commit 166e89d99dd85a856343cca51eee781b793801f2 upstream.
Fix legacy IRQ generation for RK3399 PCIe endpoint core according to
the technical reference manual (TRM). Assert and deassert legacy
interrupt (INTx) through the legacy interrupt control register
("PCIE_CLIENT_LEGACY_INT_CTRL") instead of manually generating a PCIe
message. The generation of the legacy interrupt was tested and validated
with the PCIe endpoint test driver.
Link: https://lore.kernel.org/r/20230418074700.1083505-8-rick.wertenbroek@gmail.com
Fixes: cf590b0783 ("PCI: rockchip: Add EP driver for Rockchip PCIe controller")
Tested-by: Damien Le Moal <dlemoal@kernel.org>
Signed-off-by: Rick Wertenbroek <rick.wertenbroek@gmail.com>
Signed-off-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
Reviewed-by: Damien Le Moal <dlemoal@kernel.org>
Cc: stable@vger.kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 9dd3c7c4c8c3f7f010d9cdb7c3f42506d93c9527 upstream.
The RK3399 PCIe controller should wait until the PHY PLLs are locked.
Add poll and timeout to wait for PHY PLLs to be locked. If they cannot
be locked generate error message and jump to error handler. Accessing
registers in the PHY clock domain when PLLs are not locked causes hang
The PHY PLLs status is checked through a side channel register.
This is documented in the TRM section 17.5.8.1 "PCIe Initialization
Sequence".
Link: https://lore.kernel.org/r/20230418074700.1083505-5-rick.wertenbroek@gmail.com
Fixes: cf590b0783 ("PCI: rockchip: Add EP driver for Rockchip PCIe controller")
Tested-by: Damien Le Moal <dlemoal@kernel.org>
Signed-off-by: Rick Wertenbroek <rick.wertenbroek@gmail.com>
Signed-off-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
Reviewed-by: Damien Le Moal <dlemoal@kernel.org>
Cc: stable@vger.kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit f397fd4ac1fa3afcabd8cee030f953ccaed2a364 upstream.
Assert PCI Configuration Enable bit after probe. When this bit is left to
0 in the endpoint mode, the RK3399 PCIe endpoint core will generate
configuration request retry status (CRS) messages back to the root complex.
Assert this bit after probe to allow the RK3399 PCIe endpoint core to reply
to configuration requests from the root complex.
This is documented in section 17.5.8.1.2 of the RK3399 TRM.
Link: https://lore.kernel.org/r/20230418074700.1083505-4-rick.wertenbroek@gmail.com
Fixes: cf590b0783 ("PCI: rockchip: Add EP driver for Rockchip PCIe controller")
Tested-by: Damien Le Moal <dlemoal@kernel.org>
Signed-off-by: Rick Wertenbroek <rick.wertenbroek@gmail.com>
Signed-off-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
Reviewed-by: Damien Le Moal <dlemoal@kernel.org>
Cc: stable@vger.kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit c60738de85f40b0b9f5cb23c21f9246e5a47908c ]
Smatch reported:
1. drivers/pci/controller/pci-ftpci100.c:526 faraday_pci_probe() warn:
'clk' from clk_prepare_enable() not released on lines: 442,451,462,478,512,517.
2. drivers/pci/controller/pci-ftpci100.c:526 faraday_pci_probe() warn:
'p->bus_clk' from clk_prepare_enable() not released on lines: 451,462,478,512,517.
The clock resource is obtained by devm_clk_get(), and then
clk_prepare_enable() makes the clock resource ready for use. After that,
clk_disable_unprepare() should be called to release the clock resource
when it is no longer needed. However, while doing some error handling
in faraday_pci_probe(), clk_disable_unprepare() is not called to release
clk and p->bus_clk before returning. These return lines are exactly 442,
451, 462, 478, 512, 517.
Fix this warning by replacing devm_clk_get() with devm_clk_get_enabled(),
which is equivalent to devm_clk_get() + clk_prepare_enable(). And with
devm_clk_get_enabled(), the clock will automatically be disabled,
unprepared and freed when the device is unbound from the bus.
Link: https://lore.kernel.org/r/20230508043641.23807-1-yejunyan@hust.edu.cn
Fixes: b3c433efb8 ("PCI: faraday: Fix wrong pointer passed to PTR_ERR()")
Fixes: 2eeb02b285 ("PCI: faraday: Add clock handling")
Fixes: 783a862563 ("PCI: faraday: Use pci_parse_request_of_pci_ranges()")
Fixes: d3c68e0a7e ("PCI: faraday: Add Faraday Technology FTPCI100 PCI Host Bridge driver")
Fixes: f1e8bd21e3 ("PCI: faraday: Convert IRQ masking to raw PCI config accessors")
Signed-off-by: Junyan Ye <yejunyan@hust.edu.cn>
Signed-off-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
Reviewed-by: Dongliang Mu <dzm91@hust.edu.cn>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 0e12f830236928b6fadf40d917a7527f0a048d2f ]
The Link Retraining process is initiated to account for the Gen2 defect in
the Cadence PCIe controller in J721E SoC. The errata corresponding to this
is i2085, documented at:
https://www.ti.com/lit/er/sprz455c/sprz455c.pdf
The existing workaround implemented for the errata waits for the Data Link
initialization to complete and assumes that the link retraining process
at the Physical Layer has completed. However, it is possible that the
Physical Layer training might be ongoing as indicated by the
PCI_EXP_LNKSTA_LT bit in the PCI_EXP_LNKSTA register.
