]> git.proxmox.com Git - mirror_edk2.git/commit
OvmfPkg/IoMmuDxe: abort harder on memory encryption mask failures
authorLaszlo Ersek <lersek@redhat.com>
Wed, 2 Aug 2017 17:35:38 +0000 (19:35 +0200)
committerLaszlo Ersek <lersek@redhat.com>
Fri, 4 Aug 2017 23:31:53 +0000 (01:31 +0200)
commitf1658838c267723139711c0b15d98a74980ae4c5
treeb6d9bebcfe96d2338b7aae3d933b5379a7d687dc
parent58e681406f8fd59fad3f20ec3737549d4237e847
OvmfPkg/IoMmuDxe: abort harder on memory encryption mask failures

Upon a MemEncryptSevClearPageEncMask() failure in Map(), it wouldn't be
difficult to release the bounce buffer that was implicitly allocated for
BusMasterRead[64] and BusMasterWrite[64] operations. However, undoing any
partial memory encryption mask changes -- partial page splitting and PTE
modifications -- is practically impossible. (For example, restoring the
encryption mask on the entire range has no reason to fare any better than
the MemEncryptSevClearPageEncMask() call itself.)

For this reason, keep ASSERT_EFI_ERROR(), but hang in RELEASE builds too,
if MemEncryptSevClearPageEncMask() or MemEncryptSevSetPageEncMask() fails.

Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc: Brijesh Singh <brijesh.singh@amd.com>
Cc: Jordan Justen <jordan.l.justen@intel.com>
Cc: Tom Lendacky <thomas.lendacky@amd.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Tested-by: Brijesh Singh <brijesh.singh@amd.com>
Reviewed-by: Brijesh Singh <brijesh.singh@amd.com>
OvmfPkg/IoMmuDxe/AmdSevIoMmu.c