diff mbox

target-openrisc: Add typename for CPU models.

Message ID 1372756315-17437-1-git-send-email-elta.era@gmail.com
State New
Headers show

Commit Message

Dongxue Zhang July 2, 2013, 9:11 a.m. UTC 
Make target-openrisc running OK by add typename in openrisc_cpu_class_by_name().

Signed-off-by: Dongxue Zhang <elta.era@gmail.com>
---
 target-openrisc/cpu.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

Comments

Jia Liu July 2, 2013, 9:29 a.m. UTC | #1 
Hi Dongxue,

On Tue, Jul 2, 2013 at 5:11 PM, Dongxue Zhang <elta.era@gmail.com> wrote:
>
> Make target-openrisc running OK by add typename in
openrisc_cpu_class_by_name().
>
> Signed-off-by: Dongxue Zhang <elta.era@gmail.com>
> ---
>  target-openrisc/cpu.c | 4 +++-
>  1 file changed, 3 insertions(+), 1 deletion(-)
>
> diff --git a/target-openrisc/cpu.c b/target-openrisc/cpu.c
> index fd90d37..d38c28b 100644
> --- a/target-openrisc/cpu.c
> +++ b/target-openrisc/cpu.c
> @@ -96,12 +96,14 @@ static void openrisc_cpu_initfn(Object *obj)
>  static ObjectClass *openrisc_cpu_class_by_name(const char *cpu_model)
>  {
>      ObjectClass *oc;
> +    char *typename;
>
>      if (cpu_model == NULL) {
>          return NULL;
>      }
>
> -    oc = object_class_by_name(cpu_model);
> +    typename = g_strdup_printf("%s-" TYPE_OPENRISC_CPU, cpu_model);
> +    oc = object_class_by_name(typename);
>      if (oc != NULL && (!object_class_dynamic_cast(oc, TYPE_OPENRISC_CPU)
||
>                         object_class_is_abstract(oc))) {
>          return NULL;

Thanks for your fix, it looks and test good to me.

>
> --
> 1.8.1.2
>
>

Regards,
Jia
Andreas Färber July 2, 2013, 10:18 a.m. UTC | #2 
Hi Jia,

Am 02.07.2013 11:29, schrieb Jia Liu:
> On Tue, Jul 2, 2013 at 5:11 PM, Dongxue Zhang <elta.era@gmail.com
> <mailto:elta.era@gmail.com>> wrote:
>>
>> Make target-openrisc running OK by add typename in
> openrisc_cpu_class_by_name().
>>
>> Signed-off-by: Dongxue Zhang <elta.era@gmail.com
> <mailto:elta.era@gmail.com>>
>> ---
>>  target-openrisc/cpu.c | 4 +++-
>>  1 file changed, 3 insertions(+), 1 deletion(-)
>>
>> diff --git a/target-openrisc/cpu.c b/target-openrisc/cpu.c
>> index fd90d37..d38c28b 100644
>> --- a/target-openrisc/cpu.c
>> +++ b/target-openrisc/cpu.c
>> @@ -96,12 +96,14 @@ static void openrisc_cpu_initfn(Object *obj)
>>  static ObjectClass *openrisc_cpu_class_by_name(const char *cpu_model)
>>  {
>>      ObjectClass *oc;
>> +    char *typename;
>>
>>      if (cpu_model == NULL) {
>>          return NULL;
>>      }
>>
>> -    oc = object_class_by_name(cpu_model);
>> +    typename = g_strdup_printf("%s-" TYPE_OPENRISC_CPU, cpu_model);
>> +    oc = object_class_by_name(typename);
>>      if (oc != NULL && (!object_class_dynamic_cast(oc,
> TYPE_OPENRISC_CPU) ||
>>                         object_class_is_abstract(oc))) {
>>          return NULL;
> 
> Thanks for your fix, it looks and test good to me.

Sorry for the breakage. Do you want to add a
Reviewed-by/Tested-by/Acked-by? I'd queue it for you then.

