From: Leif Lindholm <leif.lindholm@linaro.org>
To: Heyi Guo <heyi.guo@linaro.org>
Cc: linaro-uefi@lists.linaro.org, edk2-devel@lists.01.org,
graeme.gregory@linaro.org, ard.biesheuvel@linaro.org,
guoheyi@huawei.com, wanghuiqiang@huawei.com,
huangming23@huawei.com, zhangjinsong2@huawei.com, waip23@126.com,
Ming Huang <waip23@foxmail.com>
Subject: Re: [PATCH edk2-platforms v3 11/11] Hisilicon D03/D05: Enlarge iATU for RP with ARI capable device.
Date: Thu, 21 Sep 2017 14:14:50 +0100 [thread overview]
Message-ID: <20170921131450.kshvdxp6lybmgtkw@bivouac.eciton.net> (raw)
In-Reply-To: <1505991597-52989-19-git-send-email-heyi.guo@linaro.org>
On Thu, Sep 21, 2017 at 06:59:57PM +0800, Heyi Guo wrote:
> From: Ming Huang <waip23@foxmail.com>
>
> 1. Because Hi161x chip doesn't support "ARI Forwarding Enable"
> function, BIOS will enumerate 32 same devices (Device Number 0~31)
> when attach a Non-ARI capable device in the RP. Hi161x chip will
> not fix it, need BIOS patch.
> 2. Just enlarge iatu for those root port with ARI capable device
> attached, Non-ARI capable device's RP, keep iatu limitation.
> 3. Remove previous temporary solution as below commit id:
> "7d157da88852cc91df2b11b10ade2edbbfbe77da"
>
> Contributed-under: TianoCore Contribution Agreement 1.1
> Signed-off-by: Jason zhang <zhangjinsong2@huawei.com>
Please adjust + add email addresses.
The content below is fine.
/
Leif
> ---
> Silicon/Hisilicon/Drivers/PciHostBridgeDxe/PciHostBridge.c | 1 +
> Silicon/Hisilicon/Drivers/PciHostBridgeDxe/PciHostBridge.h | 7 ++
> Silicon/Hisilicon/Drivers/PciHostBridgeDxe/PciRootBridgeIo.c | 79 ++++++++++++++++++++
> 3 files changed, 87 insertions(+)
>
> diff --git a/Silicon/Hisilicon/Drivers/PciHostBridgeDxe/PciHostBridge.c b/Silicon/Hisilicon/Drivers/PciHostBridgeDxe/PciHostBridge.c
> index e3d3988..9fa3f84 100644
> --- a/Silicon/Hisilicon/Drivers/PciHostBridgeDxe/PciHostBridge.c
> +++ b/Silicon/Hisilicon/Drivers/PciHostBridgeDxe/PciHostBridge.c
> @@ -839,6 +839,7 @@ NotifyPhase(
>
> case EfiPciHostBridgeEndEnumeration:
> PCIE_DEBUG("Case EfiPciHostBridgeEndEnumeration\n");
> + EnlargeAtuConfig0 (This);
> break;
>
> case EfiPciHostBridgeBeginBusAllocation:
> diff --git a/Silicon/Hisilicon/Drivers/PciHostBridgeDxe/PciHostBridge.h b/Silicon/Hisilicon/Drivers/PciHostBridgeDxe/PciHostBridge.h
> index cddda6b..c04361f 100644
> --- a/Silicon/Hisilicon/Drivers/PciHostBridgeDxe/PciHostBridge.h
> +++ b/Silicon/Hisilicon/Drivers/PciHostBridgeDxe/PciHostBridge.h
> @@ -401,6 +401,9 @@ PreprocessController (
> #define EFI_RESOURCE_NONEXISTENT 0xFFFFFFFFFFFFFFFFULL
> #define EFI_RESOURCE_LESS 0xFFFFFFFFFFFFFFFEULL
>
> +#define INVALID_CAPABILITY_00 0x00
> +#define INVALID_CAPABILITY_FF 0xFF
> +#define PCI_CAPABILITY_POINTER_MASK 0xFC
>
> //
> // Driver Instance Data Prototypes
> @@ -518,4 +521,8 @@ RootBridgeConstructor (
> IN UINT32 Seg
> );
>
> +VOID