Fix the existing workaround, to ensure that the Physical Layer training
has also completed, in addition to the Data Link initialization.
Link: https://lore.kernel.org/r/20230315070800.1615527-1-s-vadapalli@ti.com
Fixes: 4740b969aaf5 ("PCI: cadence: Retrain Link to work around Gen2 training defect")
Signed-off-by: Siddharth Vadapalli <s-vadapalli@ti.com>
Signed-off-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
Reviewed-by: Vignesh Raghavendra <vigneshr@ti.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 2738d5ab7929a845b654cd171a1e275c37eb428e upstream.
When the host tries to remove a PCI device, the host first sends a
PCI_EJECT message to the guest, and the guest is supposed to gracefully
remove the PCI device and send a PCI_EJECTION_COMPLETE message to the host;
the host then sends a VMBus message CHANNELMSG_RESCIND_CHANNELOFFER to
the guest (when the guest receives this message, the device is already
unassigned from the guest) and the guest can do some final cleanup work;
if the guest fails to respond to the PCI_EJECT message within one minute,
the host sends the VMBus message CHANNELMSG_RESCIND_CHANNELOFFER and
removes the PCI device forcibly.
In the case of fast device addition/removal, it's possible that the PCI
device driver is still configuring MSI-X interrupts when the guest receives
the PCI_EJECT message; the channel callback calls hv_pci_eject_device(),
which sets hpdev->state to hv_pcichild_ejecting, and schedules a work
hv_eject_device_work(); if the PCI device driver is calling
pci_alloc_irq_vectors() -> ... -> hv_compose_msi_msg(), we can break the
while loop in hv_compose_msi_msg() due to the updated hpdev->state, and
leave data->chip_data with its default value of NULL; later, when the PCI
device driver calls request_irq() -> ... -> hv_irq_unmask(), the guest
crashes in hv_arch_irq_unmask() due to data->chip_data being NULL.
Fix the issue by not testing hpdev->state in the while loop: when the
guest receives PCI_EJECT, the device is still assigned to the guest, and
the guest has one minute to finish the device removal gracefully. We don't
really need to (and we should not) test hpdev->state in the loop.
Fixes: de0aa7b2f9 ("PCI: hv: Fix 2 hang issues in hv_compose_msi_msg()")
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20230615044451.5580-3-decui@microsoft.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit a847234e24d03d01a9566d1d9dcce018cc018d67 upstream.
This reverts commit d6af2ed29c.
The statement "the hv_pci_bus_exit() call releases structures of all its
child devices" in commit d6af2ed29c is not true: in the path
hv_pci_probe() -> hv_pci_enter_d0() -> hv_pci_bus_exit(hdev, true): the
parameter "keep_devs" is true, so hv_pci_bus_exit() does *not* release the
child "struct hv_pci_dev *hpdev" that is created earlier in
pci_devices_present_work() -> new_pcichild_device().
The commit d6af2ed29c was originally made in July 2020 for RHEL 7.7,
where the old version of hv_pci_bus_exit() was used; when the commit was
rebased and merged into the upstream, people didn't notice that it's
not really necessary. The commit itself doesn't cause any issue, but it
makes hv_pci_probe() more complicated. Revert it to facilitate some
upcoming changes to hv_pci_probe().
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Acked-by: Wei Hu <weh@microsoft.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20230615044451.5580-5-decui@microsoft.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 440b5e3663271b0ffbd4908115044a6a51fb938b upstream.
Since day 1 of the driver, there has been a race between
hv_pci_query_relations() and survey_child_resources(): during fast
device hotplug, hv_pci_query_relations() may error out due to
device-remove and the stack variable 'comp' is no longer valid;
however, pci_devices_present_work() -> survey_child_resources() ->
complete() may be running on another CPU and accessing the no-longer-valid
'comp'. Fix the race by flushing the workqueue before we exit from
hv_pci_query_relations().
Fixes: 4daace0d8c ("PCI: hv: Add paravirtual PCI front-end for Microsoft Hyper-V VMs")
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Acked-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20230615044451.5580-2-decui@microsoft.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 5f5ac460dfe7f4e11f99de9870f240e39189cf72 ]
commit bb38919ec5 ("PCI: imx6: Add support for i.MX6 PCIe controller")
added a fault hook to this driver in the probe function. So it was only
installed if needed.
commit bde4a5a00e ("PCI: imx6: Allow probe deferral by reset GPIO")
moved it from probe to driver init which installs the hook unconditionally
as soon as the driver is compiled into a kernel.
When this driver is compiled as a module, the hook is not registered
until after the driver has been matched with a .compatible and
loaded.
commit 415b6185c5 ("PCI: imx6: Fix config read timeout handling")
extended the fault handling code.
commit 2d8ed461db ("PCI: imx6: Add support for i.MX8MQ")
added some protection for non-ARM architectures, but this does not
protect non-i.MX ARM architectures.
