fix field order for INVPCID descriptor #508
Merged
+2
−2
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Both Intel's and AMD's manuals describe the INVPCID descriptor as a 128-bit value with the linear address stored in the upper half and the PCID stored in the lower half. x86 uses little-endian byte ordering and so the lower half (pcid) should be stored before the upper half (address). Previously, our `InvpcidDescriptor` type stored the halves in the opposite order with the address before the pcid. This patch fixes the order, so that the pcid is stored before the address. This new order is also the order used by Linux and OpenBSD: https://github.com/torvalds/linux/blob/3e5e6c9900c3d71895e8bdeacfb579462e98eba1/arch/x86/include/asm/invpcid.h#L8 https://github.com/openbsd/src/blob/4e368faebf86e9a349446b5839c333bc17bd3f9a/sys/arch/amd64/include/cpufunc.h#L173 It's beyond me how this wasn't noticed earlier. The previous incorrect layout could lead to TLB entries not being flushed and #GP faults.
|
I briefly changed the branch protection rules, so that I could merge this PR without CI passing (there's already a fix for CI in #507). |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Both Intel's and AMD's manuals describe the INVPCID descriptor as a 128-bit value with the linear address stored in the upper half and the PCID stored in the lower half. x86 uses little-endian byte ordering and so the lower half (pcid) should be stored before the upper half (address). Previously, our
InvpcidDescriptortype stored the halves in the opposite order with the address before the pcid.This patch fixes the order, so that the pcid is stored before the address. This new order is also the order used by Linux and OpenBSD:
https://github.com/torvalds/linux/blob/3e5e6c9900c3d71895e8bdeacfb579462e98eba1/arch/x86/include/asm/invpcid.h#L8
https://github.com/openbsd/src/blob/4e368faebf86e9a349446b5839c333bc17bd3f9a/sys/arch/amd64/include/cpufunc.h#L173
It's beyond me how this wasn't noticed earlier. The previous incorrect layout could lead to TLB entries not being flushed and #GP faults.