>> +#error QEMU needs a compiler that is compatible with GCC v4.1 or newer
>> +#endif
> 
> This encodes our current minimum which is fine as a first step.
> 
> I think we could reasonably increase our min version now that we
> have declared explicitly what platforms we intend to support

Do we really want to artificially limit our support here without any
further reasons? If the users want to compile QEMU on an older system,
and one of the libraries does not quite match anymore, it's often easy
enough to recompile a newer version of the library to get things going
again. Recompiling a whole compiler is way more cumbersome, though...

OTOH, we could get rid of some more #if QEMU_GNUC_PREREQ spots in the
source code if we bump the minimum version to 4.8 ... so that might be a
real reason to increase the minimum.

> We would need an explicit check for clang, however, instead of
> relying on it claiming gcc 4.2 support - that's a way inaccurate
> claim anyway so detecting a specific clang version would be
> better regardless IMHO

Fine for me if we bump the minimum Clang version to 3.4 ... then we
could get rid of the the CONFIG_INT128 hacks that we have in various
parts of the code.

 Thomas

Re: [Qemu-devel] [PATCH for-4.0] compiler.h: Add an explicit check for the compiler version

[Paolo Bonzini]

On 30/11/18 13:30, Thomas Huth wrote:
> Do we really want to artificially limit our support here without any
> further reasons? If the users want to compile QEMU on an older system,
> and one of the libraries does not quite match anymore, it's often easy
> enough to recompile a newer version of the library to get things going
> again. Recompiling a whole compiler is way more cumbersome, though...
> 
> OTOH, we could get rid of some more #if QEMU_GNUC_PREREQ spots in the
> source code if we bump the minimum version to 4.8 ... so that might be a
> real reason to increase the minimum.

Of all of them, I think, if we bump to 4.7?

Older 4.6 and 4.7 version have the PR55489 bug that we're checking for
in configure, too ("we should be able to delete this at the end of 2013"
:)).  Bumping to 4.8 would let us remove TRANSLATE_OPT_CFLAGS for good.

Paolo

>> We would need an explicit check for clang, however, instead of
>> relying on it claiming gcc 4.2 support - that's a way inaccurate
>> claim anyway so detecting a specific clang version would be
>> better regardless IMHO
> 
> Fine for me if we bump the minimum Clang version to 3.4 ... then we
> could get rid of the the CONFIG_INT128 hacks that we have in various
> parts of the code.
> 
>  Thomas
> 

Re: [Qemu-devel] [PATCH for-4.0] compiler.h: Add an explicit check for the compiler version

[Thomas Huth]

On 2018-11-30 13:46, Paolo Bonzini wrote:
> On 30/11/18 13:30, Thomas Huth wrote:
>> Do we really want to artificially limit our support here without any
>> further reasons? If the users want to compile QEMU on an older system,
>> and one of the libraries does not quite match anymore, it's often easy
>> enough to recompile a newer version of the library to get things going
>> again. Recompiling a whole compiler is way more cumbersome, though...
>>
>> OTOH, we could get rid of some more #if QEMU_GNUC_PREREQ spots in the
>> source code if we bump the minimum version to 4.8 ... so that might be a
>> real reason to increase the minimum.
> 
> Of all of them, I think, if we bump to 4.7?
> 
> Older 4.6 and 4.7 version have the PR55489 bug that we're checking for
> in configure, too ("we should be able to delete this at the end of 2013"
> :)).  Bumping to 4.8 would let us remove TRANSLATE_OPT_CFLAGS for good.

Ok, sounds like bumping to 4.8 would really help to get rid of some
legacy stuff ... so please drop this patch for compiler.h, I'll try to
come up with some other patches for checking the GCC version and Clang
version in the configure script already instead.

 Thomas

Re: [Qemu-devel] [PATCH for-4.0] compiler.h: Add an explicit check for the compiler version

[Daniel P. Berrangé]