Since fault handlers can be triggered on any architecture for different
reasons, there is no guarantee that they will be triggered only for the
assumed situation, leading to improper error handling (i.MX6-specific
imx6q_pcie_abort_handler) on foreign systems.
I had seen strange L3 imprecise external abort messages several times on
OMAP4 and OMAP5 devices and couldn't make sense of them until I realized
they were related to this unused imx6q driver because I had
CONFIG_PCI_IMX6=y.
Note that CONFIG_PCI_IMX6=y is useful for kernel binaries that are designed
to run on different ARM SoC and be differentiated only by device tree
binaries. So turning off CONFIG_PCI_IMX6 is not a solution.
Therefore we check the compatible in the init function before registering
the fault handler.
Link: https://lore.kernel.org/r/e1bcfc3078c82b53aa9b78077a89955abe4ea009.1678380991.git.hns@goldelico.com
Fixes: bde4a5a00e ("PCI: imx6: Allow probe deferral by reset GPIO")
Fixes: 415b6185c5 ("PCI: imx6: Fix config read timeout handling")
Fixes: 2d8ed461db ("PCI: imx6: Add support for i.MX8MQ")
Signed-off-by: H. Nikolaus Schaller <hns@goldelico.com>
Signed-off-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
Reviewed-by: Richard Zhu <hongxing.zhu@nxp.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 2542e16c392508800f1d9037feee881a9c444951 upstream.
Qcom PCIe IP version v2.7.0 and its derivatives don't contain the
PCIE20_PARF_AXI_MSTR_WR_ADDR_HALT register. Instead, they have the new
PCIE20_PARF_AXI_MSTR_WR_ADDR_HALT_V2 register. So fix the incorrect
register usage which is modifying a different register.
Also in this IP version, this register change doesn't depend on MSI
being enabled. So remove that check also.
Link: https://lore.kernel.org/r/20230316081117.14288-2-manivannan.sadhasivam@linaro.org
Fixes: ed8cc3b1fc ("PCI: qcom: Add support for SDM845 PCIe controller")
Signed-off-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Signed-off-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
Cc: <stable@vger.kernel.org> # 5.6+
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 8b3517f88ff2983f52698893519227c10aac90b2 ]
Except for isochronous-configured devices, software may set
Max_Read_Request_Size (MRRS) to any value up to 4096. If a device issues a
read request with size greater than the completer's Max_Payload_Size (MPS),
the completer is required to break the response into multiple completions.
Instead of correctly responding with multiple completions to a large read
request, some LS7A Root Ports respond with a Completer Abort. To prevent
this, the MRRS must be limited to an implementation-specific value.
The OS cannot detect that value, so rely on BIOS to configure MRRS before
booting, and quirk the Root Ports so we never set an MRRS larger than that
BIOS value for any downstream device.
N.B. Hot-added devices are not configured by BIOS, and they power up with
MRRS = 512 bytes, so these devices will be limited to 512 bytes. If the
LS7A limit is smaller, those hot-added devices may not work correctly, but
per [1], hotplug is not supported with this chipset revision.
[1] https://lore.kernel.org/r/073638a7-ae68-2847-ac3d-29e5e760d6af@loongson.cn
[bhelgaas: commit log]
Link: https://bugzilla.kernel.org/show_bug.cgi?id=216884
Link: https://lore.kernel.org/r/20230201043018.778499-3-chenhuacai@loongson.cn
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 66110361281b2f7da0c8bd51eaf1f152f4236035 ]
commit aeaa0bfe89 ("PCI: dwc: Move N_FTS setup to common setup")
incorrectly uses pci->link_gen in deriving the index to the
n_fts[] array also introducing the issue of accessing beyond the
boundaries of array for greater than Gen-2 speeds. This change fixes
that issue.
Link: https://lore.kernel.org/r/20220926111923.22487-1-vidyas@nvidia.com
Fixes: aeaa0bfe89 ("PCI: dwc: Move N_FTS setup to common setup")
Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
Signed-off-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
Reviewed-by: Rob Herring <robh@kernel.org>
Acked-by: Jingoo Han <jingoohan1@gmail.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 6646e99bcec627e866bc84365af37942c72b4b76 ]
As part of Root Port interrupt handling, level-0 register is read first and
based on the bits set in that, corresponding level-1 registers are read for
further interrupt processing. Since both these values are currently read
into the same 'val' variable, checking level-0 bits the second time around
is happening on the 'val' variable value of level-1 register contents
instead of freshly reading the level-0 value again.
Fix by using different variables to store level-0 and level-1 registers
contents.
Link: https://lore.kernel.org/r/20220721142052.25971-11-vidyas@nvidia.com
Fixes: 56e15a238d ("PCI: tegra: Add Tegra194 PCIe support")
Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 38f897ae3d44900f627cad708a15db498ce2ca31 ]
We currently enable clocks BEFORE we write to PARF_PHY_CTRL reg to enable
clocks and resets. This causes the driver to never set to a ready state
with the error 'Phy link never came up'.
This is caused by the PHY clock getting enabled before setting the required
bits in the PARF regs.
A workaround for this was set but with this new discovery we can drop
the workaround and use a proper solution to the problem by just enabling
the clock only AFTER the PARF_PHY_CTRL bit is set.
