| Message ID | 20220504232102.469959-1-evgreen@chromium.org (mailing list archive) |
|---|---|
| Headers | show |
| Series | Encrypted Hibernation | expand |
Hi! > We are exploring enabling hibernation in some new scenarios. However, > our security team has a few requirements, listed below: > 1. The hibernate image must be encrypted with protection derived from > both the platform (eg TPM) and user authentication data (eg > password). > 2. Hibernation must not be a vector by which a malicious userspace can > escalate to the kernel. Can you (or your security team) explain why requirement 2. is needed? On normal systems, trusted userspace handles kernel upgrades (for example), so it can escalate to kernel priviledges. Best regards, Pavel
On Fri, May 6, 2022 at 9:08 AM Pavel Machek <pavel@ucw.cz> wrote: > > Hi! > > > We are exploring enabling hibernation in some new scenarios. However, > > our security team has a few requirements, listed below: > > 1. The hibernate image must be encrypted with protection derived from > > both the platform (eg TPM) and user authentication data (eg > > password). > > 2. Hibernation must not be a vector by which a malicious userspace can > > escalate to the kernel. > > Can you (or your security team) explain why requirement 2. is needed? > > On normal systems, trusted userspace handles kernel upgrades (for example), > so it can escalate to kernel priviledges. > Our systems are a little more sealed up than a normal distro, we use Verified Boot [1]. To summarize, RO firmware with an embedded public key verifies that the kernel+commandline was signed by Google. The commandline includes the root hash of the rootfs as well (where the modules live). So when an update is applied (A/B style, including the whole rootfs), assuming the RO firmware stayed RO (which requires physical measures to defeat), we can guarantee that the kernel, commandline, and rootfs have not been tampered with. Verified boot gives us confidence that on each boot, we're at least starting from known code. This makes it more challenging for an attacker to persist an exploit across reboot. With the kernel and modules verified, we try to make it non-trivial for someone who does manage to gain root execution once from escalating to kernel execution. Hibernation would be one obvious escalation route, so we're hoping to find a way to enable it without handing out that easy primitive. [1] https://www.chromium.org/chromium-os/chromiumos-design-docs/verified-boot/ > Best regards, > Pavel > -- > (english) http://www.livejournal.com/~pavelmachek > (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
On Mon, May 9, 2022 at 6:44 PM Evan Green <evgreen@chromium.org> wrote: > > On Fri, May 6, 2022 at 9:08 AM Pavel Machek <pavel@ucw.cz> wrote: > > > > Hi! > > > > > We are exploring enabling hibernation in some new scenarios. However, > > > our security team has a few requirements, listed below: > > > 1. The hibernate image must be encrypted with protection derived from > > > both the platform (eg TPM) and user authentication data (eg > > > password). > > > 2. Hibernation must not be a vector by which a malicious userspace can > > > escalate to the kernel. > > > > Can you (or your security team) explain why requirement 2. is needed? > > > > On normal systems, trusted userspace handles kernel upgrades (for example), > > so it can escalate to kernel priviledges. > > > > Our systems are a little more sealed up than a normal distro, we use > Verified Boot [1]. To summarize, RO firmware with an embedded public > key verifies that the kernel+commandline was signed by Google. The > commandline includes the root hash of the rootfs as well (where the > modules live). So when an update is applied (A/B style, including the > whole rootfs), assuming the RO firmware stayed RO (which requires > physical measures to defeat), we can guarantee that the kernel, > commandline, and rootfs have not been tampered with. > > Verified boot gives us confidence that on each boot, we're at least > starting from known code. This makes it more challenging for an > attacker to persist an exploit across reboot. With the kernel and > modules verified, we try to make it non-trivial for someone who does > manage to gain root execution once from escalating to kernel > execution. Hibernation would be one obvious escalation route, so we're > hoping to find a way to enable it without handing out that easy > primitive. > > [1] https://www.chromium.org/chromium-os/chromiumos-design-docs/verified-boot/ So I guess this really is an RFC. Honestly, I need more time to go through this and there are pieces of it that need to be looked at other people (like the TPM-related changes). Thanks!
