* [PATCH v1] arm64:pci: fix the IOV device enabled crash issue in designware
@ 2016-08-23 6:01 ` Po Liu
0 siblings, 0 replies; 26+ messages in thread
From: Po Liu @ 2016-08-23 6:01 UTC (permalink / raw)
To: linux-pci
Cc: Roy Zang, Arnd Bergmann, Po Liu, Stuart Yoder, Yang-Leo Li,
linux-arm-kernel, Bjorn Helgaas, Mingkai Hu
When echo a number to /sys/bus/pci/devices/xxx/sriov_numvfs to enable the
VF devices. A crash log occurred. This found to be access the IOV devices
config space failure issue.
The read/write config space from host would judge the pcie device plugin
or not by:
if (bus->primary == pp->root_bus_nr && dev > 0)
return 0;
Although all pcie devices for dev(coming from the device and function
number) is zero. But the dev is not zero for VF. So remove the
condition.
Signed-off-by: Po Liu <po.liu@nxp.com>
---
drivers/pci/host/pcie-designware.c | 7 -------
1 file changed, 7 deletions(-)
diff --git a/drivers/pci/host/pcie-designware.c b/drivers/pci/host/pcie-designware.c
index 12afce1..dd20eb2 100644
--- a/drivers/pci/host/pcie-designware.c
+++ b/drivers/pci/host/pcie-designware.c
@@ -670,13 +670,6 @@ static int dw_pcie_valid_config(struct pcie_port *pp,
if (bus->number == pp->root_bus_nr && dev > 0)
return 0;
- /*
- * do not read more than one device on the bus directly attached
- * to RC's (Virtual Bridge's) DS side.
- */
- if (bus->primary == pp->root_bus_nr && dev > 0)
- return 0;
-
return 1;
}
--
2.1.0.27.g96db324
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 26+ messages in thread
* [PATCH v1] arm64:pci: fix the IOV device enabled crash issue in designware
@ 2016-08-23 6:01 ` Po Liu
0 siblings, 0 replies; 26+ messages in thread
From: Po Liu @ 2016-08-23 6:01 UTC (permalink / raw)
To: linux-arm-kernel
When echo a number to /sys/bus/pci/devices/xxx/sriov_numvfs to enable the
VF devices. A crash log occurred. This found to be access the IOV devices
config space failure issue.
The read/write config space from host would judge the pcie device plugin
or not by:
if (bus->primary == pp->root_bus_nr && dev > 0)
return 0;
Although all pcie devices for dev(coming from the device and function
number) is zero. But the dev is not zero for VF. So remove the
condition.
Signed-off-by: Po Liu <po.liu@nxp.com>
---
drivers/pci/host/pcie-designware.c | 7 -------
1 file changed, 7 deletions(-)
diff --git a/drivers/pci/host/pcie-designware.c b/drivers/pci/host/pcie-designware.c
index 12afce1..dd20eb2 100644
--- a/drivers/pci/host/pcie-designware.c
+++ b/drivers/pci/host/pcie-designware.c
@@ -670,13 +670,6 @@ static int dw_pcie_valid_config(struct pcie_port *pp,
if (bus->number == pp->root_bus_nr && dev > 0)
return 0;
- /*
- * do not read more than one device on the bus directly attached
- * to RC's (Virtual Bridge's) DS side.
- */
- if (bus->primary == pp->root_bus_nr && dev > 0)
- return 0;
-
return 1;
}
--
2.1.0.27.g96db324
^ permalink raw reply related [flat|nested] 26+ messages in thread
* Re: [PATCH v1] arm64:pci: fix the IOV device enabled crash issue in designware
2016-08-23 6:01 ` Po Liu
@ 2016-08-24 20:50 ` Bjorn Helgaas
-1 siblings, 0 replies; 26+ messages in thread
From: Bjorn Helgaas @ 2016-08-24 20:50 UTC (permalink / raw)
To: Po Liu
Cc: Sören Brinkmann, Roy Zang, Arnd Bergmann, linux-pci,
Pratyush Anand, Stuart Yoder, Yang-Leo Li, Mingkai Hu,
Jingoo Han, Bjorn Helgaas, Ley Foon Tan, Michal Simek,
linux-arm-kernel
[+cc Jingoo, Pratyush, Michal, S=F6ren, Ley]
On Tue, Aug 23, 2016 at 02:01:12PM +0800, Po Liu wrote:
> When echo a number to /sys/bus/pci/devices/xxx/sriov_numvfs to enable the
> VF devices. A crash log occurred. This found to be access the IOV devices
> config space failure issue.
What was the actual crash? The mere fact that we made a config read fail
should not cause a crash. We might erroneously prevent access to VF
devices, but it shouldn't crash. So maybe there's another bug elsewhere
that we should fix first.
> The read/write config space from host would judge the pcie device plugin
> or not by:
> =
> if (bus->primary =3D=3D pp->root_bus_nr && dev > 0)
> return 0;
I'm guessing other drivers have the same issue, e.g.,
altera_pcie_valid_config(), xilinx_pcie_valid_device().
Can you look through them and fix them all at once?
> Although all pcie devices for dev(coming from the device and function
> number) is zero. But the dev is not zero for VF. So remove the
> condition.
> =
> Signed-off-by: Po Liu <po.liu@nxp.com>
> ---
> drivers/pci/host/pcie-designware.c | 7 -------
> 1 file changed, 7 deletions(-)
> =
> diff --git a/drivers/pci/host/pcie-designware.c b/drivers/pci/host/pcie-d=
esignware.c
> index 12afce1..dd20eb2 100644
> --- a/drivers/pci/host/pcie-designware.c
> +++ b/drivers/pci/host/pcie-designware.c
> @@ -670,13 +670,6 @@ static int dw_pcie_valid_config(struct pcie_port *pp,
> if (bus->number =3D=3D pp->root_bus_nr && dev > 0)
> return 0;
> =
> - /*
> - * do not read more than one device on the bus directly attached
> - * to RC's (Virtual Bridge's) DS side.
> - */
> - if (bus->primary =3D=3D pp->root_bus_nr && dev > 0)
> - return 0;
> -
> return 1;
> }
> =
> -- =
> 2.1.0.27.g96db324
> =
> =
> _______________________________________________
> linux-arm-kernel mailing list
> linux-arm-kernel@lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 26+ messages in thread
* [PATCH v1] arm64:pci: fix the IOV device enabled crash issue in designware
@ 2016-08-24 20:50 ` Bjorn Helgaas
0 siblings, 0 replies; 26+ messages in thread
From: Bjorn Helgaas @ 2016-08-24 20:50 UTC (permalink / raw)
To: linux-arm-kernel
[+cc Jingoo, Pratyush, Michal, S?ren, Ley]
On Tue, Aug 23, 2016 at 02:01:12PM +0800, Po Liu wrote:
> When echo a number to /sys/bus/pci/devices/xxx/sriov_numvfs to enable the
> VF devices. A crash log occurred. This found to be access the IOV devices
> config space failure issue.
What was the actual crash? The mere fact that we made a config read fail
should not cause a crash. We might erroneously prevent access to VF
devices, but it shouldn't crash. So maybe there's another bug elsewhere
that we should fix first.
> The read/write config space from host would judge the pcie device plugin
> or not by:
>
> if (bus->primary == pp->root_bus_nr && dev > 0)
> return 0;
I'm guessing other drivers have the same issue, e.g.,
altera_pcie_valid_config(), xilinx_pcie_valid_device().
Can you look through them and fix them all at once?
> Although all pcie devices for dev(coming from the device and function
> number) is zero. But the dev is not zero for VF. So remove the
> condition.
>
> Signed-off-by: Po Liu <po.liu@nxp.com>
> ---
> drivers/pci/host/pcie-designware.c | 7 -------
> 1 file changed, 7 deletions(-)
>
> diff --git a/drivers/pci/host/pcie-designware.c b/drivers/pci/host/pcie-designware.c
> index 12afce1..dd20eb2 100644
> --- a/drivers/pci/host/pcie-designware.c
> +++ b/drivers/pci/host/pcie-designware.c
> @@ -670,13 +670,6 @@ static int dw_pcie_valid_config(struct pcie_port *pp,
> if (bus->number == pp->root_bus_nr && dev > 0)
> return 0;
>
> - /*
> - * do not read more than one device on the bus directly attached
> - * to RC's (Virtual Bridge's) DS side.
> - */
> - if (bus->primary == pp->root_bus_nr && dev > 0)
> - return 0;
> -
> return 1;
> }
>
> --
> 2.1.0.27.g96db324
>
>
> _______________________________________________
> linux-arm-kernel mailing list
> linux-arm-kernel at lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 26+ messages in thread
* RE: [PATCH v1] arm64:pci: fix the IOV device enabled crash issue in designware
2016-08-24 20:50 ` Bjorn Helgaas
@ 2016-08-25 4:53 ` Po Liu
-1 siblings, 0 replies; 26+ messages in thread
From: Po Liu @ 2016-08-25 4:53 UTC (permalink / raw)
To: Bjorn Helgaas
Cc: S?ren Brinkmann, Roy Zang, Arnd Bergmann, linux-pci,
Pratyush Anand, Stuart Yoder, Yang-Leo Li, Mingkai Hu,
Jingoo Han, Bjorn Helgaas, Ley Foon Tan, Michal Simek,
linux-arm-kernel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^ permalink raw reply [flat|nested] 26+ messages in thread
* [PATCH v1] arm64:pci: fix the IOV device enabled crash issue in designware
@ 2016-08-25 4:53 ` Po Liu
0 siblings, 0 replies; 26+ messages in thread
From: Po Liu @ 2016-08-25 4:53 UTC (permalink / raw)
To: linux-arm-kernel
Hi Bjorn,
Best regards,
Liu Po
> -----Original Message-----
> From: Bjorn Helgaas [mailto:helgaas at kernel.org]
> Sent: Thursday, August 25, 2016 4:51 AM
> To: Po Liu
> Cc: linux-pci at vger.kernel.org; Roy Zang; Arnd Bergmann; Stuart Yoder;
> Yang-Leo Li; linux-arm-kernel at lists.infradead.org; Bjorn Helgaas;
> Mingkai Hu; Ley Foon Tan; Michal Simek; S?ren Brinkmann; Jingoo Han;
> Pratyush Anand
> Subject: Re: [PATCH v1] arm64:pci: fix the IOV device enabled crash
> issue in designware
>
> [+cc Jingoo, Pratyush, Michal, S?ren, Ley]
>
> On Tue, Aug 23, 2016 at 02:01:12PM +0800, Po Liu wrote:
> > When echo a number to /sys/bus/pci/devices/xxx/sriov_numvfs to enable
> > the VF devices. A crash log occurred. This found to be access the IOV
> > devices config space failure issue.
