[Qemu-devel] New documentation for the LatticeMicro32 target

[Qemu-devel] New documentation for the LatticeMicro32 target

[Programmingkid]

I just made a documentation page for the LatticeMicro32 target. I need to know its current status, how much of this system is implemented, what software runs on it. If anyone could supply more information that would be appreciated. Pictures of this target running would be great as well. Thanks.

Here is the page: http://wiki.qemu.org/Documentation/Platforms/LatticeMico32

Re: [Qemu-devel] New documentation for the LatticeMicro32 target

[Michael Walle]

Am 2017-07-09 17:39, schrieb Programmingkid:
> I just made a documentation page for the LatticeMicro32 target. I need
> to know its current status, how much of this system is implemented,
> what software runs on it. If anyone could supply more information that
> would be appreciated. Pictures of this target running would be great
> as well. Thanks.
> 
> Here is the page: 
> http://wiki.qemu.org/Documentation/Platforms/LatticeMico32

Hi,

thank you for your efforts. I have to admit that the current status is 
unknown.

I've just compiled the latest head and it looks like the BQL is missing 
for the interrupt helpers. I'll post a patch soon. With the patch, the 
basic stuff is working but there seems to be an issue with OpenGL.

You can read the quickstart instructions at:
http://milkymist.walle.cc/README.qemu

The flash image isn't working because it will boot right into the 
visualization mode, which is using the OpenGL stuff. The "-kernel 
flickernoise" method is working though. You can get the flickernoise 
binary at:
http://milkymist.walle.cc/updates/current/flickernoise

Besides the milkymist emulation, there is plain lm32 emulation based on 
the Lattice reference design which should be ok, because the tests are 
working (make -C tests/tcg/lm32 check)

-michael

Re: [Qemu-devel] New documentation for the LatticeMicro32 target

[Programmingkid]

> On Jul 17, 2017, at 1:40 AM, Michael Walle <michael@walle.cc> wrote:
> 
> Am 2017-07-09 17:39, schrieb Programmingkid:
>> I just made a documentation page for the LatticeMicro32 target. I need
>> to know its current status, how much of this system is implemented,
>> what software runs on it. If anyone could supply more information that
>> would be appreciated. Pictures of this target running would be great
>> as well. Thanks.
>> Here is the page: http://wiki.qemu.org/Documentation/Platforms/LatticeMico32
> 
> Hi,
> 
> thank you for your efforts. I have to admit that the current status is unknown.
> 
> I've just compiled the latest head and it looks like the BQL is missing for the interrupt helpers. I'll post a patch soon. With the patch, the basic stuff is working but there seems to be an issue with OpenGL.
> 
> You can read the quickstart instructions at:
> http://milkymist.walle.cc/README.qemu
> 
> The flash image isn't working because it will boot right into the visualization mode, which is using the OpenGL stuff. The "-kernel flickernoise" method is working though. You can get the flickernoise binary at:
> http://milkymist.walle.cc/updates/current/flickernoise
> 
> Besides the milkymist emulation, there is plain lm32 emulation based on the Lattice reference design which should be ok, because the tests are working (make -C tests/tcg/lm32 check)
> 
> -michael

Thank you for all this information. It has been added to the wiki page. 

Re: [Qemu-devel] New documentation for the LatticeMicro32 target

[Programmingkid]

