[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-devel] [PATCH v1] x86/hvm: Generic instruction re-execution mechanism for execute faults

On 21.11.2018 13:41, Roger Pau Monné wrote:
> On Wed, Nov 21, 2018 at 10:28:18AM +0000, Alexandru Stefan ISAILA wrote:
>> On 21.11.2018 11:56, Roger Pau Monné wrote:
>>> On Mon, Nov 19, 2018 at 03:56:14PM +0000, Alexandru Stefan ISAILA wrote:
>>>> On 19.11.2018 17:08, Roger Pau Monné wrote:
>>> Also, after looking at the code I'm not sure I see why this needs to
>>> be VMX specific, AFAICT it doesn't directly call any VMX functions?
>> It is vmx specific because svm does not have single step. We talked
>> about in the past about this and it turned out that it was to much
>> trouble to make a custom single step.
> I still think this shouldn't be VMX specific, and you should just
> return -EOPNOTSUPP if single stepping is not supported, just like
> hvm_debug_op does. In fact I'm missing a helper to set single
> stepping, which would be the right place to return -EOPNOTSUPP.
> Then your rexec wouldn't need to know anything about the hardware and
> would just attempt to set single stepping, failing if it cannot be
> enabled.

There is a helper function for single step "hvm_is_singlestep_supported" 
and we have the d->arch.monitor.inguest_pagefault_disabled monitor flag. 
I can return not supported in the xc_monitor_inguest_pagefault call and 
save time on the process.

Xen-devel mailing list



Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.