user/sven/linux.git/drivers/pci, branch v5.5.11

PCI/ATS: Restore EXPORT_SYMBOL_GPL() for pci_{enable,disable}_ats()

2020-02-24T07:38:19Z

[ Upstream commit bb950bca5d522119f8b9ce3f6cbac4841c6d6517 ] Commit d355bb209783 ("PCI/ATS: Remove unnecessary EXPORT_SYMBOL_GPL()") unexported a bunch of symbols from the PCI core since the only external users were non-modular IOMMU drivers. Although most of those symbols can remain private for now, 'pci_{enable,disable_ats()' is required for the ARM SMMUv3 driver to build as a module, otherwise we get a build failure as follows: | ERROR: "pci_enable_ats" [drivers/iommu/arm-smmu-v3.ko] undefined! | ERROR: "pci_disable_ats" [drivers/iommu/arm-smmu-v3.ko] undefined! Re-export these two functions so that the ARM SMMUv3 driver can be build as a module. Cc: Bjorn Helgaas Cc: Joerg Roedel Signed-off-by: Greg Kroah-Hartman [will: rewrote commit message] Signed-off-by: Will Deacon Tested-by: John Garry # smmu v3 Reviewed-by: Greg Kroah-Hartman Acked-by: Bjorn Helgaas Signed-off-by: Joerg Roedel Signed-off-by: Sasha Levin

PCI: Add DMA alias quirk for PLX PEX NTB

2020-02-24T07:38:16Z

[ Upstream commit 7b90dfc4873b87c468cc6046538f46a531c1d785 ] The PLX PEX NTB forwards DMA transactions using Requester IDs that don't exist as PCI devices. The devfn for a transaction is used as an index into a lookup table storing the origin of a transaction on the other side of the bridge. Alias all possible devfns to the NTB device so that any transaction coming in is governed by the mappings for the NTB. Signed-off-by: James Sewart Signed-off-by: Bjorn Helgaas Reviewed-by: Logan Gunthorpe Signed-off-by: Sasha Levin

PCI: Add nr_devfns parameter to pci_add_dma_alias()

2020-02-24T07:38:16Z

[ Upstream commit 09298542cd891b43778db1f65aa3613aa5a562eb ] Add a "nr_devfns" parameter to pci_add_dma_alias() so it can be used to create DMA aliases for a range of devfns. [bhelgaas: incorporate nr_devfns fix from James, update quirk_pex_vca_alias() and setup_aliases()] Signed-off-by: James Sewart Signed-off-by: Bjorn Helgaas Signed-off-by: Sasha Levin

PCI: Increase D3 delay for AMD Ryzen5/7 XHCI controllers

2020-02-24T07:38:14Z

[ Upstream commit 3030df209aa8cf831b9963829bd9f94900ee8032 ] On Asus UX434DA (AMD Ryzen7 3700U) and Asus X512DK (AMD Ryzen5 3500U), the XHCI controller fails to resume from runtime suspend or s2idle, and USB becomes unusable from that point. xhci_hcd 0000:03:00.4: Refused to change power state, currently in D3 xhci_hcd 0000:03:00.4: enabling device (0000 -> 0002) xhci_hcd 0000:03:00.4: WARN: xHC restore state timeout xhci_hcd 0000:03:00.4: PCI post-resume error -110! xhci_hcd 0000:03:00.4: HC died; cleaning up During suspend, a transition to D3cold is attempted, however the affected platforms do not seem to cut the power to the PCI device when in this state, so the device stays in D3hot. Upon resume, the D3hot-to-D0 transition is successful only if the D3 delay is increased to 20ms. The transition failure does not appear to be detectable as a CRS condition. Add a PCI quirk to increase the delay on the affected hardware. Link: https://bugzilla.kernel.org/show_bug.cgi?id=205587 Link: http://lkml.kernel.org/r/CAD8Lp47Vh69gQjROYG69=waJgL7hs1PwnLonL9+27S_TcRhixA@mail.gmail.com Link: https://lore.kernel.org/r/20191127053836.31624-2-drake@endlessm.com Signed-off-by: Daniel Drake Signed-off-by: Bjorn Helgaas Reviewed-by: Mika Westerberg Signed-off-by: Sasha Levin