This correctly sets up the PCIe link and makes it usable even when a
bootloader leaves the PCIe link in an undefined state.
Fixes: 82a823833f ("PCI: qcom: Add Qualcomm PCIe controller driver")
Link: https://lore.kernel.org/r/20220708222743.27019-1-ansuelsmth@gmail.com
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit d1cf738f2b65a5640234e1da90a68d3523fbed83 ]
dw_pcie_disable_atu() was introduced by f8aed6ec62 ("PCI: dwc:
designware: Add EP mode support") and supported only the viewport version
of the iATU CSRs.
DW PCIe IP cores v4.80a and newer also support unrolled iATU/eDMA space.
Callers of dw_pcie_disable_atu(), including pci_epc_ops.clear_bar(),
pci_epc_ops.unmap_addr(), and dw_pcie_setup_rc(), don't work correctly when
it is enabled.
Add dw_pcie_disable_atu() support for controllers with unrolled iATU CSRs
enabled.
[bhelgaas: commit log]
Fixes: f8aed6ec62 ("PCI: dwc: designware: Add EP mode support")
Link: https://lore.kernel.org/r/20220624143428.8334-3-Sergey.Semin@baikalelectronics.ru
Tested-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Signed-off-by: Serge Semin <Sergey.Semin@baikalelectronics.ru>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit e8fbd344a5ea62663554b8546b6bf9f88b93785a ]
pm_runtime_enable() will increase power disable depth. If
dw_pcie_ep_init() fails, we should use pm_runtime_disable() to balance it
with pm_runtime_enable().
Add missing pm_runtime_disable() for tegra_pcie_config_ep().
Fixes: c57247f940 ("PCI: tegra: Add support for PCIe endpoint mode in Tegra194")
Link: https://lore.kernel.org/r/20220602031910.55859-1-linmq006@gmail.com
Signed-off-by: Miaoqian Lin <linmq006@gmail.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Vidya Sagar <vidyas@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit a2bad844a67b1c7740bda63e87453baf63c3a7f7 upstream.
According to Dexuan, the hypervisor folks beleive that multi-msi
allocations are not correct. compose_msi_msg() will allocate multi-msi
one by one. However, multi-msi is a block of related MSIs, with alignment
requirements. In order for the hypervisor to allocate properly aligned
and consecutive entries in the IOMMU Interrupt Remapping Table, there
should be a single mapping request that requests all of the multi-msi
vectors in one shot.
Dexuan suggests detecting the multi-msi case and composing a single
request related to the first MSI. Then for the other MSIs in the same
block, use the cached information. This appears to be viable, so do it.
5.10 backport - add hv_msi_get_int_vector helper function. Fixed merge
conflict due to delivery_mode name change (APIC_DELIVERY_MODE_FIXED
is the value given to dest_Fixed). Removed unused variable in
hv_compose_msi_msg. Fixed reference to msi_desc->pci to point to
the same is_msix variable. Removed changes to compose_msi_req_v3 since
it doesn't exist yet.
Suggested-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: Jeffrey Hugo <quic_jhugo@quicinc.com>
Reviewed-by: Dexuan Cui <decui@microsoft.com>
Tested-by: Michael Kelley <mikelley@microsoft.com>
Link: https://lore.kernel.org/r/1652282599-21643-1-git-send-email-quic_jhugo@quicinc.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
Signed-off-by: Carl Vanderlip <quic_carlv@quicinc.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit b4b77778ecc5bfbd4e77de1b2fd5c1dd3c655f1f upstream.
Currently if compose_msi_msg() is called multiple times, it will free any
previous IRTE allocation, and generate a new allocation. While nothing
prevents this from occurring, it is extraneous when Linux could just reuse
the existing allocation and avoid a bunch of overhead.
However, when future IRTE allocations operate on blocks of MSIs instead of
a single line, freeing the allocation will impact all of the lines. This
could cause an issue where an allocation of N MSIs occurs, then some of
the lines are retargeted, and finally the allocation is freed/reallocated.
The freeing of the allocation removes all of the configuration for the
entire block, which requires all the lines to be retargeted, which might
not happen since some lines might already be unmasked/active.
Signed-off-by: Jeffrey Hugo <quic_jhugo@quicinc.com>
Reviewed-by: Dexuan Cui <decui@microsoft.com>
Tested-by: Dexuan Cui <decui@microsoft.com>
Tested-by: Michael Kelley <mikelley@microsoft.com>
Link: https://lore.kernel.org/r/1652282582-21595-1-git-send-email-quic_jhugo@quicinc.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
Signed-off-by: Carl Vanderlip <quic_carlv@quicinc.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 455880dfe292a2bdd3b4ad6a107299fce610e64b upstream.
In the multi-MSI case, hv_arch_irq_unmask() will only operate on the first
MSI of the N allocated. This is because only the first msi_desc is cached
and it is shared by all the MSIs of the multi-MSI block. This means that
hv_arch_irq_unmask() gets the correct address, but the wrong data (always
0).
This can break MSIs.