If you could upload a Linux test image somewhere that may help avoid
breakages in the future.

Also we reported that there was no maintainer for target-openrisc/ in
MAINTAINERS file, do you want to put yourself there so that you are
CC'ed on patches?

Here's a pointer to the latest refactoring that partially affects or32:
http://lists.gnu.org/archive/html/qemu-devel/2013-06/msg05354.html

Regards,
Andreas
Jia Liu July 2, 2013, 12:30 p.m. UTC | #3 
Hi Andreas,

On Tue, Jul 2, 2013 at 6:18 PM, Andreas Färber <afaerber@suse.de> wrote:
>
> Hi Jia,
>
> Am 02.07.2013 11:29, schrieb Jia Liu:
> > On Tue, Jul 2, 2013 at 5:11 PM, Dongxue Zhang <elta.era@gmail.com
> > <mailto:elta.era@gmail.com>> wrote:
> >>
> >> Make target-openrisc running OK by add typename in
> > openrisc_cpu_class_by_name().
> >>
> >> Signed-off-by: Dongxue Zhang <elta.era@gmail.com
> > <mailto:elta.era@gmail.com>>
> >> ---
> >>  target-openrisc/cpu.c | 4 +++-
> >>  1 file changed, 3 insertions(+), 1 deletion(-)
> >>
> >> diff --git a/target-openrisc/cpu.c b/target-openrisc/cpu.c
> >> index fd90d37..d38c28b 100644
> >> --- a/target-openrisc/cpu.c
> >> +++ b/target-openrisc/cpu.c
> >> @@ -96,12 +96,14 @@ static void openrisc_cpu_initfn(Object *obj)
> >>  static ObjectClass *openrisc_cpu_class_by_name(const char *cpu_model)
> >>  {
> >>      ObjectClass *oc;
> >> +    char *typename;
> >>
> >>      if (cpu_model == NULL) {
> >>          return NULL;
> >>      }
> >>
> >> -    oc = object_class_by_name(cpu_model);
> >> +    typename = g_strdup_printf("%s-" TYPE_OPENRISC_CPU, cpu_model);
> >> +    oc = object_class_by_name(typename);
> >>      if (oc != NULL && (!object_class_dynamic_cast(oc,
> > TYPE_OPENRISC_CPU) ||
> >>                         object_class_is_abstract(oc))) {
> >>          return NULL;
> >
> > Thanks for your fix, it looks and test good to me.
>
> Sorry for the breakage. Do you want to add a
> Reviewed-by/Tested-by/Acked-by? I'd queue it for you then.

Thank you very much! May you please tell me how can I make a
Reviewed-by/Tested-by/Acked-by? I don't know too much about it.

>
> If you could upload a Linux test image somewhere that may help avoid
> breakages in the future.

I find some Linux test images at http://qemu-project.org/Testing . How can
I upload one upon to there?

>
> Also we reported that there was no maintainer for target-openrisc/ in
> MAINTAINERS file, do you want to put yourself there so that you are
> CC'ed on patches?

Thank you, I'll submit a patch to add myself into MAINTAINERS file and
review target-openrisc.

>
> Here's a pointer to the latest refactoring that partially affects or32:
> http://lists.gnu.org/archive/html/qemu-devel/2013-06/msg05354.html

Thank you for patching target-openrisc, I'll test it.

