From: "Ni, Ruiyu" <ruiyu.ni@intel.com>
To: "Ni, Ruiyu" <ruiyu.ni@intel.com>,
"edk2-devel@lists.01.org" <edk2-devel@lists.01.org>
Cc: Laszlo Ersek <lersek@redhat.com>
Subject: Re: [PATCH] MdeModulePkg/PciBus: Count multiple hotplug resource paddings
Date: Sat, 30 Sep 2017 05:12:26 +0000 [thread overview]
Message-ID: <734D49CCEBEEF84792F5B80ED585239D5BA83A43@SHSMSX104.ccr.corp.intel.com> (raw)
In-Reply-To: <20170930051030.200300-1-ruiyu.ni@intel.com>
Laszlo,
Please check whether this patch can resolve the hang issue you reported.
I don't use your suggestion to use the MAX resource paddings.
Instead, I count all the resource paddings.
If it can work, I will submit another patch to clean up the code in CalculateApertureIo16().
That function could be similar to the CalculateAperture().
Thanks/Ray
> -----Original Message-----
> From: edk2-devel [mailto:edk2-devel-bounces@lists.01.org] On Behalf Of Ruiyu
> Ni
> Sent: Saturday, September 30, 2017 1:11 PM
> To: edk2-devel@lists.01.org
> Cc: Laszlo Ersek <lersek@redhat.com>
> Subject: [edk2] [PATCH] MdeModulePkg/PciBus: Count multiple hotplug
> resource paddings
>
> The current implementation assumes there is only one hotplug resource padding
> for each resource type. It's not true considering
> DegradeResource(): MEM64 resource could be degraded to MEM32 resource.
>
> The patch treat the resource paddings using the same logic as treating
> typical/actual resources and the total resource of a bridge is set to the MAX of
> typical/actual resources and resource paddings.
>
> Contributed-under: TianoCore Contribution Agreement 1.1
> Signed-off-by: Ruiyu Ni <ruiyu.ni@intel.com>
> Cc: Laszlo Ersek <lersek@redhat.com>
> ---
> .../Bus/Pci/PciBusDxe/PciResourceSupport.c | 67 +++++++---------------
> 1 file changed, 21 insertions(+), 46 deletions(-)
>
> diff --git a/MdeModulePkg/Bus/Pci/PciBusDxe/PciResourceSupport.c
> b/MdeModulePkg/Bus/Pci/PciBusDxe/PciResourceSupport.c
> index e93134613b..f086b1732d 100644
> --- a/MdeModulePkg/Bus/Pci/PciBusDxe/PciResourceSupport.c
> +++ b/MdeModulePkg/Bus/Pci/PciBusDxe/PciResourceSupport.c
> @@ -343,14 +343,9 @@ CalculateResourceAperture (
> IN PCI_RESOURCE_NODE *Bridge
> )
> {
> - UINT64 Aperture;
> + UINT64 Aperture[2];
> LIST_ENTRY *CurrentLink;
> PCI_RESOURCE_NODE *Node;
> - UINT64 PaddingAperture;
> - UINT64 Offset;
> -
> - Aperture = 0;
> - PaddingAperture = 0;
>
> if (Bridge == NULL) {
> return ;
> @@ -362,6 +357,8 @@ CalculateResourceAperture (
> return ;
> }
>
> + Aperture[PciResUsageTypical] = 0;
> + Aperture[PciResUsagePadding] = 0;
> //
> // Assume the bridge is aligned
> //
> @@ -369,58 +366,30 @@ CalculateResourceAperture (
> ; !IsNull (&Bridge->ChildList, CurrentLink)
> ; CurrentLink = GetNextNode (&Bridge->ChildList, CurrentLink)
> ) {
> -
> Node = RESOURCE_NODE_FROM_LINK (CurrentLink);
> - if (Node->ResourceUsage == PciResUsagePadding) {
> - ASSERT (PaddingAperture == 0);
> - PaddingAperture = Node->Length;
> - continue;
> - }
>
> //
> - // Apply padding resource if available
> + // It's possible for a bridge to contain multiple padding resource
> + // nodes due to DegradeResource().
> //
> - Offset = Aperture & (Node->Alignment);
> -
> - if (Offset != 0) {
> -
> - Aperture = Aperture + (Node->Alignment + 1) - Offset;
> -
> - }
> -
> + ASSERT ((Node->ResourceUsage == PciResUsageTypical) ||
> + (Node->ResourceUsage == PciResUsagePadding));
> + ASSERT (Node->ResourceUsage < ARRAY_SIZE (Aperture));
> //
> // Recode current aperture as a offset
> - // this offset will be used in future real allocation
> + // Apply padding resource to meet alignment requirement
> + // Node offset will be used in future real allocation
> //
> - Node->Offset = Aperture;
> + Node->Offset = ALIGN_VALUE (Aperture[Node->ResourceUsage],
> + Node->Alignment + 1);
>
> //
> - // Increment aperture by the length of node
> + // Record the total aperture.
> //
> - Aperture += Node->Length;
> - }
> -
> - //
> - // At last, adjust the aperture with the bridge's
> - // alignment
> - //
> - Offset = Aperture & (Bridge->Alignment);
> - if (Offset != 0) {
> - Aperture = Aperture + (Bridge->Alignment + 1) - Offset;
> + Aperture[Node->ResourceUsage] = Node->Offset + Node->Length;
> }
>
> //
> - // If the bridge has already padded the resource and the
> - // amount of padded resource is larger, then keep the
> - // padded resource
> - //
> - if (Bridge->Length < Aperture) {
> - Bridge->Length = Aperture;
> - }
> -
> - //
> - // Adjust the bridge's alignment to the first child's alignment
> - // if the bridge has at least one child
> + // Adjust the bridge's alignment to the MAX (first) alignment of all children.
> //
> CurrentLink = Bridge->ChildList.ForwardLink;
> if (CurrentLink != &Bridge->ChildList) { @@ -431,10 +400,16 @@
> CalculateResourceAperture (
> }
>
> //
> + // At last, adjust the aperture with the bridge's alignment //
> + Aperture[PciResUsageTypical] = ALIGN_VALUE
> + (Aperture[PciResUsageTypical], Bridge->Alignment + 1);
> + Aperture[PciResUsagePadding] = ALIGN_VALUE
> + (Aperture[PciResUsagePadding], Bridge->Alignment + 1);
> +
> + //
> // Hotplug controller needs padding resources.
> // Use the larger one between the padding resource and actual occupied
> resource.
> //
> - Bridge->Length = MAX (Bridge->Length, PaddingAperture);
> + Bridge->Length = MAX (Aperture[PciResUsageTypical],
> + Aperture[PciResUsagePadding]);
> }
>
> /**
> --
> 2.12.2.windows.2
>
> _______________________________________________
> edk2-devel mailing list
> edk2-devel@lists.01.org
> https://lists.01.org/mailman/listinfo/edk2-devel
next prev parent reply other threads:[~2017-09-30 5:09 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-09-30 5:10 [PATCH] MdeModulePkg/PciBus: Count multiple hotplug resource paddings Ruiyu Ni
2017-09-30 5:12 ` Ni, Ruiyu [this message]
2017-09-30 22:34 ` Laszlo Ersek
2017-10-09 2:24 ` Zeng, Star
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=734D49CCEBEEF84792F5B80ED585239D5BA83A43@SHSMSX104.ccr.corp.intel.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