Hi Rafael, On Tue, May 17, 2022 at 9:06 AM Rafael J. Wysocki <rafael@kernel.org> wrote: > > On Mon, May 9, 2022 at 6:44 PM Evan Green <evgreen@chromium.org> wrote: > > > > On Fri, May 6, 2022 at 9:08 AM Pavel Machek <pavel@ucw.cz> wrote: > > > > > > Hi! > > > > > > > We are exploring enabling hibernation in some new scenarios. However, > > > > our security team has a few requirements, listed below: > > > > 1. The hibernate image must be encrypted with protection derived from > > > > both the platform (eg TPM) and user authentication data (eg > > > > password). > > > > 2. Hibernation must not be a vector by which a malicious userspace can > > > > escalate to the kernel. > > > > > > Can you (or your security team) explain why requirement 2. is needed? > > > > > > On normal systems, trusted userspace handles kernel upgrades (for example), > > > so it can escalate to kernel priviledges. > > > > > > > Our systems are a little more sealed up than a normal distro, we use > > Verified Boot [1]. To summarize, RO firmware with an embedded public > > key verifies that the kernel+commandline was signed by Google. The > > commandline includes the root hash of the rootfs as well (where the > > modules live). So when an update is applied (A/B style, including the > > whole rootfs), assuming the RO firmware stayed RO (which requires > > physical measures to defeat), we can guarantee that the kernel, > > commandline, and rootfs have not been tampered with. > > > > Verified boot gives us confidence that on each boot, we're at least > > starting from known code. This makes it more challenging for an > > attacker to persist an exploit across reboot. With the kernel and > > modules verified, we try to make it non-trivial for someone who does > > manage to gain root execution once from escalating to kernel > > execution. Hibernation would be one obvious escalation route, so we're > > hoping to find a way to enable it without handing out that easy > > primitive. > > > > [1] https://www.chromium.org/chromium-os/chromiumos-design-docs/verified-boot/ > > So I guess this really is an RFC. Yes, I suppose it is. > > Honestly, I need more time to go through this and there are pieces of > it that need to be looked at other people (like the TPM-related > changes). No problem, thanks for the reply to let me know. I expect some back and forth in terms of what should be hidden behind abstractions and where exactly things should live. But I wanted to get this out to upstream as early as I could, just to get initial reactions on the overall concept and design. Looking forward to hearing your thoughts when you get a chance, and let me know if there are others I should be adding that I've missed. -Evan > > Thanks!
On Tue, May 17, 2022 at 10:34 AM Evan Green <evgreen@chromium.org> wrote: > > Hi Rafael, > > On Tue, May 17, 2022 at 9:06 AM Rafael J. Wysocki <rafael@kernel.org> wrote: > > > > On Mon, May 9, 2022 at 6:44 PM Evan Green <evgreen@chromium.org> wrote: > > > > > > On Fri, May 6, 2022 at 9:08 AM Pavel Machek <pavel@ucw.cz> wrote: > > > > > > > > Hi! > > > > > > > > > We are exploring enabling hibernation in some new scenarios. However, > > > > > our security team has a few requirements, listed below: > > > > > 1. The hibernate image must be encrypted with protection derived from > > > > > both the platform (eg TPM) and user authentication data (eg > > > > > password). > > > > > 2. Hibernation must not be a vector by which a malicious userspace can > > > > > escalate to the kernel. > > > > > > > > Can you (or your security team) explain why requirement 2. is needed? > > > > > > > > On normal systems, trusted userspace handles kernel upgrades (for example), > > > > so it can escalate to kernel priviledges. > > > > > > > > > > Our systems are a little more sealed up than a normal distro, we use > > > Verified Boot [1]. To summarize, RO firmware with an embedded public > > > key verifies that the kernel+commandline was signed by Google. The > > > commandline includes the root hash of the rootfs as well (where the > > > modules live). So when an update is applied (A/B style, including the > > > whole rootfs), assuming the RO firmware stayed RO (which requires > > > physical measures to defeat), we can guarantee that the kernel, > > > commandline, and