Lets assume MSI0 is vector 34 on CPU0, and MSI1 is vector 33 on CPU0.
hv_arch_irq_unmask() is called on MSI0. It uses a hypercall to configure
the MSI address and data (0) to vector 34 of CPU0. This is correct. Then
hv_arch_irq_unmask is called on MSI1. It uses another hypercall to
configure the MSI address and data (0) to vector 33 of CPU0. This is
wrong, and results in both MSI0 and MSI1 being routed to vector 33. Linux
will observe extra instances of MSI1 and no instances of MSI0 despite the
endpoint device behaving correctly.
For the multi-MSI case, we need unique address and data info for each MSI,
but the cached msi_desc does not provide that. However, that information
can be gotten from the int_desc cached in the chip_data by
compose_msi_msg(). Fix the multi-MSI case to use that cached information
instead. Since hv_set_msi_entry_from_desc() is no longer applicable,
remove it.
5.10 backport - removed unused hv_set_msi_entry_from_desc function from
mshyperv.h instead of pci-hyperv.c. msi_entry.address/data.as_uint32
changed to direct reference (as they are u32's, just sans union).
Signed-off-by: Jeffrey Hugo <quic_jhugo@quicinc.com>
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Link: https://lore.kernel.org/r/1651068453-29588-1-git-send-email-quic_jhugo@quicinc.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
Signed-off-by: Carl Vanderlip <quic_carlv@quicinc.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 08e61e861a0e47e5e1a3fb78406afd6b0cea6b6d upstream.
If the allocation of multiple MSI vectors for multi-MSI fails in the core
PCI framework, the framework will retry the allocation as a single MSI
vector, assuming that meets the min_vecs specified by the requesting
driver.
Hyper-V advertises that multi-MSI is supported, but reuses the VECTOR
domain to implement that for x86. The VECTOR domain does not support
multi-MSI, so the alloc will always fail and fallback to a single MSI
allocation.
In short, Hyper-V advertises a capability it does not implement.
Hyper-V can support multi-MSI because it coordinates with the hypervisor
to map the MSIs in the IOMMU's interrupt remapper, which is something the
VECTOR domain does not have. Therefore the fix is simple - copy what the
x86 IOMMU drivers (AMD/Intel-IR) do by removing
X86_IRQ_ALLOC_CONTIGUOUS_VECTORS after calling the VECTOR domain's
pci_msi_prepare().
5.10 backport - adds the hv_msi_prepare wrapper function
Fixes: 4daace0d8c ("PCI: hv: Add paravirtual PCI front-end for Microsoft Hyper-V VMs")
Signed-off-by: Jeffrey Hugo <quic_jhugo@quicinc.com>
Reviewed-by: Dexuan Cui <decui@microsoft.com>
Link: https://lore.kernel.org/r/1649856981-14649-1-git-send-email-quic_jhugo@quicinc.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
Signed-off-by: Carl Vanderlip <quic_carlv@quicinc.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit a6809941c1f17f455db2cf4ca19c6d8c8746ec25 ]
According to the PCIe standard the PERST# signal (reset-gpio in
fsl,imx* compatible dts) should be kept asserted for at least 100 usec
before the PCIe refclock is stable, should be kept asserted for at
least 100 msec after the power rails are stable and the host should wait
at least 100 msec after it is de-asserted before accessing the
configuration space of any attached device.
From PCIe CEM r2.0, sec 2.6.2
T-PVPERL: Power stable to PERST# inactive - 100 msec
T-PERST-CLK: REFCLK stable before PERST# inactive - 100 usec.
From PCIe r5.0, sec 6.6.1
With a Downstream Port that does not support Link speeds greater than
5.0 GT/s, software must wait a minimum of 100 ms before sending a
Configuration Request to the device immediately below that Port.
Failure to do so could prevent PCIe devices to be working correctly,
and this was experienced with real devices.
Move reset assert to imx6_pcie_assert_core_reset(), this way we ensure
that PERST# is asserted before enabling any clock, move de-assert to the
end of imx6_pcie_deassert_core_reset() after the clock is enabled and
deemed stable and add a new delay of 100 msec just afterward.
Link: https://lore.kernel.org/all/20220211152550.286821-1-francesco.dolcini@toradex.com
Link: https://lore.kernel.org/r/20220404081509.94356-1-francesco.dolcini@toradex.com
Fixes: bb38919ec5 ("PCI: imx6: Add support for i.MX6 PCIe controller")
Signed-off-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Reviewed-by: Lucas Stach <l.stach@pengutronix.de>
Acked-by: Richard Zhu <hongxing.zhu@nxp.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 805dfc18dd3d4dd97a987d4406593b5a225b1253 upstream.
In advk_pcie_handle_msi() it is expected that when bit i in the W1C
register PCIE_MSI_STATUS_REG is cleared, the PCIE_MSI_PAYLOAD_REG is
updated to contain the MSI number corresponding to index i.
Experiments show that this is not so, and instead PCIE_MSI_PAYLOAD_REG
always contains the number of the last received MSI, overall.
Do not read PCIE_MSI_PAYLOAD_REG register for determining MSI interrupt
number. Since Aardvark already forbids more than 32 interrupts and uses
own allocated hwirq numbers, the msi_idx already corresponds to the
received MSI number.