>
> Regards,
> Andreas
>
> --
> SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany
> GF: Jeff Hawn, Jennifer Guild, Felix Imendörffer; HRB 16746 AG Nürnberg

Regards,
Jia
Jia Liu July 5, 2013, 11:48 p.m. UTC | #4 
Hi Andreas,


On Tue, Jul 2, 2013 at 6:18 PM, Andreas Färber <afaerber@suse.de> wrote:
>
> Hi Jia,
>
> Am 02.07.2013 11:29, schrieb Jia Liu:
> > On Tue, Jul 2, 2013 at 5:11 PM, Dongxue Zhang <elta.era@gmail.com
> > <mailto:elta.era@gmail.com>> wrote:
> >>
> >> Make target-openrisc running OK by add typename in
> > openrisc_cpu_class_by_name().
> >>
> >> Signed-off-by: Dongxue Zhang <elta.era@gmail.com
> > <mailto:elta.era@gmail.com>>
> >> ---
> >>  target-openrisc/cpu.c | 4 +++-
> >>  1 file changed, 3 insertions(+), 1 deletion(-)
> >>
> >> diff --git a/target-openrisc/cpu.c b/target-openrisc/cpu.c
> >> index fd90d37..d38c28b 100644
> >> --- a/target-openrisc/cpu.c
> >> +++ b/target-openrisc/cpu.c
> >> @@ -96,12 +96,14 @@ static void openrisc_cpu_initfn(Object *obj)
> >>  static ObjectClass *openrisc_cpu_class_by_name(const char *cpu_model)
> >>  {
> >>      ObjectClass *oc;
> >> +    char *typename;
> >>
> >>      if (cpu_model == NULL) {
> >>          return NULL;
> >>      }
> >>
> >> -    oc = object_class_by_name(cpu_model);
> >> +    typename = g_strdup_printf("%s-" TYPE_OPENRISC_CPU, cpu_model);
> >> +    oc = object_class_by_name(typename);
> >>      if (oc != NULL && (!object_class_dynamic_cast(oc,
> > TYPE_OPENRISC_CPU) ||
> >>                         object_class_is_abstract(oc))) {
> >>          return NULL;
> >
> > Thanks for your fix, it looks and test good to me.
>
> Sorry for the breakage. Do you want to add a
> Reviewed-by/Tested-by/Acked-by? I'd queue it for you then.
>
> If you could upload a Linux test image somewhere that may help avoid
> breakages in the future.

I've upload a Linux test image to my Google Drive.
https://docs.google.com/file/d/0BxeTrz3x0CBLbTNmU0lrV1Y0V0U/edit

>
> Also we reported that there was no maintainer for target-openrisc/ in
> MAINTAINERS file, do you want to put yourself there so that you are
> CC'ed on patches?
>
> Here's a pointer to the latest refactoring that partially affects or32:
> http://lists.gnu.org/archive/html/qemu-devel/2013-06/msg05354.html
>
> Regards,
> Andreas
>
> --
> SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany
> GF: Jeff Hawn, Jennifer Guild, Felix Imendörffer; HRB 16746 AG Nürnberg

Regards,
Jia
Andreas Färber July 7, 2013, 4:08 p.m. UTC | #5 
Hi,

Am 02.07.2013 14:30, schrieb Jia Liu:
> On Tue, Jul 2, 2013 at 6:18 PM, Andreas Färber <afaerber@suse.de
> <mailto:afaerber@suse.de>> wrote:
>> Am 02.07.2013 11:29, schrieb Jia Liu:
>> > On Tue, Jul 2, 2013 at 5:11 PM, Dongxue Zhang <elta.era@gmail.com
> <mailto:elta.era@gmail.com>
>> > <mailto:elta.era@gmail.com <mailto:elta.era@gmail.com>>> wrote:
>> >>
>> >> Make target-openrisc running OK by add typename in
>> > openrisc_cpu_class_by_name().
>> >>
>> >> Signed-off-by: Dongxue Zhang <elta.era@gmail.com
> <mailto:elta.era@gmail.com>
>> > <mailto:elta.era@gmail.com <mailto:elta.era@gmail.com>>>
>> >> ---
>> >>  target-openrisc/cpu.c | 4 +++-
>> >>  1 file changed, 3 insertions(+), 1 deletion(-)
>> >>
>> >> diff --git a/target-openrisc/cpu.c b/target-openrisc/cpu.c
>> >> index fd90d37..d38c28b 100644
>> >> --- a/target-openrisc/cpu.c
>> >> +++ b/target-openrisc/cpu.c
>> >> @@ -96,12 +96,14 @@ static void openrisc_cpu_initfn(Object *obj)
>> >>  static ObjectClass *openrisc_cpu_class_by_name(const char *cpu_model)
>> >>  {
>> >>      ObjectClass *oc;
>> >> +    char *typename;
>> >>
>> >>      if (cpu_model == NULL) {
>> >>          return NULL;
>> >>      }
>> >>
>> >> -    oc = object_class_by_name(cpu_model);
>> >> +    typename = g_strdup_printf("%s-" TYPE_OPENRISC_CPU, cpu_model);
>> >> +    oc = object_class_by_name(typename);
>> >>      if (oc != NULL && (!object_class_dynamic_cast(oc,
>> > TYPE_OPENRISC_CPU) ||
>> >>                         object_class_is_abstract(oc))) {
>> >>          return NULL;
>> >
>> > Thanks for your fix, it looks and test good to me.
>>
>> Sorry for the breakage. Do you want to add a
>> Reviewed-by/Tested-by/Acked-by? I'd queue it for you then.