rootfs have not been tampered with. > > > > > > Verified boot gives us confidence that on each boot, we're at least > > > starting from known code. This makes it more challenging for an > > > attacker to persist an exploit across reboot. With the kernel and > > > modules verified, we try to make it non-trivial for someone who does > > > manage to gain root execution once from escalating to kernel > > > execution. Hibernation would be one obvious escalation route, so we're > > > hoping to find a way to enable it without handing out that easy > > > primitive. > > > > > > [1] https://www.chromium.org/chromium-os/chromiumos-design-docs/verified-boot/ > > > > So I guess this really is an RFC. > > Yes, I suppose it is. > > > > > Honestly, I need more time to go through this and there are pieces of > > it that need to be looked at other people (like the TPM-related > > changes). > > No problem, thanks for the reply to let me know. I expect some back > and forth in terms of what should be hidden behind abstractions and > where exactly things should live. But I wanted to get this out to > upstream as early as I could, just to get initial reactions on the > overall concept and design. Looking forward to hearing your thoughts > when you get a chance, and let me know if there are others I should be > adding that I've missed. Gentle bump in case this dropped off of radars, I'd still appreciate any feedback folks had on this series. -Evan > > -Evan > > > > > Thanks!
On Thu, Jun 16, 2022 at 8:42 AM Evan Green <evgreen@chromium.org> wrote: > > On Tue, May 17, 2022 at 10:34 AM Evan Green <evgreen@chromium.org> wrote: > > > > Hi Rafael, > > > > On Tue, May 17, 2022 at 9:06 AM Rafael J. Wysocki <rafael@kernel.org> wrote: > > > > > > On Mon, May 9, 2022 at 6:44 PM Evan Green <evgreen@chromium.org> wrote: > > > > > > > > On Fri, May 6, 2022 at 9:08 AM Pavel Machek <pavel@ucw.cz> wrote: > > > > > > > > > > Hi! > > > > > > > > > > > We are exploring enabling hibernation in some new scenarios. However, > > > > > > our security team has a few requirements, listed below: > > > > > > 1. The hibernate image must be encrypted with protection derived from > > > > > > both the platform (eg TPM) and user authentication data (eg > > > > > > password). > > > > > > 2. Hibernation must not be a vector by which a malicious userspace can > > > > > > escalate to the kernel. > > > > > > > > > > Can you (or your security team) explain why requirement 2. is needed? > > > > > > > > > > On normal systems, trusted userspace handles kernel upgrades (for example), > > > > > so it can escalate to kernel priviledges. > > > > > > > > > > > > > Our systems are a little more sealed up than a normal distro, we use > > > > Verified Boot [1]. To summarize, RO firmware with an embedded public > > > > key verifies that the kernel+commandline was signed by Google. The > > > > commandline includes the root hash of the rootfs as well (where the > > > > modules live). So when an update is applied (A/B style, including the > > > > whole rootfs), assuming the RO firmware stayed RO (which requires > > > > physical measures to defeat), we can guarantee that the kernel, > > > > commandline, and rootfs have not been tampered with. > > > > > > > > Verified boot gives us confidence that on each boot, we're at least > > > > starting from known code. This makes it more challenging for an > > > > attacker to persist an exploit across reboot. With the kernel and > > > > modules verified, we try to make it non-trivial for someone who does > > > > manage to gain root execution once from escalating to kernel > > > > execution. Hibernation would be one obvious escalation route, so we're > > > > hoping to find a way to enable it without handing out that easy > > > > primitive. > > > > > > > > [1] https://www.chromium.org/chromium-os/chromiumos-design-docs/verified-boot/ > > > > > > So I guess this really is an RFC. > > > > Yes, I suppose it is. > > > > > > > > Honestly, I need more time to go through this and there are pieces of > > > it that need to be looked at other people (like the TPM-related > > > changes). > > > > No problem, thanks for the reply to let me know. I expect some back > > and forth in terms of what should be hidden behind abstractions and > > where