summaryrefslogtreecommitdiff
path: root/arch/arm64/kvm/check-res-bits.h
AgeCommit message (Collapse)Author
2024-03-15Revert "KVM: arm64: Snapshot all non-zero RES0/RES1 sysreg fields for later ↵Oliver Upton
checking" This reverts commits 99101dda29e3186b1356b0dc4dbb835c02c71ac9 and b80b701d5a67d07f4df4a21e09cb31f6bc1feeca. Linus reports that the sysreg reserved bit checks in KVM have led to build failures, arising from commit fdd867fe9b32 ("arm64/sysreg: Add register fields for ID_AA64DFR1_EL1") giving meaning to fields that were previously RES0. Of course, this is a genuine issue, since KVM's sysreg emulation depends heavily on the definition of reserved fields. But at this point the build breakage is far more offensive, and the right course of action is to revert and retry later. All of these build-time assertions were on by default before commit 99101dda29e3 ("KVM: arm64: Make build-time check of RES0/RES1 bits optional"), so deliberately revert it all atomically to avoid introducing further breakage of bisection. Link: https://lore.kernel.org/all/CAHk-=whCvkhc8BbFOUf1ddOsgSGgEjwoKv77=HEY1UiVCydGqw@mail.gmail.com/ Acked-by: Marc Zyngier <maz@kernel.org> Signed-off-by: Oliver Upton <oliver.upton@linux.dev> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2024-02-22KVM: arm64: Make build-time check of RES0/RES1 bits optionalMarc Zyngier
In order to ease the transition towards a state of absolute paranoia where all RES0/RES1 bits gets checked against what KVM know of them, make the checks optional and guarded by a config symbol (CONFIG_KVM_ARM64_RES_BITS_PARANOIA) default to n. Signed-off-by: Marc Zyngier <maz@kernel.org> Link: https://lore.kernel.org/kvm/87frxka7ud.wl-maz@kernel.org/ Signed-off-by: Oliver Upton <oliver.upton@linux.dev>
2024-02-19KVM: arm64: Snapshot all non-zero RES0/RES1 sysreg fields for later checkingMarc Zyngier
As KVM now strongly relies on accurately handling the RES0/RES1 bits on a number of paths, add a compile-time checker that will blow in the face of the innocent bystander, should they try to sneak in an update that changes any of these RES0/RES1 fields. It is expected that such an update will come with the relevant KVM update if needed. Signed-off-by: Marc Zyngier <maz@kernel.org> Link: https://lore.kernel.org/r/20240214131827.2856277-26-maz@kernel.org Signed-off-by: Oliver Upton <oliver.upton@linux.dev>