> +EnlargeAtuConfig0 (
> + IN EFI_PCI_HOST_BRIDGE_RESOURCE_ALLOCATION_PROTOCOL *This
> + );
> #endif
> diff --git a/Silicon/Hisilicon/Drivers/PciHostBridgeDxe/PciRootBridgeIo.c b/Silicon/Hisilicon/Drivers/PciHostBridgeDxe/PciRootBridgeIo.c
> index 10d766a..b57bd51 100644
> --- a/Silicon/Hisilicon/Drivers/PciHostBridgeDxe/PciRootBridgeIo.c
> +++ b/Silicon/Hisilicon/Drivers/PciHostBridgeDxe/PciRootBridgeIo.c
> @@ -14,6 +14,7 @@
> **/
>
> #include "PciHostBridge.h"
> +#include <IndustryStandard/PciExpress30.h>
> #include <Library/DevicePathLib.h>
> #include <Library/DmaLib.h>
> #include <Library/PciExpressLib.h>
> @@ -2322,3 +2323,81 @@ RootBridgeIoConfiguration (
> return EFI_SUCCESS;
> }
>
> +BOOLEAN
> +PcieCheckAriFwdEn (
> + UINTN PciBaseAddr
> + )
> +{
> + UINT8 PciPrimaryStatus;
> + UINT8 CapabilityOffset;
> + UINT8 CapId;
> + UINT8 TempData;
> +
> + PciPrimaryStatus = MmioRead16 (PciBaseAddr + PCI_PRIMARY_STATUS_OFFSET);
> +
> + if (PciPrimaryStatus & EFI_PCI_STATUS_CAPABILITY) {
> + CapabilityOffset = MmioRead8 (PciBaseAddr + PCI_CAPBILITY_POINTER_OFFSET);
> + CapabilityOffset &= PCI_CAPABILITY_POINTER_MASK;
> +
> + while ((CapabilityOffset != INVALID_CAPABILITY_00) && (CapabilityOffset != INVALID_CAPABILITY_FF)) {
> + CapId = MmioRead8 (PciBaseAddr + CapabilityOffset);
> + if (CapId == EFI_PCI_CAPABILITY_ID_PCIEXP) {
> + break;
> + }
> + CapabilityOffset = MmioRead8 (PciBaseAddr + CapabilityOffset + 1);
> + CapabilityOffset &= PCI_CAPABILITY_POINTER_MASK;
> + }
> + } else {
> + PCIE_DEBUG ("[%a:%d] - No PCIE Capability.\n", __FUNCTION__, __LINE__);
> + return FALSE;
> + }
> +
> + if ((CapabilityOffset == INVALID_CAPABILITY_FF) || (CapabilityOffset == INVALID_CAPABILITY_00)) {
> + PCIE_DEBUG ("[%a:%d] - No PCIE Capability.\n", __FUNCTION__, __LINE__);
> + return FALSE;
> + }
> +
> + TempData = MmioRead16 (PciBaseAddr + CapabilityOffset +
> + EFI_PCIE_CAPABILITY_DEVICE_CONTROL_2_OFFSET);
> + TempData &= EFI_PCIE_CAPABILITY_DEVICE_CAPABILITIES_2_ARI_FORWARDING;
> +
> + if (TempData == EFI_PCIE_CAPABILITY_DEVICE_CAPABILITIES_2_ARI_FORWARDING) {
> + return TRUE;
> + } else {
> + return FALSE;
> + }
> +}
> +
> +VOID
> +EnlargeAtuConfig0 (
> + IN EFI_PCI_HOST_BRIDGE_RESOURCE_ALLOCATION_PROTOCOL *This
> + )
> +{
> + UINTN RbPciBase;
> + UINT64 MemLimit;
> + LIST_ENTRY *List;
> + PCI_HOST_BRIDGE_INSTANCE *HostBridgeInstance;
> + PCI_ROOT_BRIDGE_INSTANCE *RootBridgeInstance;
> +
> + PCIE_DEBUG ("In Enlarge RP iatu Config 0.\n");
> +
> + HostBridgeInstance = INSTANCE_FROM_RESOURCE_ALLOCATION_THIS (This);
> + List = HostBridgeInstance->Head.ForwardLink;
> +
> + while (List != &HostBridgeInstance->Head) {
> + PCIE_DEBUG ("HostBridge has data.\n");
> + RootBridgeInstance = DRIVER_INSTANCE_FROM_LIST_ENTRY (List);
> +
> + RbPciBase = RootBridgeInstance->RbPciBar;
> +
> + // Those ARI FWD Enable Root Bridge, need enlarge iatu window.
