Re: [PATCH v3 0/4] Intel MPX support

2014-01-28 Thread H. Peter Anvin
On 01/27/2014 11:01 PM, Ren Qiaowei wrote: > > Yes. Though all non-MPX threads are slowed down, the whole process > benefit from MPX. > > Anyway, HPA suggest these syscalls, which use MMU notifier, should be > not needed, we can do what they do in userspace runtime. What do you > think about it?

Re: [PATCH v3 0/4] Intel MPX support

2014-01-28 Thread H. Peter Anvin
On 01/27/2014 11:01 PM, Ren Qiaowei wrote: Yes. Though all non-MPX threads are slowed down, the whole process benefit from MPX. Anyway, HPA suggest these syscalls, which use MMU notifier, should be not needed, we can do what they do in userspace runtime. What do you think about it? I

Re: [PATCH v3 0/4] Intel MPX support

2014-01-27 Thread Ren Qiaowei
On 01/28/2014 02:42 PM, Ingo Molnar wrote: * Ren Qiaowei wrote: MPX kernel code, namely this patchset, has mainly the 2 responsibilities: provide handlers for bounds faults (#BR), and manage bounds memory. AFAICS the kernel side implementation causes no runtime overhead for non-MPX

Re: [PATCH v3 0/4] Intel MPX support

2014-01-27 Thread Ingo Molnar
* Ren Qiaowei wrote: > >> MPX kernel code, namely this patchset, has mainly the 2 > >> responsibilities: provide handlers for bounds faults (#BR), and > >> manage bounds memory. > > > > AFAICS the kernel side implementation causes no runtime overhead > > for non-MPX workloads, and also

Re: [PATCH v3 0/4] Intel MPX support

2014-01-27 Thread Ingo Molnar
* Ren Qiaowei qiaowei@intel.com wrote: MPX kernel code, namely this patchset, has mainly the 2 responsibilities: provide handlers for bounds faults (#BR), and manage bounds memory. AFAICS the kernel side implementation causes no runtime overhead for non-MPX workloads, and also

Re: [PATCH v3 0/4] Intel MPX support

2014-01-27 Thread Ren Qiaowei
On 01/28/2014 02:42 PM, Ingo Molnar wrote: * Ren Qiaowei qiaowei@intel.com wrote: MPX kernel code, namely this patchset, has mainly the 2 responsibilities: provide handlers for bounds faults (#BR), and manage bounds memory. AFAICS the kernel side implementation causes no runtime

Re: [PATCH v3 0/4] Intel MPX support

2014-01-26 Thread Ren Qiaowei
On 01/26/2014 04:19 PM, Ingo Molnar wrote: * Qiaowei Ren wrote: This patchset adds support for the Memory Protection Extensions (MPX) feature found in future Intel processors. MPX can be used in conjunction with compiler changes to check memory references, for those references whose

Re: [PATCH v3 0/4] Intel MPX support

2014-01-26 Thread Ingo Molnar
* Qiaowei Ren wrote: > This patchset adds support for the Memory Protection Extensions > (MPX) feature found in future Intel processors. > > MPX can be used in conjunction with compiler changes to check memory > references, for those references whose compile-time normal intentions > are

Re: [PATCH v3 0/4] Intel MPX support

2014-01-26 Thread Ingo Molnar
* Qiaowei Ren qiaowei@intel.com wrote: This patchset adds support for the Memory Protection Extensions (MPX) feature found in future Intel processors. MPX can be used in conjunction with compiler changes to check memory references, for those references whose compile-time normal

Re: [PATCH v3 0/4] Intel MPX support

2014-01-26 Thread Ren Qiaowei
On 01/26/2014 04:19 PM, Ingo Molnar wrote: * Qiaowei Ren qiaowei@intel.com wrote: This patchset adds support for the Memory Protection Extensions (MPX) feature found in future Intel processors. MPX can be used in conjunction with compiler changes to check memory references, for those

[PATCH v3 0/4] Intel MPX support

2014-01-25 Thread Qiaowei Ren
This patchset adds support for the Memory Protection Extensions (MPX) feature found in future Intel processors. MPX can be used in conjunction with compiler changes to check memory references, for those references whose compile-time normal intentions are usurped at runtime due to buffer overflow

[PATCH v3 0/4] Intel MPX support

2014-01-25 Thread Qiaowei Ren
This patchset adds support for the Memory Protection Extensions (MPX) feature found in future Intel processors. MPX can be used in conjunction with compiler changes to check memory references, for those references whose compile-time normal intentions are usurped at runtime due to buffer overflow