>
> What was the actual crash? The mere fact that we made a config read
> fail should not cause a crash. We might erroneously prevent access to
> VF devices, but it shouldn't crash. So maybe there's another bug
> elsewhere that we should fix first.
I built with CONFIG_PCI_IOV=y and notice a crash when I use it:
centqds-60 cd /sys/class/net/
centqds-61 ls
enP1p1s0@ enP2p1s0f0@ enP2p1s0f1@ lo@ sit0@
centqds-62 cd enP2p1s0f1/device
centqds-63 ls
broken_parity_status driver_override msi_irqs/ sriov_numvfs
class enable net/ sriov_totalvfs
config iommu_group@ power/ subsystem@
consistent_dma_mask_bits irq remove subsystem_device
device local_cpulist rescan subsystem_vendor
devspec local_cpus reset uevent
dma_mask_bits modalias resource vendor
driver@ msi_bus rom
centqds-64 zcat /proc/config.gz | grep _IOV
CONFIG_PCI_IOV=y
centqds-65 sudo su
[root at centqds 0002:01:00.1]# echo 2 > sriov_numvfs
[ 317.604543] ixgbe 0002:01:00.1 enP2p1s0f1: SR-IOV enabled with 2 VFs
[ 317.714431] (null): of_irq_parse_pci() failed with rc=134
[ 317.719906] -----------[ cut here ]-----------
[ 317.724525] WARNING: CPU: 6 PID: 3179 at drivers/pci/probe.c:1555 pci_device_add+0x144/0x148()
[ 317.733123] Modules linked in:
[ 317.736175] CPU: 6 PID: 3179 Comm: bash Not tainted 4.1.8-00024-g0a32d65-dirty #32
[ 317.743731] Hardware name: Freescale Layerscape 2088a QDS Board (DT)
[ 317.750077] Call trace:
[ 317.752516] [<fffffe0000096cf8>] dump_backtrace+0x0/0x12c
Message from[ 317.757916] [<fffffe0000096e34>] show_stack+0x10/0x1c
syslogd at centqds[ 317.764341] [<fffffe00008937fc>] dump_stack+0x84/0xd4
at Jul 26 15:51[ 317.770770] [<fffffe00000be590>] warn_slowpath_common+0x94/0xcc
:10 ...
kerne[ 317.778067] [<fffffe00000be68c>] warn_slowpath_null+0x14/0x20
l:Call trace:
[ 317.785192] [<fffffe00003c5fa4>] pci_device_add+0x140/0x148
[ 317.792133] [<fffffe00003de784>] pci_enable_sriov+0x470/0x7a0
[ 317.797873] [<fffffe00005fa9dc>] ixgbe_pci_sriov_configure+0x8c/0x148
[ 317.804302] [<fffffe00003cf354>] sriov_numvfs_store+0x78/0x11c
[ 317.810129] [<fffffe0000443d80>] dev_attr_store+0x14/0x28
[ 317.815521] [<fffffe0000200164>] sysfs_kf_write+0x40/0x4c
[ 317.820908] [<fffffe00001ff5ec>] kernfs_fop_write+0xb8/0x180
[ 317.826561] [<fffffe000019a568>] __vfs_write+0x28/0x10c
[ 317.831775] [<fffffe000019ad00>] vfs_write+0x90/0x1a0
[ 317.836819] [<fffffe000019b584>] SyS_write+0x40/0xa0
[ 317.841772] --[ end trace 83725a9784fd702a ]--
[ 317.846393] BUG: failure at fs/sysfs/file.c:481/sysfs_create_bin_file()!
[ 317.853081] Kernel panic - not syncing: BUG!
[ 317.857339] CPU: 6 PID: 3179 Comm: bash Tainted: G W 4.1.8-00024-g0a32d65-dirty #32
[ 317.866110] Hardware name: Freescale Layerscape 2088a QDS Board (DT)
[ 317.872451] Call trace:
[ 317.874887] [<fffffe0000096cf8>] dump_backtrace+0x0/0x12c
[ 317.880274] [<fffffe0000096e34>] show_stack+0x10/0x1c
[ 317.885315] [<fffffe00008937fc>] dump_stack+0x84/0xd4
[ 317.890354] [<fffffe00008909e8>] panic+0xe4/0x21c
[ 317.895047] [<fffffe0000200988>] sysfs_create_bin_file+0x60/0x64
[ 317.901041] [<fffffe00003cf52c>] pci_create_sysfs_dev_files+0x48/0x2a8
[ 317.907556] [<fffffe00003c4238>] pci_bus_add_device+0x20/0x6c
The code process is that: "echo 2 > sriov_numvf" makes driver load .sriov_configure. At last to load pci_enable_sriov().
The first time vf device operate the config space in the pci_setup_device() (this function was load in the virtfn_add()) is pci_read_config_byte(dev, PCI_HEADER_TYPE, &hdr_type) return failure. So the virtfn didn't initialized proper.
This found to be "bus->primary == pp->root_bus_nr && dev > 0" then return failure in host controller. The dev came from devfn must not zero(is about 0x10).
then read config space failure. This makes the dev->bus is NULL. Lead to upper crash.
>
> > The read/write config space from host would judge the pcie device
> > plugin or not by:
> >
> > if (bus->primary == pp->root_bus_nr && dev > 0)
> > return 0;
>
> I'm guessing other drivers have the same issue, e.g.,
> altera_pcie_valid_config(), xilinx_pcie_valid_device().
>
> Can you look through them and fix them all at once?
Ok, if the problem is same.
>
> > Although all pcie devices for dev(coming from the device and function
> > number) is zero. But the dev is not zero for VF. So remove the
> > condition.
> >
> > Signed-off-by: Po Liu <po.liu@nxp.com>
> > ---
> > drivers/pci/host/pcie-designware.c | 7 -------
> > 1 file changed, 7 deletions(-)
> >
> > diff --git a/drivers/pci/host/pcie-designware.c
> > b/drivers/pci/host/pcie-designware.c
> > index 12afce1..dd20eb2 100644
> > --- a/drivers/pci/host/pcie-designware.c
> > +++ b/drivers/pci/host/pcie-designware.c
> > @@ -670,13 +670,6 @@ static int dw_pcie_valid_config(struct pcie_port
> *pp,
> > if (bus->number == pp->root_bus_nr && dev > 0)
> > return 0;
> >
> > - /*
> > - * do not read more than one device on the bus directly attached
> > - * to RC's (Virtual Bridge's) DS side.
> > - */
> > - if (bus->primary == pp->root_bus_nr && dev > 0)
> > - return 0;
> > -
> > return 1;
> > }
> >
> > --
> > 2.1.0.27.g96db324
> >
> >
> > _______________________________________________
> > linux-arm-kernel mailing list
> > linux-arm-kernel at lists.infradead.org
> > http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH v1] arm64:pci: fix the IOV device enabled crash issue in designware
2016-08-25 4:53 ` Po Liu
@ 2016-08-25 18:09 ` Bjorn Helgaas
-1 siblings, 0 replies; 26+ messages in thread
From: Bjorn Helgaas @ 2016-08-25 18:09 UTC (permalink / raw)
To: Po Liu
Cc: linux-arm-kernel, Roy Zang, Arnd Bergmann, linux-pci,
Pratyush Anand, Stuart Yoder, Yang-Leo Li, Mingkai Hu,
Jingoo Han, Bjorn Helgaas, Ley Foon Tan, Michal Simek,
S?ren Brinkmann
On Thu, Aug 25, 2016 at 04:53:19AM +0000, Po Liu wrote:
> > -----Original Message-----
> > From: Bjorn Helgaas [mailto:helgaas@kernel.org]
> > Sent: Thursday, August 25, 2016 4:51 AM
> > To: Po Liu
> > Cc: linux-pci@vger.kernel.org; Roy Zang; Arnd Bergmann; Stuart Yoder;
> > Yang-Leo Li; linux-arm-kernel@lists.infradead.org; Bjorn Helgaas;
> > Mingkai Hu; Ley Foon Tan; Michal Simek; S=F6ren Brinkmann; Jingoo Han;
> > Pratyush Anand
> > Subject: Re: [PATCH v1] arm64:pci: fix the IOV device enabled crash
> > issue in designware
> > =
> > [+cc Jingoo, Pratyush, Michal, S=F6ren, Ley]
> > =
> > On Tue, Aug 23, 2016 at 02:01:12PM +0800, Po Liu wrote:
> > > When echo a number to /sys/bus/pci/devices/xxx/sriov_numvfs to enable
> > > the VF devices. A crash log occurred. This found to be access the IOV
> > > devices config space failure issue.
