From: "Laszlo Ersek" <lersek@redhat.com>
To: devel@edk2.groups.io, cenjiahui@huawei.com
Cc: Jordan Justen <jordan.l.justen@intel.com>,
Ard Biesheuvel <ard.biesheuvel@arm.com>,
Rebecca Cran <rebecca@bsdio.com>,
Peter Grehan <grehan@freebsd.org>,
Anthony Perard <anthony.perard@citrix.com>,
Julien Grall <julien@xen.org>, Leif Lindholm <leif@nuviainc.com>,
Sami Mujawar <sami.mujawar@arm.com>,
xieyingtai@huawei.com, wu.wubin@huawei.com
Subject: Re: [edk2-devel] [PATCH v6 09/11] OvmfPkg/PciHostBridgeUtilityLib: Extend GetRootBridges() with BusMin/BusMax
Date: Wed, 20 Jan 2021 15:01:43 +0100 [thread overview]
Message-ID: <b6259389-1232-29f9-702c-983533d43b71@redhat.com> (raw)
In-Reply-To: <20210119011302.10908-10-cenjiahui@huawei.com>
On 01/19/21 02:13, Jiahui Cen via groups.io wrote:
> Extend parameter list of PciHostBridgeUtilityGetRootBridges() with BusMin/
> BusMax, so that the utility function could be compatible with ArmVirtPkg
> who uses mutable bus range [BusMin, BusMax] insteand of [0, PCI_MAX_BUS].
>
> BZ: https://bugzilla.tianocore.org/show_bug.cgi?id=3059
>
> Cc: Jordan Justen <jordan.l.justen@intel.com>
> Cc: Laszlo Ersek <lersek@redhat.com>
> Cc: Ard Biesheuvel <ard.biesheuvel@arm.com>
> Signed-off-by: Jiahui Cen <cenjiahui@huawei.com>
> ---
> OvmfPkg/Include/Library/PciHostBridgeUtilityLib.h | 6 ++++
> OvmfPkg/Library/PciHostBridgeLib/PciHostBridgeLib.c | 2 ++
> OvmfPkg/Library/PciHostBridgeUtilityLib/PciHostBridgeUtilityLib.c | 30 ++++++++++++++++----
> 3 files changed, 33 insertions(+), 5 deletions(-)
>
> diff --git a/OvmfPkg/Include/Library/PciHostBridgeUtilityLib.h b/OvmfPkg/Include/Library/PciHostBridgeUtilityLib.h
> index a0ea44d96a67..d2dc18a1afad 100644
> --- a/OvmfPkg/Include/Library/PciHostBridgeUtilityLib.h
> +++ b/OvmfPkg/Include/Library/PciHostBridgeUtilityLib.h
> @@ -112,6 +112,10 @@ PciHostBridgeUtilityUninitRootBridge (
>
> @param[in] NoExtendedConfigSpace No Extended Config Space.
>
> + @param[in] BusMin Minimum Bus number, inclusive.
> +
> + @param[in] BusMax Maximum Bus number, inclusive.
> +
> @param[in] Io IO aperture.
>
> @param[in] Mem MMIO aperture.
> @@ -132,6 +136,8 @@ PciHostBridgeUtilityGetRootBridges (
> IN UINT64 AllocationAttributes,
> IN BOOLEAN DmaAbove4G,
> IN BOOLEAN NoExtendedConfigSpace,
> + IN UINTN BusMin,
> + IN UINTN BusMax,
> IN PCI_ROOT_BRIDGE_APERTURE *Io,
> IN PCI_ROOT_BRIDGE_APERTURE *Mem,
> IN PCI_ROOT_BRIDGE_APERTURE *MemAbove4G,
> diff --git a/OvmfPkg/Library/PciHostBridgeLib/PciHostBridgeLib.c b/OvmfPkg/Library/PciHostBridgeLib/PciHostBridgeLib.c
> index 91b9e6baa1e8..7d9fb0fb293a 100644
> --- a/OvmfPkg/Library/PciHostBridgeLib/PciHostBridgeLib.c
> +++ b/OvmfPkg/Library/PciHostBridgeLib/PciHostBridgeLib.c
> @@ -85,6 +85,8 @@ PciHostBridgeGetRootBridges (
> AllocationAttributes,
> FALSE,
> PcdGet16 (PcdOvmfHostBridgePciDevId) != INTEL_Q35_MCH_DEVICE_ID,
> + 0,
> + PCI_MAX_BUS,
> &Io,
> &Mem,
> &MemAbove4G,
> diff --git a/OvmfPkg/Library/PciHostBridgeUtilityLib/PciHostBridgeUtilityLib.c b/OvmfPkg/Library/PciHostBridgeUtilityLib/PciHostBridgeUtilityLib.c
> index 1d78984b83ad..69bed5c7843f 100644
> --- a/OvmfPkg/Library/PciHostBridgeUtilityLib/PciHostBridgeUtilityLib.c
> +++ b/OvmfPkg/Library/PciHostBridgeUtilityLib/PciHostBridgeUtilityLib.c
> @@ -200,6 +200,10 @@ PciHostBridgeUtilityUninitRootBridge (
>
> @param[in] NoExtendedConfigSpace No Extended Config Space.
