On Thursday 09 June 2022 11:22:55 Bjorn Helgaas wrote: > [+cc Guilherme, Michael, Ben (author of 63a72284b159 and PPC folks), thread: > https://lore.kernel.org/r/20220504175718.29011-1-pali@xxxxxxxxxx] > > On Fri, May 06, 2022 at 12:33:02AM +0200, Pali Rohár wrote: > > On Thursday 05 May 2022 15:10:01 Tyrel Datwyler wrote: > > > On 5/5/22 02:31, Pali Rohár wrote: > > > > On Thursday 05 May 2022 07:16:40 Christophe Leroy wrote: > > > >> Le 04/05/2022 à 19:57, Pali Rohár a écrit : > > > >>> Since commit 63a72284b159 ("powerpc/pci: Assign fixed PHB > > > >>> number based on device-tree properties"), powerpc kernel > > > >>> always fallback to PCI domain assignment from OF / Device Tree > > > >>> 'reg' property of the PCI controller. > > > >>> > > > >>> PCI code for other Linux architectures use increasing > > > >>> assignment of the PCI domain for individual controllers > > > >>> (assign the first free number), like it was also for powerpc > > > >>> prior mentioned commit. > > > >>> > > > >>> Upgrading powerpc kernels from LTS 4.4 version (which does not > > > >>> contain mentioned commit) to new LTS versions brings a > > > >>> regression in domain assignment. > > > >> > > > >> Can you elaborate why it is a regression ? > > > >> 63a72284b159 That commit says 'no functionnal changes', I'm > > > >> having hard time understanding how a nochange can be a > > > >> regression. > > > > > > > > It is not 'no functional change'. That commit completely changed > > > > PCI domain assignment in a way that is incompatible with other > > > > architectures and also incompatible with the way how it was done > > > > prior that commit. > > > > > > I agree that the "no functional change" statement is incorrect. > > > However, for most powerpc platforms it ended up being simply a > > > cosmetic behavior change. As far as I can tell there is nothing > > > requiring domain ids to increase montonically from zero or that > > > each architecture is required to use the same domain numbering > > > scheme. > > > > That is truth. But it looks really suspicious why domains are not > > assigned monotonically. Some scripts / applications are using PCI > > location (domain:bus:dev:func) for remembering PCI device and domain > > change can cause issue for config files. And some (older) applications > > expects existence of domain zero. In systems without hot plug support > > with small number of domains (e.g. 3) it means that there are always > > domains 0, 1 and 2. > > > > > Its hard to call this a true regression unless it actually broke > > > something. The commit in question has been in the kernel since 4.8 > > > which was released over 5 1/2 years ago. > > > > I agree, it really depends on how you look at it. > > > > The important is that lot of people are using LTS versions and are > > doing upgrades when LTS support is dropped. Which for 4.4 now > > happened. So not all smaller or "cosmetic" changes could be detected > > until longer LTS period pass. > > > > > With all that said looking closer at the code in question I think > > > it is fair to assume that the author only intended this change for > > > powernv and pseries platforms and not every powerpc platform. That > > > change was done to make persistent naming easier to manage in > > > userspace. > > > > I agree that this behavior change may be useful in some situations > > and I do not object this need. > > > > > Your change defaults back to the old behavior which will now break > > > both powernv and pseries platforms with regard to hotplugging and > > > persistent naming. > > > > I was aware of it, that change could cause issues. And that is why I > > added config option for choosing behavior. So users would be able to > > choose what they need. > > > > > We could properly limit it to powernv and pseries by using > > > ibm,fw-phb-id instead of reg property in the look up that follows > > > a failed ibm,opal-phbid lookup. I think this is acceptable as long > > > as no other powerpc platforms have started using this behavior for > > > persistent naming. > > > > And what about setting that new config option to enabled by default > > for those series? > > > > Or is there issue with introduction of the new config option? > > > > One of the point is that it is really a good idea to have > > similar/same behavior for all linux platforms. And if it cannot be > > enabled by default (for backward compatibility) add at least some > > option, so new platforms can start using it or users can decide to > > switch behavior. > > This is a powerpc thing so I'm just kibbitzing a little. > > This basically looks like a new config option to selectively revert > 63a72284b159. That seems hard to maintain and doesn't seem like > something that needs to be baked into the kernel at compile-time. > > The 63a72284b159 commit log says persistent NIC names are tied to PCI > domain/bus/dev/fn addresses, which seems like something we should > discourage because we can't predict PCI addresses in general. I > assume other platforms typically use udev with MAC addresses or > something? This is not about ethernet NIC cards only. But affects also WiFi cards (which registers phy dev, not netdev) and also all other PCIe cards which do not have to be network-based. Hence MAC address or udev does not play role there. > > > > For example, prior that commit on P2020 RDB board were PCI > > > > domains 0, 1 and 2. > > > > > > > > $ lspci > > > > 0000:00:00.0 PCI bridge: Freescale Semiconductor Inc P2020E (rev 21) > > > > 0000:01:00.0 USB controller: Texas Instruments TUSB73x0 SuperSpeed USB 3.0 xHCI Host Controller (rev 02) > > > > 0001:02:00.0 PCI bridge: Freescale Semiconductor Inc P2020E (rev 21) > > > > 0001:03:00.0 Network controller: Qualcomm Atheros AR93xx Wireless Network Adapter (rev 01) > > > > 0002:04:00.0 PCI bridge: Freescale Semiconductor Inc P2020E (rev 21) > > > > 0002:05:00.0 Network controller: Qualcomm Atheros QCA986x/988x 802.11ac Wireless Network Adapter > > > > > > > > After that commit on P2020 RDB board are PCI domains 0x8000, > > > > 0x9000 and 0xa000. > > > > > > > > $ lspci > > > > 8000:00:00.0 PCI bridge: Freescale Semiconductor Inc P2020E (rev 21) > > > > 8000:01:00.0 USB controller: Texas Instruments TUSB73x0 SuperSpeed USB 3.0 xHCI Host Controller (rev 02) > > > > 9000:02:00.0 PCI bridge: Freescale Semiconductor Inc P2020E (rev 21) > > > > 9000:03:00.0 Network controller: Qualcomm Atheros AR93xx Wireless Network Adapter (rev 01) > > > > a000:04:00.0 PCI bridge: Freescale Semiconductor Inc P2020E (rev 21) > > > > a000:05:00.0 Network controller: Qualcomm Atheros QCA986x/988x 802.11ac Wireless Network Adapter > > > > > > > > It is somehow strange that PCI domains are not indexed one by one and > > > > also that there is no domain 0 > > > > > > > > With my patch when CONFIG_PPC_PCI_DOMAIN_FROM_OF_REG is not set, then > > > > previous behavior used and PCI domains are again 0, 1 and 2. > > > > > > > >> Usually we don't commit regressions to mainline ... > > > >> > > > >>> Fix this issue by introducing a new option > > > >>> CONFIG_PPC_PCI_DOMAIN_FROM_OF_REG When this options is > > > >>> disabled then powerpc kernel would assign PCI domains in the > > > >>> similar way like it is doing kernel for other architectures > > > >>> and also how it was done prior that commit. > > > >> > > > >> You don't define CONFIG_PPC_PCI_DOMAIN_FROM_OF_REG on by > > > >> default, it means this commit will change the behaviour. Is > > > >> that expected ? > > > >> > > > >> Is that really worth a user selectable option ? Is the user > > > >> able to decide what he needs ? > > > > > > > > Well, I hope that maintainers of that code answer to these > > > > questions. > > > > > > > > In any case, I think that it could be a user