"Pixel 11 series is a lot more expensive for an incremental improvement to the CPU, the same underpowered GPU and reduced RAM for the Pro base models. They finally caught up to the last generation of Qualcomm cellular radio. It's overpriced, the upgrades aren't impressive and losing MTE is appalling."
Lost any remaining respect for Pixel with the decisions they've made for Pixel 11. The product design and hardware teams have both tried to one up each other and who can deliver the bigger pile of steaming crap.
What, MTE is such a promising tech and for security and now that we need all security we can get. WTF this is terrible development. You cant go backwards when the world needs getting even more secure.
I don't understand much from MTE/Android or MIE/iOS but the explanation is also confusing when:
- They claim Apple did a great job integrating MIE in iOS
- iOS doesn't encourage to opt into MTE ... Apple's docs warn developers of performance and stability issues
So it seems like even for iOS, this special security feature is only available for Apple own iOS app (at most?).
Then also:
> Even Signal doesn't opt-in. Our approach enables forcing using MTE in the standard allocators regardless.
So does that mean that Signal and any other apps are enrolled in MTE on Graphene?
iOS uses MTE for nearly all of the important parts of the OS. It uses MTE within the kernel and for a substantial portion of the base OS processes. They focused on deploying it to the most security relevant processes first but it's deployed for a large portion of the OS beyond those. Apple has done a very good job protecting the OS with it. They've done a very poor job getting the app ecosystem to adopt it. Android has more app ecosystem adoption due to having a better open source app ecosystem where GrapheneOS users have asked apps to enable it by default.
GrapheneOS uses MTE for the kernel and nearly every userspace process including all the base OS apps. We've had to fix many upstream Linux kernel and Pixel kernel driver bugs found by MTE. However, we aren't trying to fix the Pixel userspace drivers code ourselves so we have a few userspace processes excluded from MTE caused by userspace driver library/service bugs.
We recommend users enable our toggle for enabling MTE by default for every user installed app not explicitly marked as incompatible in our compatibility database. GrapheneOS has user-facing notifications for invalid memory accesses caught by MTE providing a traceback to share with the app developers. Users can use the per-app toggle to work around it if it makes an app unusable. We take the same approach for other aggressive exploit protections provided by GrapheneOS. Exploit protections with only rare compatibility issues are enabled by default for apps with a per-app toggle to opt-out and no toggle for the global default.
I see a significant difference between making a security feature available but opt-in during a teething phase, and removing the feature altogether. Am I missing something?
Such a shame. Google themselves barely utilised it in their stock builds. I wonder if there is some silicon or development overhead they decided to cut.
I'm optimistic about the Motorola partnership and when my 9 Pro is EOL I'll definitely look at how that's played out.
Google likely cut MTE to reduce the die space required by CPU cache. It saves them money on a feature they never deployed by default and only used for a few processes as part of Android Advanced Protection Mode (AAPM) without enabling it in the Linux kernel. Their security team should have gotten it deployed by default for a large portion of the OS by now and then it would have been much harder to justify removing it. The small performance impact of asymmetric mode is irrelevant for the vast majority of the OS and it can also be used in the near zero cost asynchronous mode. We use synchronous in the kernel for security reasons but they didn't have to do that.
It's a completely different story for GrapheneOS where it means losing one of the main kernel and userspace security protections. This is one of the only ways we can significantly harden the Linux kernel with existing security features. The Linux kernel has always been a huge security liability for Android and AI models are making that much more obvious to everyone.
Google will likely end up heavily using MTE in the future. Pixel 11 devices won't be able to benefit from it. They're at the start of 7 years of updates but they won't be getting the benefit of future updates enabling MTE. Pixel 8 and later will benefit from Google likely expanding use of MTE for AAPM and eventually beginning to use it by default. It's unlikely Google will stop working on expanding MTE due to the Pixel 11 hardware decision. Multiple other OEMs are interested in MTE for devices made for businesses and governments even if Google decided it wasn't worth the cost for Pixels.
This isn't terrible development, it's simply that GrapheneOS devs can't do miracles
They can't just continue supporting phones if they have security downgrades, especially if it's an important feature.
I think they mean "this is a terrible development" in the sense that this is terrible news/turn of events. They arent criticizing GrapheneOSs software development.
Pixel 11 has an incremental upgrade to the CPU, the same underpowered GPU and less RAM for the base Pro models. It has a new major version of the secure element (Titan M3) which likely greatly improves Before First Unlock state security for users without a strong passphrase. It has a better radio from a performance and efficiency perspective but it's still a more than yearly generation behind Qualcomm. The cellular radio firmware is also likely a security downgrade but the userspace code is more secure due to moving to AOSP IMS code. If these devices hadn't removed MTE, then they would have been a sidegrade with marginally better security. Removing MTE ruins these devices for GrapheneOS. We could still support the Pixel 11 series but it would be greatly rolling back overall security to before the Pixel 8. It does at least have PAC and BTI but MTE was the main improvement with the Pixel 8 rather than those.
"Pixel 11 series is a lot more expensive for an incremental improvement to the CPU, the same underpowered GPU and reduced RAM for the Pro base models. They finally caught up to the last generation of Qualcomm cellular radio. It's overpriced, the upgrades aren't impressive and losing MTE is appalling."
