From: "Philippe Mathieu-Daudé" <philmd@redhat.com>
To: devel@edk2.groups.io, lersek@redhat.com
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>,
Gerd Hoffmann <kraxel@redhat.com>,
Jordan Justen <jordan.l.justen@intel.com>
Subject: Re: [edk2-devel] [PATCH for-edk2-stable201905 6/6] OvmfPkg/PlatformPei: set 32-bit UC area at PciBase / PciExBarBase (pc/q35)
Date: Mon, 3 Jun 2019 13:07:22 +0200 [thread overview]
Message-ID: <c257e1e5-29f4-ce9f-644b-a5b191483577@redhat.com> (raw)
In-Reply-To: <20190529151209.17503-7-lersek@redhat.com>
On 5/29/19 5:12 PM, Laszlo Ersek wrote:
> (This is a replacement for commit 39b9a5ffe661 ("OvmfPkg/PlatformPei: fix
> MTRR for low-RAM sizes that have many bits clear", 2019-05-16).)
>
> Reintroduce the same logic as seen in commit 39b9a5ffe661 for the pc
> (i440fx) board type.
>
> For q35, the same approach doesn't work any longer, given that (a) we'd
> like to keep the PCIEXBAR in the platform DSC a fixed-at-build PCD, and
> (b) QEMU expects the PCIEXBAR to reside at a lower address than the 32-bit
> PCI MMIO aperture.
>
> Therefore, introduce a helper function for determining the 32-bit
> "uncacheable" (MMIO) area base address:
>
> - On q35, this function behaves statically. Furthermore, the MTRR setup
> exploits that the range [0xB000_0000, 0xFFFF_FFFF] can be marked UC with
> just two variable MTRRs (one at 0xB000_0000 (size 256MB), another at
> 0xC000_0000 (size 1GB)).
>
> - On pc (i440fx), the function behaves dynamically, implementing the same
> logic as commit 39b9a5ffe661 did. The PciBase value is adjusted to the
> value calculated, similarly to commit 39b9a5ffe661. A further
> simplification is that we show that the UC32 area size truncation to a
> whole power of two automatically guarantees a >=2GB base address.
>
> Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
> Cc: Gerd Hoffmann <kraxel@redhat.com>
> Cc: Jordan Justen <jordan.l.justen@intel.com>
> Ref: https://bugzilla.tianocore.org/show_bug.cgi?id=1859
> Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Philippe Mathieu-Daude <philmd@redhat.com>
> ---
> OvmfPkg/PlatformPei/Platform.h | 7 +++
> OvmfPkg/PlatformPei/MemDetect.c | 59 ++++++++++++++++++--
> OvmfPkg/PlatformPei/Platform.c | 5 +-
> 3 files changed, 66 insertions(+), 5 deletions(-)
>
> diff --git a/OvmfPkg/PlatformPei/Platform.h b/OvmfPkg/PlatformPei/Platform.h
> index 81af8b71480f..2f3cebcd3a6a 100644
> --- a/OvmfPkg/PlatformPei/Platform.h
> +++ b/OvmfPkg/PlatformPei/Platform.h
> @@ -62,6 +62,11 @@ GetSystemMemorySizeBelow4gb (
> VOID
> );
>
> +VOID
> +QemuUc32BaseInitialization (
> + VOID
> + );
> +
> VOID
> InitializeRamRegions (
> VOID
> @@ -114,4 +119,6 @@ extern UINT32 mMaxCpuCount;
>
> extern UINT16 mHostBridgeDevId;
>
> +extern UINT32 mQemuUc32Base;
> +
> #endif // _PLATFORM_PEI_H_INCLUDED_
> diff --git a/OvmfPkg/PlatformPei/MemDetect.c b/OvmfPkg/PlatformPei/MemDetect.c
> index e890e36408a6..d451989f31c9 100644
> --- a/OvmfPkg/PlatformPei/MemDetect.c
> +++ b/OvmfPkg/PlatformPei/MemDetect.c
> @@ -14,6 +14,7 @@ Module Name:
> // The package level header files this module uses
> //
> #include <IndustryStandard/E820.h>
> +#include <IndustryStandard/I440FxPiix4.h>
> #include <IndustryStandard/Q35MchIch9.h>
> #include <PiPei.h>
>
> @@ -42,6 +43,8 @@ STATIC UINT32 mS3AcpiReservedMemorySize;
>
> STATIC UINT16 mQ35TsegMbytes;
>
> +UINT32 mQemuUc32Base;
> +
> VOID
> Q35TsegMbytesInitialization (
> VOID
> @@ -98,6 +101,54 @@ Q35TsegMbytesInitialization (
> }
>
>
> +VOID
> +QemuUc32BaseInitialization (
> + VOID
> + )
> +{
> + UINT32 LowerMemorySize;
> + UINT32 Uc32Size;
> +
> + if (mXen) {
> + return;
> + }
> +
> + if (mHostBridgeDevId == INTEL_Q35_MCH_DEVICE_ID) {
> + //
> + // On q35, the 32-bit area that we'll mark as UC, through variable MTRRs,
> + // starts at PcdPciExpressBaseAddress. The platform DSC is responsible for
> + // setting PcdPciExpressBaseAddress such that describing the
> + // [PcdPciExpressBaseAddress, 4GB) range require a very small number of
> + // variable MTRRs (preferably 1 or 2).
