On 01/11/19 06:46, Eric Dong wrote:
> ((Facs->Flags & EFI_ACPI_4_0_OSPM_64BIT_WAKE__F) != 0))
>
> In above code, Facs->OspmFlags should be used instead.
> EFI_ACPI_4_0_OSPM_64BIT_WAKE__F is a bit in OSPM Enabled Firmware
> Control Structure Flags field, not in Firmware Control Structure
> Feature Flags.
>
> BZ: https://bugzilla.tianocore.org/show_bug.cgi?id=1430
I've now checked the ACPI 6.2 spec, and the bug report is correct.
(1) The subject line of the patch is unhelpful. Instead, please use:
MdeModulePkg/BootScriptExecuteorDxe: check 64BIT_WAKE_F in FACS.OSPMFlags
This is 73 characters, uses names from the ACPI spec, and describes the
change better.
> Cc: Aleksiy <oleks...@ami.com>
> Cc: Jian Wang <jian.j.w...@intel.com>
> Contributed-under: TianoCore Contribution Agreement 1.1
> signed-off-by: Eric Dong <eric.d...@intel.com>
(2) Please remove this line; it is both mis-spelled (due to the
lower-case "s"), and the next line makes it superfluous.
> Signed-off-by: Eric Dong <eric.d...@intel.com>
> ---
> MdeModulePkg/Universal/Acpi/BootScriptExecutorDxe/ScriptExecute.c | 4 ++--
> MdeModulePkg/Universal/Acpi/BootScriptExecutorDxe/X64/SetIdtEntry.c | 4 ++--
> 2 files changed, 4 insertions(+), 4 deletions(-)
>
> diff --git
> a/MdeModulePkg/Universal/Acpi/BootScriptExecutorDxe/ScriptExecute.c
> b/MdeModulePkg/Universal/Acpi/BootScriptExecutorDxe/ScriptExecute.c
> index e76abb7b7b..13af957a4d 100644
> --- a/MdeModulePkg/Universal/Acpi/BootScriptExecutorDxe/ScriptExecute.c
> +++ b/MdeModulePkg/Universal/Acpi/BootScriptExecutorDxe/ScriptExecute.c
> @@ -4,7 +4,7 @@
> This driver is dispatched by Dxe core and the driver will reload itself to
> ACPI reserved memory
> in the entry point. The functionality is to interpret and restore the S3
> boot script
>
> -Copyright (c) 2006 - 2018, Intel Corporation. All rights reserved.<BR>
> +Copyright (c) 2006 - 2019, Intel Corporation. All rights reserved.<BR>
> Copyright (c) 2017, AMD Incorporated. All rights reserved.<BR>
>
> This program and the accompanying materials
> @@ -156,7 +156,7 @@ S3BootScriptExecutorEntryFunction (
> TempStackTop = (UINTN)&TempStack + sizeof(TempStack);
> if ((Facs->Version ==
> EFI_ACPI_4_0_FIRMWARE_ACPI_CONTROL_STRUCTURE_VERSION) &&
> ((Facs->Flags & EFI_ACPI_4_0_64BIT_WAKE_SUPPORTED_F) != 0) &&
> - ((Facs->Flags & EFI_ACPI_4_0_OSPM_64BIT_WAKE__F) != 0)) {
> + ((Facs->OspmFlags & EFI_ACPI_4_0_OSPM_64BIT_WAKE__F) != 0)) {
> //
> // X64 long mode waking vector
> //
> diff --git
> a/MdeModulePkg/Universal/Acpi/BootScriptExecutorDxe/X64/SetIdtEntry.c
> b/MdeModulePkg/Universal/Acpi/BootScriptExecutorDxe/X64/SetIdtEntry.c
> index 1c6bb47b60..7b9627d579 100644
> --- a/MdeModulePkg/Universal/Acpi/BootScriptExecutorDxe/X64/SetIdtEntry.c
> +++ b/MdeModulePkg/Universal/Acpi/BootScriptExecutorDxe/X64/SetIdtEntry.c
> @@ -3,7 +3,7 @@
>
> Set a IDT entry for interrupt vector 3 for debug purpose for x64 platform
>
> -Copyright (c) 2006 - 2018, Intel Corporation. All rights reserved.<BR>
> +Copyright (c) 2006 - 2019, Intel Corporation. All rights reserved.<BR>
> Copyright (c) 2017, AMD Incorporated. All rights reserved.<BR>
>
>
> @@ -117,7 +117,7 @@ IsLongModeWakingVector (
> if (Facs->XFirmwareWakingVector != 0) {
> if ((Facs->Version ==
> EFI_ACPI_4_0_FIRMWARE_ACPI_CONTROL_STRUCTURE_VERSION) &&
> ((Facs->Flags & EFI_ACPI_4_0_64BIT_WAKE_SUPPORTED_F) != 0) &&
> - ((Facs->Flags & EFI_ACPI_4_0_OSPM_64BIT_WAKE__F) != 0)) {
> + ((Facs->OspmFlags & EFI_ACPI_4_0_OSPM_64BIT_WAKE__F) != 0)) {
> // Both BIOS and OS wants 64bit vector
> if (FeaturePcdGet (PcdDxeIplSwitchToLongMode)) {
> return TRUE;
>
Otherwise, the patch looks correct; all 64BIT_WAKE_F mentions are
covered, under BootScriptExecuteorDxe. (In fact, under all of
MdeModulePkg.)
The patch is expected to make no difference on QEMU / OVMF, because QEMU
exports a revision 1 FACS table for OVMF to install, and the expressions
that are modified by the patch are never evaluated. But, I'll report
back with test results too.
With (1) and (2) addressed:
Reviewed-by: Laszlo Ersek <ler...@redhat.com>
Thanks,
Laszlo
_______________________________________________
edk2-devel mailing list
edk2-devel@lists.01.org
https://lists.01.org/mailman/listinfo/edk2-devel