> > =
> > What was the actual crash? The mere fact that we made a config read
> > fail should not cause a crash. We might erroneously prevent access to
> > VF devices, but it shouldn't crash. So maybe there's another bug
> > elsewhere that we should fix first.
> =
> I built with CONFIG_PCI_IOV=3Dy and notice a crash when I use it:
> =
> centqds-60 cd /sys/class/net/
> centqds-61 ls
> enP1p1s0@ enP2p1s0f0@ enP2p1s0f1@ lo@ sit0@
> centqds-62 cd enP2p1s0f1/device
> centqds-63 ls
> broken_parity_status driver_override msi_irqs/ sriov_numvfs
> class enable net/ sriov_totalvfs
> config iommu_group@ power/ subsystem@
> consistent_dma_mask_bits irq remove subsystem_device
> device local_cpulist rescan subsystem_vendor
> devspec local_cpus reset uevent
> dma_mask_bits modalias resource vendor
> driver@ msi_bus rom
> centqds-64 zcat /proc/config.gz | grep _IOV
> CONFIG_PCI_IOV=3Dy
> centqds-65 sudo su
> [root@centqds 0002:01:00.1]# echo 2 > sriov_numvfs
> [ 317.604543] ixgbe 0002:01:00.1 enP2p1s0f1: SR-IOV enabled with 2 VFs
> [ 317.714431] (null): of_irq_parse_pci() failed with rc=3D134
> [ 317.719906] -----------[ cut here ]-----------
> [ 317.724525] WARNING: CPU: 6 PID: 3179 at drivers/pci/probe.c:1555 pci_d=
evice_add+0x144/0x148()
> [ 317.733123] Modules linked in:
> [ 317.736175] CPU: 6 PID: 3179 Comm: bash Not tainted 4.1.8-00024-g0a32d6=
5-dirty #32
> [ 317.743731] Hardware name: Freescale Layerscape 2088a QDS Board (DT)
> [ 317.750077] Call trace:
> [ 317.752516] [<fffffe0000096cf8>] dump_backtrace+0x0/0x12c
> Message from[ 317.757916] [<fffffe0000096e34>] show_stack+0x10/0x1c
> syslogd@centqds[ 317.764341] [<fffffe00008937fc>] dump_stack+0x84/0xd4
> at Jul 26 15:51[ 317.770770] [<fffffe00000be590>] warn_slowpath_common+0x=
94/0xcc
> :10 ...
> kerne[ 317.778067] [<fffffe00000be68c>] warn_slowpath_null+0x14/0x20
> l:Call trace:
> [ 317.785192] [<fffffe00003c5fa4>] pci_device_add+0x140/0x148
> [ 317.792133] [<fffffe00003de784>] pci_enable_sriov+0x470/0x7a0
> [ 317.797873] [<fffffe00005fa9dc>] ixgbe_pci_sriov_configure+0x8c/0x148
> [ 317.804302] [<fffffe00003cf354>] sriov_numvfs_store+0x78/0x11c
> [ 317.810129] [<fffffe0000443d80>] dev_attr_store+0x14/0x28
> [ 317.815521] [<fffffe0000200164>] sysfs_kf_write+0x40/0x4c
> [ 317.820908] [<fffffe00001ff5ec>] kernfs_fop_write+0xb8/0x180
> [ 317.826561] [<fffffe000019a568>] __vfs_write+0x28/0x10c
> [ 317.831775] [<fffffe000019ad00>] vfs_write+0x90/0x1a0
> [ 317.836819] [<fffffe000019b584>] SyS_write+0x40/0xa0
> [ 317.841772] --[ end trace 83725a9784fd702a ]--
> [ 317.846393] BUG: failure at fs/sysfs/file.c:481/sysfs_create_bin_file()!
> [ 317.853081] Kernel panic - not syncing: BUG!
> [ 317.857339] CPU: 6 PID: 3179 Comm: bash Tainted: G W 4.1.8-00024-g0a32d=
65-dirty #32
> [ 317.866110] Hardware name: Freescale Layerscape 2088a QDS Board (DT)
> [ 317.872451] Call trace:
> [ 317.874887] [<fffffe0000096cf8>] dump_backtrace+0x0/0x12c
> [ 317.880274] [<fffffe0000096e34>] show_stack+0x10/0x1c
> [ 317.885315] [<fffffe00008937fc>] dump_stack+0x84/0xd4
> [ 317.890354] [<fffffe00008909e8>] panic+0xe4/0x21c
> [ 317.895047] [<fffffe0000200988>] sysfs_create_bin_file+0x60/0x64
> [ 317.901041] [<fffffe00003cf52c>] pci_create_sysfs_dev_files+0x48/0x2a8
> [ 317.907556] [<fffffe00003c4238>] pci_bus_add_device+0x20/0x6c
> =
> The code process is that: "echo 2 > sriov_numvf" makes driver load .sriov=
_configure. At last to load pci_enable_sriov().
> The first time vf device operate the config space in the pci_setup_device=
() (this function was load in the virtfn_add()) is pci_read_config_byte(dev=
, PCI_HEADER_TYPE, &hdr_type) return failure. So the virtfn didn't initiali=
zed proper.
> =
> This found to be "bus->primary =3D=3D pp->root_bus_nr && dev > 0" then re=
turn failure in host controller. The dev came from devfn must not zero(is a=
bout 0x10).
> =
> then read config space failure. This makes the dev->bus is NULL. Lead to =
upper crash.
I think the crash (BUG: failure at fs/sysfs/file.c:481/sysfs_create_bin_fil=
e())
happens in this path:
sriov_numvfs_store
ixgbe_pci_sriov_configure
ixgbe_pci_sriov_enable
pci_enable_sriov
sriov_enable
pci_iov_add_virtfn
virtfn =3D pci_alloc_dev()
pci_setup_device(virtfn)
if (pci_read_config_byte(dev, PCI_HEADER_TYPE, &hdr_type))
return -EIO
pci_bus_add_device(virtfn)
pci_create_sysfs_dev_files
sysfs_create_bin_file
BUG_ON(!kobj)
If the config read of PCI_HEADER_TYPE fails, pci_setup_device() returns
-EIO, but pci_iov_add_virtfn() doesn't check it. Can you update
pci_iov_add_virtfn() so it checks that return value? That should fix the
crash, even without your designware patch.
Obviously, it won't make SR-IOV work, so we still need both patches.
Bjorn
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 26+ messages in thread
* [PATCH v1] arm64:pci: fix the IOV device enabled crash issue in designware
@ 2016-08-25 18:09 ` Bjorn Helgaas
0 siblings, 0 replies; 26+ messages in thread
From: Bjorn Helgaas @ 2016-08-25 18:09 UTC (permalink / raw)
To: linux-arm-kernel
On Thu, Aug 25, 2016 at 04:53:19AM +0000, Po Liu wrote:
> > -----Original Message-----
> > From: Bjorn Helgaas [mailto:helgaas at kernel.org]
> > Sent: Thursday, August 25, 2016 4:51 AM
> > To: Po Liu
> > Cc: linux-pci at vger.kernel.org; Roy Zang; Arnd Bergmann; Stuart Yoder;
> > Yang-Leo Li; linux-arm-kernel at lists.infradead.org; Bjorn Helgaas;
> > Mingkai Hu; Ley Foon Tan; Michal Simek; S?ren Brinkmann; Jingoo Han;
> > Pratyush Anand
> > Subject: Re: [PATCH v1] arm64:pci: fix the IOV device enabled crash
> > issue in designware
> >
> > [+cc Jingoo, Pratyush, Michal, S?ren, Ley]
> >
> > On Tue, Aug 23, 2016 at 02:01:12PM +0800, Po Liu wrote:
> > > When echo a number to /sys/bus/pci/devices/xxx/sriov_numvfs to enable
> > > the VF devices. A crash log occurred. This found to be access the IOV
> > > devices config space failure issue.
> >
> > What was the actual crash? The mere fact that we made a config read
> > fail should not cause a crash. We might erroneously prevent access to
> > VF devices, but it shouldn't crash. So maybe there's another bug
> > elsewhere that we should fix first.
>
> I built with CONFIG_PCI_IOV=y and notice a crash when I use it:
>
> centqds-60 cd /sys/class/net/
> centqds-61 ls
> enP1p1s0@ enP2p1s0f0@ enP2p1s0f1@ lo@ sit0@
> centqds-62 cd enP2p1s0f1/device
> centqds-63 ls
> broken_parity_status driver_override msi_irqs/ sriov_numvfs
> class enable net/ sriov_totalvfs
> config iommu_group@ power/ subsystem@
> consistent_dma_mask_bits irq remove subsystem_device
> device local_cpulist rescan subsystem_vendor
> devspec local_cpus reset uevent
> dma_mask_bits modalias resource vendor
> driver@ msi_bus rom
> centqds-64 zcat /proc/config.gz | grep _IOV
> CONFIG_PCI_IOV=y
> centqds-65 sudo su
> [root at centqds 0002:01:00.1]# echo 2 > sriov_numvfs
> [ 317.604543] ixgbe 0002:01:00.1 enP2p1s0f1: SR-IOV enabled with 2 VFs
> [ 317.714431] (null): of_irq_parse_pci() failed with rc=134
> [ 317.719906] -----------[ cut here ]-----------
> [ 317.724525] WARNING: CPU: 6 PID: 3179 at drivers/pci/probe.c:1555 pci_device_add+0x144/0x148()
> [ 317.733123] Modules linked in:
> [ 317.736175] CPU: 6 PID: 3179 Comm: bash Not tainted 4.1.8-00024-g0a32d65-dirty #32
> [ 317.743731] Hardware name: Freescale Layerscape 2088a QDS Board (DT)
> [ 317.750077] Call trace:
> [ 317.752516] [<fffffe0000096cf8>] dump_backtrace+0x0/0x12c
> Message from[ 317.757916] [<fffffe0000096e34>] show_stack+0x10/0x1c
> syslogd at centqds[ 317.764341] [<fffffe00008937fc>] dump_stack+0x84/0xd4
> at Jul 26 15:51[ 317.770770] [<fffffe00000be590>] warn_slowpath_common+0x94/0xcc
> :10 ...
