* [PATCH v3] vga: don't abort when adding a duplicate isa-vga device
@ 2021-08-16 13:55 Jose R. Ziviani
2021-08-17 7:25 ` Thomas Huth
2022-03-10 17:06 ` Thomas Huth
0 siblings, 2 replies; 6+ messages in thread
From: Jose R. Ziviani @ 2021-08-16 13:55 UTC (permalink / raw)
To: qemu-devel; +Cc: thuth, philmd, kraxel, Jose R. Ziviani
If users try to add an isa-vga device that was already registered,
still in command line, qemu will crash:
$ qemu-system-mips64el -M pica61 -device isa-vga
RAMBlock "vga.vram" already registered, abort!
Aborted (core dumped)
That particular board registers the device automaticaly, so it's
not obvious that a VGA device already exists. This patch changes
this behavior by displaying a message and exiting without crashing.
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/44
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Jose R. Ziviani <jziviani@suse.de>
---
v2 to v3: Improved error message
v1 to v2: Use error_setg instead of error_report
hw/display/vga-isa.c | 10 ++++++++++
1 file changed, 10 insertions(+)
diff --git a/hw/display/vga-isa.c b/hw/display/vga-isa.c
index 90851e730b..30d55b41c3 100644
--- a/hw/display/vga-isa.c
+++ b/hw/display/vga-isa.c
@@ -33,6 +33,7 @@
#include "hw/loader.h"
#include "hw/qdev-properties.h"
#include "qom/object.h"
+#include "qapi/error.h"
#define TYPE_ISA_VGA "isa-vga"
OBJECT_DECLARE_SIMPLE_TYPE(ISAVGAState, ISA_VGA)
@@ -61,6 +62,15 @@ static void vga_isa_realizefn(DeviceState *dev, Error **errp)
MemoryRegion *vga_io_memory;
const MemoryRegionPortio *vga_ports, *vbe_ports;
+ /*
+ * make sure this device is not being added twice, if so
+ * exit without crashing qemu
+ */
+ if (qemu_ram_block_by_name("vga.vram")) {
+ error_setg(errp, "'isa-vga' device already registered");
+ return;
+ }
+
s->global_vmstate = true;
vga_common_init(s, OBJECT(dev));
s->legacy_address_space = isa_address_space(isadev);
--
2.32.0
^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [PATCH v3] vga: don't abort when adding a duplicate isa-vga device
2021-08-16 13:55 [PATCH v3] vga: don't abort when adding a duplicate isa-vga device Jose R. Ziviani
@ 2021-08-17 7:25 ` Thomas Huth
2021-08-17 7:36 ` Mark Cave-Ayland
2022-03-10 17:06 ` Thomas Huth
1 sibling, 1 reply; 6+ messages in thread
From: Thomas Huth @ 2021-08-17 7:25 UTC (permalink / raw)
To: Jose R. Ziviani, qemu-devel; +Cc: QEMU Trivial, philmd, kraxel
On 16/08/2021 15.55, Jose R. Ziviani wrote:
> If users try to add an isa-vga device that was already registered,
> still in command line, qemu will crash:
>
> $ qemu-system-mips64el -M pica61 -device isa-vga
> RAMBlock "vga.vram" already registered, abort!
> Aborted (core dumped)
>
> That particular board registers the device automaticaly, so it's
> not obvious that a VGA device already exists. This patch changes
> this behavior by displaying a message and exiting without crashing.
>
> Resolves: https://gitlab.com/qemu-project/qemu/-/issues/44
> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
> Signed-off-by: Jose R. Ziviani <jziviani@suse.de>
> ---
> v2 to v3: Improved error message
> v1 to v2: Use error_setg instead of error_report
>
> hw/display/vga-isa.c | 10 ++++++++++
> 1 file changed, 10 insertions(+)
>
> diff --git a/hw/display/vga-isa.c b/hw/display/vga-isa.c
> index 90851e730b..30d55b41c3 100644
> --- a/hw/display/vga-isa.c
> +++ b/hw/display/vga-isa.c
> @@ -33,6 +33,7 @@
> #include "hw/loader.h"
> #include "hw/qdev-properties.h"
> #include "qom/object.h"
> +#include "qapi/error.h"
>
> #define TYPE_ISA_VGA "isa-vga"
> OBJECT_DECLARE_SIMPLE_TYPE(ISAVGAState, ISA_VGA)
> @@ -61,6 +62,15 @@ static void vga_isa_realizefn(DeviceState *dev, Error **errp)
> MemoryRegion *vga_io_memory;
> const MemoryRegionPortio *vga_ports, *vbe_ports;
>
> + /*
> + * make sure this device is not being added twice, if so
> + * exit without crashing qemu
> + */
> + if (qemu_ram_block_by_name("vga.vram")) {
> + error_setg(errp, "'isa-vga' device already registered");
> + return;
> + }
> +
> s->global_vmstate = true;
> vga_common_init(s, OBJECT(dev));
> s->legacy_address_space = isa_address_space(isadev);
>
Reviewed-by: Thomas Huth <thuth@redhat.com>
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH v3] vga: don't abort when adding a duplicate isa-vga device
2021-08-17 7:25 ` Thomas Huth
@ 2021-08-17 7:36 ` Mark Cave-Ayland
2021-08-17 8:07 ` Philippe Mathieu-Daudé
0 siblings, 1 reply; 6+ messages in thread
From: Mark Cave-Ayland @ 2021-08-17 7:36 UTC (permalink / raw)
To: Thomas Huth, Jose R. Ziviani, qemu-devel; +Cc: QEMU Trivial, philmd, kraxel
On 17/08/2021 08:25, Thomas Huth wrote:
> On 16/08/2021 15.55, Jose R. Ziviani wrote:
>> If users try to add an isa-vga device that was already registered,
>> still in command line, qemu will crash:
>>
>> $ qemu-system-mips64el -M pica61 -device isa-vga
>> RAMBlock "vga.vram" already registered, abort!
