On Thu, Jul 06, 2017 at 02:55:02PM +0200, John Paul Adrian Glaubitz wrote:
> This fixes the segfaults for me with newer chroots.
Ok, that was too fast. While some segfaults are now gone, there are
still some other commands segfaulting. Again, reverting patch 5
and Laurent's patch makes these
Le 06/07/2017 à 02:23, Richard Henderson a écrit :
> kernel also checks PC < gUSA region end point,
> try this:
>
> diff --git a/linux-user/signal.c b/linux-user/signal.c
> index 1e716a9..4e1e4f0 100644
> --- a/linux-user/signal.c
> +++ b/linux-user/signal.c
> @@ -3477,7 +3477,8 @@ static
Le 06/07/2017 à 02:23, Richard Henderson a écrit :
> We translate gUSA regions atomically in a parallel context.
> But in a serial context a gUSA region may be interrupted.
> In that case, restart the region as the kernel would.
>
> Signed-off-by: Richard Henderson
> ---
>
On Thu, Jul 06, 2017 at 03:09:59AM +0200, Laurent Vivier wrote:
> Reviewed-by: Laurent Vivier
I have identified this patch as reponsible for the
segfaults. Reverting this patch fixes the issue. The crashes are
random. Sometimes it crashes directly when entering the chroot,
Le 06/07/2017 à 11:13, John Paul Adrian Glaubitz a écrit :
> On Thu, Jul 06, 2017 at 10:35:15AM +0200, Laurent Vivier wrote:
>> I think it would be interesting if you generate a qemu binary with it
>> and use it in your buildds to test it.
>>
>> I guess the series can be pulled directly from
On Thu, Jul 06, 2017 at 10:35:15AM +0200, Laurent Vivier wrote:
> I think it would be interesting if you generate a qemu binary with it
> and use it in your buildds to test it.
>
> I guess the series can be pulled directly from Richard's branch:
>
> git://github.com/rth7680/qemu.git tgt-sh4
On Thu, Jul 06, 2017 at 10:35:15AM +0200, Laurent Vivier wrote:
> This is a patch series proposed by Richard, it is not included in qemu
> at the moment:
>
> http://lists.nongnu.org/archive/html/qemu-devel/2017-07/msg01196.html
Aye, awesome!
> I think it would be interesting if you generate a
Le 06/07/2017 à 10:10, John Paul Adrian Glaubitz a écrit :
> Hi!
>
> Wow, great to see that there is actually now support for handling gUSA
> in qemu-user on sh4. I wonder whether this will help resolve the
> lock-up issues on qemu-sh4-user when building Debian packages for
> sh4.
>
> The
Hi!
Wow, great to see that there is actually now support for handling gUSA
in qemu-user on sh4. I wonder whether this will help resolve the
lock-up issues on qemu-sh4-user when building Debian packages for
sh4.
The lockups occur fairly often when any process uses
multi-threading. Would this help
Le 06/07/2017 à 02:23, Richard Henderson a écrit :
> We translate gUSA regions atomically in a parallel context.
> But in a serial context a gUSA region may be interrupted.
> In that case, restart the region as the kernel would.
>
> Signed-off-by: Richard Henderson
> ---
>
We translate gUSA regions atomically in a parallel context.
But in a serial context a gUSA region may be interrupted.
In that case, restart the region as the kernel would.
Signed-off-by: Richard Henderson
---
linux-user/signal.c | 21 +
1 file changed, 21
11 matches
Mail list logo