> kerne[ 317.778067] [<fffffe00000be68c>] warn_slowpath_null+0x14/0x20
> l:Call trace:
> [ 317.785192] [<fffffe00003c5fa4>] pci_device_add+0x140/0x148
> [ 317.792133] [<fffffe00003de784>] pci_enable_sriov+0x470/0x7a0
> [ 317.797873] [<fffffe00005fa9dc>] ixgbe_pci_sriov_configure+0x8c/0x148
> [ 317.804302] [<fffffe00003cf354>] sriov_numvfs_store+0x78/0x11c
> [ 317.810129] [<fffffe0000443d80>] dev_attr_store+0x14/0x28
> [ 317.815521] [<fffffe0000200164>] sysfs_kf_write+0x40/0x4c
> [ 317.820908] [<fffffe00001ff5ec>] kernfs_fop_write+0xb8/0x180
> [ 317.826561] [<fffffe000019a568>] __vfs_write+0x28/0x10c
> [ 317.831775] [<fffffe000019ad00>] vfs_write+0x90/0x1a0
> [ 317.836819] [<fffffe000019b584>] SyS_write+0x40/0xa0
> [ 317.841772] --[ end trace 83725a9784fd702a ]--
> [ 317.846393] BUG: failure at fs/sysfs/file.c:481/sysfs_create_bin_file()!
> [ 317.853081] Kernel panic - not syncing: BUG!
> [ 317.857339] CPU: 6 PID: 3179 Comm: bash Tainted: G W 4.1.8-00024-g0a32d65-dirty #32
> [ 317.866110] Hardware name: Freescale Layerscape 2088a QDS Board (DT)
> [ 317.872451] Call trace:
> [ 317.874887] [<fffffe0000096cf8>] dump_backtrace+0x0/0x12c
> [ 317.880274] [<fffffe0000096e34>] show_stack+0x10/0x1c
> [ 317.885315] [<fffffe00008937fc>] dump_stack+0x84/0xd4
> [ 317.890354] [<fffffe00008909e8>] panic+0xe4/0x21c
> [ 317.895047] [<fffffe0000200988>] sysfs_create_bin_file+0x60/0x64
> [ 317.901041] [<fffffe00003cf52c>] pci_create_sysfs_dev_files+0x48/0x2a8
> [ 317.907556] [<fffffe00003c4238>] pci_bus_add_device+0x20/0x6c
>
> The code process is that: "echo 2 > sriov_numvf" makes driver load .sriov_configure. At last to load pci_enable_sriov().
> The first time vf device operate the config space in the pci_setup_device() (this function was load in the virtfn_add()) is pci_read_config_byte(dev, PCI_HEADER_TYPE, &hdr_type) return failure. So the virtfn didn't initialized proper.
>
> This found to be "bus->primary == pp->root_bus_nr && dev > 0" then return failure in host controller. The dev came from devfn must not zero(is about 0x10).
>
> then read config space failure. This makes the dev->bus is NULL. Lead to upper crash.
I think the crash (BUG: failure at fs/sysfs/file.c:481/sysfs_create_bin_file())
happens in this path:
sriov_numvfs_store
ixgbe_pci_sriov_configure
ixgbe_pci_sriov_enable
pci_enable_sriov
sriov_enable
pci_iov_add_virtfn
virtfn = pci_alloc_dev()
pci_setup_device(virtfn)
if (pci_read_config_byte(dev, PCI_HEADER_TYPE, &hdr_type))
return -EIO
pci_bus_add_device(virtfn)
pci_create_sysfs_dev_files
sysfs_create_bin_file
BUG_ON(!kobj)
If the config read of PCI_HEADER_TYPE fails, pci_setup_device() returns
-EIO, but pci_iov_add_virtfn() doesn't check it. Can you update
pci_iov_add_virtfn() so it checks that return value? That should fix the
crash, even without your designware patch.
Obviously, it won't make SR-IOV work, so we still need both patches.
Bjorn
^ permalink raw reply [flat|nested] 26+ messages in thread
* RE: [PATCH v1] arm64:pci: fix the IOV device enabled crash issue in designware
2016-08-25 18:09 ` Bjorn Helgaas
@ 2016-08-26 8:17 ` Po Liu
-1 siblings, 0 replies; 26+ messages in thread
From: Po Liu @ 2016-08-26 8:17 UTC (permalink / raw)
To: Bjorn Helgaas
Cc: linux-arm-kernel, Roy Zang, Arnd Bergmann, linux-pci,
Pratyush Anand, Stuart Yoder, Yang-Leo Li, Mingkai Hu,
Jingoo Han, Bjorn Helgaas, Ley Foon Tan, Michal Simek,
S?ren Brinkmann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^ permalink raw reply [flat|nested] 26+ messages in thread
* [PATCH v1] arm64:pci: fix the IOV device enabled crash issue in designware
@ 2016-08-26 8:17 ` Po Liu
0 siblings, 0 replies; 26+ messages in thread
From: Po Liu @ 2016-08-26 8:17 UTC (permalink / raw)
To: linux-arm-kernel
Hi Bjorn,
Best regards,
Liu Po
> -----Original Message-----
> From: Bjorn Helgaas [mailto:helgaas at kernel.org]
> Sent: Friday, August 26, 2016 2:10 AM
> To: Po Liu
> Cc: S?ren Brinkmann; Roy Zang; Arnd Bergmann; linux-pci at vger.kernel.org;
> Pratyush Anand; Stuart Yoder; Yang-Leo Li; Mingkai Hu; Jingoo Han; Bjorn
> Helgaas; Ley Foon Tan; Michal Simek; linux-arm-
> kernel at lists.infradead.org
> Subject: Re: [PATCH v1] arm64:pci: fix the IOV device enabled crash
> issue in designware
>
> On Thu, Aug 25, 2016 at 04:53:19AM +0000, Po Liu wrote:
> > > -----Original Message-----
> > > From: Bjorn Helgaas [mailto:helgaas at kernel.org]
> > > Sent: Thursday, August 25, 2016 4:51 AM
> > > To: Po Liu
> > > Cc: linux-pci at vger.kernel.org; Roy Zang; Arnd Bergmann; Stuart
> > > Yoder; Yang-Leo Li; linux-arm-kernel at lists.infradead.org; Bjorn
> > > Helgaas; Mingkai Hu; Ley Foon Tan; Michal Simek; S?ren Brinkmann;
> > > Jingoo Han; Pratyush Anand
> > > Subject: Re: [PATCH v1] arm64:pci: fix the IOV device enabled crash
> > > issue in designware
> > >
> > > [+cc Jingoo, Pratyush, Michal, S?ren, Ley]
> > >
> > > On Tue, Aug 23, 2016 at 02:01:12PM +0800, Po Liu wrote:
> > > > When echo a number to /sys/bus/pci/devices/xxx/sriov_numvfs to
> > > enable > the VF devices. A crash log occurred. This found to be
> > > access the IOV > devices config space failure issue.
> > >
> > > What was the actual crash? The mere fact that we made a config
> > > read fail should not cause a crash. We might erroneously prevent
> > > access to VF devices, but it shouldn't crash. So maybe there's
> > > another bug elsewhere that we should fix first.
> >
> > I built with CONFIG_PCI_IOV=y and notice a crash when I use it:
> >
> > centqds-60 cd /sys/class/net/
> > centqds-61 ls
> > enP1p1s0@ enP2p1s0f0@ enP2p1s0f1@ lo@ sit0@
> > centqds-62 cd enP2p1s0f1/device
> > centqds-63 ls
> > broken_parity_status driver_override msi_irqs/ sriov_numvfs class
> > enable net/ sriov_totalvfs config iommu_group@ power/ subsystem@
> > consistent_dma_mask_bits irq remove subsystem_device device
> > local_cpulist rescan subsystem_vendor devspec local_cpus reset uevent
> > dma_mask_bits modalias resource vendor driver@ msi_bus rom
> > centqds-64 zcat /proc/config.gz | grep _IOV CONFIG_PCI_IOV=y
> > centqds-65 sudo su
> > [root at centqds 0002:01:00.1]# echo 2 > sriov_numvfs [ 317.604543] ixgbe
> > 0002:01:00.1 enP2p1s0f1: SR-IOV enabled with 2 VFs [ 317.714431]
> > (null): of_irq_parse_pci() failed with rc=134
>
>
> > [ 317.719906] -----------[ cut here ]----------- [ 317.724525]
> > WARNING: CPU: 6 PID: 3179 at drivers/pci/probe.c:1555
> > pci_device_add+0x144/0x148() [ 317.733123] Modules linked in:
> > [ 317.736175] CPU: 6 PID: 3179 Comm: bash Not tainted
> > 4.1.8-00024-g0a32d65-dirty #32 [ 317.743731] Hardware name: Freescale
> > Layerscape 2088a QDS Board (DT) [ 317.750077] Call trace:
> > [ 317.752516] [<fffffe0000096cf8>] dump_backtrace+0x0/0x12c Message
> > from[ 317.757916] [<fffffe0000096e34>] show_stack+0x10/0x1c
> > syslogd at centqds[ 317.764341] [<fffffe00008937fc>] dump_stack+0x84/0xd4
> > at Jul 26 15:51[ 317.770770] [<fffffe00000be590>]
> > warn_slowpath_common+0x94/0xcc
> > :10 ...