> On Jul 17, 2017, at 10:56 AM, Programmingkid <programmingkidx@gmail.com> wrote:
> 
> 
>> On Jul 17, 2017, at 1:40 AM, Michael Walle <michael@walle.cc> wrote:
>> 
>> Am 2017-07-09 17:39, schrieb Programmingkid:
>>> I just made a documentation page for the LatticeMicro32 target. I need
>>> to know its current status, how much of this system is implemented,
>>> what software runs on it. If anyone could supply more information that
>>> would be appreciated. Pictures of this target running would be great
>>> as well. Thanks.
>>> Here is the page: http://wiki.qemu.org/Documentation/Platforms/LatticeMico32
>> 
>> Hi,
>> 
>> thank you for your efforts. I have to admit that the current status is unknown.
>> 
>> I've just compiled the latest head and it looks like the BQL is missing for the interrupt helpers. I'll post a patch soon. With the patch, the basic stuff is working but there seems to be an issue with OpenGL.
>> 
>> You can read the quickstart instructions at:
>> http://milkymist.walle.cc/README.qemu
>> 
>> The flash image isn't working because it will boot right into the visualization mode, which is using the OpenGL stuff. The "-kernel flickernoise" method is working though. You can get the flickernoise binary at:
>> http://milkymist.walle.cc/updates/current/flickernoise
>> 
>> Besides the milkymist emulation, there is plain lm32 emulation based on the Lattice reference design which should be ok, because the tests are working (make -C tests/tcg/lm32 check)
>> 
>> -michael
> 
> Thank you for all this information. It has been added to the wiki page. 
> 

I just tried qemu-system-lm32 and it crashed.

Command-line: qemu-system-lm32 -M milkymist -kernel flickernoise

Error message: audio: Failed to create voice `mm_ac97.in'
**
ERROR: /accel/tcg/tcg-all.c:42:tcg_handle_interrupt: assertion failed: (qemu_mutex_iothread_locked())
Abort trap: 6

Host: Mac OS 10.12.5

Re: [Qemu-devel] New documentation for the LatticeMicro32 target

[Michael Walle]

Am 2017-07-17 17:18, schrieb Programmingkid:
>> On Jul 17, 2017, at 10:56 AM, Programmingkid 
>> <programmingkidx@gmail.com> wrote:
>> 
>> 
>>> On Jul 17, 2017, at 1:40 AM, Michael Walle <michael@walle.cc> wrote:
>>> 
>>> Am 2017-07-09 17:39, schrieb Programmingkid:
>>>> I just made a documentation page for the LatticeMicro32 target. I 
>>>> need
>>>> to know its current status, how much of this system is implemented,
>>>> what software runs on it. If anyone could supply more information 
>>>> that
>>>> would be appreciated. Pictures of this target running would be great
>>>> as well. Thanks.
>>>> Here is the page: 
>>>> http://wiki.qemu.org/Documentation/Platforms/LatticeMico32
>>> 
>>> Hi,
>>> 
>>> thank you for your efforts. I have to admit that the current status 
>>> is unknown.
>>> 
>>> I've just compiled the latest head and it looks like the BQL is 
>>> missing for the interrupt helpers. I'll post a patch soon. With the 
>>> patch, the basic stuff is working but there seems to be an issue with 
>>> OpenGL.
>>> 
>>> You can read the quickstart instructions at:
>>> http://milkymist.walle.cc/README.qemu
>>> 
>>> The flash image isn't working because it will boot right into the 
>>> visualization mode, which is using the OpenGL stuff. The "-kernel 
>>> flickernoise" method is working though. You can get the flickernoise 
>>> binary at:
>>> http://milkymist.walle.cc/updates/current/flickernoise
>>> 
>>> Besides the milkymist emulation, there is plain lm32 emulation based 
>>> on the Lattice reference design which should be ok, because the tests 
>>> are working (make -C tests/tcg/lm32 check)
>>> 
>>> -michael
>> 
>> Thank you for all this information. It has been added to the wiki 
>> page.
>> 
> 
> I just tried qemu-system-lm32 and it crashed.
> 
> Command-line: qemu-system-lm32 -M milkymist -kernel flickernoise
> 
> Error message: audio: Failed to create voice `mm_ac97.in'
> **
> ERROR: /accel/tcg/tcg-all.c:42:tcg_handle_interrupt: assertion failed:
> (qemu_mutex_iothread_locked())
> Abort trap: 6
> 
> Host: Mac OS 10.12.5