>> Aborted (core dumped)
>>
>> That particular board registers the device automaticaly, so it's
>> not obvious that a VGA device already exists. This patch changes
>> this behavior by displaying a message and exiting without crashing.
>>
>> Resolves: https://gitlab.com/qemu-project/qemu/-/issues/44
>> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
>> Signed-off-by: Jose R. Ziviani <jziviani@suse.de>
>> ---
>> v2 to v3: Improved error message
>> v1 to v2: Use error_setg instead of error_report
>>
>> hw/display/vga-isa.c | 10 ++++++++++
>> 1 file changed, 10 insertions(+)
>>
>> diff --git a/hw/display/vga-isa.c b/hw/display/vga-isa.c
>> index 90851e730b..30d55b41c3 100644
>> --- a/hw/display/vga-isa.c
>> +++ b/hw/display/vga-isa.c
>> @@ -33,6 +33,7 @@
>> #include "hw/loader.h"
>> #include "hw/qdev-properties.h"
>> #include "qom/object.h"
>> +#include "qapi/error.h"
>> #define TYPE_ISA_VGA "isa-vga"
>> OBJECT_DECLARE_SIMPLE_TYPE(ISAVGAState, ISA_VGA)
>> @@ -61,6 +62,15 @@ static void vga_isa_realizefn(DeviceState *dev, Error **errp)
>> MemoryRegion *vga_io_memory;
>> const MemoryRegionPortio *vga_ports, *vbe_ports;
>> + /*
>> + * make sure this device is not being added twice, if so
>> + * exit without crashing qemu
>> + */
>> + if (qemu_ram_block_by_name("vga.vram")) {
>> + error_setg(errp, "'isa-vga' device already registered");
>> + return;
>> + }
>> +
>> s->global_vmstate = true;
>> vga_common_init(s, OBJECT(dev));
>> s->legacy_address_space = isa_address_space(isadev);
>>
>
> Reviewed-by: Thomas Huth <thuth@redhat.com>
Instead of checking for the presence of the vga.vram block, would it be better to use
the standard object_resolve_path_type() method to check for the presence of the
existing isa-vga device instead? See
https://lists.gnu.org/archive/html/qemu-devel/2021-07/msg00717.html for how this was
done for virgl.
ATB,
Mark.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH v3] vga: don't abort when adding a duplicate isa-vga device
2021-08-17 7:36 ` Mark Cave-Ayland
@ 2021-08-17 8:07 ` Philippe Mathieu-Daudé
2021-08-17 18:51 ` Jose R. Ziviani
0 siblings, 1 reply; 6+ messages in thread
From: Philippe Mathieu-Daudé @ 2021-08-17 8:07 UTC (permalink / raw)
To: Mark Cave-Ayland, Thomas Huth, Jose R. Ziviani, qemu-devel
Cc: QEMU Trivial, kraxel
On 8/17/21 9:36 AM, Mark Cave-Ayland wrote:
> On 17/08/2021 08:25, Thomas Huth wrote:
>
>> On 16/08/2021 15.55, Jose R. Ziviani wrote:
>>> If users try to add an isa-vga device that was already registered,
>>> still in command line, qemu will crash:
>>>
>>> $ qemu-system-mips64el -M pica61 -device isa-vga
>>> RAMBlock "vga.vram" already registered, abort!
>>> Aborted (core dumped)
>>>
>>> That particular board registers the device automaticaly, so it's
>>> not obvious that a VGA device already exists. This patch changes
>>> this behavior by displaying a message and exiting without crashing.