I guess that's a thumbs down then?
Can someone from AOSP/Pixel hardware security shed some light? This is strange timing, especially when the approach has been validated by Apple also.
Lost any remaining respect for Pixel with the decisions they've made for Pixel 11. The product design and hardware teams have both tried to one up each other and who can deliver the bigger pile of steaming crap.
What, MTE is such a promising tech and for security and now that we need all security we can get. WTF this is terrible development. You cant go backwards when the world needs getting even more secure.
I don't understand much from MTE/Android or MIE/iOS but the explanation is also confusing when:
- They claim Apple did a great job integrating MIE in iOS - iOS doesn't encourage to opt into MTE ... Apple's docs warn developers of performance and stability issues
So it seems like even for iOS, this special security feature is only available for Apple own iOS app (at most?).
Then also:
> Even Signal doesn't opt-in. Our approach enables forcing using MTE in the standard allocators regardless.
So does that mean that Signal and any other apps are enrolled in MTE on Graphene?
iOS uses MTE for nearly all of the important parts of the OS. It uses MTE within the kernel and for a substantial portion of the base OS processes. They focused on deploying it to the most security relevant processes first but it's deployed for a large portion of the OS beyond those. Apple has done a very good job protecting the OS with it. They've done a very poor job getting the app ecosystem to adopt it. Android has more app ecosystem adoption due to having a better open source app ecosystem where GrapheneOS users have asked apps to enable it by default.
GrapheneOS uses MTE for the kernel and nearly every userspace process including all the base OS apps. We've had to fix many upstream Linux kernel and Pixel kernel driver bugs found by MTE. However, we aren't trying to fix the Pixel userspace drivers code ourselves so we have a few userspace processes excluded from MTE caused by userspace driver library/service bugs.
We recommend users enable our toggle for enabling MTE by default for every user installed app not explicitly marked as incompatible in our compatibility database. GrapheneOS has user-facing notifications for invalid memory accesses caught by MTE providing a traceback to share with the app developers. Users can use the per-app toggle to work around it if it makes an app unusable. We take the same approach for other aggressive exploit protections provided by GrapheneOS. Exploit protections with only rare compatibility issues are enabled by default for apps with a per-app toggle to opt-out and no toggle for the global default.
I see a significant difference between making a security feature available but opt-in during a teething phase, and removing the feature altogether. Am I missing something?
Such a shame. Google themselves barely utilised it in their stock builds. I wonder if there is some silicon or development overhead they decided to cut.
I'm optimistic about the Motorola partnership and when my 9 Pro is EOL I'll definitely look at how that's played out.
Google likely cut MTE to reduce the die space required by CPU cache. It saves them money on a feature they never deployed by default and only used for a few processes as part of Android Advanced Protection Mode (AAPM) without enabling it in the Linux kernel. Their security team should have gotten it deployed by default for a large portion of the OS by now and then it would have been much harder to justify removing it. The small performance impact of asymmetric mode is irrelevant for the vast majority of the OS and it can also be used in the near zero cost asynchronous mode. We use synchronous in the kernel for security reasons but they didn't have to do that.
It's a completely different story for GrapheneOS where it means losing one of the main kernel and userspace security protections. This is one of the only ways we can significantly harden the Linux kernel with existing security features. The Linux kernel has always been a huge security liability for Android and AI models are making that much more obvious to everyone.
Google will likely end up heavily using MTE in the future. Pixel 11 devices won't be able to benefit from it. They're at the start of 7 years of updates but they won't be getting the benefit of future updates enabling MTE. Pixel 8 and later will benefit from Google likely expanding use of MTE for AAPM and eventually beginning to use it by default. It's unlikely Google will stop working on expanding MTE due to the Pixel 11 hardware decision. Multiple other OEMs are interested in MTE for devices made for businesses and governments even if Google decided it wasn't worth the cost for Pixels.
This isn't terrible development, it's simply that GrapheneOS devs can't do miracles They can't just continue supporting phones if they have security downgrades, especially if it's an important feature.
I think they mean "this is a terrible development" in the sense that this is terrible news/turn of events. They arent criticizing GrapheneOSs software development.
They also say that the Pixel 11 is a very very tiny improvement wrt the Pixel 10, with less RAM and more expensive.
So basically they say don't buy it, wait for the Motorola phones.
Pixel 11 has an incremental upgrade to the CPU, the same underpowered GPU and less RAM for the base Pro models. It has a new major version of the secure element (Titan M3) which likely greatly improves Before First Unlock state security for users without a strong passphrase. It has a better radio from a performance and efficiency perspective but it's still a more than yearly generation behind Qualcomm. The cellular radio firmware is also likely a security downgrade but the userspace code is more secure due to moving to AOSP IMS code. If these devices hadn't removed MTE, then they would have been a sidegrade with marginally better security. Removing MTE ruins these devices for GrapheneOS. We could still support the Pixel 11 series but it would be greatly rolling back overall security to before the Pixel 8. It does at least have PAC and BTI but MTE was the main improvement with the Pixel 8 rather than those.
They have their reasons. Pixels are only getting worse and worse after the 9