Thanks, queued on qom-cpu (with extended commit message):
https://github.com/afaerber/qemu-cpu/commits/qom-cpu

CC'ing qemu-stable. I also note that my mistake made it possible to use
-cpu or1200-or32-cpu or -cpu any-or32-cpu instead of -cpu or1200 or -cpu
any, which this patch reverts. I believe that to be intentional.
Otherwise we can extend the patch to allow both.

> Thank you very much! May you please tell me how can I make a
> Reviewed-by/Tested-by/Acked-by? I don't know too much about it.

You can see how it's used on the list almost every day. ;)

Reviewed-by: Your Name <your@email> means that you have reviewed the
patch to be sensible, functionally correct and containing no style issues.

Tested-by should be obvious. Based on your email I've added that one for
now.

Acked-by is short for acknowledged and, depending on interpretation,
means either that the change/concept looks okay or that it has been
reviewed and compile-tested. Anthony prefers to get Reviewed-bys.

>> If you could upload a Linux test image somewhere that may help avoid
>> breakages in the future.
> 
> I find some Linux test images at http://qemu-project.org/Testing . How
> can I upload one upon to there?

We can't. For one the Wiki has an upload file size limit, for another
hosting GPL'ed binaries requires to either host or supply on request the
corresponding source code (but IANAL).

Regards,
Andreas
Jia Liu July 8, 2013, 2:06 a.m. UTC | #6 
Hi Andreas,


On Mon, Jul 8, 2013 at 12:08 AM, Andreas Färber <afaerber@suse.de> wrote:
>
> Hi,
>
> Am 02.07.2013 14:30, schrieb Jia Liu:
> > On Tue, Jul 2, 2013 at 6:18 PM, Andreas Färber <afaerber@suse.de
> > <mailto:afaerber@suse.de>> wrote:
> >> Am 02.07.2013 11:29, schrieb Jia Liu:
> >> > On Tue, Jul 2, 2013 at 5:11 PM, Dongxue Zhang <elta.era@gmail.com
> > <mailto:elta.era@gmail.com>
> >> > <mailto:elta.era@gmail.com <mailto:elta.era@gmail.com>>> wrote:
> >> >>
> >> >> Make target-openrisc running OK by add typename in
> >> > openrisc_cpu_class_by_name().
> >> >>
> >> >> Signed-off-by: Dongxue Zhang <elta.era@gmail.com
> > <mailto:elta.era@gmail.com>
> >> > <mailto:elta.era@gmail.com <mailto:elta.era@gmail.com>>>
> >> >> ---
> >> >>  target-openrisc/cpu.c | 4 +++-
> >> >>  1 file changed, 3 insertions(+), 1 deletion(-)
> >> >>
> >> >> diff --git a/target-openrisc/cpu.c b/target-openrisc/cpu.c
> >> >> index fd90d37..d38c28b 100644
> >> >> --- a/target-openrisc/cpu.c
> >> >> +++ b/target-openrisc/cpu.c
> >> >> @@ -96,12 +96,14 @@ static void openrisc_cpu_initfn(Object *obj)
> >> >>  static ObjectClass *openrisc_cpu_class_by_name(const char
*cpu_model)
> >> >>  {
> >> >>      ObjectClass *oc;
> >> >> +    char *typename;
> >> >>
> >> >>      if (cpu_model == NULL) {
> >> >>          return NULL;
> >> >>      }
> >> >>
> >> >> -    oc = object_class_by_name(cpu_model);
> >> >> +    typename = g_strdup_printf("%s-" TYPE_OPENRISC_CPU, cpu_model);
> >> >> +    oc = object_class_by_name(typename);
> >> >>      if (oc != NULL && (!object_class_dynamic_cast(oc,
> >> > TYPE_OPENRISC_CPU) ||
> >> >>                         object_class_is_abstract(oc))) {
> >> >>          return NULL;
> >> >
> >> > Thanks for your fix, it looks and test good to me.
> >>
> >> Sorry for the breakage. Do you want to add a
> >> Reviewed-by/Tested-by/Acked-by? I'd queue it for you then.
>
> Thanks, queued on qom-cpu (with extended commit message):
> https://github.com/afaerber/qemu-cpu/commits/qom-cpu
>
> CC'ing qemu-stable. I also note that my mistake made it possible to use
> -cpu or1200-or32-cpu or -cpu any-or32-cpu instead of -cpu or1200 or -cpu
> any, which this patch reverts. I believe that to be intentional.
> Otherwise we can extend the patch to allow both.


Thanks, we will working on this.

>
>
> > Thank you very much! May you please tell me how can I make a
> > Reviewed-by/Tested-by/Acked-by? I don't know too much about it.
>
> You can see how it's used on the list almost every day. ;)
>
> Reviewed-by: Your Name <your@email> means that you have reviewed the
> patch to be sensible, functionally correct and containing no style issues.
>
> Tested-by should be obvious. Based on your email I've added that one for
> now.
>
> Acked-by is short for acknowledged and, depending on interpretation,
> means either that the change/concept looks okay or that it has been
> reviewed and compile-tested. Anthony prefers to get Reviewed-bys.
>
> >> If you could upload a Linux test image somewhere that may help avoid
> >> breakages in the future.
> >
> > I find some Linux test images at http://qemu-project.org/Testing . How
> > can I upload one upon to there?
>
> We can't. For one the Wiki has an upload file size limit, for another
> hosting GPL'ed binaries requires to either host or supply on request the
> corresponding source code (but IANAL).
>
> Regards,
> Andreas
>
> --
> SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany
> GF: Jeff Hawn, Jennifer Guild, Felix Imendörffer; HRB 16746 AG Nürnberg

Regards,
Jia
diff mbox

Patch

diff --git a/target-openrisc/cpu.c b/target-openrisc/cpu.c
index fd90d37..d38c28b 100644
--- a/target-openrisc/cpu.c
+++ b/target-openrisc/cpu.c
@@ -96,12 +96,14 @@  static void openrisc_cpu_initfn(Object *obj)
 static ObjectClass *openrisc_cpu_class_by_name(const char *cpu_model)
 {
     ObjectClass *oc;
+    char *typename;
 
     if (cpu_model == NULL) {
         return NULL;
     }
 
-    oc = object_class_by_name(cpu_model);
+    typename = g_strdup_printf("%s-" TYPE_OPENRISC_CPU, cpu_model);
+    oc = object_class_by_name(typename);
     if (oc != NULL && (!object_class_dynamic_cast(oc, TYPE_OPENRISC_CPU) ||
                        object_class_is_abstract(oc))) {
         return NULL;