PCI: Add generic quirk for increasing D3hot delay

2020-02-24T07:38:14Z

[ Upstream commit 62fe23df067715a21c4aef44068efe7ceaa8f627 ] Separate the D3 delay increase functionality out of quirk_radeon_pm() into its own function so that it can be shared with other quirks, including the AMD Ryzen XHCI quirk that will be introduced in a followup commit. Tweak the function name and message to indicate more clearly that the delay relates to a D3hot-to-D0 transition. Link: https://lore.kernel.org/r/20191127053836.31624-1-drake@endlessm.com Signed-off-by: Daniel Drake Signed-off-by: Bjorn Helgaas Reviewed-by: Mika Westerberg Signed-off-by: Sasha Levin

PCI: iproc: Apply quirk_paxc_bridge() for module as well as built-in

2020-02-24T07:38:14Z

[ Upstream commit 574f29036fce385e28617547955dd6911d375025 ] Previously quirk_paxc_bridge() was applied when the iproc driver was built-in, but not when it was compiled as a module. This happened because it was under #ifdef CONFIG_PCIE_IPROC_PLATFORM: PCIE_IPROC_PLATFORM=y causes CONFIG_PCIE_IPROC_PLATFORM to be defined, but PCIE_IPROC_PLATFORM=m causes CONFIG_PCIE_IPROC_PLATFORM_MODULE to be defined. Move quirk_paxc_bridge() to pcie-iproc.c and drop the #ifdef so the quirk is always applied, whether iproc is built-in or a module. [bhelgaas: commit log, move to pcie-iproc.c, not pcie-iproc-platform.c] Link: https://lore.kernel.org/r/20191211174511.89713-1-wei.liu@kernel.org Signed-off-by: Wei Liu Signed-off-by: Bjorn Helgaas Signed-off-by: Sasha Levin

PCI: Fix pci_add_dma_alias() bitmask size

2020-02-24T07:38:01Z

[ Upstream commit f8bf2aeb651b3460a4b36fd7ba1ba1d31777d35c ] The number of possible devfns is 256, but pci_add_dma_alias() allocated a bitmap of size 255. Fix this off-by-one error. This fixes commits 338c3149a221 ("PCI: Add support for multiple DMA aliases") and c6635792737b ("PCI: Allocate dma_alias_mask with bitmap_zalloc()"), but I doubt it was possible to see a problem because it takes 4 64-bit longs (or 8 32-bit longs) to hold 255 bits, and bitmap_zalloc() doesn't save the 255-bit size anywhere. [bhelgaas: commit log, move #define to drivers/pci/pci.h, include loop limit fix from Qian Cai: https://lore.kernel.org/r/20191218170004.5297-1-cai@lca.pw] Signed-off-by: James Sewart Signed-off-by: Bjorn Helgaas Reviewed-by: Logan Gunthorpe Signed-off-by: Sasha Levin

PCI/AER: Initialize aer_fifo

2020-02-14T21:52:57Z

commit d95f20c4f07020ebc605f3b46af4b6db9eb5fc99 upstream. Previously we did not call INIT_KFIFO() for aer_fifo. This leads to kfifo_put() sometimes returning 0 (queue full) when in fact it is not. It is easy to reproduce the problem by using aer-inject: $ aer-inject -s :82:00.0 multiple-corr-nonfatal The content of the multiple-corr-nonfatal file is as below: AER COR RCVR HL 0 1 2 3 AER UNCOR POISON_TLP HL 4 5 6 7 Fixes: 27c1ce8bbed7 ("PCI/AER: Use kfifo for tracking events instead of reimplementing it") Link: https://lore.kernel.org/r/1579767991-103898-1-git-send-email-liudongdong3@huawei.com Signed-off-by: Dongdong Liu Signed-off-by: Bjorn Helgaas Signed-off-by: Greg Kroah-Hartman

