[Qemu-devel] [PATCH v2 0/3] pc: compat macroses fixes/cleanups

[Qemu-devel] [PATCH v2 0/3] pc: compat macroses fixes/cleanups

[Igor Mammedov]

Changes since v1:
 - drop 'pc: fix regression introduced by adding 2.8 machine'
   it's not needed
 - cleanup old style compat chaining
 - add missing HW_COMPAT_2_8 to PC_COMPAT_2_8 machine type

ref to v1:
  https://lists.gnu.org/archive/html/qemu-devel/2016-09/msg03140.html


Igor Mammedov (3):
  pc: clean up COMPAT macroses chaining
  target-i386: turn off CPU.l3-cache only for 2.7 and older machine
    types
  add stub HW_COMPAT_2_8 to PC_COMPAT_2_8

 include/hw/compat.h  |  2 ++
 include/hw/i386/pc.h | 10 ++++------
 2 files changed, 6 insertions(+), 6 deletions(-)

-- 
2.7.4

[Qemu-devel] [PATCH v2 1/3] pc: clean up COMPAT macroses chaining

[Igor Mammedov]

Since commit
 bacc344c ("machine: add properties to compat_props incrementaly")
there is no need to chain per machine type compat macroses.

Clean up places where it was done anyway so it will be
consistent and won't confuse contributors during addtion
of new machine types.

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
---
 include/hw/i386/pc.h | 2 --
 1 file changed, 2 deletions(-)

diff --git a/include/hw/i386/pc.h b/include/hw/i386/pc.h
index ab8e319..b0a61f3 100644
--- a/include/hw/i386/pc.h
+++ b/include/hw/i386/pc.h
@@ -375,7 +375,6 @@ bool e820_get_entry(int, uint32_t, uint64_t *, uint64_t *);
 
 
 #define PC_COMPAT_2_7 \
-    PC_COMPAT_2_8 \
     HW_COMPAT_2_7
 
 #define PC_COMPAT_2_6 \
@@ -405,7 +404,6 @@ bool e820_get_entry(int, uint32_t, uint64_t *, uint64_t *);
     },
 
 #define PC_COMPAT_2_5 \
-    PC_COMPAT_2_6 \
     HW_COMPAT_2_5
 
 /* Helper for setting model-id for CPU models that changed model-id
-- 
2.7.4

[Qemu-devel] [PATCH v2 2/3] target-i386: turn off CPU.l3-cache only for 2.7 and older machine types

[Igor Mammedov]

commit (14c985cff target-i386: present virtual L3 cache info for vcpus)
misplaced compat property putting it in new 2.8 machine type
which would effectively to disable feature until 2.9 is released.
Intent of commit probably should be to disable feature for 2.7
and older while allowing not yet released 2.8 to have feature
enabled by default.

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Marcel Apfelbaum <marcel@redhat.com>
---
 include/hw/i386/pc.h | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/include/hw/i386/pc.h b/include/hw/i386/pc.h
index b0a61f3..29a6c9b 100644
--- a/include/hw/i386/pc.h
+++ b/include/hw/i386/pc.h
@@ -367,16 +367,15 @@ int e820_get_num_entries(void);
 bool e820_get_entry(int, uint32_t, uint64_t *, uint64_t *);
 
 #define PC_COMPAT_2_8 \
+
+#define PC_COMPAT_2_7 \
+    HW_COMPAT_2_7 \
     {\
         .driver   = TYPE_X86_CPU,\
         .property = "l3-cache",\
         .value    = "off",\
     },
 
-
-#define PC_COMPAT_2_7 \
-    HW_COMPAT_2_7
-
 #define PC_COMPAT_2_6 \
     HW_COMPAT_2_6 \
     {\
-- 
2.7.4

[Qemu-devel] [PATCH v2 3/3] add stub HW_COMPAT_2_8 to PC_COMPAT_2_8

[Igor Mammedov]

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
---
 include/hw/compat.h  | 2 ++
 include/hw/i386/pc.h | 1 +
 2 files changed, 3 insertions(+)

diff --git a/include/hw/compat.h b/include/hw/compat.h
index a1d6694..ddbb2c4 100644
--- a/include/hw/compat.h
+++ b/include/hw/compat.h
@@ -1,6 +1,8 @@
 #ifndef HW_COMPAT_H
 #define HW_COMPAT_H
 
