public inbox for devel@edk2.groups.io
 help / color / mirror / Atom feed
From: Igor Druzhinin <igor.druzhinin@citrix.com>
To: Anthony PERARD <anthony.perard@citrix.com>
Cc: <edk2-devel@lists.01.org>, <jordan.l.justen@intel.com>,
	<lersek@redhat.com>, <ard.biesheuvel@linaro.org>,
	<julien.grall@arm.com>, <xen-devel@lists.xenproject.org>
Subject: Re: [PATCH RESEND 3/3] OvmfPkg/XenSupport: turn off address decoding before BAR sizing
Date: Thu, 14 Mar 2019 20:09:41 +0000	[thread overview]
Message-ID: <da362720-afea-7af5-af48-1136cfd9bd5a@citrix.com> (raw)
In-Reply-To: <20190314175527.GD11621@perard.uk.xensource.com>

On 14/03/2019 17:55, Anthony PERARD wrote:
> On Wed, Mar 06, 2019 at 12:40:56PM +0000, Igor Druzhinin wrote:
>> On Xen, hvmloader firmware leaves address decoding enabled for
>> enumerated PCI device before jumping into OVMF. OVMF seems to
>> expect it to be disabled and tries to size PCI BARs in several places
>> without disabling it which causes BAR64, for example, being
>> incorrectly placed by QEMU.
>>
>> Fix it by disabling PCI address decoding explicitly before the
>> first attempt to size BARs on Xen.
>>
>> Contributed-under: TianoCore Contribution Agreement 1.1
>> Signed-off-by: Igor Druzhinin <igor.druzhinin@citrix.com>
>> ---
>>  OvmfPkg/Library/PciHostBridgeLib/XenSupport.c | 34 +++++++++++++++++++++++++++
>>  1 file changed, 34 insertions(+)
>>
>> diff --git a/OvmfPkg/Library/PciHostBridgeLib/XenSupport.c b/OvmfPkg/Library/PciHostBridgeLib/XenSupport.c
>> index 408fb24..9940335 100644
>> --- a/OvmfPkg/Library/PciHostBridgeLib/XenSupport.c
>> +++ b/OvmfPkg/Library/PciHostBridgeLib/XenSupport.c
>> @@ -55,6 +55,33 @@ PcatPciRootBridgeBarExisted (
>>    EnableInterrupts ();
>>  }
>>  
>> +#define EFI_PCI_COMMAND_DECODE ((UINT16)(EFI_PCI_COMMAND_IO_SPACE | \
>> +                                         EFI_PCI_COMMAND_MEMORY_SPACE))
>> +STATIC
>> +VOID
>> +PcatPciRootBridgeDecoding (
>> +  IN  UINTN                          Address,
>> +  IN  BOOLEAN                        Enabled,
>> +  OUT UINT16                         *OriginalValue
>> +  )
>> +{
>> +  UINT16                            Value;
>> +
>> +  //
>> +  // Preserve the original value
>> +  //
>> +  Value = *OriginalValue;
>> +  *OriginalValue = PciRead16 (Address);
>> +
>> +  if (!Enabled) {
>> +    if (*OriginalValue & EFI_PCI_COMMAND_DECODE)
>> +       PciWrite16 (Address, *OriginalValue & ~EFI_PCI_COMMAND_DECODE);
> 
> The edk2 coding style ask to always use { } with if:
> https://github.com/tianocore/tianocore.github.io/wiki/Code-Style-C
> 
>> +  } else {
>> +    if (Value & EFI_PCI_COMMAND_DECODE)
>> +      PciWrite16 (Address, Value);
> 
> here too.
> 
>> +  }
>> +}
>> +
>>  STATIC
>>  VOID
>>  PcatPciRootBridgeParseBars (
>> @@ -76,6 +103,7 @@ PcatPciRootBridgeParseBars (
>>    UINT32                            Value;
>>    UINT32                            OriginalUpperValue;
>>    UINT32                            UpperValue;
>> +  UINT16                            OriginalCommand;
>>    UINT64                            Mask;
>>    UINTN                             Offset;
>>    UINT64                            Base;
>> @@ -83,6 +111,12 @@ PcatPciRootBridgeParseBars (
>>    UINT64                            Limit;
>>    PCI_ROOT_BRIDGE_APERTURE          *MemAperture;
>>  
>> +  // Disable address decoding for every device before OVMF starts sizing it
>> +  PcatPciRootBridgeDecoding (
>> +    PCI_LIB_ADDRESS (Bus, Device, Function, PCI_COMMAND_OFFSET),
>> +    FALSE, &OriginalCommand
> 
> Is it on purpose that the original register value is stored in
> OriginalCommand, but never used again?

Those are probably remnants of extended patch version for XenServer -
will remove together with other suggestions from you and Laszlo.

Igor


      reply	other threads:[~2019-03-14 20:09 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-03-06 12:40 [PATCH RESEND 0/3] Xen PCI passthrough fixes Igor Druzhinin
2019-03-06 12:40 ` [PATCH RESEND 1/3] OvmfPkg/XenSupport: remove usage of prefetchable PCI host bridge aperture Igor Druzhinin
2019-03-14 17:41   ` Anthony PERARD
2019-03-14 19:45     ` Igor Druzhinin
2019-03-19 14:03       ` Anthony PERARD
2019-03-20 11:15         ` Laszlo Ersek
     [not found]   ` <20190322082650.tk65vju74g4gt7vj@MacBook-Air-de-Roger.local>
2019-03-22  9:06     ` [Xen-devel] " Laszlo Ersek
     [not found]       ` <20190324035053.xs3yccnpmn5dy6cl@MacBook-Air-de-Roger.local>
2019-03-25 14:32         ` Igor Druzhinin
2019-03-06 12:40 ` [PATCH RESEND 2/3] OvmfPkg/XenSupport: use a correct PCI host bridge aperture for BAR64 Igor Druzhinin
2019-03-14 17:44   ` Anthony PERARD
2019-03-06 12:40 ` [PATCH RESEND 3/3] OvmfPkg/XenSupport: turn off address decoding before BAR sizing Igor Druzhinin
2019-03-06 13:22   ` Laszlo Ersek
2019-03-06 14:26     ` Igor Druzhinin
2019-03-06 17:39       ` Laszlo Ersek
2019-03-14 17:55   ` Anthony PERARD
2019-03-14 20:09     ` Igor Druzhinin [this message]

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=da362720-afea-7af5-af48-1136cfd9bd5a@citrix.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