Link: https://lore.kernel.org/r/20220110015018.26359-3-kabel@kernel.org
Fixes: 8c39d71036 ("PCI: aardvark: Add Aardvark PCI host controller driver")
Signed-off-by: Pali Rohár <pali@kernel.org>
Signed-off-by: Marek Behún <kabel@kernel.org>
Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Signed-off-by: Marek Behún <kabel@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 7d8dc1f7cd007a7ce94c5b4c20d63a8b8d6d7751 upstream.
We already clear all the other interrupts (ISR0, ISR1, HOST_CTRL_INT).
Define a new macro PCIE_MSI_ALL_MASK and do the same clearing for MSIs,
to ensure that we don't start receiving spurious interrupts.
Use this new mask in advk_pcie_handle_msi();
Link: https://lore.kernel.org/r/20211130172913.9727-5-kabel@kernel.org
Signed-off-by: Pali Rohár <pali@kernel.org>
Signed-off-by: Marek Behún <kabel@kernel.org>
Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit b0b0b8b897f8e12b2368e868bd7cdc5742d5c5a9 ]
Aardvark hardware supports Multi-MSI and MSI_FLAG_MULTI_PCI_MSI is already
set for the MSI chip. But when allocating MSI interrupt numbers for
Multi-MSI, the numbers need to be properly aligned, otherwise endpoint
devices send MSI interrupt with incorrect numbers.
Fix this issue by using function bitmap_find_free_region() instead of
bitmap_find_next_zero_area().
To ensure that aligned MSI interrupt numbers are used by endpoint devices,
we cannot use Linux virtual irq numbers (as they are random and not
properly aligned). Instead we need to use the aligned hwirq numbers.
This change fixes receiving MSI interrupts on Armada 3720 boards and
allows using NVMe disks which use Multi-MSI feature with 3 interrupts.
Without this NVMe disks freeze booting as linux nvme-core.c is waiting
60s for an interrupt.
Link: https://lore.kernel.org/r/20220110015018.26359-4-kabel@kernel.org
Signed-off-by: Pali Rohár <pali@kernel.org>
Signed-off-by: Marek Behún <kabel@kernel.org>
Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 735f5ae49e1b44742cc63ca9b5c1ffde3e94ba91 ]
The emulated bridge returns incorrect value for PCI_EXP_RTSTA register
during readout in advk_pci_bridge_emul_pcie_conf_read() function: the
correct bit is BIT(16), but we are setting BIT(23), because the code
does
*value = (isr0 & PCIE_MSG_PM_PME_MASK) << 16
where
PCIE_MSG_PM_PME_MASK
is
BIT(7).
The code should probably have been something like
*value = (!!(isr0 & PCIE_MSG_PM_PME_MASK)) << 16,
but we are better of using an if() and using the proper macro for this
bit.
Link: https://lore.kernel.org/r/20220110015018.26359-15-kabel@kernel.org
Fixes: 8a3ebd8de3 ("PCI: aardvark: Implement emulated root PCI bridge config space")
Signed-off-by: Pali Rohár <pali@kernel.org>
Signed-off-by: Marek Behún <kabel@kernel.org>
Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 1f1050c5e1fefb34ac90a506b43e9da803b5f8f7 upstream.
Older mvebu hardware provides PCIe Capability structure only in version 1.
New mvebu and aardvark hardware provides it in version 2. So do not force
version to 2 in pci_bridge_emul_init() and rather allow drivers to set
correct version. Drivers need to set version in pcie_conf.cap field without
overwriting PCI_CAP_LIST_ID register. Both drivers (mvebu and aardvark) do
not provide slot support yet, so do not set PCI_EXP_FLAGS_SLOT flag.
Link: https://lore.kernel.org/r/20211124155944.1290-6-pali@kernel.org
Fixes: 23a5fba4d9 ("PCI: Introduce PCI bridge emulated config space common logic")
Signed-off-by: Pali Rohár <pali@kernel.org>
Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Cc: stable@vger.kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 39bd54d43b3f8b3c7b3a75f5d868d8bb858860e7 upstream.
This reverts commit 239edf686c14a9ff926dec2f350289ed7adfefe2.
239edf686c14 ("PCI: aardvark: Fix support for PCI_ROM_ADDRESS1 on emulated
bridge") added support for the Type 1 Expansion ROM BAR at config offset
0x38, based on the register being listed in the Marvell Armada A3720 spec.
But the spec doesn't document it at all for RC mode, and there is no ROM in
the SOC, so remove this emulation for now.
The PCI bridge which represents aardvark's PCIe Root Port has an Expansion
ROM Base Address register at offset 0x30, but its meaning is different than
PCI's Expansion ROM BAR register, although the layout is the same. (This
is why we thought it does the same thing.)
First: there is no ROM (or part of BootROM) in the A3720 SOC dedicated for
PCIe Root Port (or controller in RC mode) containing executable code that
would initialize the Root Port, suitable for execution in bootloader (this
is how Expansion ROM BAR is used on x86).
Second: in A3720 spec the register (address 0xD0070030) is not documented
at all for Root Complex mode, but similar to other BAR registers, it has an
"entangled partner" in register 0xD0075920, which does address translation
for the BAR in 0xD0070030:
- the BAR register sets the address from the view of PCIe bus
- the translation register sets the address from the view of the CPU
The other BAR registers also have this entangled partner, and they can be
used to:
- in RC mode: address-checking on the receive side of the RC (they can
define address ranges for memory accesses from remote Endpoints to the
RC)
- in Endpoint mode: allow the remote CPU to access memory on A3720
The Expansion ROM BAR has only the Endpoint part documented, but from the
similarities we think that it can also be used in RC mode in that way.