As I said before, I guess the BQL is missing here. Try the following 
patch:

diff --git a/target/lm32/op_helper.c b/target/lm32/op_helper.c
index 2177c8ad12..93b8d09794 100644
--- a/target/lm32/op_helper.c
+++ b/target/lm32/op_helper.c
@@ -102,12 +102,16 @@ void HELPER(wcsr_dc)(CPULM32State *env, uint32_t 
dc)

  void HELPER(wcsr_im)(CPULM32State *env, uint32_t im)
  {
+    qemu_mutex_lock_iothread();
      lm32_pic_set_im(env->pic_state, im);
+    qemu_mutex_unlock_iothread();
  }

  void HELPER(wcsr_ip)(CPULM32State *env, uint32_t im)
  {
+    qemu_mutex_lock_iothread();
      lm32_pic_set_ip(env->pic_state, im);
+    qemu_mutex_unlock_iothread();
  }

  void HELPER(wcsr_jtx)(CPULM32State *env, uint32_t jtx)



-michael

Re: [Qemu-devel] New documentation for the LatticeMicro32 target

[Programmingkid]

> On Jul 18, 2017, at 7:44 AM, Michael Walle <michael@walle.cc> wrote:
> 
> Am 2017-07-17 17:18, schrieb Programmingkid:
>>> On Jul 17, 2017, at 10:56 AM, Programmingkid <programmingkidx@gmail.com> wrote:
>>>> On Jul 17, 2017, at 1:40 AM, Michael Walle <michael@walle.cc> wrote:
>>>> Am 2017-07-09 17:39, schrieb Programmingkid:
>>>>> I just made a documentation page for the LatticeMicro32 target. I need
>>>>> to know its current status, how much of this system is implemented,
>>>>> what software runs on it. If anyone could supply more information that
>>>>> would be appreciated. Pictures of this target running would be great
>>>>> as well. Thanks.
>>>>> Here is the page: http://wiki.qemu.org/Documentation/Platforms/LatticeMico32
>>>> Hi,
>>>> thank you for your efforts. I have to admit that the current status is unknown.
>>>> I've just compiled the latest head and it looks like the BQL is missing for the interrupt helpers. I'll post a patch soon. With the patch, the basic stuff is working but there seems to be an issue with OpenGL.
>>>> You can read the quickstart instructions at:
>>>> http://milkymist.walle.cc/README.qemu
>>>> The flash image isn't working because it will boot right into the visualization mode, which is using the OpenGL stuff. The "-kernel flickernoise" method is working though. You can get the flickernoise binary at:
>>>> http://milkymist.walle.cc/updates/current/flickernoise
>>>> Besides the milkymist emulation, there is plain lm32 emulation based on the Lattice reference design which should be ok, because the tests are working (make -C tests/tcg/lm32 check)
>>>> -michael
>>> Thank you for all this information. It has been added to the wiki page.
>> I just tried qemu-system-lm32 and it crashed.
>> Command-line: qemu-system-lm32 -M milkymist -kernel flickernoise
>> Error message: audio: Failed to create voice `mm_ac97.in'
>> **
>> ERROR: /accel/tcg/tcg-all.c:42:tcg_handle_interrupt: assertion failed:
>> (qemu_mutex_iothread_locked())
>> Abort trap: 6
>> Host: Mac OS 10.12.5
> 
> As I said before, I guess the BQL is missing here. Try the following patch:
> 
> diff --git a/target/lm32/op_helper.c b/target/lm32/op_helper.c
> index 2177c8ad12..93b8d09794 100644
> --- a/target/lm32/op_helper.c
> +++ b/target/lm32/op_helper.c
> @@ -102,12 +102,16 @@ void HELPER(wcsr_dc)(CPULM32State *env, uint32_t dc)
> 
> void HELPER(wcsr_im)(CPULM32State *env, uint32_t im)
> {
> +    qemu_mutex_lock_iothread();
>     lm32_pic_set_im(env->pic_state, im);
> +    qemu_mutex_unlock_iothread();
> }
> 
> void HELPER(wcsr_ip)(CPULM32State *env, uint32_t im)
> {
> +    qemu_mutex_lock_iothread();
>     lm32_pic_set_ip(env->pic_state, im);
> +    qemu_mutex_unlock_iothread();
> }
> 
> void HELPER(wcsr_jtx)(CPULM32State *env, uint32_t jtx)
> 
> 
> 
> -michael
> 

