mbox

[PULL,0/2] OpenRISC patch queue for 2.3

Message ID 1422929995-33807-1-git-send-email-proljc@gmail.com
State New
Headers show

Pull-request

git://github.com/J-Liu/qemu.git or32

Message

Jia Liu Feb. 3, 2015, 2:19 a.m. UTC
Hi Anthony,

This is my OpenRISC patch queue for 2.3, it have been well tested, please pull.

Thanks to Christian and Sebastian, they made the LD/ST updated.


Regards,
Jia



The following changes since commit 16017c48547960539fcadb1f91d252124f442482:

  softfloat: Clarify license status (2015-01-29 16:45:45 +0000)

are available in the git repository at:

  git://github.com/J-Liu/qemu.git or32

for you to fetch changes up to bcf7d6e965d5323cae32f18c5e6c8af173c40d41:

  target-openrisc: Add l.lwa/l.swa support (2015-02-03 09:55:41 +0800)

----------------------------------------------------------------
Christian Svensson (1):
      target-openrisc: Add l.lwa/l.swa support

Sebastian Macke (1):
      target-openrisc: Separate of load/store instructions

 target-openrisc/cpu.h       |   3 +
 target-openrisc/interrupt.c |   3 +
 target-openrisc/translate.c | 188 ++++++++++++++++++++++++++++++++++----------
 3 files changed, 153 insertions(+), 41 deletions(-)

Comments

Peter Maydell Feb. 3, 2015, 10:40 a.m. UTC | #1
On 3 February 2015 at 02:19, Jia Liu <proljc@gmail.com> wrote:
> Hi Anthony,
>
> This is my OpenRISC patch queue for 2.3, it have been well tested, please pull.

...it can't have been very well tested, because it doesn't
compile:

target-openrisc/interrupt.c: In function ‘openrisc_cpu_do_interrupt’:
target-openrisc/interrupt.c:58:8: error: ‘CPUOpenRISCState’ has no
member named ‘cpu_lock_addr’

thanks
-- PMM
Sebastian Macke Feb. 3, 2015, 1:04 p.m. UTC | #2
Hi Peter,

unfortunately you are right.

The correct line is this:

      /* invalidate lock */
-    env->cpu_lock_addr = -1;
+    env->lock_addr = -1;

I am sorry. It was most likely the last line which I added. But I 
forgot, that I disabled the system emulation already.
Therefore my make process didn't complain.
Should I send an updated patch, or can you do a hot-fix?

Sebastian


On 2/3/2015 11:40 AM, Peter Maydell wrote:
> On 3 February 2015 at 02:19, Jia Liu <proljc@gmail.com> wrote:
>> Hi Anthony,
>>
>> This is my OpenRISC patch queue for 2.3, it have been well tested, please pull.
> ...it can't have been very well tested, because it doesn't
> compile:
>
> target-openrisc/interrupt.c: In function ‘openrisc_cpu_do_interrupt’:
> target-openrisc/interrupt.c:58:8: error: ‘CPUOpenRISCState’ has no
> member named ‘cpu_lock_addr’
>
> thanks
> -- PMM
Peter Maydell Feb. 3, 2015, 1:07 p.m. UTC | #3
On 3 February 2015 at 13:04, Sebastian Macke <sebastian@macke.de> wrote:
> Hi Peter,
>
> unfortunately you are right.
>
> The correct line is this:
>
>      /* invalidate lock */
> -    env->cpu_lock_addr = -1;
> +    env->lock_addr = -1;
>
> I am sorry. It was most likely the last line which I added. But I forgot,
> that I disabled the system emulation already.
> Therefore my make process didn't complain.
> Should I send an updated patch, or can you do a hot-fix?

You should send an updated patch, and then Jia needs to re-test
and send a new pull request.

Somebody ought to be testing these instructions in system
emulation mode as well as linux-user...

thanks
-- PMM