On Fri, Nov 30, 2018 at 01:30:12PM +0100, Thomas Huth wrote:
> On 2018-11-30 12:15, Daniel P. Berrangé wrote:
> > On Fri, Nov 30, 2018 at 11:43:40AM +0100, Thomas Huth wrote:
> >> The questions about our minimum compiler requirement pops up every
> >> couple of months, and we then have to recall the details each time.
> >> So let's document this in a proper way, by adding a comment and
> >> check for the right compiler version to our compiler.h header.
> >>
> >> Signed-off-by: Thomas Huth <thuth@redhat.com>
> >> ---
> >>  include/qemu/compiler.h | 8 ++++++++
> >>  1 file changed, 8 insertions(+)
> >>
> >> diff --git a/include/qemu/compiler.h b/include/qemu/compiler.h
> >> index ca9bc85..775446b 100644
> >> --- a/include/qemu/compiler.h
> >> +++ b/include/qemu/compiler.h
> >> @@ -22,6 +22,14 @@
> >>  # define QEMU_GNUC_PREREQ(maj, min) 0
> >>  #endif
> >>  
> >> +/*
> >> + * We need at least GCC 4.1 for atomics support. Clang also supports these,
> >> + * and reports itself as GCC 4.2, so it passes this check, too.
> >> + */
> >> +#if !QEMU_GNUC_PREREQ(4, 1)
> >> +#error QEMU needs a compiler that is compatible with GCC v4.1 or newer
> >> +#endif
> > 
> > This encodes our current minimum which is fine as a first step.
> > 
> > I think we could reasonably increase our min version now that we
> > have declared explicitly what platforms we intend to support
> 
> Do we really want to artificially limit our support here without any
> further reasons? If the users want to compile QEMU on an older system,
> and one of the libraries does not quite match anymore, it's often easy
> enough to recompile a newer version of the library to get things going
> again. Recompiling a whole compiler is way more cumbersome, though...

The whole point of declaring our supported platforms is that we'll
explicitly not care about whether we can build on older platforms,
regardless of whether the incompatibility is with libraries or
toolchains or something else. It gives us a clear rule on when we're
able to drop back compat support in the code for older tools chains
or libraries, reducing our own maint burden long term.

We fully expect some users will be inconvenienced by this, but the
benefits to our maint workload will outweigh this cost. This lets
us deliver a better project to the vast majority of our userbase
who are not on ancient platforms.

> OTOH, we could get rid of some more #if QEMU_GNUC_PREREQ spots in the
> source code if we bump the minimum version to 4.8 ... so that might be a
> real reason to increase the minimum.

Yes, exactly

> > We would need an explicit check for clang, however, instead of
> > relying on it claiming gcc 4.2 support - that's a way inaccurate
> > claim anyway so detecting a specific clang version would be
> > better regardless IMHO
> 
> Fine for me if we bump the minimum Clang version to 3.4 ... then we
> could get rid of the the CONFIG_INT128 hacks that we have in various
> parts of the code.

Indeed that would be good too.

Regards,
Daniel
-- 
|: https://berrange.com      -o-    https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org         -o-            https://fstop138.berrange.com :|
|: https://entangle-photo.org    -o-    https://www.instagram.com/dberrange :|

Re: [Qemu-devel] [PATCH for-4.0] compiler.h: Add an explicit check for the compiler version

[Daniel P. Berrangé]

On Fri, Nov 30, 2018 at 01:52:05PM +0100, Thomas Huth wrote:
> On 2018-11-30 13:46, Paolo Bonzini wrote:
> > On 30/11/18 13:30, Thomas Huth wrote:
> >> Do we really want to artificially limit our support here without any
> >> further reasons? If the users want to compile QEMU on an older system,
> >> and one of the libraries does not quite match anymore, it's often easy
> >> enough to recompile a newer version of the library to get things going
> >> again. Recompiling a whole compiler is way more cumbersome, though...
> >>
> >> OTOH, we could get rid of some more #if QEMU_GNUC_PREREQ spots in the
> >> source code if we bump the minimum version to 4.8 ... so that might be a
> >> real reason to increase the minimum.
> > 
> > Of all of them, I think, if we bump to 4.7?
> > 
> > Older 4.6 and 4.7 version have the PR55489 bug that we're checking for
> > in configure, too ("we should be able to delete this at the end of 2013"
> > :)).  Bumping to 4.8 would let us remove TRANSLATE_OPT_CFLAGS for good.
> 
> Ok, sounds like bumping to 4.8 would really help to get rid of some
> legacy stuff ... so please drop this patch for compiler.h, I'll try to
> come up with some other patches for checking the GCC version and Clang
> version in the configure script already instead.

BTW, when checking clang version there's some strangeness with macOS
which reports completely different versions from the clang it was
derived from, so we'd require separate checks for normal clang vs
apple's clang.


Regards,
Daniel
-- 
|: https://berrange.com      -o-    https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org         -o-            https://fstop138.berrange.com :|
|: https://entangle-photo.org    -o-    https://www.instagram.com/dberrange :|