Thank you very much for the patch. It fixed the problem. Now I can take a picture of the screen and add it to the wiki page. 

Re: [Qemu-devel] New documentation for the LatticeMicro32 target

[Peter Maydell]

On 18 July 2017 at 12:44, Michael Walle <michael@walle.cc> wrote:
> Am 2017-07-17 17:18, schrieb Programmingkid:
>> I just tried qemu-system-lm32 and it crashed.
>>
>> Command-line: qemu-system-lm32 -M milkymist -kernel flickernoise
>>
>> Error message: audio: Failed to create voice `mm_ac97.in'
>> **
>> ERROR: /accel/tcg/tcg-all.c:42:tcg_handle_interrupt: assertion failed:
>> (qemu_mutex_iothread_locked())
>> Abort trap: 6
>>
>> Host: Mac OS 10.12.5
>
>
> As I said before, I guess the BQL is missing here. Try the following patch:
>
> diff --git a/target/lm32/op_helper.c b/target/lm32/op_helper.c
> index 2177c8ad12..93b8d09794 100644
> --- a/target/lm32/op_helper.c
> +++ b/target/lm32/op_helper.c
> @@ -102,12 +102,16 @@ void HELPER(wcsr_dc)(CPULM32State *env, uint32_t dc)
>
>  void HELPER(wcsr_im)(CPULM32State *env, uint32_t im)
>  {
> +    qemu_mutex_lock_iothread();
>      lm32_pic_set_im(env->pic_state, im);
> +    qemu_mutex_unlock_iothread();
>  }
>
>  void HELPER(wcsr_ip)(CPULM32State *env, uint32_t im)
>  {
> +    qemu_mutex_lock_iothread();
>      lm32_pic_set_ip(env->pic_state, im);
> +    qemu_mutex_unlock_iothread();
>  }
>
>  void HELPER(wcsr_jtx)(CPULM32State *env, uint32_t jtx)

I just fell over this, it looks like this fix never made it into master.

Michael, could you submit this as a full patch with a commit message
and signed-off-by, please? cc: qemu-stable@nongnu.org as well, since
it fixes a "breaks the board completely" bug.

thanks
-- PMM

[Qemu-devel] [PATCH] lm32: take BQL before writing IP/IM register

[Michael Walle]

Writing to these registers may raise an interrupt request. Actually,
this prevents the milkymist board from starting.

Cc: qemu-stable@nongnu.org
Signed-off-by: Michael Walle <michael@walle.cc>
---
 target/lm32/op_helper.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/target/lm32/op_helper.c b/target/lm32/op_helper.c
index 30f670eee8..a766a1ece4 100644
--- a/target/lm32/op_helper.c
+++ b/target/lm32/op_helper.c
@@ -102,12 +102,16 @@ void HELPER(wcsr_dc)(CPULM32State *env, uint32_t dc)
 
 void HELPER(wcsr_im)(CPULM32State *env, uint32_t im)
 {
+    qemu_mutex_lock_iothread();
     lm32_pic_set_im(env->pic_state, im);
+    qemu_mutex_unlock_iothread();
 }
 
 void HELPER(wcsr_ip)(CPULM32State *env, uint32_t im)
 {
+    qemu_mutex_lock_iothread();
     lm32_pic_set_ip(env->pic_state, im);
+    qemu_mutex_unlock_iothread();
 }
 
 void HELPER(wcsr_jtx)(CPULM32State *env, uint32_t jtx)
-- 
2.11.0

Re: [Qemu-devel] [PATCH] lm32: take BQL before writing IP/IM register

[Michael Walle]

Hi Peter,

do you apply this patch? Or do I have to send a pull request?