>
> + @param[in] BusMin Minimum Bus number, inclusive.
> +
> + @param[in] BusMax Maximum Bus number, inclusive.
> +
> @param[in] Io IO aperture.
>
> @param[in] Mem MMIO aperture.
> @@ -220,6 +224,8 @@ PciHostBridgeUtilityGetRootBridges (
> IN UINT64 AllocationAttributes,
> IN BOOLEAN DmaAbove4G,
> IN BOOLEAN NoExtendedConfigSpace,
> + IN UINTN BusMin,
> + IN UINTN BusMax,
> IN PCI_ROOT_BRIDGE_APERTURE *Io,
> IN PCI_ROOT_BRIDGE_APERTURE *Mem,
> IN PCI_ROOT_BRIDGE_APERTURE *MemAbove4G,
> @@ -236,6 +242,13 @@ PciHostBridgeUtilityGetRootBridges (
> UINTN LastRootBridgeNumber;
> UINTN RootBridgeNumber;
>
> + if (BusMin > BusMax || BusMax > PCI_MAX_BUS) {
> + DEBUG ((DEBUG_ERROR, "%a: invalid bus range with BusMin %d and BusMax %d\n",
> + __FUNCTION__, BusMin, BusMax));
(1) The "%d" format specifier is appropriate for INT32 values. But UINTN
is either UINT32 or UINT64 (dependent on the edk2 architecture the
platform is built for).
The safe way to print a UINTN value is to (a) cast it to UINT64, and (b)
use the "%Lu" (or "%Lx") format specifier.
I'll fix this up for you.
> + *Count = 0;
(2) As I said under patch v6 07/11, this zeroing should be centralized
at the top of the function. I'll fix it up for you.
> + return NULL;
> + }
> +
> //
> // QEMU provides the number of extra root buses, shortening the exhaustive
> // search below. If there is no hint, the feature is missing.
> @@ -247,7 +260,14 @@ PciHostBridgeUtilityGetRootBridges (
> QemuFwCfgSelectItem (FwCfgItem);
> QemuFwCfgReadBytes (FwCfgSize, &ExtraRootBridges);
>
> - if (ExtraRootBridges > PCI_MAX_BUS) {
> + //
> + // Validate the number of extra root bridges. As BusMax is inclusive, the
> + // max bus count is (BusMax - BusMin + 1). From that, the "main" root bus
> + // is always givin, so the max count for the "extra" root bridges is one
(3) s/givin/a given/
I'll fix it.
> + // less, i.e. (BusMax - BusMin). If QEME hint exceeds that, we have invalid
(4) s/QEME/the QEMU/
I'll fix it.
> + // behavior.
> + //
> + if (ExtraRootBridges > BusMax - BusMin) {
> DEBUG ((DEBUG_ERROR, "%a: invalid count of extra root buses (%Lu) "
> "reported by QEMU\n", __FUNCTION__, ExtraRootBridges));
> *Count = 0;
> @@ -271,15 +291,15 @@ PciHostBridgeUtilityGetRootBridges (
> //
> // The "main" root bus is always there.
> //
> - LastRootBridgeNumber = 0;
> + LastRootBridgeNumber = BusMin;
Side remark (nothing to fix): I'm slightly curious what would happen if
the DTB on the arm/aarch64 "virt" machine reported a BusMin different
from 0. What does that mean for the MMCONFIG offsets? If BusMin is
nonzero, does that mean we have a unused slice at the start of MMCONFIG?