selectable option > > > > as in that commit is explained that in some situation is makes > > > > sense to do PCI domain numbering based on DT reg. > > > > > > > > But as I pointed above, upgrading from 4.4 TLS kernel to some > > > > new TLS kernel brings above regression, so I think that there > > > > should be a way to disable this behavior. > > > > > > > > In my opinion, for people who are upgrading from 4.4 TLS kernel, > > > > this option should be turned off by default (= do not change > > > > behavior). For people who want same behaviour on powerpc as on > > > > other platforms, also it should be turned off by default. > > > > > > > >>> > > > >>> Fixes: 63a72284b159 ("powerpc/pci: Assign fixed PHB number based on device-tree properties") > > > >> > > > >> Is that really a fix ? What is the problem really ? > > > > > > > > Problem is that PCI domains were changed in a way that is not > > > > compatible neither with version prior that commit and neither > > > > with how other linux platforms assign PCI domains for > > > > controllers. > > > > > > > >>> Signed-off-by: Pali Rohár <pali@xxxxxxxxxx> > > > >>> --- > > > >>> arch/powerpc/Kconfig | 10 ++++++++++ > > > >>> arch/powerpc/kernel/pci-common.c | 4 ++-- > > > >>> 2 files changed, 12 insertions(+), 2 deletions(-) > > > >>> > > > >>> diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig > > > >>> index 174edabb74fa..4dd3e3acddda 100644 > > > >>> --- a/arch/powerpc/Kconfig > > > >>> +++ b/arch/powerpc/Kconfig > > > >>> @@ -375,6 +375,16 @@ config PPC_OF_PLATFORM_PCI > > > >>> depends on PCI > > > >>> depends on PPC64 # not supported on 32 bits yet > > > >>> > > > >>> +config PPC_PCI_DOMAIN_FROM_OF_REG > > > >>> + bool "Use OF reg property for PCI domain" > > > >>> + depends on PCI > > > >> > > > >> Should it depend on PPC_OF_PLATFORM_PCI instead ? > > > > > > > > No, PPC_OF_PLATFORM_PCI has line "depends on PPC64 # not supported on 32 > > > > bits yet". But it is already used also for e.g. P2020 which is 32-bit > > > > platform. > > > > > > > >>> + help > > > >>> + By default PCI domain for host bridge during its registration is > > > >>> + chosen as the lowest unused PCI domain number. > > > >>> + > > > >>> + When this option is enabled then PCI domain is determined from > > > >>> + the OF / Device Tree 'reg' property. > > > >>> + > > > >>> config ARCH_SUPPORTS_UPROBES > > > >>> def_bool y > > > >>> > > > >>> diff --git a/arch/powerpc/kernel/pci-common.c b/arch/powerpc/kernel/pci-common.c > > > >>> index 8bc9cf62cd93..8cb6fc5302ae 100644 > > > >>> --- a/arch/powerpc/kernel/pci-common.c > > > >>> +++ b/arch/powerpc/kernel/pci-common.c > > > >>> @@ -74,7 +74,6 @@ void __init set_pci_dma_ops(const struct dma_map_ops *dma_ops) > > > >>> static int get_phb_number(struct device_node *dn) > > > >>> { > > > >>> int ret, phb_id = -1; > > > >>> - u32 prop_32; > > > >>> u64 prop; > > > >>> > > > >>> /* > > > >>> @@ -83,7 +82,8 @@ static int get_phb_number(struct device_node *dn) > > > >>> * reading "ibm,opal-phbid", only present in OPAL environment. > > > >>> */ > > > >>> ret = of_property_read_u64(dn, "ibm,opal-phbid", &prop); > > > >> > > > >> This looks like very specific, it is not reflected in the commit log. > > > > > > > > I have not changed nor touched this "ibm,opal-phbid" setting. And it was > > > > not also touched in that mentioned patch. I see that no DTS file in > > > > kernel use this option (so probably only DTS files supplied by > > > > bootloader use it). So I thought that there is not reason to mention in > > > > commit message. > > > > > > > > But if you think so, I can add some info to commit message about it. > > > > > > > >>> - if (ret) { > > > >>> + if (ret && IS_ENABLED(CONFIG_PPC_PCI_DOMAIN_FROM_OF_REG)) { > > > >>> + u32 prop_32; > > > >>> ret = of_property_read_u32_index(dn, "reg", 1, &prop_32); > > > >>> prop = prop_32; > > > >>> } > > >