> > kerne[ 317.778067] [<fffffe00000be68c>] warn_slowpath_null+0x14/0x20
> > l:Call trace:
> > [ 317.785192] [<fffffe00003c5fa4>] pci_device_add+0x140/0x148 [
> > 317.792133] [<fffffe00003de784>] pci_enable_sriov+0x470/0x7a0 [
> > 317.797873] [<fffffe00005fa9dc>] ixgbe_pci_sriov_configure+0x8c/0x148
> > [ 317.804302] [<fffffe00003cf354>] sriov_numvfs_store+0x78/0x11c [
> > 317.810129] [<fffffe0000443d80>] dev_attr_store+0x14/0x28 [
> > 317.815521] [<fffffe0000200164>] sysfs_kf_write+0x40/0x4c [
> > 317.820908] [<fffffe00001ff5ec>] kernfs_fop_write+0xb8/0x180 [
> > 317.826561] [<fffffe000019a568>] __vfs_write+0x28/0x10c [ 317.831775]
> > [<fffffe000019ad00>] vfs_write+0x90/0x1a0 [ 317.836819]
> > [<fffffe000019b584>] SyS_write+0x40/0xa0 [ 317.841772] --[ end trace
> > 83725a9784fd702a ]-- [ 317.846393] BUG: failure at
> > fs/sysfs/file.c:481/sysfs_create_bin_file()!
> > [ 317.853081] Kernel panic - not syncing: BUG!
> > [ 317.857339] CPU: 6 PID: 3179 Comm: bash Tainted: G W
> > 4.1.8-00024-g0a32d65-dirty #32 [ 317.866110] Hardware name: Freescale
> > Layerscape 2088a QDS Board (DT) [ 317.872451] Call trace:
> > [ 317.874887] [<fffffe0000096cf8>] dump_backtrace+0x0/0x12c [
> > 317.880274] [<fffffe0000096e34>] show_stack+0x10/0x1c [ 317.885315]
> > [<fffffe00008937fc>] dump_stack+0x84/0xd4 [ 317.890354]
> > [<fffffe00008909e8>] panic+0xe4/0x21c [ 317.895047]
> > [<fffffe0000200988>] sysfs_create_bin_file+0x60/0x64 [ 317.901041]
> > [<fffffe00003cf52c>] pci_create_sysfs_dev_files+0x48/0x2a8
> > [ 317.907556] [<fffffe00003c4238>] pci_bus_add_device+0x20/0x6c
> >
> > The code process is that: "echo 2 > sriov_numvf" makes driver
> load .sriov_configure. At last to load pci_enable_sriov().
> > The first time vf device operate the config space in the
> pci_setup_device() (this function was load in the virtfn_add()) is
> pci_read_config_byte(dev, PCI_HEADER_TYPE, &hdr_type) return failure. So
> the virtfn didn't initialized proper.
> >
> > This found to be "bus->primary == pp->root_bus_nr && dev > 0" then
> return failure in host controller. The dev came from devfn must not
> zero(is about 0x10).
> >
> > then read config space failure. This makes the dev->bus is NULL. Lead
> to upper crash.
>
> I think the crash (BUG: failure at
> fs/sysfs/file.c:481/sysfs_create_bin_file())
> happens in this path:
>
> sriov_numvfs_store
> ixgbe_pci_sriov_configure
> ixgbe_pci_sriov_enable
> pci_enable_sriov
> sriov_enable
> pci_iov_add_virtfn
> virtfn = pci_alloc_dev()
> pci_setup_device(virtfn)
> if (pci_read_config_byte(dev, PCI_HEADER_TYPE,
> &hdr_type))
> return -EIO
> pci_bus_add_device(virtfn)
> pci_create_sysfs_dev_files
> sysfs_create_bin_file
> BUG_ON(!kobj)
>
> If the config read of PCI_HEADER_TYPE fails, pci_setup_device() returns
> -EIO, but pci_iov_add_virtfn() doesn't check it. Can you update
> pci_iov_add_virtfn() so it checks that return value? That should fix
> the crash, even without your designware patch.
>
> Obviously, it won't make SR-IOV work, so we still need both patches.
Thanks, I'll provide patch in V2.
>
> Bjorn
^ permalink raw reply [flat|nested] 26+ messages in thread
* [PATCH v2] arm64:pci: fix the IOV device access config space valid condition
2016-08-23 6:01 ` Po Liu
@ 2016-08-29 7:26 ` Po Liu
-1 siblings, 0 replies; 26+ messages in thread
From: Po Liu @ 2016-08-29 7:26 UTC (permalink / raw)
To: linux-pci
Cc: soren.brinkmann, Roy Zang, Arnd Bergmann, jingoohan1, Po Liu,
pratyush.anand, Stuart Yoder, Yang-Leo Li, linux-arm-kernel,
Bjorn Helgaas, lftan, michal.simek, Mingkai Hu
When echo a number to /sys/bus/pci/devices/xxx/sriov_numvfs to enable the
VF devices. A crash log occurred. This found to be access the IOV devices
config space failure issue.
The read/write config space from host would judge the pcie device plugin
or not by(Designware platform as example):
if (bus->primary == pp->root_bus_nr && dev > 0)
return 0;
Although all PCIe devices for dev(coming from the device and function
number) is zero. But the dev is not zero for VF devices. So remove the
condition.
These PCI hosts were changed: designware, altera, xilinx.
Signed-off-by: Po Liu <po.liu@nxp.com>
---
changes for v2:
- add pci hosts: altera, xilinx;
drivers/pci/host/pcie-altera.c | 7 -------
drivers/pci/host/pcie-designware.c | 7 -------
drivers/pci/host/pcie-xilinx.c | 7 -------
3 files changed, 21 deletions(-)
diff --git a/drivers/pci/host/pcie-altera.c b/drivers/pci/host/pcie-altera.c
index 2b78376..edbe0a7 100644
--- a/drivers/pci/host/pcie-altera.c
+++ b/drivers/pci/host/pcie-altera.c
@@ -171,13 +171,6 @@ static bool altera_pcie_valid_config(struct altera_pcie *pcie,
if (bus->number == pcie->root_bus_nr && dev > 0)
return false;
- /*
- * Do not read more than one device on the bus directly attached
- * to root port, root port can only attach to one downstream port.
- */
- if (bus->primary == pcie->root_bus_nr && dev > 0)
- return false;
-
return true;
}
diff --git a/drivers/pci/host/pcie-designware.c b/drivers/pci/host/pcie-designware.c
index 12afce1..dd20eb2 100644
--- a/drivers/pci/host/pcie-designware.c
+++ b/drivers/pci/host/pcie-designware.c
@@ -670,13 +670,6 @@ static int dw_pcie_valid_config(struct pcie_port *pp,
if (bus->number == pp->root_bus_nr && dev > 0)
return 0;
- /*
- * do not read more than one device on the bus directly attached
- * to RC's (Virtual Bridge's) DS side.
- */
- if (bus->primary == pp->root_bus_nr && dev > 0)
- return 0;
-
return 1;
}
diff --git a/drivers/pci/host/pcie-xilinx.c b/drivers/pci/host/pcie-xilinx.c
index a30e016..75c89db 100644
--- a/drivers/pci/host/pcie-xilinx.c
+++ b/drivers/pci/host/pcie-xilinx.c
@@ -168,13 +168,6 @@ static bool xilinx_pcie_valid_device(struct pci_bus *bus, unsigned int devfn)
if (bus->number == port->root_busno && devfn > 0)
return false;
- /*
- * Do not read more than one device on the bus directly attached
- * to RC.
- */
- if (bus->primary == port->root_busno && devfn > 0)
- return false;
-
return true;
}
--
2.1.0.27.g96db324
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 26+ messages in thread
* [PATCH v2] arm64:pci: fix the IOV device access config space valid condition
@ 2016-08-29 7:26 ` Po Liu
0 siblings, 0 replies; 26+ messages in thread
From: Po Liu @ 2016-08-29 7:26 UTC (permalink / raw)
To: linux-arm-kernel
When echo a number to /sys/bus/pci/devices/xxx/sriov_numvfs to enable the
VF devices. A crash log occurred. This found to be access the IOV devices
config space failure issue.
The read/write config space from host would judge the pcie device plugin
or not by(Designware platform as example):
if (bus->primary == pp->root_bus_nr && dev > 0)
return 0;
Although all PCIe devices for dev(coming from the device and function
number) is zero. But the dev is not zero for VF devices. So remove the
condition.
These PCI hosts were changed: designware, altera, xilinx.
Signed-off-by: Po Liu <po.liu@nxp.com>
---
changes for v2:
- add pci hosts: altera, xilinx;
drivers/pci/host/pcie-altera.c | 7 -------
drivers/pci/host/pcie-designware.c | 7 -------
drivers/pci/host/pcie-xilinx.c | 7 -------
3 files changed, 21 deletions(-)
diff --git a/drivers/pci/host/pcie-altera.c b/drivers/pci/host/pcie-altera.c
index 2b78376..edbe0a7 100644
--- a/drivers/pci/host/pcie-altera.c
+++ b/drivers/pci/host/pcie-altera.c
@@ -171,13 +171,6 @@ static bool altera_pcie_valid_config(struct altera_pcie *pcie,
if (bus->number == pcie->root_bus_nr && dev > 0)
return false;
- /*
- * Do not read more than one device on the bus directly attached
- * to root port, root port can only attach to one downstream port.