exactly things should live. But I wanted to get this out to > > upstream as early as I could, just to get initial reactions on the > > overall concept and design. Looking forward to hearing your thoughts > > when you get a chance, and let me know if there are others I should be > > adding that I've missed. > > Gentle bump in case this dropped off of radars, I'd still appreciate > any feedback folks had on this series. One more bump here, as we'd really love to get encrypted hibernation to a form upstream would accept if at all possible. We were considering landing this in our Chrome OS tree for now, then coming back in a couple months with a "we've been baking this ourselves and it's going so great, oooh yeah". I'm not sure if upstream would find that compelling or not. But in any case, some guidance towards making this more upstream friendly would be well appreciated. One thing I realized in attempting to pick this myself is that the trusted key blob format has moved to ASN.1. So I should really move the creation ticket to the new ASN.1 format (if I can figure out the right OID for that piece), which would allow me to drop a lot of the ugly stuff in tpm2_unpack_blob(). Maybe if I get no other comments I'll work on that and resend. -Evan
On Mon, Aug 1, 2022 at 3:33 PM Evan Green <evgreen@chromium.org> wrote: > One more bump here, as we'd really love to get encrypted hibernation > to a form upstream would accept if at all possible. We were > considering landing this in our Chrome OS tree for now, then coming > back in a couple months with a "we've been baking this ourselves and > it's going so great, oooh yeah". I'm not sure if upstream would find > that compelling or not. But in any case, some guidance towards making > this more upstream friendly would be well appreciated. > > One thing I realized in attempting to pick this myself is that the > trusted key blob format has moved to ASN.1. So I should really move > the creation ticket to the new ASN.1 format (if I can figure out the > right OID for that piece), which would allow me to drop a lot of the > ugly stuff in tpm2_unpack_blob(). Maybe if I get no other comments > I'll work on that and resend. I've been revamping my TPM-backed verified hibernation implementation based on this work, so I'd definitely be enthusiastic about it being mergeable.
On Tue, Aug 02, 2022 at 11:36:43AM -0700, Matthew Garrett wrote: > On Mon, Aug 1, 2022 at 3:33 PM Evan Green <evgreen@chromium.org> wrote: > > > One more bump here, as we'd really love to get encrypted hibernation > > to a form upstream would accept if at all possible. We were > > considering landing this in our Chrome OS tree for now, then coming > > back in a couple months with a "we've been baking this ourselves and > > it's going so great, oooh yeah". I'm not sure if upstream would find > > that compelling or not. But in any case, some guidance towards making > > this more upstream friendly would be well appreciated. > > > > One thing I realized in attempting to pick this myself is that the > > trusted key blob format has moved to ASN.1. So I should really move > > the creation ticket to the new ASN.1 format (if I can figure out the > > right OID for that piece), which would allow me to drop a lot of the > > ugly stuff in tpm2_unpack_blob(). Maybe if I get no other comments > > I'll work on that and resend. > > I've been revamping my TPM-backed verified hibernation implementation > based on this work, so I'd definitely be enthusiastic about it being > mergeable. BTW, is it tested with QEMU + swtpm? BR, Jarkko
On Wed, Aug 3, 2022 at 5:59 PM Jarkko Sakkinen <jarkko@kernel.org> wrote: > > On Tue, Aug 02, 2022 at 11:36:43AM -0700, Matthew Garrett wrote: > > On Mon, Aug 1, 2022 at 3:33 PM Evan Green <evgreen@chromium.org> wrote: > > > > > One more bump here, as we'd really love to get encrypted hibernation > > > to a form upstream would accept if at all possible. We were > > > considering landing this in our Chrome OS tree for now, then coming > > > back in a couple months with a "we've been baking this ourselves and > > > it's going so great, oooh yeah". I'm not sure if upstream would find > > > that compelling or not. But in any case, some guidance towards making > > > this more upstream friendly would be