Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

On 14/09/2016 20:36, Andy Lutomirski wrote: > On Wed, Sep 14, 2016 at 12:28 PM, Andrew Cooper > wrote: >> On 14/09/2016 20:23, Boris Ostrovsky wrote: >>> On 09/14/2016 02:52 PM, Andy Lutomirski wrote: On Tue, Sep 13, 2016 at 11:13 PM, Kyle Huey

Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

On 14/09/2016 20:36, Andy Lutomirski wrote: > On Wed, Sep 14, 2016 at 12:28 PM, Andrew Cooper > wrote: >> On 14/09/2016 20:23, Boris Ostrovsky wrote: >>> On 09/14/2016 02:52 PM, Andy Lutomirski wrote: On Tue, Sep 13, 2016 at 11:13 PM, Kyle Huey wrote: > On Mon, Sep 12, 2016 at 9:56 AM,

Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

On Wed, Sep 14, 2016 at 12:28 PM, Andrew Cooper wrote: > On 14/09/2016 20:23, Boris Ostrovsky wrote: >> On 09/14/2016 02:52 PM, Andy Lutomirski wrote: >>> On Tue, Sep 13, 2016 at 11:13 PM, Kyle Huey wrote: On Mon, Sep 12, 2016 at 9:56 AM, Andy

Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

On Wed, Sep 14, 2016 at 12:28 PM, Andrew Cooper wrote: > On 14/09/2016 20:23, Boris Ostrovsky wrote: >> On 09/14/2016 02:52 PM, Andy Lutomirski wrote: >>> On Tue, Sep 13, 2016 at 11:13 PM, Kyle Huey wrote: On Mon, Sep 12, 2016 at 9:56 AM, Andy Lutomirski wrote: > You should

Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

On 14/09/2016 19:52, Andy Lutomirski wrote: > On Tue, Sep 13, 2016 at 11:13 PM, Kyle Huey wrote: >> On Mon, Sep 12, 2016 at 9:56 AM, Andy Lutomirski wrote: >>> You should explicitly check that, if the >>> feature is set under Xen PV, then the MSR actually

Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

On 14/09/2016 19:52, Andy Lutomirski wrote: > On Tue, Sep 13, 2016 at 11:13 PM, Kyle Huey wrote: >> On Mon, Sep 12, 2016 at 9:56 AM, Andy Lutomirski wrote: >>> You should explicitly check that, if the >>> feature is set under Xen PV, then the MSR actually works as >>> advertised. This may

Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

On 14/09/2016 20:23, Boris Ostrovsky wrote: > On 09/14/2016 02:52 PM, Andy Lutomirski wrote: >> On Tue, Sep 13, 2016 at 11:13 PM, Kyle Huey wrote: >>> On Mon, Sep 12, 2016 at 9:56 AM, Andy Lutomirski >>> wrote: You should explicitly check that, if

Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

On 14/09/2016 20:23, Boris Ostrovsky wrote: > On 09/14/2016 02:52 PM, Andy Lutomirski wrote: >> On Tue, Sep 13, 2016 at 11:13 PM, Kyle Huey wrote: >>> On Mon, Sep 12, 2016 at 9:56 AM, Andy Lutomirski >>> wrote: You should explicitly check that, if the feature is set under Xen PV, then

Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

On 09/14/2016 02:52 PM, Andy Lutomirski wrote: > On Tue, Sep 13, 2016 at 11:13 PM, Kyle Huey wrote: >> On Mon, Sep 12, 2016 at 9:56 AM, Andy Lutomirski wrote: >>> You should explicitly check that, if the >>> feature is set under Xen PV, then the MSR

Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

On 09/14/2016 02:52 PM, Andy Lutomirski wrote: > On Tue, Sep 13, 2016 at 11:13 PM, Kyle Huey wrote: >> On Mon, Sep 12, 2016 at 9:56 AM, Andy Lutomirski wrote: >>> You should explicitly check that, if the >>> feature is set under Xen PV, then the MSR actually works as >>> advertised. This may

Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

On Tue, Sep 13, 2016 at 11:13 PM, Kyle Huey wrote: > On Mon, Sep 12, 2016 at 9:56 AM, Andy Lutomirski wrote: >> You should explicitly check that, if the >> feature is set under Xen PV, then the MSR actually works as >> advertised. This may require talking

Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

On Tue, Sep 13, 2016 at 11:13 PM, Kyle Huey wrote: > On Mon, Sep 12, 2016 at 9:56 AM, Andy Lutomirski wrote: >> You should explicitly check that, if the >> feature is set under Xen PV, then the MSR actually works as >> advertised. This may require talking to the Xen folks to make sure >> you're

Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

On Mon, Sep 12, 2016 at 9:56 AM, Andy Lutomirski wrote: > You should explicitly check that, if the > feature is set under Xen PV, then the MSR actually works as > advertised. This may require talking to the Xen folks to make sure > you're testing the right configuration.

Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

On Mon, Sep 12, 2016 at 9:56 AM, Andy Lutomirski wrote: > You should explicitly check that, if the > feature is set under Xen PV, then the MSR actually works as > advertised. This may require talking to the Xen folks to make sure > you're testing the right configuration. This is interesting.

Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

On Mon, Sep 12, 2016 at 7:15 AM, Kyle Huey wrote: > On Mon, Sep 12, 2016 at 2:07 AM, Borislav Petkov wrote: >> On Sun, Sep 11, 2016 at 05:29:23PM -0700, Kyle Huey wrote: >>> @@ -2162,6 +2168,12 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, >>> arg2,

Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

On Mon, Sep 12, 2016 at 7:15 AM, Kyle Huey wrote: > On Mon, Sep 12, 2016 at 2:07 AM, Borislav Petkov wrote: >> On Sun, Sep 11, 2016 at 05:29:23PM -0700, Kyle Huey wrote: >>> @@ -2162,6 +2168,12 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, >>> arg2, unsigned long, arg3, >>> case

Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

On Sep 12, 2016 10:57 AM, "Jann Horn" wrote: > > On Mon, Sep 12, 2016 at 09:56:11AM -0700, Andy Lutomirski wrote: > > On Sep 11, 2016 5:29 PM, "Kyle Huey" wrote: > > > > > > rr (http://rr-project.org/), a userspace record-and-replay reverse- > > > execution

Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

On Sep 12, 2016 10:57 AM, "Jann Horn" wrote: > > On Mon, Sep 12, 2016 at 09:56:11AM -0700, Andy Lutomirski wrote: > > On Sep 11, 2016 5:29 PM, "Kyle Huey" wrote: > > > > > > rr (http://rr-project.org/), a userspace record-and-replay reverse- > > > execution debugger, would like to trap and

Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

On Mon, 12 Sep 2016, Andi Kleen wrote: > The cpuid char driver does this, other code may too. Such as anything that calls sync_core(). That includes processor hotplug paths, and who knows what else... > You probably would need to protect these CPUIDs with an exception > handler that temporarily

Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

On Mon, 12 Sep 2016, Andi Kleen wrote: > The cpuid char driver does this, other code may too. Such as anything that calls sync_core(). That includes processor hotplug paths, and who knows what else... > You probably would need to protect these CPUIDs with an exception > handler that temporarily

Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

On Mon, Sep 12, 2016 at 09:56:11AM -0700, Andy Lutomirski wrote: > On Sep 11, 2016 5:29 PM, "Kyle Huey" wrote: > > > > rr (http://rr-project.org/), a userspace record-and-replay reverse- > > execution debugger, would like to trap and emulate the CPUID instruction. > > This

Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

On Mon, Sep 12, 2016 at 09:56:11AM -0700, Andy Lutomirski wrote: > On Sep 11, 2016 5:29 PM, "Kyle Huey" wrote: > > > > rr (http://rr-project.org/), a userspace record-and-replay reverse- > > execution debugger, would like to trap and emulate the CPUID instruction. > > This would allow us to a)

Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

Kyle Huey writes: > rr (http://rr-project.org/), a userspace record-and-replay reverse- > execution debugger, would like to trap and emulate the CPUID instruction. > This would allow us to a) mask away certain hardware features that rr does > not support (e.g. RDRAND) and b) enable trace

Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

Kyle Huey writes: > rr (http://rr-project.org/), a userspace record-and-replay reverse- > execution debugger, would like to trap and emulate the CPUID instruction. > This would allow us to a) mask away certain hardware features that rr does > not support (e.g. RDRAND) and b) enable trace

Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

On Mon, Sep 12, 2016 at 09:56:11AM -0700, Andy Lutomirski wrote: > This should have a cpufeature bit and show up in /proc/cpuinfo. That Haha, test a cpufeature with a faulting CPUID :-) Yeah, we need a synthetic flag and query MSR_PLATFORM_INFO[31] which denotes whether the other MSR -

Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

On Mon, Sep 12, 2016 at 09:56:11AM -0700, Andy Lutomirski wrote: > This should have a cpufeature bit and show up in /proc/cpuinfo. That Haha, test a cpufeature with a faulting CPUID :-) Yeah, we need a synthetic flag and query MSR_PLATFORM_INFO[31] which denotes whether the other MSR -

Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

On Sep 11, 2016 5:29 PM, "Kyle Huey" wrote: > > rr (http://rr-project.org/), a userspace record-and-replay reverse- > execution debugger, would like to trap and emulate the CPUID instruction. > This would allow us to a) mask away certain hardware features that rr does > not

Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

On Sep 11, 2016 5:29 PM, "Kyle Huey" wrote: > > rr (http://rr-project.org/), a userspace record-and-replay reverse- > execution debugger, would like to trap and emulate the CPUID instruction. > This would allow us to a) mask away certain hardware features that rr does > not support (e.g. RDRAND)

Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

On Mon, Sep 12, 2016 at 07:15:16AM -0700, Kyle Huey wrote: > Copied from PR_SET_TSC. Would you prefer something like > disable_cpuid/disable_cpuid_and_set_flag for > hard_disable_CPUID/disable_CPUID? Maybe something like this: switch_cpuid_faulting(bool on) { if (on)

Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

On Mon, Sep 12, 2016 at 07:15:16AM -0700, Kyle Huey wrote: > Copied from PR_SET_TSC. Would you prefer something like > disable_cpuid/disable_cpuid_and_set_flag for > hard_disable_CPUID/disable_CPUID? Maybe something like this: switch_cpuid_faulting(bool on) { if (on)

Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

Thanks for the review! On Mon, Sep 12, 2016 at 2:07 AM, Borislav Petkov wrote: > On Sun, Sep 11, 2016 at 05:29:23PM -0700, Kyle Huey wrote: >> rr (http://rr-project.org/), a userspace record-and-replay reverse- >> execution debugger, would like to trap and emulate the CPUID

Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

Thanks for the review! On Mon, Sep 12, 2016 at 2:07 AM, Borislav Petkov wrote: > On Sun, Sep 11, 2016 at 05:29:23PM -0700, Kyle Huey wrote: >> rr (http://rr-project.org/), a userspace record-and-replay reverse- >> execution debugger, would like to trap and emulate the CPUID instruction. >> This

Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

On Sun, Sep 11, 2016 at 05:29:23PM -0700, Kyle Huey wrote: > rr (http://rr-project.org/), a userspace record-and-replay reverse- > execution debugger, would like to trap and emulate the CPUID instruction. > This would allow us to a) mask away certain hardware features that rr does > not support

Re: [PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

On Sun, Sep 11, 2016 at 05:29:23PM -0700, Kyle Huey wrote: > rr (http://rr-project.org/), a userspace record-and-replay reverse- > execution debugger, would like to trap and emulate the CPUID instruction. > This would allow us to a) mask away certain hardware features that rr does > not support

[PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

rr (http://rr-project.org/), a userspace record-and-replay reverse- execution debugger, would like to trap and emulate the CPUID instruction. This would allow us to a) mask away certain hardware features that rr does not support (e.g. RDRAND) and b) enable trace portability across machines by

[PATCH] prctl,x86 Add PR_[GET|SET]_CPUID for controlling the CPUID instruction.

rr (http://rr-project.org/), a userspace record-and-replay reverse- execution debugger, would like to trap and emulate the CPUID instruction. This would allow us to a) mask away certain hardware features that rr does not support (e.g. RDRAND) and b) enable trace portability across machines by