So either Expansion ROM BAR has different meaning (if the hypothesis above
is true), or we don't know it's meaning (since it is not documented for RC
mode).
Remove the register from the emulated bridge accessing functions.
[bhelgaas: summarize reason for removal (first paragraph)]
Fixes: 239edf686c14 ("PCI: aardvark: Fix support for PCI_ROM_ADDRESS1 on emulated bridge")
Link: https://lore.kernel.org/r/20211125160148.26029-3-kabel@kernel.org
Signed-off-by: Marek Behún <kabel@kernel.org>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Pali Rohár <pali@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit f76b36d40beee0a13aa8f6aa011df0d7cbbb8a7f upstream.
Fix multiple link training issues in aardvark driver. The main reason of
these issues was misunderstanding of what certain registers do, since their
names and comments were misleading: before commit 96be36dbff ("PCI:
aardvark: Replace custom macros by standard linux/pci_regs.h macros"), the
pci-aardvark.c driver used custom macros for accessing standard PCIe Root
Bridge registers, and misleading comments did not help to understand what
the code was really doing.
After doing more tests and experiments I've come to the conclusion that the
SPEED_GEN register in aardvark sets the PCIe revision / generation
compliance and forces maximal link speed. Both GEN3 and GEN2 values set the
read-only PCI_EXP_FLAGS_VERS bits (PCIe capabilities version of Root
Bridge) to value 2, while GEN1 value sets PCI_EXP_FLAGS_VERS to 1, which
matches with PCI Express specifications revisions 3, 2 and 1 respectively.
Changing SPEED_GEN also sets the read-only bits PCI_EXP_LNKCAP_SLS and
PCI_EXP_LNKCAP2_SLS to corresponding speed.
(Note that PCI Express rev 1 specification does not define PCI_EXP_LNKCAP2
and PCI_EXP_LNKCTL2 registers and when SPEED_GEN is set to GEN1 (which
also sets PCI_EXP_FLAGS_VERS set to 1), lspci cannot access
PCI_EXP_LNKCAP2 and PCI_EXP_LNKCTL2 registers.)
Changing PCIe link speed can be done via PCI_EXP_LNKCTL2_TLS bits of
PCI_EXP_LNKCTL2 register. Armada 3700 Functional Specifications says that
the default value of PCI_EXP_LNKCTL2_TLS is based on SPEED_GEN value, but
tests showed that the default value is always 8.0 GT/s, independently of
speed set by SPEED_GEN. So after setting SPEED_GEN, we must also set value
in PCI_EXP_LNKCTL2 register via PCI_EXP_LNKCTL2_TLS bits.
Triggering PCI_EXP_LNKCTL_RL bit immediately after setting LINK_TRAINING_EN
bit actually doesn't do anything. Tests have shown that a delay is needed
after enabling LINK_TRAINING_EN bit. As triggering PCI_EXP_LNKCTL_RL
currently does nothing, remove it.
Commit 43fc679ced ("PCI: aardvark: Improve link training") introduced
code which sets SPEED_GEN register based on negotiated link speed from
PCI_EXP_LNKSTA_CLS bits of PCI_EXP_LNKSTA register. This code was added to
fix detection of Compex WLE900VX (Atheros QCA9880) WiFi GEN1 PCIe cards, as
otherwise these cards were "invisible" on PCIe bus (probably because they
crashed). But apparently more people reported the same issues with these
cards also with other PCIe controllers [1] and I was able to reproduce this
issue also with other "noname" WiFi cards based on Atheros QCA9890 chip
(with the same PCI vendor/device ids as Atheros QCA9880). So this is not an
issue in aardvark but rather an issue in Atheros QCA98xx chips. Also, this
issue only exists if the kernel is compiled with PCIe ASPM support, and a
generic workaround for this is to change PCIe Bridge to 2.5 GT/s link speed
via PCI_EXP_LNKCTL2_TLS_2_5GT bits in PCI_EXP_LNKCTL2 register [2], before
triggering PCI_EXP_LNKCTL_RL bit. This workaround also works when SPEED_GEN
is set to value GEN2 (5 GT/s). So remove this hack completely in the
aardvark driver and always set SPEED_GEN to value from 'max-link-speed' DT
property. Fix for Atheros QCA98xx chips is handled separately by patch [2].
These two things (code for triggering PCI_EXP_LNKCTL_RL bit and changing
SPEED_GEN value) also explain why commit 6964494582 ("PCI: aardvark:
Train link immediately after enabling training") somehow fixed detection of
those problematic Compex cards with Atheros chips: if triggering link
retraining (via PCI_EXP_LNKCTL_RL bit) was done immediately after enabling
link training (via LINK_TRAINING_EN), it did nothing. If there was a
specific delay, aardvark HW already initialized PCIe link and therefore
triggering link retraining caused the above issue. Compex cards triggered
link down event and disappeared from the PCIe bus.
