Re: [PATCH v20 08/25] x86/mm: Introduce _PAGE_COW

2021-02-15 Thread Yu, Yu-cheng
On 2/10/2021 12:28 PM, Yu, Yu-cheng wrote: On 2/10/2021 11:42 AM, Kees Cook wrote: On Wed, Feb 10, 2021 at 09:56:46AM -0800, Yu-cheng Yu wrote: There is essentially no room left in the x86 hardware PTEs on some OSes (not Linux).  That left the hardware architects looking for a way to represent

Re: [PATCH v20 08/25] x86/mm: Introduce _PAGE_COW

2021-02-10 Thread Yu, Yu-cheng
On 2/10/2021 11:42 AM, Kees Cook wrote: On Wed, Feb 10, 2021 at 09:56:46AM -0800, Yu-cheng Yu wrote: There is essentially no room left in the x86 hardware PTEs on some OSes (not Linux). That left the hardware architects looking for a way to represent a new memory type (shadow stack) within the

Re: [PATCH v20 08/25] x86/mm: Introduce _PAGE_COW

2021-02-10 Thread Kees Cook
On Wed, Feb 10, 2021 at 09:56:46AM -0800, Yu-cheng Yu wrote: > There is essentially no room left in the x86 hardware PTEs on some OSes > (not Linux). That left the hardware architects looking for a way to > represent a new memory type (shadow stack) within the existing bits. > They chose to repurp

[PATCH v20 08/25] x86/mm: Introduce _PAGE_COW

2021-02-10 Thread Yu-cheng Yu
There is essentially no room left in the x86 hardware PTEs on some OSes (not Linux). That left the hardware architects looking for a way to represent a new memory type (shadow stack) within the existing bits. They chose to repurpose a lightly-used state: Write=0, Dirty=1. The reason it's lightly