well appreciated. > > > > > > One thing I realized in attempting to pick this myself is that the > > > trusted key blob format has moved to ASN.1. So I should really move > > > the creation ticket to the new ASN.1 format (if I can figure out the > > > right OID for that piece), which would allow me to drop a lot of the > > > ugly stuff in tpm2_unpack_blob(). Maybe if I get no other comments > > > I'll work on that and resend. > > > > I've been revamping my TPM-backed verified hibernation implementation > > based on this work, so I'd definitely be enthusiastic about it being > > mergeable. > > BTW, is it tested with QEMU + swtpm? For myself, so far I've been testing on a recent Intel Chromebook. The H1 (aka cr50) security chip on modern chromebooks implements a subset [1] of TPM2.0, and is exposed through the standard TPM APIs in the kernel. I can make sure to test on Qemu as well, is there anything in particular I should look out for? -Evan [1] https://chromium-review.googlesource.com/c/chromiumos/third_party/tpm2/+/3373466 > > BR, Jarkko
On Thu, Aug 04, 2022 at 02:55:35PM -0700, Evan Green wrote: > On Wed, Aug 3, 2022 at 5:59 PM Jarkko Sakkinen <jarkko@kernel.org> wrote: > > > > On Tue, Aug 02, 2022 at 11:36:43AM -0700, Matthew Garrett wrote: > > > On Mon, Aug 1, 2022 at 3:33 PM Evan Green <evgreen@chromium.org> wrote: > > > > > > > One more bump here, as we'd really love to get encrypted hibernation > > > > to a form upstream would accept if at all possible. We were > > > > considering landing this in our Chrome OS tree for now, then coming > > > > back in a couple months with a "we've been baking this ourselves and > > > > it's going so great, oooh yeah". I'm not sure if upstream would find > > > > that compelling or not. But in any case, some guidance towards making > > > > this more upstream friendly would be well appreciated. > > > > > > > > One thing I realized in attempting to pick this myself is that the > > > > trusted key blob format has moved to ASN.1. So I should really move > > > > the creation ticket to the new ASN.1 format (if I can figure out the > > > > right OID for that piece), which would allow me to drop a lot of the > > > > ugly stuff in tpm2_unpack_blob(). Maybe if I get no other comments > > > > I'll work on that and resend. > > > > > > I've been revamping my TPM-backed verified hibernation implementation > > > based on this work, so I'd definitely be enthusiastic about it being > > > mergeable. > > > > BTW, is it tested with QEMU + swtpm? > > For myself, so far I've been testing on a recent Intel Chromebook. The > H1 (aka cr50) security chip on modern chromebooks implements a subset > [1] of TPM2.0, and is exposed through the standard TPM APIs in the > kernel. I can make sure to test on Qemu as well, is there anything in > particular I should look out for? I was just thinking what I could use for testing BR, Jarkko
We are exploring enabling hibernation in some new scenarios. However, our security team has a few requirements, listed below: 1. The hibernate image must be encrypted with protection derived from both the platform (eg TPM) and user authentication data (eg password). 2. Hibernation must not be a vector by which a malicious userspace can escalate to the kernel. Requirement #1 can be achieved solely with uswsusp, however requirement 2 necessitates mechanisms in the kernel to guarantee integrity of the hibernate image. The kernel needs a way to authenticate that it generated the hibernate image being loaded, and that the image has not been tampered with. Adding support for in-kernel AEAD encryption with a TPM-sealed key allows us to achieve both requirements with a single computation pass. Matthew Garrett published a series [1] that aligns closely with this goal. His series utilized the fact that PCR23 is a resettable PCR that can be blocked from access by usermode. The TPM can create a sealed key tied to PCR23 in two ways. First, the TPM can attest to the value of PCR23 when the key was created, which the kernel can use on resume to verify that the kernel must have created the key (since it is the only one capable of modifying PCR23). It can also create a policy that enforces PCR23 be set to a specific value as a condition of unsealing the key, preventing usermode from