-michael

Am 2018-01-09 18:01, schrieb Michael Walle:
> Writing to these registers may raise an interrupt request. Actually,
> this prevents the milkymist board from starting.
> 
> Cc: qemu-stable@nongnu.org
> Signed-off-by: Michael Walle <michael@walle.cc>
> ---
>  target/lm32/op_helper.c | 4 ++++
>  1 file changed, 4 insertions(+)
> 
> diff --git a/target/lm32/op_helper.c b/target/lm32/op_helper.c
> index 30f670eee8..a766a1ece4 100644
> --- a/target/lm32/op_helper.c
> +++ b/target/lm32/op_helper.c
> @@ -102,12 +102,16 @@ void HELPER(wcsr_dc)(CPULM32State *env, uint32_t 
> dc)
> 
>  void HELPER(wcsr_im)(CPULM32State *env, uint32_t im)
>  {
> +    qemu_mutex_lock_iothread();
>      lm32_pic_set_im(env->pic_state, im);
> +    qemu_mutex_unlock_iothread();
>  }
> 
>  void HELPER(wcsr_ip)(CPULM32State *env, uint32_t im)
>  {
> +    qemu_mutex_lock_iothread();
>      lm32_pic_set_ip(env->pic_state, im);
> +    qemu_mutex_unlock_iothread();
>  }
> 
>  void HELPER(wcsr_jtx)(CPULM32State *env, uint32_t jtx)

Re: [Qemu-devel] [PATCH] lm32: take BQL before writing IP/IM register

[Philippe Mathieu-Daudé]

On 02/01/2018 06:09 AM, Michael Walle wrote:
> 
> Hi Peter,
> 
> do you apply this patch? Or do I have to send a pull request?

Cc'ing Paolo.

> 
> -michael
> 
> Am 2018-01-09 18:01, schrieb Michael Walle:
>> Writing to these registers may raise an interrupt request. Actually,
>> this prevents the milkymist board from starting.
>>
>> Cc: qemu-stable@nongnu.org
>> Signed-off-by: Michael Walle <michael@walle.cc>

Tested-by: Philippe Mathieu-Daudé <f4bug@amsat.org>

>> ---
>>  target/lm32/op_helper.c | 4 ++++
>>  1 file changed, 4 insertions(+)
>>
>> diff --git a/target/lm32/op_helper.c b/target/lm32/op_helper.c
>> index 30f670eee8..a766a1ece4 100644
>> --- a/target/lm32/op_helper.c
>> +++ b/target/lm32/op_helper.c
>> @@ -102,12 +102,16 @@ void HELPER(wcsr_dc)(CPULM32State *env, uint32_t
>> dc)
>>
>>  void HELPER(wcsr_im)(CPULM32State *env, uint32_t im)
>>  {
>> +    qemu_mutex_lock_iothread();
>>      lm32_pic_set_im(env->pic_state, im);
>> +    qemu_mutex_unlock_iothread();
>>  }
>>
>>  void HELPER(wcsr_ip)(CPULM32State *env, uint32_t im)
>>  {
>> +    qemu_mutex_lock_iothread();
>>      lm32_pic_set_ip(env->pic_state, im);
>> +    qemu_mutex_unlock_iothread();
>>  }
>>
>>  void HELPER(wcsr_jtx)(CPULM32State *env, uint32_t jtx)
> 

Re: [Qemu-devel] [PATCH] lm32: take BQL before writing IP/IM register

[Paolo Bonzini]

On 09/05/2018 21:45, Philippe Mathieu-Daudé wrote:
> On 02/01/2018 06:09 AM, Michael Walle wrote:
>>
>> Hi Peter,
>>
>> do you apply this patch? Or do I have to send a pull request?
> 
> Cc'ing Paolo.

