Hi all, This is sixth iteration of the patch series trying to solve the problem reported by Chris Chiu [1]. In summary the current resource distribution code does not cover the initial device enumeration so if we find unconfigured bridges they get the bare minimum. In addition to that it turned out the current resource distribution code does not take into account possible multifunction devices and/or other devices on the bus. The patch 1/2 tries to make it more generic. I've tested it on QEMU following the topology Jonathan is using and also in a a couple of systems with Thunderbolt controller and complex topologies to make sure it still keeps working. The previous versions of the series can be found: v5: https://lore.kernel.org/linux-pci/20230112110000.59974-1-mika.westerberg@xxxxxxxxxxxxxxx/ v4: https://lore.kernel.org/linux-pci/20230104091635.63331-1-mika.westerberg@xxxxxxxxxxxxxxx/ v3: https://lore.kernel.org/linux-pci/20221130112221.66612-1-mika.westerberg@xxxxxxxxxxxxxxx/ v2: https://lore.kernel.org/linux-pci/20221114115953.40236-1-mika.westerberg@xxxxxxxxxxxxxxx/ v1: https://lore.kernel.org/linux-pci/20221103103254.30497-1-mika.westerberg@xxxxxxxxxxxxxxx/ Changes from v5: * Correct typo in mmio calculation (io_per_hp -> mmio_per_hp) * Add missing -1 to the mmio_pref end when align == 0. Changes from v4: * Split the alignment fix into separate patch * Create helper for removing resources * Skip VF BARs when removing * Add check for 32-bit/64-bit prefetchable resource so that we account them correctly. * Update comments and commit log slightly according to review comments. * Did not add the "lookup for hotplug bridges below non-hotplug ones". It turned out to be non-trivial. The current code works for the "common" case of Thunderbolt/USB4 PCIe switches but we may need to revisit this if there is a real need. Changes from v3: * Make it more generic and not depend on how many bridges there are on the bus. Changes from v2: * Make both patches to work with PCI devices too (do not expect that the bridge is always first device on the bus). * Allow distribution with bridges that do not have all resource windows programmed (therefore the patch 2/2 is not revert anymore) * I did not add the tags from Rafael and Jonathan because the code is not exactly the same anymore so was not sure if they still apply. Changes from v1: * Re-worded the commit message to hopefully explain the problem better * Added Link: to the bug report * Update the comment according to Bjorn's suggestion * Dropped the ->multifunction check * Use %#llx in log format. [1] https://bugzilla.kernel.org/show_bug.cgi?id=216000 Mika Westerberg (3): PCI: Align extra resources for hotplug bridges properly PCI: Take other bus devices into account when distributing resources PCI: Distribute available resources for root buses too drivers/pci/setup-bus.c | 232 ++++++++++++++++++++++++++++------------ 1 file changed, 166 insertions(+), 66 deletions(-) -- 2.39.0
