Re: [PATCH v3] xen/arm: Convert runstate address during hypcall

2020-08-14 Thread Bertrand Marquis
> On 1 Aug 2020, at 00:03, Stefano Stabellini wrote: > > On Fri, 31 Jul 2020, Bertrand Marquis wrote: >> Sorry missed some points in my previous answer. >> >>> On 30 Jul 2020, at 22:50, Julien Grall wrote: >>> >>> Hi Bertrand, >>> >>> To avoid extra work on your side, I would recommend to

Re: [PATCH v3] xen/arm: Convert runstate address during hypcall

2020-08-14 Thread Bertrand Marquis
> On 13 Aug 2020, at 18:28, Julien Grall wrote: > > Hi Bertrand, > > On 31/07/2020 14:16, Bertrand Marquis wrote: >>> On 30 Jul 2020, at 22:50, Julien Grall wrote: >>> On 30/07/2020 11:24, Bertrand Marquis wrote: At the moment on Arm, a Linux guest running with KTPI enabled will ca

Re: [PATCH v3] xen/arm: Convert runstate address during hypcall

2020-08-14 Thread Bertrand Marquis
> On 13 Aug 2020, at 18:35, Julien Grall wrote: > > Hi Bertrand, > > On 31/07/2020 14:26, Bertrand Marquis wrote: +{ +gprintk(XENLOG_WARNING, "Cannot map runstate pointer at 0x%"PRIvaddr +": Invalid offset\n", vaddr); >>> >>> We usually enforc

Re: [PATCH v3] xen/arm: Convert runstate address during hypcall

2020-08-13 Thread Julien Grall
Hi Bertrand, On 31/07/2020 14:26, Bertrand Marquis wrote: +{ +gprintk(XENLOG_WARNING, "Cannot map runstate pointer at 0x%"PRIvaddr +": Invalid offset\n", vaddr); We usually enforce 80 character per lines except for format string. So it is easier to grep them in the

Re: [PATCH v3] xen/arm: Convert runstate address during hypcall

2020-08-13 Thread Julien Grall
Hi Bertrand, On 31/07/2020 14:16, Bertrand Marquis wrote: On 30 Jul 2020, at 22:50, Julien Grall wrote: On 30/07/2020 11:24, Bertrand Marquis wrote: At the moment on Arm, a Linux guest running with KTPI enabled will cause the following error when a context switch happens in user mode: (XEN)

Re: [PATCH v3] xen/arm: Convert runstate address during hypcall

2020-07-31 Thread Stefano Stabellini
On Fri, 31 Jul 2020, Bertrand Marquis wrote: > Sorry missed some points in my previous answer. > > > On 30 Jul 2020, at 22:50, Julien Grall wrote: > > > > Hi Bertrand, > > > > To avoid extra work on your side, I would recommend to wait a bit before > > sending a new version. It would be good t

Re: [PATCH v3] xen/arm: Convert runstate address during hypcall

2020-07-31 Thread Julien Grall
Hi Bertrand, On 31/07/2020 14:09, Bertrand Marquis wrote: On 31 Jul 2020, at 14:19, Jan Beulich wrote: On 30.07.2020 22:50, Julien Grall wrote: On 30/07/2020 11:24, Bertrand Marquis wrote: At the moment on Arm, a Linux guest running with KTPI enabled will cause the following error when a

Re: [PATCH v3] xen/arm: Convert runstate address during hypcall

2020-07-31 Thread Bertrand Marquis
Sorry missed some points in my previous answer. > On 30 Jul 2020, at 22:50, Julien Grall wrote: > > Hi Bertrand, > > To avoid extra work on your side, I would recommend to wait a bit before > sending a new version. It would be good to at least settle the conversation > in v2 regarding the app

Re: [PATCH v3] xen/arm: Convert runstate address during hypcall

2020-07-31 Thread Bertrand Marquis
> On 31 Jul 2020, at 03:18, Stefano Stabellini wrote: > > On Thu, 30 Jul 2020, Julien Grall wrote: >> Hi Bertrand, >> >> To avoid extra work on your side, I would recommend to wait a bit before >> sending a new version. It would be good to at least settle the conversation >> in >> v2 regardi

Re: [PATCH v3] xen/arm: Convert runstate address during hypcall

2020-07-31 Thread Bertrand Marquis
> On 30 Jul 2020, at 22:50, Julien Grall wrote: > > Hi Bertrand, > > To avoid extra work on your side, I would recommend to wait a bit before > sending a new version. It would be good to at least settle the conversation > in v2 regarding the approach taken. Thanks for the advice. I thought

Re: [PATCH v3] xen/arm: Convert runstate address during hypcall

2020-07-31 Thread Bertrand Marquis
> On 31 Jul 2020, at 14:19, Jan Beulich wrote: > > On 30.07.2020 22:50, Julien Grall wrote: >> On 30/07/2020 11:24, Bertrand Marquis wrote: >>> At the moment on Arm, a Linux guest running with KTPI enabled will >>> cause the following error when a context switch happens in user mode: >>> (XEN)

Re: [PATCH v3] xen/arm: Convert runstate address during hypcall

2020-07-31 Thread Jan Beulich
On 30.07.2020 22:50, Julien Grall wrote: > On 30/07/2020 11:24, Bertrand Marquis wrote: >> At the moment on Arm, a Linux guest running with KTPI enabled will >> cause the following error when a context switch happens in user mode: >> (XEN) p2m.c:1890: d1v0: Failed to walk page-table va 0xff837e

Re: [PATCH v3] xen/arm: Convert runstate address during hypcall

2020-07-30 Thread Stefano Stabellini
On Thu, 30 Jul 2020, Julien Grall wrote: > Hi Bertrand, > > To avoid extra work on your side, I would recommend to wait a bit before > sending a new version. It would be good to at least settle the conversation in > v2 regarding the approach taken. > > On 30/07/2020 11:24, Bertrand Marquis wrote:

Re: [PATCH v3] xen/arm: Convert runstate address during hypcall

2020-07-30 Thread Julien Grall
Hi Bertrand, To avoid extra work on your side, I would recommend to wait a bit before sending a new version. It would be good to at least settle the conversation in v2 regarding the approach taken. On 30/07/2020 11:24, Bertrand Marquis wrote: At the moment on Arm, a Linux guest running with

[PATCH v3] xen/arm: Convert runstate address during hypcall

2020-07-30 Thread Bertrand Marquis
At the moment on Arm, a Linux guest running with KTPI enabled will cause the following error when a context switch happens in user mode: (XEN) p2m.c:1890: d1v0: Failed to walk page-table va 0xff837ebe0cd0 The error is caused by the virtual address for the runstate area registered by the guest