Please send a pull request.

Paolo

>>
>> -michael
>>
>> Am 2018-01-09 18:01, schrieb Michael Walle:
>>> Writing to these registers may raise an interrupt request. Actually,
>>> this prevents the milkymist board from starting.
>>>
>>> Cc: qemu-stable@nongnu.org
>>> Signed-off-by: Michael Walle <michael@walle.cc>
> 
> Tested-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
> 
>>> ---
>>>  target/lm32/op_helper.c | 4 ++++
>>>  1 file changed, 4 insertions(+)
>>>
>>> diff --git a/target/lm32/op_helper.c b/target/lm32/op_helper.c
>>> index 30f670eee8..a766a1ece4 100644
>>> --- a/target/lm32/op_helper.c
>>> +++ b/target/lm32/op_helper.c
>>> @@ -102,12 +102,16 @@ void HELPER(wcsr_dc)(CPULM32State *env, uint32_t
>>> dc)
>>>
>>>  void HELPER(wcsr_im)(CPULM32State *env, uint32_t im)
>>>  {
>>> +    qemu_mutex_lock_iothread();
>>>      lm32_pic_set_im(env->pic_state, im);
>>> +    qemu_mutex_unlock_iothread();
>>>  }
>>>
>>>  void HELPER(wcsr_ip)(CPULM32State *env, uint32_t im)
>>>  {
>>> +    qemu_mutex_lock_iothread();
>>>      lm32_pic_set_ip(env->pic_state, im);
>>> +    qemu_mutex_unlock_iothread();
>>>  }
>>>
>>>  void HELPER(wcsr_jtx)(CPULM32State *env, uint32_t jtx)
>>

Re: [Qemu-devel] [PATCH] lm32: take BQL before writing IP/IM register

[Alex Bennée]

Michael Walle <michael@walle.cc> writes:

> Writing to these registers may raise an interrupt request. Actually,
> this prevents the milkymist board from starting.
>
> Cc: qemu-stable@nongnu.org
> Signed-off-by: Michael Walle <michael@walle.cc>

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

> ---
>  target/lm32/op_helper.c | 4 ++++
>  1 file changed, 4 insertions(+)
>
> diff --git a/target/lm32/op_helper.c b/target/lm32/op_helper.c
> index 30f670eee8..a766a1ece4 100644
> --- a/target/lm32/op_helper.c
> +++ b/target/lm32/op_helper.c
> @@ -102,12 +102,16 @@ void HELPER(wcsr_dc)(CPULM32State *env, uint32_t dc)
>
>  void HELPER(wcsr_im)(CPULM32State *env, uint32_t im)
>  {
> +    qemu_mutex_lock_iothread();
>      lm32_pic_set_im(env->pic_state, im);
> +    qemu_mutex_unlock_iothread();
>  }
>
>  void HELPER(wcsr_ip)(CPULM32State *env, uint32_t im)
>  {
> +    qemu_mutex_lock_iothread();
>      lm32_pic_set_ip(env->pic_state, im);
> +    qemu_mutex_unlock_iothread();
>  }
>
>  void HELPER(wcsr_jtx)(CPULM32State *env, uint32_t jtx)


--
Alex Bennée

Re: [Qemu-devel] New documentation for the LatticeMicro32 target

[Michael Walle]

Am 2018-01-09 12:58, schrieb Peter Maydell:
> I just fell over this, it looks like this fix never made it into 
> master.
> 
> Michael, could you submit this as a full patch with a commit message
> and signed-off-by, please? cc: qemu-stable@nongnu.org as well, since
> it fixes a "breaks the board completely" bug.

oops, sorry about that :(

-michael