- */
- if (bus->primary == pcie->root_bus_nr && dev > 0)
- return false;
-
return true;
}
diff --git a/drivers/pci/host/pcie-designware.c b/drivers/pci/host/pcie-designware.c
index 12afce1..dd20eb2 100644
--- a/drivers/pci/host/pcie-designware.c
+++ b/drivers/pci/host/pcie-designware.c
@@ -670,13 +670,6 @@ static int dw_pcie_valid_config(struct pcie_port *pp,
if (bus->number == pp->root_bus_nr && dev > 0)
return 0;
- /*
- * do not read more than one device on the bus directly attached
- * to RC's (Virtual Bridge's) DS side.
- */
- if (bus->primary == pp->root_bus_nr && dev > 0)
- return 0;
-
return 1;
}
diff --git a/drivers/pci/host/pcie-xilinx.c b/drivers/pci/host/pcie-xilinx.c
index a30e016..75c89db 100644
--- a/drivers/pci/host/pcie-xilinx.c
+++ b/drivers/pci/host/pcie-xilinx.c
@@ -168,13 +168,6 @@ static bool xilinx_pcie_valid_device(struct pci_bus *bus, unsigned int devfn)
if (bus->number == port->root_busno && devfn > 0)
return false;
- /*
- * Do not read more than one device on the bus directly attached
- * to RC.
- */
- if (bus->primary == port->root_busno && devfn > 0)
- return false;
-
return true;
}
--
2.1.0.27.g96db324
^ permalink raw reply related [flat|nested] 26+ messages in thread
* Re: [PATCH v2] arm64:pci: fix the IOV device access config space valid condition
2016-08-29 7:26 ` Po Liu
@ 2016-08-30 15:43 ` Jingoo Han
-1 siblings, 0 replies; 26+ messages in thread
From: Jingoo Han @ 2016-08-30 15:43 UTC (permalink / raw)
To: 'Po Liu', linux-pci
Cc: soren.brinkmann, 'Roy Zang', 'Arnd Bergmann',
pratyush.anand, 'Stuart Yoder', 'Yang-Leo Li',
linux-arm-kernel, 'Bjorn Helgaas',
lftan, michal.simek, 'Mingkai Hu'
On Monday, August 29, 2016 3:27 AM, Po Liu wrote:
>
> When echo a number to /sys/bus/pci/devices/xxx/sriov_numvfs to enable the
> VF devices. A crash log occurred. This found to be access the IOV devices
> config space failure issue.
>
> The read/write config space from host would judge the pcie device plugin
> or not by(Designware platform as example):
>
> if (bus->primary == pp->root_bus_nr && dev > 0)
> return 0;
>
> Although all PCIe devices for dev(coming from the device and function
> number) is zero. But the dev is not zero for VF devices. So remove the
> condition.
>
> These PCI hosts were changed: designware, altera, xilinx.
>
> Signed-off-by: Po Liu <po.liu@nxp.com>
For 'pcie-designware.c',
Acked-by: Jingoo Han <jingoohan1@gmail.com>
I really don't like to modify host driver stuffs.
But, if there is no alternatives, this patch looks good.
Best regards,
Jingoo Han
> ---
> changes for v2:
> - add pci hosts: altera, xilinx;
>
> drivers/pci/host/pcie-altera.c | 7 -------
> drivers/pci/host/pcie-designware.c | 7 -------
> drivers/pci/host/pcie-xilinx.c | 7 -------
> 3 files changed, 21 deletions(-)
>
> diff --git a/drivers/pci/host/pcie-altera.c b/drivers/pci/host/pcie-
> altera.c index 2b78376..edbe0a7 100644
> --- a/drivers/pci/host/pcie-altera.c
> +++ b/drivers/pci/host/pcie-altera.c
> @@ -171,13 +171,6 @@ static bool altera_pcie_valid_config(struct
> altera_pcie *pcie,
> if (bus->number == pcie->root_bus_nr && dev > 0)
> return false;
>
> - /*
> - * Do not read more than one device on the bus directly attached
> - * to root port, root port can only attach to one downstream port.
> - */
> - if (bus->primary == pcie->root_bus_nr && dev > 0)
> - return false;
> -
> return true;
> }
>
> diff --git a/drivers/pci/host/pcie-designware.c b/drivers/pci/host/pcie-
> designware.c
> index 12afce1..dd20eb2 100644
> --- a/drivers/pci/host/pcie-designware.c
> +++ b/drivers/pci/host/pcie-designware.c
> @@ -670,13 +670,6 @@ static int dw_pcie_valid_config(struct pcie_port *pp,
> if (bus->number == pp->root_bus_nr && dev > 0)
> return 0;
>
> - /*
> - * do not read more than one device on the bus directly attached
> - * to RC's (Virtual Bridge's) DS side.
> - */
> - if (bus->primary == pp->root_bus_nr && dev > 0)
> - return 0;
> -
> return 1;
> }
>
> diff --git a/drivers/pci/host/pcie-xilinx.c b/drivers/pci/host/pcie-
> xilinx.c index a30e016..75c89db 100644
> --- a/drivers/pci/host/pcie-xilinx.c
> +++ b/drivers/pci/host/pcie-xilinx.c
> @@ -168,13 +168,6 @@ static bool xilinx_pcie_valid_device(struct pci_bus
> *bus, unsigned int devfn)
> if (bus->number == port->root_busno && devfn > 0)
> return false;
>
> - /*
> - * Do not read more than one device on the bus directly attached
> - * to RC.
> - */
> - if (bus->primary == port->root_busno && devfn > 0)
> - return false;
> -
> return true;
> }
>
> --
> 2.1.0.27.g96db324
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 26+ messages in thread
* [PATCH v2] arm64:pci: fix the IOV device access config space valid condition
@ 2016-08-30 15:43 ` Jingoo Han
0 siblings, 0 replies; 26+ messages in thread
From: Jingoo Han @ 2016-08-30 15:43 UTC (permalink / raw)
To: linux-arm-kernel
On Monday, August 29, 2016 3:27 AM, Po Liu wrote:
>
> When echo a number to /sys/bus/pci/devices/xxx/sriov_numvfs to enable the
> VF devices. A crash log occurred. This found to be access the IOV devices
> config space failure issue.
>
> The read/write config space from host would judge the pcie device plugin
> or not by(Designware platform as example):
>
> if (bus->primary == pp->root_bus_nr && dev > 0)
> return 0;
>
> Although all PCIe devices for dev(coming from the device and function
> number) is zero. But the dev is not zero for VF devices. So remove the
> condition.
>
> These PCI hosts were changed: designware, altera, xilinx.
>
> Signed-off-by: Po Liu <po.liu@nxp.com>
For 'pcie-designware.c',
Acked-by: Jingoo Han <jingoohan1@gmail.com>
I really don't like to modify host driver stuffs.
But, if there is no alternatives, this patch looks good.
Best regards,
Jingoo Han
> ---
> changes for v2:
> - add pci hosts: altera, xilinx;
>
> drivers/pci/host/pcie-altera.c | 7 -------
> drivers/pci/host/pcie-designware.c | 7 -------
> drivers/pci/host/pcie-xilinx.c | 7 -------
> 3 files changed, 21 deletions(-)
>
> diff --git a/drivers/pci/host/pcie-altera.c b/drivers/pci/host/pcie-
> altera.c index 2b78376..edbe0a7 100644
> --- a/drivers/pci/host/pcie-altera.c
> +++ b/drivers/pci/host/pcie-altera.c
> @@ -171,13 +171,6 @@ static bool altera_pcie_valid_config(struct
> altera_pcie *pcie,
> if (bus->number == pcie->root_bus_nr && dev > 0)
> return false;
>
> - /*
> - * Do not read more than one device on the bus directly attached
> - * to root port, root port can only attach to one downstream port.
> - */
> - if (bus->primary == pcie->root_bus_nr && dev > 0)
> - return false;
> -
> return true;
> }
>
> diff --git a/drivers/pci/host/pcie-designware.c b/drivers/pci/host/pcie-
> designware.c
> index 12afce1..dd20eb2 100644
> --- a/drivers/pci/host/pcie-designware.c
> +++ b/drivers/pci/host/pcie-designware.c
> @@ -670,13 +670,6 @@ static int dw_pcie_valid_config(struct pcie_port *pp,
> if (bus->number == pp->root_bus_nr && dev > 0)
> return 0;
>
> - /*
> - * do not read more than one device on the bus directly attached
> - * to RC's (Virtual Bridge's) DS side.
> - */
> - if (bus->primary == pp->root_bus_nr && dev > 0)
> - return 0;
> -
> return 1;
> }
>
> diff --git a/drivers/pci/host/pcie-xilinx.c b/drivers/pci/host/pcie-
> xilinx.c index a30e016..75c89db 100644
> --- a/drivers/pci/host/pcie-xilinx.c
> +++ b/drivers/pci/host/pcie-xilinx.c
> @@ -168,13 +168,6 @@ static bool xilinx_pcie_valid_device(struct pci_bus
> *bus, unsigned int devfn)
> if (bus->number == port->root_busno && devfn > 0)
> return false;
>
> - /*
> - * Do not read more than one device on the bus directly attached
> - * to RC.
> - */
> - if (bus->primary == port->root_busno && devfn > 0)
> - return false;
> -
> return true;
> }
>
> --
> 2.1.0.27.g96db324
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH v2] arm64:pci: fix the IOV device access config space valid condition
2016-08-30 15:43 ` Jingoo Han
@ 2016-08-30 15:54 ` Roy Zang
-1 siblings, 0 replies; 26+ messages in thread
From: Roy Zang @ 2016-08-30 15:54 UTC (permalink / raw)
To: Jingoo Han, Po Liu, linux-pci
Cc: soren.brinkmann, 'Arnd Bergmann',
pratyush.anand, Stuart Yoder, Yang-Leo Li, linux-arm-kernel,
'Bjorn Helgaas',
lftan, michal.simek, Mingkai Hu
On 08/30/2016 10:43 AM, Jingoo Han wrote:
> I really don't like to modify host driver stuffs.