PCI: Don't disable bridge BARs when assigning bus resources

2020-02-14T21:52:57Z

commit 9db8dc6d0785225c42a37be7b44d1b07b31b8957 upstream. Some PCI bridges implement BARs in addition to bridge windows. For example, here's a PLX switch: 04:00.0 PCI bridge: PLX Technology, Inc. PEX 8724 24-Lane, 6-Port PCI Express Gen 3 (8 GT/s) Switch, 19 x 19mm FCBGA (rev ca) (prog-if 00 [Normal decode]) Flags: bus master, fast devsel, latency 0, IRQ 30, NUMA node 0 Memory at 90a00000 (32-bit, non-prefetchable) [size=256K] Bus: primary=04, secondary=05, subordinate=0a, sec-latency=0 I/O behind bridge: 00002000-00003fff Memory behind bridge: 90000000-909fffff Prefetchable memory behind bridge: 0000380000800000-0000380000bfffff Previously, when the kernel assigned resource addresses (with the pci=realloc command line parameter, for example) it could clear the struct resource corresponding to the BAR. When this happened, lspci would report this BAR as "ignored": Region 0: Memory at (32-bit, non-prefetchable) [size=256K] This is because the kernel reports a zero start address and zero flags in the corresponding sysfs resource file and in /proc/bus/pci/devices. Investigation with 'lspci -x', however, shows the BIOS-assigned address will still be programmed in the device's BAR registers. It's clearly a bug that the kernel lost track of the BAR value, but in most cases, this still won't result in a visible issue because nothing uses the memory, so nothing is affected. However, when an IOMMU is in use, it will not reserve this space in the IOVA because the kernel no longer thinks the range is valid. (See dmar_init_reserved_ranges() for the Intel implementation of this.) Without the proper reserved range, a DMA mapping may allocate an IOVA that matches a bridge BAR, which results in DMA accesses going to the BAR instead of the intended RAM. The problem was in pci_assign_unassigned_root_bus_resources(). When any resource from a bridge device fails to get assigned, the code set the resource's flags to zero. This makes sense for bridge windows, as they will be re-enabled later, but for regular BARs, it makes the kernel permanently lose track of the fact that they decode address space. Change pci_assign_unassigned_root_bus_resources() and pci_assign_unassigned_bridge_resources() so they only clear "res->flags" for bridge *windows*, not bridge BARs. Fixes: da7822e5ad71 ("PCI: update bridge resources to get more big ranges when allocating space (again)") Link: https://lore.kernel.org/r/20200108213208.4612-1-logang@deltatee.com [bhelgaas: commit log, check for pci_is_bridge()] Reported-by: Kit Chow Signed-off-by: Logan Gunthorpe Signed-off-by: Bjorn Helgaas Signed-off-by: Greg Kroah-Hartman

PCI: tegra: Fix afi_pex2_ctrl reg offset for Tegra30

2020-02-14T21:52:57Z

commit 21a92676e1fe292acb077b13106b08c22ed36b14 upstream. Fix AFI_PEX2_CTRL reg offset for Tegra30 by moving it from the Tegra20 SoC struct where it erroneously got added. This fixes the AFI_PEX2_CTRL reg offset being uninitialised subsequently failing to bring up the third PCIe port. Fixes: adb2653b3d2e ("PCI: tegra: Add AFI_PEX2_CTRL reg offset as part of SoC struct") Signed-off-by: Marcel Ziswiler Signed-off-by: Lorenzo Pieralisi Reviewed-by: Andrew Murray Acked-by: Thierry Reding Signed-off-by: Greg Kroah-Hartman