>>>
>>> Resolves: https://gitlab.com/qemu-project/qemu/-/issues/44
>>> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
>>> Signed-off-by: Jose R. Ziviani <jziviani@suse.de>
>>> ---
>>> v2 to v3: Improved error message
>>> v1 to v2: Use error_setg instead of error_report
>>>
>>> hw/display/vga-isa.c | 10 ++++++++++
>>> 1 file changed, 10 insertions(+)
>>>
>>> diff --git a/hw/display/vga-isa.c b/hw/display/vga-isa.c
>>> index 90851e730b..30d55b41c3 100644
>>> --- a/hw/display/vga-isa.c
>>> +++ b/hw/display/vga-isa.c
>>> @@ -33,6 +33,7 @@
>>> #include "hw/loader.h"
>>> #include "hw/qdev-properties.h"
>>> #include "qom/object.h"
>>> +#include "qapi/error.h"
>>> #define TYPE_ISA_VGA "isa-vga"
>>> OBJECT_DECLARE_SIMPLE_TYPE(ISAVGAState, ISA_VGA)
>>> @@ -61,6 +62,15 @@ static void vga_isa_realizefn(DeviceState *dev,
>>> Error **errp)
>>> MemoryRegion *vga_io_memory;
>>> const MemoryRegionPortio *vga_ports, *vbe_ports;
>>> + /*
>>> + * make sure this device is not being added twice, if so
>>> + * exit without crashing qemu
>>> + */
>>> + if (qemu_ram_block_by_name("vga.vram")) {
>>> + error_setg(errp, "'isa-vga' device already registered");
>>> + return;
>>> + }
>>> +
>>> s->global_vmstate = true;
>>> vga_common_init(s, OBJECT(dev));
>>> s->legacy_address_space = isa_address_space(isadev);
>>>
>>
>> Reviewed-by: Thomas Huth <thuth@redhat.com>
>
> Instead of checking for the presence of the vga.vram block, would it be
> better to use the standard object_resolve_path_type() method to check
> for the presence of the existing isa-vga device instead? See
> https://lists.gnu.org/archive/html/qemu-devel/2021-07/msg00717.html for
> how this was done for virgl.
I remembered there was a nicer way but couldn't find it.
If this patch were for 6.1, it was good enough. Now it
will be merged in 6.2, I prefer Mark's suggestion.
Jose, do you mind a v4?
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH v3] vga: don't abort when adding a duplicate isa-vga device
2021-08-17 8:07 ` Philippe Mathieu-Daudé
@ 2021-08-17 18:51 ` Jose R. Ziviani
0 siblings, 0 replies; 6+ messages in thread
From: Jose R. Ziviani @ 2021-08-17 18:51 UTC (permalink / raw)
To: Philippe Mathieu-Daudé
Cc: QEMU Trivial, Thomas Huth, Mark Cave-Ayland, qemu-devel, kraxel
[-- Attachment #1: Type: text/plain, Size: 3007 bytes --]
On Tue, Aug 17, 2021 at 10:07:55AM +0200, Philippe Mathieu-Daudé wrote:
> On 8/17/21 9:36 AM, Mark Cave-Ayland wrote:
> > On 17/08/2021 08:25, Thomas Huth wrote:
> >
> >> On 16/08/2021 15.55, Jose R. Ziviani wrote:
> >>> If users try to add an isa-vga device that was already registered,
> >>> still in command line, qemu will crash:
> >>>
> >>> $ qemu-system-mips64el -M pica61 -device isa-vga
> >>> RAMBlock "vga.vram" already registered, abort!
> >>> Aborted (core dumped)
> >>>
> >>> That particular board registers the device automaticaly, so it's
> >>> not obvious that a VGA device already exists. This patch changes
> >>> this behavior by displaying a message and exiting without crashing.