> But, if there is no alternatives, this patch looks good.
I do not see other good place to fix the bug.
Roy Zang <roy.zang@nxp.com>
Roy
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 26+ messages in thread
* [PATCH v2] arm64:pci: fix the IOV device access config space valid condition
@ 2016-08-30 15:54 ` Roy Zang
0 siblings, 0 replies; 26+ messages in thread
From: Roy Zang @ 2016-08-30 15:54 UTC (permalink / raw)
To: linux-arm-kernel
On 08/30/2016 10:43 AM, Jingoo Han wrote:
> I really don't like to modify host driver stuffs.
> But, if there is no alternatives, this patch looks good.
I do not see other good place to fix the bug.
Roy Zang <roy.zang@nxp.com>
Roy
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH v2] arm64:pci: fix the IOV device access config space valid condition
2016-08-30 15:54 ` Roy Zang
@ 2016-08-30 16:32 ` Roy Zang
-1 siblings, 0 replies; 26+ messages in thread
From: Roy Zang @ 2016-08-30 16:32 UTC (permalink / raw)
To: Jingoo Han, Po Liu, linux-pci
Cc: soren.brinkmann, 'Arnd Bergmann',
pratyush.anand, Stuart Yoder, Yang-Leo Li, linux-arm-kernel,
'Bjorn Helgaas',
lftan, michal.simek, Mingkai Hu
Acked-by: Roy Zang <roy.zang@nxp.com>
Roy
On 08/30/2016 10:54 AM, Roy Zang wrote:
> On 08/30/2016 10:43 AM, Jingoo Han wrote:
>> I really don't like to modify host driver stuffs.
>> But, if there is no alternatives, this patch looks good.
> I do not see other good place to fix the bug.
>
> Roy Zang <roy.zang@nxp.com>
>
> Roy
>
>
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 26+ messages in thread
* [PATCH v2] arm64:pci: fix the IOV device access config space valid condition
@ 2016-08-30 16:32 ` Roy Zang
0 siblings, 0 replies; 26+ messages in thread
From: Roy Zang @ 2016-08-30 16:32 UTC (permalink / raw)
To: linux-arm-kernel
Acked-by: Roy Zang <roy.zang@nxp.com>
Roy
On 08/30/2016 10:54 AM, Roy Zang wrote:
> On 08/30/2016 10:43 AM, Jingoo Han wrote:
>> I really don't like to modify host driver stuffs.
>> But, if there is no alternatives, this patch looks good.
> I do not see other good place to fix the bug.
>
> Roy Zang <roy.zang@nxp.com>
>
> Roy
>
>
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH v2] arm64:pci: fix the IOV device access config space valid condition
2016-08-30 15:54 ` Roy Zang
@ 2016-08-30 18:15 ` Jingoo Han
-1 siblings, 0 replies; 26+ messages in thread
From: Jingoo Han @ 2016-08-30 18:15 UTC (permalink / raw)
To: 'Roy Zang', 'Po Liu', linux-pci
Cc: soren.brinkmann, 'Arnd Bergmann',
pratyush.anand, 'Stuart Yoder', 'Yang-Leo Li',
linux-arm-kernel, 'Bjorn Helgaas',
lftan, michal.simek, 'Mingkai Hu'
On Tuesday, August 30, 2016 11:55 AM, Roy Zang wrote:
>
> On 08/30/2016 10:43 AM, Jingoo Han wrote:
> > I really don't like to modify host driver stuffs.
> > But, if there is no alternatives, this patch looks good.
>
> I do not see other good place to fix the bug.
Hi Roy Zang,
I didn't find other good place, too.
At that time, I did IOV test with the similar patch like this.
Best regards,
Jingoo Han
>
> Roy Zang <roy.zang@nxp.com>
>
> Roy
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 26+ messages in thread
* [PATCH v2] arm64:pci: fix the IOV device access config space valid condition
@ 2016-08-30 18:15 ` Jingoo Han
0 siblings, 0 replies; 26+ messages in thread
From: Jingoo Han @ 2016-08-30 18:15 UTC (permalink / raw)
To: linux-arm-kernel
On Tuesday, August 30, 2016 11:55 AM, Roy Zang wrote:
>
> On 08/30/2016 10:43 AM, Jingoo Han wrote:
> > I really don't like to modify host driver stuffs.
> > But, if there is no alternatives, this patch looks good.
>
> I do not see other good place to fix the bug.
Hi Roy Zang,
I didn't find other good place, too.
At that time, I did IOV test with the similar patch like this.
Best regards,
Jingoo Han
>
> Roy Zang <roy.zang@nxp.com>
>
> Roy
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH v2] arm64:pci: fix the IOV device access config space valid condition
2016-08-30 18:15 ` Jingoo Han
@ 2016-08-30 18:25 ` Roy Zang
-1 siblings, 0 replies; 26+ messages in thread
From: Roy Zang @ 2016-08-30 18:25 UTC (permalink / raw)
To: Jingoo Han, Po Liu, linux-pci
Cc: soren.brinkmann, 'Arnd Bergmann',
pratyush.anand, Stuart Yoder, Yang-Leo Li, linux-arm-kernel,
'Bjorn Helgaas',
lftan, michal.simek, Mingkai Hu
On 08/30/2016 01:15 PM, Jingoo Han wrote:
> On Tuesday, August 30, 2016 11:55 AM, Roy Zang wrote:
>> >
>> > On 08/30/2016 10:43 AM, Jingoo Han wrote:
>>> > > I really don't like to modify host driver stuffs.
>>> > > But, if there is no alternatives, this patch looks good.
>> >
>> > I do not see other good place to fix the bug.
> Hi Roy Zang,
>
> I didn't find other good place, too.
> At that time, I did IOV test with the similar patch like this.
>
> Best regards,
> Jingoo Han
>
I did the similar with SR-IOV capability card. the patch fix the issue.
Thanks for your confirmation and ack.
Roy
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 26+ messages in thread
* [PATCH v2] arm64:pci: fix the IOV device access config space valid condition
@ 2016-08-30 18:25 ` Roy Zang
0 siblings, 0 replies; 26+ messages in thread
From: Roy Zang @ 2016-08-30 18:25 UTC (permalink / raw)
To: linux-arm-kernel
On 08/30/2016 01:15 PM, Jingoo Han wrote:
> On Tuesday, August 30, 2016 11:55 AM, Roy Zang wrote:
>> >
>> > On 08/30/2016 10:43 AM, Jingoo Han wrote:
>>> > > I really don't like to modify host driver stuffs.
>>> > > But, if there is no alternatives, this patch looks good.
>> >
>> > I do not see other good place to fix the bug.
> Hi Roy Zang,
>
> I didn't find other good place, too.
> At that time, I did IOV test with the similar patch like this.
>
> Best regards,
> Jingoo Han
>
I did the similar with SR-IOV capability card. the patch fix the issue.
Thanks for your confirmation and ack.
Roy
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH v2] arm64:pci: fix the IOV device access config space valid condition
2016-08-29 7:26 ` Po Liu
@ 2016-09-01 0:58 ` Ley Foon Tan
-1 siblings, 0 replies; 26+ messages in thread
From: Ley Foon Tan @ 2016-09-01 0:58 UTC (permalink / raw)
To: Po Liu
Cc: Sören Brinkmann, Roy Zang, Arnd Bergmann, linux-pci,
Pratyush Anand Thakur, Stuart Yoder, Yang-Leo Li, linux-arm,
Jingoo Han, Bjorn Helgaas, Michal Simek, Mingkai Hu
On Mon, Aug 29, 2016 at 3:26 PM, Po Liu <po.liu@nxp.com> wrote:
> When echo a number to /sys/bus/pci/devices/xxx/sriov_numvfs to enable the
> VF devices. A crash log occurred. This found to be access the IOV devices
> config space failure issue.
>
> The read/write config space from host would judge the pcie device plugin
> or not by(Designware platform as example):
>
> if (bus->primary == pp->root_bus_nr && dev > 0)
> return 0;
>
> Although all PCIe devices for dev(coming from the device and function
> number) is zero. But the dev is not zero for VF devices. So remove the
> condition.
>
> These PCI hosts were changed: designware, altera, xilinx.
>
> Signed-off-by: Po Liu <po.liu@nxp.com>
> ---
For pcie-altera.c:
Acked-by: Ley Foon Tan <lftan@altera.com>
Thanks.
Regards
Ley Foon
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 26+ messages in thread
* [PATCH v2] arm64:pci: fix the IOV device access config space valid condition
@ 2016-09-01 0:58 ` Ley Foon Tan
0 siblings, 0 replies; 26+ messages in thread
From: Ley Foon Tan @ 2016-09-01 0:58 UTC (permalink / raw)
To: linux-arm-kernel
On Mon, Aug 29, 2016 at 3:26 PM, Po Liu <po.liu@nxp.com> wrote:
> When echo a number to /sys/bus/pci/devices/xxx/sriov_numvfs to enable the
> VF devices. A crash log occurred. This found to be access the IOV devices
> config space failure issue.
>
> The read/write config space from host would judge the pcie device plugin
> or not by(Designware platform as example):
>
> if (bus->primary == pp->root_bus_nr && dev > 0)
> return 0;
>
> Although all PCIe devices for dev(coming from the device and function
> number) is zero. But the dev is not zero for VF devices. So remove the
> condition.