+#define HW_COMPAT_2_8 \
+
 #define HW_COMPAT_2_7 \
     {\
         .driver   = "virtio-pci",\
diff --git a/include/hw/i386/pc.h b/include/hw/i386/pc.h
index 29a6c9b..012dc2c 100644
--- a/include/hw/i386/pc.h
+++ b/include/hw/i386/pc.h
@@ -367,6 +367,7 @@ int e820_get_num_entries(void);
 bool e820_get_entry(int, uint32_t, uint64_t *, uint64_t *);
 
 #define PC_COMPAT_2_8 \
+    HW_COMPAT_2_8
 
 #define PC_COMPAT_2_7 \
     HW_COMPAT_2_7 \
-- 
2.7.4

Re: [Qemu-devel] [PATCH v2 1/3] pc: clean up COMPAT macroses chaining

[Eduardo Habkost]

On Mon, Sep 19, 2016 at 10:32:33AM +0200, Igor Mammedov wrote:
> Since commit
>  bacc344c ("machine: add properties to compat_props incrementaly")
> there is no need to chain per machine type compat macroses.
> 
> Clean up places where it was done anyway so it will be
> consistent and won't confuse contributors during addtion
> of new machine types.
> 
> Signed-off-by: Igor Mammedov <imammedo@redhat.com>

Reviewed-by: Eduardo Habkost <ehabkost@redhat.com>

-- 
Eduardo

Re: [Qemu-devel] [PATCH v2 2/3] target-i386: turn off CPU.l3-cache only for 2.7 and older machine types

[Eduardo Habkost]

On Mon, Sep 19, 2016 at 10:32:34AM +0200, Igor Mammedov wrote:
> commit (14c985cff target-i386: present virtual L3 cache info for vcpus)
> misplaced compat property putting it in new 2.8 machine type
> which would effectively to disable feature until 2.9 is released.
> Intent of commit probably should be to disable feature for 2.7
> and older while allowing not yet released 2.8 to have feature
> enabled by default.
> 
> Signed-off-by: Igor Mammedov <imammedo@redhat.com>
> Reviewed-by: Marcel Apfelbaum <marcel@redhat.com>
> ---
>  include/hw/i386/pc.h | 7 +++----
>  1 file changed, 3 insertions(+), 4 deletions(-)
> 
> diff --git a/include/hw/i386/pc.h b/include/hw/i386/pc.h
> index b0a61f3..29a6c9b 100644
> --- a/include/hw/i386/pc.h
> +++ b/include/hw/i386/pc.h
> @@ -367,16 +367,15 @@ int e820_get_num_entries(void);
>  bool e820_get_entry(int, uint32_t, uint64_t *, uint64_t *);
>  
>  #define PC_COMPAT_2_8 \

PC_COMPAT_2_8 isn't even used by any code (and shouldn't be used
until we release QEMU 2.8). Let's just delete it by now?

> +
> +#define PC_COMPAT_2_7 \
> +    HW_COMPAT_2_7 \
>      {\
>          .driver   = TYPE_X86_CPU,\
>          .property = "l3-cache",\
>          .value    = "off",\
>      },
>  
> -
> -#define PC_COMPAT_2_7 \
> -    HW_COMPAT_2_7
> -
>  #define PC_COMPAT_2_6 \
>      HW_COMPAT_2_6 \
>      {\
> -- 
> 2.7.4
> 

-- 
Eduardo

Re: [Qemu-devel] [PATCH v2 2/3] target-i386: turn off CPU.l3-cache only for 2.7 and older machine types

[Michael S. Tsirkin]

On Mon, Sep 19, 2016 at 01:58:48PM -0300, Eduardo Habkost wrote:
> On Mon, Sep 19, 2016 at 10:32:34AM +0200, Igor Mammedov wrote:
> > commit (14c985cff target-i386: present virtual L3 cache info for vcpus)
> > misplaced compat property putting it in new 2.8 machine type
> > which would effectively to disable feature until 2.9 is released.
> > Intent of commit probably should be to disable feature for 2.7
> > and older while allowing not yet released 2.8 to have feature
> > enabled by default.
> > 
> > Signed-off-by: Igor Mammedov <imammedo@redhat.com>
> > Reviewed-by: Marcel Apfelbaum <marcel@redhat.com>
> > ---
> >  include/hw/i386/pc.h | 7 +++----
> >  1 file changed, 3 insertions(+), 4 deletions(-)
> > 
> > diff --git a/include/hw/i386/pc.h b/include/hw/i386/pc.h
> > index b0a61f3..29a6c9b 100644
> > --- a/include/hw/i386/pc.h
> > +++ b/include/hw/i386/pc.h
> > @@ -367,16 +367,15 @@ int e820_get_num_entries(void);
> >  bool e820_get_entry(int, uint32_t, uint64_t *, uint64_t *);
> >  
> >  #define PC_COMPAT_2_8 \
> 
> PC_COMPAT_2_8 isn't even used by any code (and shouldn't be used
> until we release QEMU 2.8). Let's just delete it by now?

it's unrelated to this patch though.