Commit f4c7d053d7 ("PCI: aardvark: Wait for endpoint to be ready before
training link") added 100ms sleep before calling 'Start link training'
command and explained that it is a requirement of PCI Express
specification. But the code after this 100ms sleep was not doing 'Start
link training', rather it triggered PCI_EXP_LNKCTL_RL bit via PCIe Root
Bridge to put link into Recovery state.
The required delay after fundamental reset is already done in function
advk_pcie_wait_for_link() which also checks whether PCIe link is up.
So after removing the code which triggers PCI_EXP_LNKCTL_RL bit on PCIe
Root Bridge, there is no need to wait 100ms again. Remove the extra
msleep() call and update comment about the delay required by the PCI
Express specification.
According to Marvell Armada 3700 Functional Specifications, Link training
should be enabled via aardvark register LINK_TRAINING_EN after selecting
PCIe generation and x1 lane. There is no need to disable it prior resetting
card via PERST# signal. This disabling code was introduced in commit
5169a9851d ("PCI: aardvark: Issue PERST via GPIO") as a workaround for
some Atheros cards. It turns out that this also is Atheros specific issue
and affects any PCIe controller, not only aardvark. Moreover this Atheros
issue was triggered by juggling with PCI_EXP_LNKCTL_RL, LINK_TRAINING_EN
and SPEED_GEN bits interleaved with sleeps. Now, after removing triggering
PCI_EXP_LNKCTL_RL, there is no need to explicitly disable LINK_TRAINING_EN
bit. So remove this code too. The problematic Compex cards described in
previous git commits are correctly detected in advk_pcie_train_link()
function even after applying all these changes.
Note that with this patch, and also prior this patch, some NVMe disks which
support PCIe GEN3 with 8 GT/s speed are negotiated only at the lowest link
speed 2.5 GT/s, independently of SPEED_GEN value. After manually triggering
PCI_EXP_LNKCTL_RL bit (e.g. from userspace via setpci), these NVMe disks
change link speed to 5 GT/s when SPEED_GEN was configured to GEN2. This
issue first needs to be properly investigated. I will send a fix in the
future.
On the other hand, some other GEN2 PCIe cards with 5 GT/s speed are
autonomously by HW autonegotiated at full 5 GT/s speed without need of any
software interaction.
Armada 3700 Functional Specifications describes the following steps for
link training: set SPEED_GEN to GEN2, enable LINK_TRAINING_EN, poll until
link training is complete, trigger PCI_EXP_LNKCTL_RL, poll until signal
rate is 5 GT/s, poll until link training is complete, enable ASPM L0s.
The requirement for triggering PCI_EXP_LNKCTL_RL can be explained by the
need to achieve 5 GT/s speed (as changing link speed is done by throw to
recovery state entered by PCI_EXP_LNKCTL_RL) or maybe as a part of enabling
ASPM L0s (but in this case ASPM L0s should have been enabled prior
PCI_EXP_LNKCTL_RL).
It is unknown why the original pci-aardvark.c driver was triggering
PCI_EXP_LNKCTL_RL bit before waiting for the link to be up. This does not
align with neither PCIe base specifications nor with Armada 3700 Functional
Specification. (Note that in older versions of aardvark, this bit was
called incorrectly PCIE_CORE_LINK_TRAINING, so this may be the reason.)
It is also unknown why Armada 3700 Functional Specification says that it is
needed to trigger PCI_EXP_LNKCTL_RL for GEN2 mode, as according to PCIe
base specification 5 GT/s speed negotiation is supposed to be entirely
autonomous, even if initial speed is 2.5 GT/s.
[1] - https://lore.kernel.org/linux-pci/87h7l8axqp.fsf@toke.dk/
[2] - https://lore.kernel.org/linux-pci/20210326124326.21163-1-pali@kernel.org/
Link: https://lore.kernel.org/r/20211005180952.6812-12-kabel@kernel.org
Signed-off-by: Pali Rohár <pali@kernel.org>
Signed-off-by: Marek Behún <kabel@kernel.org>
Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Reviewed-by: Marek Behún <kabel@kernel.org>
Signed-off-by: Marek Behún <kabel@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 223dec14a05337a4155f1deed46d2becce4d00fd upstream.
Commit 43f5c77bcbd2 ("PCI: aardvark: Fix reporting CRS value") fixed
handling of CRS response and when CRSSVE flag was not enabled it marked CRS
response as failed transaction (due to simplicity).
But pci-aardvark.c driver is already waiting up to the PIO_RETRY_CNT count
for PIO config response and so we can with a small change implement
re-issuing of config requests as described in PCIe base specification.
This change implements re-issuing of config requests when response is CRS.
Set upper bound of wait cycles to around PIO_RETRY_CNT, afterwards the
transaction is marked as failed and an all-ones value is returned as
before.
We do this by returning appropriate error codes from function
advk_pcie_check_pio_status(). On CRS we return -EAGAIN and caller then
reissues transaction.
Link: https://lore.kernel.org/r/20211005180952.6812-10-kabel@kernel.org
Signed-off-by: Pali Rohár <pali@kernel.org>
Signed-off-by: Marek Behún <kabel@kernel.org>
Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Reviewed-by: Marek Behún <kabel@kernel.org>
Signed-off-by: Marek Behún <kabel@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