Or does it mean that we need to shift down the bus references so that
BusMin does map to offset 0 in MMCONFIG? (See also the
"PCI_ROOT_BRIDGE.Bus.Translation" field.)
Anyway, we can look at this later, if such a DTB ever occurs in
practice. (I doubt that it ever will.)
>
> //
> // Scan all other root buses. If function 0 of any device on a bus returns a
> // VendorId register value different from all-bits-one, then that bus is
> // alive.
> //
> - for (RootBridgeNumber = 1;
> - RootBridgeNumber <= PCI_MAX_BUS && Initialized < ExtraRootBridges;
> + for (RootBridgeNumber = BusMin + 1;
> + RootBridgeNumber <= BusMax && Initialized < ExtraRootBridges;
> ++RootBridgeNumber) {
> UINTN Device;
>
> @@ -329,7 +349,7 @@ PciHostBridgeUtilityGetRootBridges (
> DmaAbove4G,
> NoExtendedConfigSpace,
> (UINT8) LastRootBridgeNumber,
> - PCI_MAX_BUS,
> + (UINT8) BusMax,
> Io,
> Mem,
> MemAbove4G,
>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Thanks
Laszlo
next prev parent reply other threads:[~2021-01-20 14:01 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-01-19 1:12 [PATCH v6 00/11] Add extra pci roots support for Arm Jiahui Cen
2021-01-19 1:12 ` [PATCH v6 01/11] OvmfPkg: Introduce PciHostBridgeUtilityLib class Jiahui Cen
2021-01-20 11:47 ` [edk2-devel] " Laszlo Ersek
2021-01-19 1:12 ` [PATCH v6 02/11] ArmVirtPkg: Refactor with PciHostBridgeUtilityLib Jiahui Cen
2021-01-19 1:12 ` [PATCH v6 03/11] OvmfPkg/PciHostBridgeLib: List missing PcdLib dependency Jiahui Cen
2021-01-20 12:14 ` [edk2-devel] " Laszlo Ersek
2021-01-20 13:29 ` Laszlo Ersek
2021-01-19 1:12 ` [PATCH v6 04/11] OvmfPkg/PciHostBridgeLib: Extract InitRootBridge() / UninitRootBridge() Jiahui Cen
2021-01-20 12:36 ` [edk2-devel] " Laszlo Ersek
2021-01-19 1:12 ` [PATCH v6 05/11] OvmfPkg/PciHostBridgeUtilityLib: Extend parameters of InitRootBridge() Jiahui Cen
2021-01-20 12:41 ` [edk2-devel] " Laszlo Ersek
2021-01-19 1:12 ` [PATCH v6 06/11] ArmVirtPkg/FdtPciHostBridgeLib: Refactor init/uninit of root bridge Jiahui Cen
2021-01-20 12:46 ` [edk2-devel] " Laszlo Ersek
2021-01-19 1:12 ` [PATCH v6 07/11] OvmfPkg/PciHostBridgeLib: Extract GetRootBridges() / FreeRootBridges() Jiahui Cen
2021-01-20 13:07 ` [edk2-devel] " Laszlo Ersek
2021-01-19 1:12 ` [PATCH v6 08/11] OvmfPkg/PciHostBridgeUtilityLib: Extend parameter list of GetRootBridges Jiahui Cen
2021-01-20 13:34 ` [edk2-devel] " Laszlo Ersek
2021-01-19 1:13 ` [PATCH v6 09/11] OvmfPkg/PciHostBridgeUtilityLib: Extend GetRootBridges() with BusMin/BusMax Jiahui Cen
2021-01-20 14:01 ` Laszlo Ersek [this message]
2021-01-19 1:13 ` [PATCH v6 10/11] ArmVirtPkg/FdtPciHostBridgeLib: Add extra pci root buses support Jiahui Cen
2021-01-20 14:06 ` [edk2-devel] " Laszlo Ersek
2021-01-19 1:13 ` [PATCH v6 11/11] ArmVirtPkg/ArmVirtQemu: Add support for HotPlug Jiahui Cen
2021-01-20 14:09 ` [edk2-devel] " Laszlo Ersek
2021-01-20 16:26 ` [edk2-devel] [PATCH v6 00/11] Add extra pci roots support for Arm Laszlo Ersek
2021-01-21 1:26 ` Jiahui Cen
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=b6259389-1232-29f9-702c-983533d43b71@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