> > +
> > +#define PC_COMPAT_2_7 \
> > +    HW_COMPAT_2_7 \
> >      {\
> >          .driver   = TYPE_X86_CPU,\
> >          .property = "l3-cache",\
> >          .value    = "off",\
> >      },
> >  
> > -
> > -#define PC_COMPAT_2_7 \
> > -    HW_COMPAT_2_7
> > -
> >  #define PC_COMPAT_2_6 \
> >      HW_COMPAT_2_6 \
> >      {\
> > -- 
> > 2.7.4
> > 
> 
> -- 
> Eduardo

Re: [Qemu-devel] [PATCH v2 2/3] target-i386: turn off CPU.l3-cache only for 2.7 and older machine types

[Igor Mammedov]

On Mon, 19 Sep 2016 13:58:48 -0300
Eduardo Habkost <ehabkost@redhat.com> wrote:

> On Mon, Sep 19, 2016 at 10:32:34AM +0200, Igor Mammedov wrote:
> > commit (14c985cff target-i386: present virtual L3 cache info for vcpus)
> > misplaced compat property putting it in new 2.8 machine type
> > which would effectively to disable feature until 2.9 is released.
> > Intent of commit probably should be to disable feature for 2.7
> > and older while allowing not yet released 2.8 to have feature
> > enabled by default.
> > 
> > Signed-off-by: Igor Mammedov <imammedo@redhat.com>
> > Reviewed-by: Marcel Apfelbaum <marcel@redhat.com>
> > ---
> >  include/hw/i386/pc.h | 7 +++----
> >  1 file changed, 3 insertions(+), 4 deletions(-)
> > 
> > diff --git a/include/hw/i386/pc.h b/include/hw/i386/pc.h
> > index b0a61f3..29a6c9b 100644
> > --- a/include/hw/i386/pc.h
> > +++ b/include/hw/i386/pc.h
> > @@ -367,16 +367,15 @@ int e820_get_num_entries(void);
> >  bool e820_get_entry(int, uint32_t, uint64_t *, uint64_t *);
> >  
> >  #define PC_COMPAT_2_8 \  
> 
> PC_COMPAT_2_8 isn't even used by any code (and shouldn't be used
> until we release QEMU 2.8). Let's just delete it by now?
Since it's already there in the tree, I'd leave it there.

> 
> > +
> > +#define PC_COMPAT_2_7 \
> > +    HW_COMPAT_2_7 \
> >      {\
> >          .driver   = TYPE_X86_CPU,\
> >          .property = "l3-cache",\
> >          .value    = "off",\
> >      },
> >  
> > -
> > -#define PC_COMPAT_2_7 \
> > -    HW_COMPAT_2_7
> > -
> >  #define PC_COMPAT_2_6 \
> >      HW_COMPAT_2_6 \
> >      {\
> > -- 
> > 2.7.4
> >   
> 

Re: [Qemu-devel] [PATCH v2 2/3] target-i386: turn off CPU.l3-cache only for 2.7 and older machine types

[Eduardo Habkost]

