Reversing the AMD Secure Processor (PSP) – Part 2: Cryptographic Co-Processor
1–10 of 31 posts
Re: Reversing the AMD Secure Processor (PSP) – Part 2: Cryptographic Co-Processor
#2I never understood why AMD is not at least making the source of these available. I would actually really like a secure cryptographic processor that's been extensively vetted and trustworthy.
Re: Reversing the AMD Secure Processor (PSP) – Part 2: Cryptographic Co-Processor
#3I never understood why AMD is not at least making the source of these available. I would actually really like a secure cryptographic processor that's been extensively vetted and trustworthy.
The common assumption is that their hands are tied on that matter.
Re: Reversing the AMD Secure Processor (PSP) – Part 2: Cryptographic Co-Processor
#4How popular is AMDs PSP, or Intels IME for that matter, in the real world. I've never seen either used at all.
Re: Reversing the AMD Secure Processor (PSP) – Part 2: Cryptographic Co-Processor
#5I never understood why AMD is not at least making the source of these available. I would actually really like a secure cryptographic processor that's been extensively vetted and trustworthy.
The common assumption is that their hands are tied on that matter.
Like OpenSIL, AMD could start a long-term project to replace the closed ASP with an open alternative. The industry now has Calpitra (AMD contributes), OpenTitan, TockOS (used by Pluton) and other open hardware and software projects for security enclaves.
Re: Reversing the AMD Secure Processor (PSP) – Part 2: Cryptographic Co-Processor
#6How popular is AMDs PSP, or Intels IME for that matter, in the real world. I've never seen either used at all.
In addition to system launch integrity, they are likely used by streaming video DRM, e.g. Netflix.
Re: Reversing the AMD Secure Processor (PSP) – Part 2: Cryptographic Co-Processor
#7Re: Reversing the AMD Secure Processor (PSP) – Part 2: Cryptographic Co-Processor
#8How popular is AMDs PSP, or Intels IME for that matter, in the real world. I've never seen either used at all.
In addition to system launch integrity, they are likely used by streaming video DRM, e.g. Netflix.
SGX was used by video DRM on intel platforms. As SGX no longer exists in modern intel processors, its not really doable anymore. netflix drm and the like are probably done on gpu, not on cpu (but I could be wrong)
It's actually impossible to have a "legal"/commercial 4k bluray setup today on modern PCs/CPUs, as they will only license it to players that can use SGX and as noted SGX no longer exists. (of course this doesn't prevent one from using vlc / libaacs and the like).
Re: Reversing the AMD Secure Processor (PSP) – Part 2: Cryptographic Co-Processor
#9I never understood why AMD is not at least making the source of these available. I would actually really like a secure cryptographic processor that's been extensively vetted and trustworthy.
They did publish the source:
https://github.com/amd/AMD-ASPFW
This was the PR(!): https://ir.amd.com/news-events/press-releases/detail/1154/am...
I wonder if the reported exploits forced them to publish the source.
Re: Reversing the AMD Secure Processor (PSP) – Part 2: Cryptographic Co-Processor
#10I never understood why AMD is not at least making the source of these available. I would actually really like a secure cryptographic processor that's been extensively vetted and trustworthy.
They did publish the source: https://github.com/amd/AMD-ASPFW This was the PR(!): https://ir.amd.com/news-events/press-releases/detail/1154/am... I wonder if the reported exploits forced them to publish the source.
That is only the part that allows encrypted virtual machines on EPYC cpus. The PSP in some form is on all AMD processors since about 2013/2014.