>
> These PCI hosts were changed: designware, altera, xilinx.
>
> Signed-off-by: Po Liu <po.liu@nxp.com>
> ---
For pcie-altera.c:
Acked-by: Ley Foon Tan <lftan@altera.com>
Thanks.
Regards
Ley Foon
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH v2] arm64:pci: fix the IOV device access config space valid condition
2016-08-29 7:26 ` Po Liu
@ 2016-09-12 21:44 ` Bjorn Helgaas
-1 siblings, 0 replies; 26+ messages in thread
From: Bjorn Helgaas @ 2016-09-12 21:44 UTC (permalink / raw)
To: Po Liu
Cc: linux-pci, soren.brinkmann, Roy Zang, Arnd Bergmann, jingoohan1,
pratyush.anand, Stuart Yoder, Yang-Leo Li, linux-arm-kernel,
Bjorn Helgaas, lftan, michal.simek, Mingkai Hu
On Mon, Aug 29, 2016 at 03:26:58PM +0800, Po Liu wrote:
> When echo a number to /sys/bus/pci/devices/xxx/sriov_numvfs to enable the
> VF devices. A crash log occurred. This found to be access the IOV devices
> config space failure issue.
>
> The read/write config space from host would judge the pcie device plugin
> or not by(Designware platform as example):
>
> if (bus->primary == pp->root_bus_nr && dev > 0)
> return 0;
>
> Although all PCIe devices for dev(coming from the device and function
> number) is zero. But the dev is not zero for VF devices. So remove the
> condition.
>
> These PCI hosts were changed: designware, altera, xilinx.
>
> Signed-off-by: Po Liu <po.liu@nxp.com>
Applied to pci/virtualization for v4.9, thanks!
I split it into three patches for backporting and reversion purposes.
> ---
> changes for v2:
> - add pci hosts: altera, xilinx;
>
> drivers/pci/host/pcie-altera.c | 7 -------
> drivers/pci/host/pcie-designware.c | 7 -------
> drivers/pci/host/pcie-xilinx.c | 7 -------
> 3 files changed, 21 deletions(-)
>
> diff --git a/drivers/pci/host/pcie-altera.c b/drivers/pci/host/pcie-altera.c
> index 2b78376..edbe0a7 100644
> --- a/drivers/pci/host/pcie-altera.c
> +++ b/drivers/pci/host/pcie-altera.c
> @@ -171,13 +171,6 @@ static bool altera_pcie_valid_config(struct altera_pcie *pcie,
> if (bus->number == pcie->root_bus_nr && dev > 0)
> return false;
>
> - /*
> - * Do not read more than one device on the bus directly attached
> - * to root port, root port can only attach to one downstream port.
> - */
> - if (bus->primary == pcie->root_bus_nr && dev > 0)
> - return false;
> -
> return true;
> }
>
> diff --git a/drivers/pci/host/pcie-designware.c b/drivers/pci/host/pcie-designware.c
> index 12afce1..dd20eb2 100644
> --- a/drivers/pci/host/pcie-designware.c
> +++ b/drivers/pci/host/pcie-designware.c
> @@ -670,13 +670,6 @@ static int dw_pcie_valid_config(struct pcie_port *pp,
> if (bus->number == pp->root_bus_nr && dev > 0)
> return 0;
>
> - /*
> - * do not read more than one device on the bus directly attached
> - * to RC's (Virtual Bridge's) DS side.
> - */
> - if (bus->primary == pp->root_bus_nr && dev > 0)
> - return 0;
> -
> return 1;
> }
>
> diff --git a/drivers/pci/host/pcie-xilinx.c b/drivers/pci/host/pcie-xilinx.c
> index a30e016..75c89db 100644
> --- a/drivers/pci/host/pcie-xilinx.c
> +++ b/drivers/pci/host/pcie-xilinx.c
> @@ -168,13 +168,6 @@ static bool xilinx_pcie_valid_device(struct pci_bus *bus, unsigned int devfn)
> if (bus->number == port->root_busno && devfn > 0)
> return false;
>
> - /*
> - * Do not read more than one device on the bus directly attached
> - * to RC.
> - */
> - if (bus->primary == port->root_busno && devfn > 0)
> - return false;
> -
> return true;
> }
>
> --
> 2.1.0.27.g96db324
>
>
> _______________________________________________
> linux-arm-kernel mailing list
> linux-arm-kernel@lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 26+ messages in thread
* [PATCH v2] arm64:pci: fix the IOV device access config space valid condition
@ 2016-09-12 21:44 ` Bjorn Helgaas
0 siblings, 0 replies; 26+ messages in thread
From: Bjorn Helgaas @ 2016-09-12 21:44 UTC (permalink / raw)
To: linux-arm-kernel
On Mon, Aug 29, 2016 at 03:26:58PM +0800, Po Liu wrote:
> When echo a number to /sys/bus/pci/devices/xxx/sriov_numvfs to enable the
> VF devices. A crash log occurred. This found to be access the IOV devices
> config space failure issue.
>
> The read/write config space from host would judge the pcie device plugin
> or not by(Designware platform as example):
>
> if (bus->primary == pp->root_bus_nr && dev > 0)
> return 0;
>
> Although all PCIe devices for dev(coming from the device and function
> number) is zero. But the dev is not zero for VF devices. So remove the
> condition.
>
> These PCI hosts were changed: designware, altera, xilinx.
>
> Signed-off-by: Po Liu <po.liu@nxp.com>
Applied to pci/virtualization for v4.9, thanks!
I split it into three patches for backporting and reversion purposes.
> ---
> changes for v2:
> - add pci hosts: altera, xilinx;
>
> drivers/pci/host/pcie-altera.c | 7 -------
> drivers/pci/host/pcie-designware.c | 7 -------
> drivers/pci/host/pcie-xilinx.c | 7 -------
> 3 files changed, 21 deletions(-)
>
> diff --git a/drivers/pci/host/pcie-altera.c b/drivers/pci/host/pcie-altera.c
> index 2b78376..edbe0a7 100644
> --- a/drivers/pci/host/pcie-altera.c
> +++ b/drivers/pci/host/pcie-altera.c
> @@ -171,13 +171,6 @@ static bool altera_pcie_valid_config(struct altera_pcie *pcie,
> if (bus->number == pcie->root_bus_nr && dev > 0)
> return false;
>
> - /*
> - * Do not read more than one device on the bus directly attached
> - * to root port, root port can only attach to one downstream port.
> - */
> - if (bus->primary == pcie->root_bus_nr && dev > 0)
> - return false;
> -
> return true;
> }
>
> diff --git a/drivers/pci/host/pcie-designware.c b/drivers/pci/host/pcie-designware.c
> index 12afce1..dd20eb2 100644
> --- a/drivers/pci/host/pcie-designware.c
> +++ b/drivers/pci/host/pcie-designware.c
> @@ -670,13 +670,6 @@ static int dw_pcie_valid_config(struct pcie_port *pp,
> if (bus->number == pp->root_bus_nr && dev > 0)
> return 0;
>
> - /*
> - * do not read more than one device on the bus directly attached
> - * to RC's (Virtual Bridge's) DS side.
> - */
> - if (bus->primary == pp->root_bus_nr && dev > 0)
> - return 0;
> -
> return 1;
> }
>
> diff --git a/drivers/pci/host/pcie-xilinx.c b/drivers/pci/host/pcie-xilinx.c
> index a30e016..75c89db 100644
> --- a/drivers/pci/host/pcie-xilinx.c
> +++ b/drivers/pci/host/pcie-xilinx.c
> @@ -168,13 +168,6 @@ static bool xilinx_pcie_valid_device(struct pci_bus *bus, unsigned int devfn)
> if (bus->number == port->root_busno && devfn > 0)
> return false;
>
> - /*
> - * Do not read more than one device on the bus directly attached
> - * to RC.
> - */
> - if (bus->primary == port->root_busno && devfn > 0)
> - return false;
> -
> return true;
> }
>
> --
> 2.1.0.27.g96db324
>
>
> _______________________________________________
> linux-arm-kernel mailing list
> linux-arm-kernel at lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 26+ messages in thread
end of thread, other threads:[~2016-09-12 21:44 UTC | newest]
Thread overview: 26+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2016-08-23 6:01 [PATCH v1] arm64:pci: fix the IOV device enabled crash issue in designware Po Liu
2016-08-23 6:01 ` Po Liu
2016-08-24 20:50 ` Bjorn Helgaas
2016-08-24 20:50 ` Bjorn Helgaas
2016-08-25 4:53 ` Po Liu
2016-08-25 4:53 ` Po Liu
2016-08-25 18:09 ` Bjorn Helgaas
2016-08-25 18:09 ` Bjorn Helgaas
2016-08-26 8:17 ` Po Liu
2016-08-26 8:17 ` Po Liu
2016-08-29 7:26 ` [PATCH v2] arm64:pci: fix the IOV device access config space valid condition Po Liu
2016-08-29 7:26 ` Po Liu
2016-08-30 15:43 ` Jingoo Han
2016-08-30 15:43 ` Jingoo Han
2016-08-30 15:54 ` Roy Zang
2016-08-30 15:54 ` Roy Zang
2016-08-30 16:32 ` Roy Zang
2016-08-30 16:32 ` Roy Zang
2016-08-30 18:15 ` Jingoo Han
2016-08-30 18:15 ` Jingoo Han
2016-08-30 18:25 ` Roy Zang
2016-08-30 18:25 ` Roy Zang
2016-09-01 0:58 ` Ley Foon Tan
2016-09-01 0:58 ` Ley Foon Tan
2016-09-12 21:44 ` Bjorn Helgaas
2016-09-12 21:44 ` Bjorn Helgaas
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.