On Tue, Sep 20, 2016 at 09:37:36AM +0200, Igor Mammedov wrote:
> On Mon, 19 Sep 2016 13:58:48 -0300
> Eduardo Habkost <ehabkost@redhat.com> wrote:
> 
> > On Mon, Sep 19, 2016 at 10:32:34AM +0200, Igor Mammedov wrote:
> > > commit (14c985cff target-i386: present virtual L3 cache info for vcpus)
> > > misplaced compat property putting it in new 2.8 machine type
> > > which would effectively to disable feature until 2.9 is released.
> > > Intent of commit probably should be to disable feature for 2.7
> > > and older while allowing not yet released 2.8 to have feature
> > > enabled by default.
> > > 
> > > Signed-off-by: Igor Mammedov <imammedo@redhat.com>
> > > Reviewed-by: Marcel Apfelbaum <marcel@redhat.com>
> > > ---
> > >  include/hw/i386/pc.h | 7 +++----
> > >  1 file changed, 3 insertions(+), 4 deletions(-)
> > > 
> > > diff --git a/include/hw/i386/pc.h b/include/hw/i386/pc.h
> > > index b0a61f3..29a6c9b 100644
> > > --- a/include/hw/i386/pc.h
> > > +++ b/include/hw/i386/pc.h
> > > @@ -367,16 +367,15 @@ int e820_get_num_entries(void);
> > >  bool e820_get_entry(int, uint32_t, uint64_t *, uint64_t *);
> > >  
> > >  #define PC_COMPAT_2_8 \  
> > 
> > PC_COMPAT_2_8 isn't even used by any code (and shouldn't be used
> > until we release QEMU 2.8). Let's just delete it by now?
> Since it's already there in the tree, I'd leave it there.

I'd remove to avoid causing confusion during the 2.8 cycle. But
as it is something independent from this patch:

Reviewed-by: Eduardo Habkost <ehabkost@redhat.com>

> 
> > 
> > > +
> > > +#define PC_COMPAT_2_7 \
> > > +    HW_COMPAT_2_7 \
> > >      {\
> > >          .driver   = TYPE_X86_CPU,\
> > >          .property = "l3-cache",\
> > >          .value    = "off",\
> > >      },
> > >  
> > > -
> > > -#define PC_COMPAT_2_7 \
> > > -    HW_COMPAT_2_7
> > > -
> > >  #define PC_COMPAT_2_6 \
> > >      HW_COMPAT_2_6 \
> > >      {\
> > > -- 
> > > 2.7.4
> > >   
> > 
> 

-- 
Eduardo

Re: [Qemu-devel] [PATCH v2 0/3] pc: compat macroses fixes/cleanups

[Igor Mammedov]

On Mon, 19 Sep 2016 10:32:32 +0200
Igor Mammedov <imammedo@redhat.com> wrote:

> Changes since v1:
>  - drop 'pc: fix regression introduced by adding 2.8 machine'
>    it's not needed
>  - cleanup old style compat chaining
>  - add missing HW_COMPAT_2_8 to PC_COMPAT_2_8 machine type

Eduardo could you take it through your tree?

> 
> ref to v1:
>   https://lists.gnu.org/archive/html/qemu-devel/2016-09/msg03140.html
> 
> 
> Igor Mammedov (3):
>   pc: clean up COMPAT macroses chaining
>   target-i386: turn off CPU.l3-cache only for 2.7 and older machine
>     types
>   add stub HW_COMPAT_2_8 to PC_COMPAT_2_8
> 
>  include/hw/compat.h  |  2 ++
>  include/hw/i386/pc.h | 10 ++++------
>  2 files changed, 6 insertions(+), 6 deletions(-)
> 

Re: [Qemu-devel] [PATCH v2 0/3] pc: compat macroses fixes/cleanups

[Eduardo Habkost]

On Tue, Sep 27, 2016 at 11:32:13AM +0200, Igor Mammedov wrote:
> On Mon, 19 Sep 2016 10:32:32 +0200
> Igor Mammedov <imammedo@redhat.com> wrote:
> 
> > Changes since v1:
> >  - drop 'pc: fix regression introduced by adding 2.8 machine'
> >    it's not needed
> >  - cleanup old style compat chaining
> >  - add missing HW_COMPAT_2_8 to PC_COMPAT_2_8 machine type
> 
> Eduardo could you take it through your tree?

I see it was already merged to master through Michael's tree.
(Except for patch 3/3, which is not needed because we should
delete PC_COMPAT_2_8).

> 
> > 
> > ref to v1:
> >   https://lists.gnu.org/archive/html/qemu-devel/2016-09/msg03140.html
> > 
> > 
> > Igor Mammedov (3):
> >   pc: clean up COMPAT macroses chaining
> >   target-i386: turn off CPU.l3-cache only for 2.7 and older machine
> >     types
> >   add stub HW_COMPAT_2_8 to PC_COMPAT_2_8
> > 
> >  include/hw/compat.h  |  2 ++
> >  include/hw/i386/pc.h | 10 ++++------
> >  2 files changed, 6 insertions(+), 6 deletions(-)
> > 

-- 
Eduardo