> + if (PcieCheckAriFwdEn (RbPciBase)) {
> + MemLimit = GetPcieCfgAddress (RootBridgeInstance->Ecam,
> + RootBridgeInstance->BusBase + 2, 0, 0, 0)
> + - 1;
> + MmioWrite32 (RbPciBase + IATU_OFFSET + IATU_VIEW_POINT, 1);
> + MmioWrite32 (RbPciBase + IATU_OFFSET + IATU_REGION_BASE_LIMIT, (UINT32) MemLimit);
> + }
> + List = List->ForwardLink;
> + }
> +}
> --
> 1.9.1
>
next prev parent reply other threads:[~2017-09-21 13:11 UTC|newest]
Thread overview: 41+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-09-21 10:59 [PATCH edk2-platforms v3 00/11] Update D03/D05 binary for edk2 update and bug fix Heyi Guo
2017-09-21 10:59 ` [PATCH edk2-non-osi v3 1/7] Hisilicon/D03/Net: Update Snp driver Heyi Guo
2017-09-21 12:51 ` Leif Lindholm
2017-09-21 10:59 ` [PATCH edk2-platforms v3 01/11] Hisilicon/D05: Modify dsc and fdf file Heyi Guo
2017-09-21 13:01 ` Leif Lindholm
2017-09-21 10:59 ` [PATCH edk2-platforms v3 02/11] Hisilicon/D03: " Heyi Guo
2017-09-21 13:02 ` Leif Lindholm
2017-09-21 10:59 ` [PATCH edk2-non-osi v3 2/7] Hisilicon/D03/Sas: Add SasPlatform Heyi Guo
2017-09-21 12:49 ` Leif Lindholm
[not found] ` <3A622B96E322004395454DF73A38DDFA75577DB2@dggemm508-mbx.china.huawei.com>
2017-09-23 16:52 ` 答复: " Leif Lindholm
2017-09-21 10:59 ` [PATCH edk2-non-osi v3 3/7] Hisilicon/D03: Update binary file Heyi Guo
2017-09-21 12:54 ` Leif Lindholm
2017-09-21 10:59 ` [PATCH edk2-platforms v3 03/11] Hisilicon: Fix the drivers use the same GUID issue Heyi Guo
2017-09-21 10:59 ` [PATCH edk2-platforms v3 04/11] Hisilicon D03/D05: get firmware version from FIRMWARE_VER Heyi Guo
2017-09-21 13:03 ` Leif Lindholm
2017-09-21 10:59 ` [PATCH edk2-non-osi v3 4/7] Hisilicon/D05/Net: Update Snp driver Heyi Guo
2017-09-21 12:56 ` Leif Lindholm
2017-09-21 10:59 ` [PATCH edk2-non-osi v3 5/7] Hisilicon/D05/Sas: Add SasPlatform Heyi Guo
2017-09-21 13:00 ` Leif Lindholm
2017-09-21 10:59 ` [PATCH edk2-platforms v3 05/11] Hisilicon/PciHostBridgeDxe: Assign BAR resource from PciRegionBase Heyi Guo
2017-09-21 12:57 ` Leif Lindholm
2017-09-21 10:59 ` [PATCH edk2-platforms v3 06/11] Hisilicon/D05/Pcie: fix bug of size definition Heyi Guo
2017-09-21 13:04 ` Leif Lindholm
2017-09-21 10:59 ` [PATCH edk2-non-osi v3 6/7] Hisilicon/D05: Update binary file Heyi Guo
2017-09-21 12:59 ` Leif Lindholm
[not found] ` <3A622B96E322004395454DF73A38DDFA75577D70@dggemm508-mbx.china.huawei.com>
2017-09-22 9:55 ` 答复: " Leif Lindholm
2017-09-21 10:59 ` [PATCH edk2-platforms v3 07/11] D05/PCIe: Modify PcieRegionBase of secondary chip Heyi Guo
2017-09-21 13:04 ` Leif Lindholm
2017-09-21 10:59 ` [PATCH edk2-non-osi v3 7/7] Hisilicon: Fix the drivers use the same GUID issue Heyi Guo
2017-09-21 10:59 ` [PATCH edk2-platforms v3 08/11] Hisilicon/D03: Disable the function of PerfTuning Heyi Guo
2017-09-21 13:07 ` Leif Lindholm
2017-09-21 10:59 ` [PATCH edk2-platforms v3 09/11] D05/ACPI: Disable D05 SAS0 and SAS2 Heyi Guo
2017-09-21 13:11 ` Leif Lindholm
2017-09-21 10:59 ` [PATCH edk2-platforms v3 10/11] D05/ACPI: Modify I2C device Heyi Guo
2017-09-21 13:12 ` Leif Lindholm
2017-09-21 10:59 ` [PATCH edk2-platforms v3 11/11] Hisilicon D03/D05: Enlarge iATU for RP with ARI capable device Heyi Guo
2017-09-21 13:14 ` Leif Lindholm [this message]
2017-09-21 12:11 ` [PATCH edk2-platforms v3 00/11] Update D03/D05 binary for edk2 update and bug fix graeme.gregory
2017-09-21 12:40 ` Leif Lindholm
2017-09-21 13:32 ` Ard Biesheuvel
2017-09-22 3:20 ` Heyi Guo
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=20170921131450.kshvdxp6lybmgtkw@bivouac.eciton.net \
--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