> >>>
> >>> Resolves: https://gitlab.com/qemu-project/qemu/-/issues/44
> >>> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
> >>> Signed-off-by: Jose R. Ziviani <jziviani@suse.de>
> >>> ---
> >>> v2 to v3: Improved error message
> >>> v1 to v2: Use error_setg instead of error_report
> >>>
> >>> hw/display/vga-isa.c | 10 ++++++++++
> >>> 1 file changed, 10 insertions(+)
> >>>
> >>> diff --git a/hw/display/vga-isa.c b/hw/display/vga-isa.c
> >>> index 90851e730b..30d55b41c3 100644
> >>> --- a/hw/display/vga-isa.c
> >>> +++ b/hw/display/vga-isa.c
> >>> @@ -33,6 +33,7 @@
> >>> #include "hw/loader.h"
> >>> #include "hw/qdev-properties.h"
> >>> #include "qom/object.h"
> >>> +#include "qapi/error.h"
> >>> #define TYPE_ISA_VGA "isa-vga"
> >>> OBJECT_DECLARE_SIMPLE_TYPE(ISAVGAState, ISA_VGA)
> >>> @@ -61,6 +62,15 @@ static void vga_isa_realizefn(DeviceState *dev,
> >>> Error **errp)
> >>> MemoryRegion *vga_io_memory;
> >>> const MemoryRegionPortio *vga_ports, *vbe_ports;
> >>> + /*
> >>> + * make sure this device is not being added twice, if so
> >>> + * exit without crashing qemu
> >>> + */
> >>> + if (qemu_ram_block_by_name("vga.vram")) {
> >>> + error_setg(errp, "'isa-vga' device already registered");
> >>> + return;
> >>> + }
> >>> +
> >>> s->global_vmstate = true;
> >>> vga_common_init(s, OBJECT(dev));
> >>> s->legacy_address_space = isa_address_space(isadev);
> >>>
> >>
> >> Reviewed-by: Thomas Huth <thuth@redhat.com>
> >
> > Instead of checking for the presence of the vga.vram block, would it be
> > better to use the standard object_resolve_path_type() method to check
> > for the presence of the existing isa-vga device instead? See
> > https://lists.gnu.org/archive/html/qemu-devel/2021-07/msg00717.html for
> > how this was done for virgl.
>
> I remembered there was a nicer way but couldn't find it.
> If this patch were for 6.1, it was good enough. Now it
> will be merged in 6.2, I prefer Mark's suggestion.
> Jose, do you mind a v4?
>
Hello people! Thanks for reviewing it.
Sure, I'll send a v4. It's not for 6.1 anyway.
Thank you very much
[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 833 bytes --]
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH v3] vga: don't abort when adding a duplicate isa-vga device
2021-08-16 13:55 [PATCH v3] vga: don't abort when adding a duplicate isa-vga device Jose R. Ziviani
2021-08-17 7:25 ` Thomas Huth
@ 2022-03-10 17:06 ` Thomas Huth
1 sibling, 0 replies; 6+ messages in thread
From: Thomas Huth @ 2022-03-10 17:06 UTC (permalink / raw)
To: qemu-devel, kraxel, Philippe Mathieu-Daudé; +Cc: Jose R. Ziviani
On 16/08/2021 15.55, Jose R. Ziviani wrote:
> If users try to add an isa-vga device that was already registered,
> still in command line, qemu will crash:
>
> $ qemu-system-mips64el -M pica61 -device isa-vga
> RAMBlock "vga.vram" already registered, abort!
> Aborted (core dumped)
>
> That particular board registers the device automaticaly, so it's
> not obvious that a VGA device already exists. This patch changes
> this behavior by displaying a message and exiting without crashing.
>
> Resolves: https://gitlab.com/qemu-project/qemu/-/issues/44
> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
> Signed-off-by: Jose R. Ziviani <jziviani@suse.de>
> ---
> v2 to v3: Improved error message
> v1 to v2: Use error_setg instead of error_report
>
> hw/display/vga-isa.c | 10 ++++++++++
> 1 file changed, 10 insertions(+)
>
> diff --git a/hw/display/vga-isa.c b/hw/display/vga-isa.c
> index 90851e730b..30d55b41c3 100644
> --- a/hw/display/vga-isa.c
> +++ b/hw/display/vga-isa.c
> @@ -33,6 +33,7 @@
> #include "hw/loader.h"
> #include "hw/qdev-properties.h"
> #include "qom/object.h"
> +#include "qapi/error.h"
>
> #define TYPE_ISA_VGA "isa-vga"
> OBJECT_DECLARE_SIMPLE_TYPE(ISAVGAState, ISA_VGA)
> @@ -61,6 +62,15 @@ static void vga_isa_realizefn(DeviceState *dev, Error **errp)
> MemoryRegion *vga_io_memory;
> const MemoryRegionPortio *vga_ports, *vbe_ports;
>
> + /*
> + * make sure this device is not being added twice, if so
> + * exit without crashing qemu
> + */
> + if (qemu_ram_block_by_name("vga.vram")) {
> + error_setg(errp, "'isa-vga' device already registered");
> + return;
> + }
> +
> s->global_vmstate = true;
> vga_common_init(s, OBJECT(dev));
> s->legacy_address_space = isa_address_space(isadev);
v4 of this patch had other issues ... Philippe's rework of the vga-mmio code
also did not fix the crash ... shall we just go with this v3 of the patch
again to fix the issue?
Thomas
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2022-03-10 17:08 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-08-16 13:55 [PATCH v3] vga: don't abort when adding a duplicate isa-vga device Jose R. Ziviani
2021-08-17 7:25 ` Thomas Huth
2021-08-17 7:36 ` Mark Cave-Ayland
2021-08-17 8:07 ` Philippe Mathieu-Daudé
2021-08-17 18:51 ` Jose R. Ziviani
2022-03-10 17:06 ` Thomas Huth
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).