unsealing the key by talking directly to the TPM. This series adopts that primitive as a foundation, tweaking and building on it a bit. Where Matthew's series used the TPM-backed key to encrypt a hash of the image, this series uses the key directly as a gcm(aes) encryption key, which the kernel uses to encrypt and decrypt the hibernate image in chunks of 16 pages. This provides both encryption and integrity, which turns out to be a noticeable performance improvement over separate passes for encryption and hashing. The series also introduces the concept of mixing user key material into the encryption key. This allows usermode to introduce key material based on unspecified external authentication data (in our case derived from something like the user password or PIN), without requiring usermode to do a separate encryption pass. Matthew also documented issues his series had [2] related to generating fake images by booting alternate kernels without the PCR23 limiting. With access to PCR23 on the same machine, usermode can create fake hibernate images that are indistinguishable to the new kernel from genuine ones. His post outlines a solution that involves adding more PCRs into the creation data and policy, with some gyrations to make this work well on a standard PC. Our approach would be similar: on our machines PCR 0 indicates whether the system is booted in secure/verified mode or developer mode. By adding PCR0 to the policy, we can reject hibernate images made in developer mode while in verified mode (or vice versa). Additionally, mixing in the user authentication data limits both data exfiltration attacks (eg a stolen laptop) and forged hibernation image attacks to attackers that already know the authentication data (eg user's password). This, combined with our relatively sealed userspace (dm-verity on the rootfs), and some judicious clearing of the hibernate image (such as across an OS update) further reduce the risk of an online attack. The remaining attack space of a forgery from someone with physical access to the device and knowledge of the authentication data is out of scope for us, given that flipping to developer mode or reflashing RO firmware trivially achieves the same thing. A couple of patches still need to be written on top of this series. The generalized functionality to OR in additional PCRs via Kconfig (like PCR 0 or 5) still needs to be added. We'll also need a patch that disallows unencrypted forms of resume from hibernation, to fully close the door to malicious userspace. However, I wanted to get this series out first and get reactions from upstream before continuing to add to it. [1] https://patchwork.kernel.org/project/linux-pm/cover/20210220013255.1083202-1-matthewgarrett@google.com/ [2] https://mjg59.dreamwidth.org/58077.html Evan Green (6): security: keys: trusted: Verify creation data PM: hibernate: Add kernel-based encryption PM: hibernate: Use TPM-backed keys to encrypt image PM: hibernate: Mix user key in encrypted hibernate PM: hibernate: Verify the digest encryption key PM: hibernate: seal the encryption key with a PCR policy Matthew Garrett (4): tpm: Add support for in-kernel resetting of PCRs tpm: Allow PCR 23 to be restricted to kernel-only use security: keys: trusted: Parse out individual components of the key blob security: keys: trusted: Allow storage of PCR values in creation data Documentation/power/userland-swsusp.rst | 8 + .../security/keys/trusted-encrypted.rst | 4 + drivers/char/tpm/Kconfig | 10 + drivers/char/tpm/tpm-dev-common.c | 8 + drivers/char/tpm/tpm-interface.c | 28 + drivers/char/tpm/tpm.h | 23 + drivers/char/tpm/tpm1-cmd.c | 69 ++ drivers/char/tpm/tpm2-cmd.c | 58 + drivers/char/tpm/tpm2-space.c | 2 +- include/keys/trusted-type.h | 9 + include/linux/tpm.h | 12 + include/uapi/linux/suspend_ioctls.h | 28 +- kernel/power/Kconfig | 15 + kernel/power/Makefile | 1 + kernel/power/power.h | 1 + kernel/power/snapenc.c | 1076 +++++++++++++++++ kernel/power/snapshot.c | 5 + kernel/power/user.c | 44 +- kernel/power/user.h | 114 ++ security/keys/trusted-keys/trusted_tpm1.c | 9 + security/keys/trusted-keys/trusted_tpm2.c | 164 ++- 21 files changed, 1670 insertions(+), 18 deletions(-) create mode 100644 kernel/power/snapenc.c create mode 100644 kernel/power/user.h