> + //
> + ASSERT (FixedPcdGet64 (PcdPciExpressBaseAddress) <= MAX_UINT32);
> + mQemuUc32Base = (UINT32)FixedPcdGet64 (PcdPciExpressBaseAddress);
> + return;
> + }
> +
> + ASSERT (mHostBridgeDevId == INTEL_82441_DEVICE_ID);
> + //
> + // On i440fx, start with the [LowerMemorySize, 4GB) range. Make sure one
> + // variable MTRR suffices by truncating the size to a whole power of two,
> + // while keeping the end affixed to 4GB. This will round the base up.
> + //
> + LowerMemorySize = GetSystemMemorySizeBelow4gb ();
> + Uc32Size = GetPowerOfTwo32 ((UINT32)(SIZE_4GB - LowerMemorySize));
> + mQemuUc32Base = (UINT32)(SIZE_4GB - Uc32Size);
> + //
> + // Assuming that LowerMemorySize is at least 1 byte, Uc32Size is at most 2GB.
> + // Therefore mQemuUc32Base is at least 2GB.
> + //
> + ASSERT (mQemuUc32Base >= BASE_2GB);
> +
> + if (mQemuUc32Base != LowerMemorySize) {
> + DEBUG ((DEBUG_VERBOSE, "%a: rounded UC32 base from 0x%x up to 0x%x, for "
> + "an UC32 size of 0x%x\n", __FUNCTION__, LowerMemorySize, mQemuUc32Base,
> + Uc32Size));
> + }
> +}
> +
> +
> /**
> Iterate over the RAM entries in QEMU's fw_cfg E820 RAM map that start outside
> of the 32-bit address range.
> @@ -688,11 +739,11 @@ QemuInitializeRam (
> ASSERT_EFI_ERROR (Status);
>
> //
> - // Set memory range from the "top of lower RAM" (RAM below 4GB) to 4GB as
> - // uncacheable
> + // Set the memory range from the start of the 32-bit MMIO area (32-bit PCI
> + // MMIO aperture on i440fx, PCIEXBAR on q35) to 4GB as uncacheable.
> //
> - Status = MtrrSetMemoryAttribute (LowerMemorySize,
> - SIZE_4GB - LowerMemorySize, CacheUncacheable);
> + Status = MtrrSetMemoryAttribute (mQemuUc32Base, SIZE_4GB - mQemuUc32Base,
> + CacheUncacheable);
> ASSERT_EFI_ERROR (Status);
> }
> }
> diff --git a/OvmfPkg/PlatformPei/Platform.c b/OvmfPkg/PlatformPei/Platform.c
> index 0876316eefbc..3ba2459872d9 100644
> --- a/OvmfPkg/PlatformPei/Platform.c
> +++ b/OvmfPkg/PlatformPei/Platform.c
> @@ -191,7 +191,8 @@ MemMapInitialization (
> ASSERT (PciExBarBase <= MAX_UINT32 - SIZE_256MB);
> PciBase = (UINT32)(PciExBarBase + SIZE_256MB);
> } else {
> - PciBase = (TopOfLowRam < BASE_2GB) ? BASE_2GB : TopOfLowRam;
> + ASSERT (TopOfLowRam <= mQemuUc32Base);
> + PciBase = mQemuUc32Base;
> }
>
> //
> @@ -650,6 +651,8 @@ InitializePlatform (
>
> PublishPeiMemory ();
>
> + QemuUc32BaseInitialization ();
> +
> InitializeRamRegions ();
>
> if (mXen) {
>
next prev parent reply other threads:[~2019-06-03 11:07 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-05-29 15:12 [PATCH for-edk2-stable201905 0/6] work around a QEMU issue triggered by the original TianoCore#1814 fix Laszlo Ersek
2019-05-29 15:12 ` [PATCH for-edk2-stable201905 1/6] Revert "OvmfPkg/PlatformPei: fix MTRR for low-RAM sizes that have many bits clear" Laszlo Ersek
2019-05-29 15:24 ` [edk2-devel] " Philippe Mathieu-Daudé
2019-05-29 15:12 ` [PATCH for-edk2-stable201905 2/6] Revert "OvmfPkg/PlatformPei: reorder the 32-bit PCI window vs. the PCIEXBAR on q35" Laszlo Ersek
2019-05-29 15:24 ` [edk2-devel] " Philippe Mathieu-Daudé
2019-05-29 15:12 ` [PATCH for-edk2-stable201905 3/6] Revert "OvmfPkg/PlatformPei: hoist PciBase assignment above the i440fx/q35 branching" Laszlo Ersek
2019-05-29 15:25 ` [edk2-devel] " Philippe Mathieu-Daudé
2019-05-29 15:12 ` [PATCH for-edk2-stable201905 4/6] Revert "OvmfPkg/PlatformPei: assign PciSize on both i440fx/q35 branches explicitly" Laszlo Ersek
2019-05-29 15:25 ` [edk2-devel] " Philippe Mathieu-Daudé
2019-05-29 15:12 ` [PATCH for-edk2-stable201905 5/6] OvmfPkg: raise the PCIEXBAR base to 2816 MB on Q35 Laszlo Ersek
2019-05-29 16:36 ` [edk2-devel] " Philippe Mathieu-Daudé
2019-05-29 15:12 ` [PATCH for-edk2-stable201905 6/6] OvmfPkg/PlatformPei: set 32-bit UC area at PciBase / PciExBarBase (pc/q35) Laszlo Ersek
2019-06-03 11:07 ` Philippe Mathieu-Daudé [this message]
2019-05-29 15:20 ` [PATCH for-edk2-stable201905 0/6] work around a QEMU issue triggered by the original TianoCore#1814 fix Ard Biesheuvel
2019-06-03 18:10 ` [edk2-devel] " Laszlo Ersek
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-list from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=c257e1e5-29f4-ce9f-644b-a5b191483577@redhat.com \
--to=devel@edk2.groups.io \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox