Message ID | 20241111134523.2796699-1-sthotton@marvell.com |
---|---|
State | New |
Headers | show |
Series | [v4] PCI: hotplug: Add OCTEON PCI hotplug controller driver | expand |
On Mon, Nov 11, 2024 at 07:15:11PM +0530, Shijith Thotton wrote: > This patch introduces a PCI hotplug controller driver for the OCTEON > PCIe device. The OCTEON PCIe device is a multi-function device where the > first function serves as the PCI hotplug controller. > > +--------------------------------+ > | Root Port | > +--------------------------------+ > | > PCIe > | > +---------------------------------------------------------------+ > | OCTEON PCIe Multifunction Device | > +---------------------------------------------------------------+ > | | | | > | | | | > +---------------------+ +----------------+ +-----+ +----------------+ > | Function 0 | | Function 1 | | ... | | Function 7 | > | (Hotplug controller)| | (Hotplug slot) | | | | (Hotplug slot) | > +---------------------+ +----------------+ +-----+ +----------------+ > | > | > +-------------------------+ > | Controller Firmware | > +-------------------------+ > > The hotplug controller driver enables hotplugging of non-controller > functions within the same device. During probing, the driver removes > the non-controller functions and registers them as PCI hotplug slots. > These slots are added back by the driver, only upon request from the > device firmware. > > The controller uses MSI-X interrupts to notify the host of hotplug > events initiated by the OCTEON firmware. Additionally, the driver > allows users to enable or disable individual functions via sysfs slot > entries, as provided by the PCI hotplug framework. Can we say something here about what the benefit of this driver is? For example, does it save power? What causes the function 0 firmware to request a hot-add or hot-removal of another function? Bjorn
>> This patch introduces a PCI hotplug controller driver for the OCTEON >> PCIe device. The OCTEON PCIe device is a multi-function device where the >> first function serves as the PCI hotplug controller. >> >> +--------------------------------+ >> | Root Port | >> +--------------------------------+ >> | >> PCIe >> | >> +---------------------------------------------------------------+ >> | OCTEON PCIe Multifunction Device | >> +---------------------------------------------------------------+ >> | | | | >> | | | | >> +---------------------+ +----------------+ +-----+ +----------------+ >> | Function 0 | | Function 1 | | ... | | Function 7 | >> | (Hotplug controller)| | (Hotplug slot) | | | | (Hotplug slot) | >> +---------------------+ +----------------+ +-----+ +----------------+ >> | >> | >> +-------------------------+ >> | Controller Firmware | >> +-------------------------+ >> >> The hotplug controller driver enables hotplugging of non-controller >> functions within the same device. During probing, the driver removes >> the non-controller functions and registers them as PCI hotplug slots. >> These slots are added back by the driver, only upon request from the >> device firmware. >> >> The controller uses MSI-X interrupts to notify the host of hotplug >> events initiated by the OCTEON firmware. Additionally, the driver >> allows users to enable or disable individual functions via sysfs slot >> entries, as provided by the PCI hotplug framework. > >Can we say something here about what the benefit of this driver is? >For example, does it save power? The driver enables hotplugging of non-controller functions within the device without requiring a fully implemented switch, reducing both power consumption and product cost. >What causes the function 0 firmware to request a hot-add or >hot-removal of another function? The firmware will enable the required number of non-controller functions based on runtime demand, allowing control over these functions. For example, in a vDPA scenario, each function could act as a different type of device (such as net, crypto, or storage) depending on the firmware configuration. Hot removal is useful in cases of live firmware updates. I will update the commit log with above details. Thanks, Shijith
On Tue, Nov 12, 2024 at 09:25:46AM +0000, Shijith Thotton wrote: > >> This patch introduces a PCI hotplug controller driver for the OCTEON > >> PCIe device. The OCTEON PCIe device is a multi-function device where the > >> first function serves as the PCI hotplug controller. > >> > >> +--------------------------------+ > >> | Root Port | > >> +--------------------------------+ > >> | > >> PCIe > >> | > >> +---------------------------------------------------------------+ > >> | OCTEON PCIe Multifunction Device | > >> +---------------------------------------------------------------+ > >> | | | | > >> | | | | > >> +---------------------+ +----------------+ +-----+ +----------------+ > >> | Function 0 | | Function 1 | | ... | | Function 7 | > >> | (Hotplug controller)| | (Hotplug slot) | | | | (Hotplug slot) | > >> +---------------------+ +----------------+ +-----+ +----------------+ > >> | > >> | > >> +-------------------------+ > >> | Controller Firmware | > >> +-------------------------+ > >> > >> The hotplug controller driver enables hotplugging of non-controller > >> functions within the same device. During probing, the driver removes > >> the non-controller functions and registers them as PCI hotplug slots. > >> These slots are added back by the driver, only upon request from the > >> device firmware. > >> > >> The controller uses MSI-X interrupts to notify the host of hotplug > >> events initiated by the OCTEON firmware. Additionally, the driver > >> allows users to enable or disable individual functions via sysfs slot > >> entries, as provided by the PCI hotplug framework. > > > >Can we say something here about what the benefit of this driver is? > >For example, does it save power? > > The driver enables hotplugging of non-controller functions within the device > without requiring a fully implemented switch, reducing both power consumption > and product cost. Reduced product cost is motivation for the hardware design, not for this hotplug driver. You didn't explicitly say that when function 0 hot-removes another function, it reduces overall power consumption. But I assume that's the case? > >What causes the function 0 firmware to request a hot-add or > >hot-removal of another function? > > The firmware will enable the required number of non-controller > functions based on runtime demand, allowing control over these > functions. For example, in a vDPA scenario, each function could act > as a different type of device (such as net, crypto, or storage) > depending on the firmware configuration. What is the path for this runtime demand? I assume function 0 provides some interface to request a specific kind of functionality (net, crypo, storage, etc)? I don't know anything about vDPA, so if that's important here, it needs a little more context. > Hot removal is useful in cases of live firmware updates. So the idea is that function X is hot-removed, which forces the driver to let go of it, the firmware is updated, and X is hot-added again, and the driver binds to it again? And somewhere in there is a reset of function X, and after the reset X is running the new firmware? Who/what initiates this whole path? Some request to function 0, saying "please remove function X"? But I guess maybe it doesn't go through function 0, since octeon_hp claims function 0, and it doesn't provide that functionality. Maybe the individual drivers for *other* functions know how to initiate these things, and those functions internally communicate with function 0 to ask it to start a hot-remove/hot-add sequence? That wouldn't explain the power reduction plan, though. A driver for function X could conceivably tell its device "I'm no longer needed" and function X could tell function 0 to remove it. That might enable some power savings. But that doesn't have a path to *re-enable* function X, since function X has been removed and there's no driver to ask for it to be hot-added again. Maybe there's some out-of-band management path that can tell function 0 to do things, independent of PCIe? So confused, Bjorn
>> >> This patch introduces a PCI hotplug controller driver for the OCTEON >> >> PCIe device. The OCTEON PCIe device is a multi-function device where the >> >> first function serves as the PCI hotplug controller. >> >> >> >> +--------------------------------+ >> >> | Root Port | >> >> +--------------------------------+ >> >> | >> >> PCIe >> >> | >> >> +---------------------------------------------------------------+ >> >> | OCTEON PCIe Multifunction Device | >> >> +---------------------------------------------------------------+ >> >> | | | | >> >> | | | | >> >> +---------------------+ +----------------+ +-----+ +----------------+ >> >> | Function 0 | | Function 1 | | ... | | Function 7 | >> >> | (Hotplug controller)| | (Hotplug slot) | | | | (Hotplug slot) | >> >> +---------------------+ +----------------+ +-----+ +----------------+ >> >> | >> >> | >> >> +-------------------------+ >> >> | Controller Firmware | >> >> +-------------------------+ >> >> >> >> The hotplug controller driver enables hotplugging of non-controller >> >> functions within the same device. During probing, the driver removes >> >> the non-controller functions and registers them as PCI hotplug slots. >> >> These slots are added back by the driver, only upon request from the >> >> device firmware. >> >> >> >> The controller uses MSI-X interrupts to notify the host of hotplug >> >> events initiated by the OCTEON firmware. Additionally, the driver >> >> allows users to enable or disable individual functions via sysfs slot >> >> entries, as provided by the PCI hotplug framework. >> > >> >Can we say something here about what the benefit of this driver is? >> >For example, does it save power? >> >> The driver enables hotplugging of non-controller functions within the device >> without requiring a fully implemented switch, reducing both power >consumption >> and product cost. > >Reduced product cost is motivation for the hardware design, not for >this hotplug driver. > >You didn't explicitly say that when function 0 hot-removes another >function, it reduces overall power consumption. But I assume that's >the case? > Yes, I will explain it in detail below >> >What causes the function 0 firmware to request a hot-add or >> >hot-removal of another function? >> >> The firmware will enable the required number of non-controller >> functions based on runtime demand, allowing control over these >> functions. For example, in a vDPA scenario, each function could act >> as a different type of device (such as net, crypto, or storage) >> depending on the firmware configuration. > >What is the path for this runtime demand? I assume function 0 >provides some interface to request a specific kind of functionality >(net, crypo, storage, etc)? > Right now, it done via firmware management console. >I don't know anything about vDPA, so if that's important here, it >needs a little more context. > >> Hot removal is useful in cases of live firmware updates. > >So the idea is that function X is hot-removed, which forces the driver >to let go of it, the firmware is updated, and X is hot-added again, >and the driver binds to it again? > I will explain the process in detail, which should also address the questions below. >And somewhere in there is a reset of function X, and after the reset >X is running the new firmware? > >Who/what initiates this whole path? Some request to function 0, >saying "please remove function X"? > >But I guess maybe it doesn't go through function 0, since octeon_hp >claims function 0, and it doesn't provide that functionality. Maybe >the individual drivers for *other* functions know how to initiate >these things, and those functions internally communicate with function >0 to ask it to start a hot-remove/hot-add sequence? > >That wouldn't explain the power reduction plan, though. A driver for >function X could conceivably tell its device "I'm no longer needed" >and function X could tell function 0 to remove it. That might enable >some power savings. But that doesn't have a path to *re-enable* >function X, since function X has been removed and there's no driver to >ask for it to be hot-added again. > >Maybe there's some out-of-band management path that can tell function >0 to do things, independent of PCIe? > Our implementation aims to achieve two main objectives: 1. Enable changing a function's personality at runtime. 2. Reduce power consumption. The OCTEON PCI device has multiple ARM cores running Linux, with its firmware composed of multiple components. For example, the firmware includes components like Virtio-net, NVMe, and Virtio-Crypto, which can be assigned to any function at runtime. The device firmware is accessible via a management console, allowing components to be started or stopped. For each component, an associated function is hot-added on the host to expose its functionality. Initially, after boot, only Function 0 and the controller firmware are active. Here's a breakdown: At Time 0: - Linux boots on the device, starting the controller firmware. At Time 1: - The hotplug driver loads on the host, temporarily removing other functions. At Time 2: - A network device firmware component starts on an ARM core (initiated through a console command). - This component sets up the Function 1 configuration space, data, and other request handlers for network processing. - The firmware issues a hot-add request to Function 0 (hotplug driver) on the host to enable Function 1. At Time 3: - The Function 0 hotplug driver on the host receives the hot-add request and enables Function 1 on the host. - A network driver binds to Function 1 based on device class and ID. At Time 4: - The network device firmware component receives a stop signal. - The firmware issues a hot-remove request for Function 1 on the host. - The firmware component halts, reducing the device's power consumption. At Time 5: - The Function 0 hotplug driver on the host receives the hot-remove request and disables Function 1 on the host. At Time 6: - A crypto device firmware component starts on an ARM core. - This component configures the Function 1 configuration space for crypto processing and sets up the required firmware handlers. - The firmware issues a hot-add request to enable Function 1 on the host. At Time 7: - The Function 0 hotplug driver on the host receives the hot-add request and enables Function 1 on the host. - A crypto driver binds to Function 1 based on device class and ID. The firmware component for each function only runs and is hot-added when needed. Only Function 0 and the controller firmware remain active continuously. This dynamic control reduces power usage by keeping unnecessary components off. Additionally, a single function can adapt its personality based on the associated firmware component, enhancing flexibility. I hope this clarifies the implementation. Let me know if you have any questions. Thanks, Shijith
On Wed, Nov 13, 2024 at 12:20:20PM +0000, Shijith Thotton wrote: > >> >> This patch introduces a PCI hotplug controller driver for the OCTEON > >> >> PCIe device. The OCTEON PCIe device is a multi-function device where the > >> >> first function serves as the PCI hotplug controller. > >> >> > >> >> +--------------------------------+ > >> >> | Root Port | > >> >> +--------------------------------+ > >> >> | > >> >> PCIe > >> >> | > >> >> +---------------------------------------------------------------+ > >> >> | OCTEON PCIe Multifunction Device | > >> >> +---------------------------------------------------------------+ > >> >> | | | | > >> >> | | | | > >> >> +---------------------+ +----------------+ +-----+ +----------------+ > >> >> | Function 0 | | Function 1 | | ... | | Function 7 | > >> >> | (Hotplug controller)| | (Hotplug slot) | | | | (Hotplug slot) | > >> >> +---------------------+ +----------------+ +-----+ +----------------+ > >> >> | > >> >> | > >> >> +-------------------------+ > >> >> | Controller Firmware | > >> >> +-------------------------+ > >> >> > >> >> The hotplug controller driver enables hotplugging of non-controller > >> >> functions within the same device. During probing, the driver removes > >> >> the non-controller functions and registers them as PCI hotplug slots. > >> >> These slots are added back by the driver, only upon request from the > >> >> device firmware. > >> >> > >> >> The controller uses MSI-X interrupts to notify the host of hotplug > >> >> events initiated by the OCTEON firmware. Additionally, the driver > >> >> allows users to enable or disable individual functions via sysfs slot > >> >> entries, as provided by the PCI hotplug framework. > >> > > >> >Can we say something here about what the benefit of this driver is? > >> >For example, does it save power? > >> > >> The driver enables hotplugging of non-controller functions within the device > >> without requiring a fully implemented switch, reducing both power > >consumption > >> and product cost. > > > >Reduced product cost is motivation for the hardware design, not for > >this hotplug driver. > > > >You didn't explicitly say that when function 0 hot-removes another > >function, it reduces overall power consumption. But I assume that's > >the case? > > > > Yes, I will explain it in detail below > > >> >What causes the function 0 firmware to request a hot-add or > >> >hot-removal of another function? > >> > >> The firmware will enable the required number of non-controller > >> functions based on runtime demand, allowing control over these > >> functions. For example, in a vDPA scenario, each function could act > >> as a different type of device (such as net, crypto, or storage) > >> depending on the firmware configuration. > > > >What is the path for this runtime demand? I assume function 0 > >provides some interface to request a specific kind of functionality > >(net, crypo, storage, etc)? > > > > Right now, it done via firmware management console. > > >I don't know anything about vDPA, so if that's important here, it > >needs a little more context. > > > >> Hot removal is useful in cases of live firmware updates. > > > >So the idea is that function X is hot-removed, which forces the driver > >to let go of it, the firmware is updated, and X is hot-added again, > >and the driver binds to it again? > > > > I will explain the process in detail, which should also address the questions > below. > > >And somewhere in there is a reset of function X, and after the reset > >X is running the new firmware? > > > >Who/what initiates this whole path? Some request to function 0, > >saying "please remove function X"? > > > >But I guess maybe it doesn't go through function 0, since octeon_hp > >claims function 0, and it doesn't provide that functionality. Maybe > >the individual drivers for *other* functions know how to initiate > >these things, and those functions internally communicate with function > >0 to ask it to start a hot-remove/hot-add sequence? > > > >That wouldn't explain the power reduction plan, though. A driver for > >function X could conceivably tell its device "I'm no longer needed" > >and function X could tell function 0 to remove it. That might enable > >some power savings. But that doesn't have a path to *re-enable* > >function X, since function X has been removed and there's no driver to > >ask for it to be hot-added again. > > > >Maybe there's some out-of-band management path that can tell function > >0 to do things, independent of PCIe? > > > > Our implementation aims to achieve two main objectives: > > 1. Enable changing a function's personality at runtime. > 2. Reduce power consumption. > > The OCTEON PCI device has multiple ARM cores running Linux, with its firmware > composed of multiple components. For example, the firmware includes components > like Virtio-net, NVMe, and Virtio-Crypto, which can be assigned to any function > at runtime. The device firmware is accessible via a management console, allowing > components to be started or stopped. For each component, an associated function > is hot-added on the host to expose its functionality. Initially, after boot, only > Function 0 and the controller firmware are active. > > Here's a breakdown: > > At Time 0: > - Linux boots on the device, starting the controller firmware. > > At Time 1: > - The hotplug driver loads on the host, temporarily removing other functions. > > At Time 2: > - A network device firmware component starts on an ARM core (initiated through > a console command). > - This component sets up the Function 1 configuration space, data, and other > request handlers for network processing. > - The firmware issues a hot-add request to Function 0 (hotplug driver) on the > host to enable Function 1. > > At Time 3: > - The Function 0 hotplug driver on the host receives the hot-add request and > enables Function 1 on the host. > - A network driver binds to Function 1 based on device class and ID. > > At Time 4: > - The network device firmware component receives a stop signal. > - The firmware issues a hot-remove request for Function 1 on the host. > - The firmware component halts, reducing the device's power consumption. > > At Time 5: > - The Function 0 hotplug driver on the host receives the hot-remove request and > disables Function 1 on the host. > > At Time 6: > - A crypto device firmware component starts on an ARM core. > - This component configures the Function 1 configuration space for crypto > processing and sets up the required firmware handlers. > - The firmware issues a hot-add request to enable Function 1 on the host. > > At Time 7: > - The Function 0 hotplug driver on the host receives the hot-add request and enables Function 1 on the host. > - A crypto driver binds to Function 1 based on device class and ID. > > The firmware component for each function only runs and is hot-added when > needed. Only Function 0 and the controller firmware remain active > continuously. This dynamic control reduces power usage by keeping unnecessary > components off. Additionally, a single function can adapt its personality based > on the associated firmware component, enhancing flexibility. > > I hope this clarifies the implementation. Let me know if you have any > questions. Thanks very much! I propose adding text like this to the commit log: There is an out-of-band management console interface to firmware running on function 0 whereby an administrator can disable functions to save power or enable them with one of several personalities (virtio-net, virtio-crypto, NVMe, etc) for the other functions. Function 0 initiates hotplug events handled by this driver when the other functions are enabled or disabled. I provisionally applied this to pci/hotplug-octeon, but will be happy to update the text if necessary. Bjorn
>> >> >> This patch introduces a PCI hotplug controller driver for the OCTEON >> >> >> PCIe device. The OCTEON PCIe device is a multi-function device where >the >> >> >> first function serves as the PCI hotplug controller. >> >> >> >> >> >> +--------------------------------+ >> >> >> | Root Port | >> >> >> +--------------------------------+ >> >> >> | >> >> >> PCIe >> >> >> | >> >> >> +---------------------------------------------------------------+ >> >> >> | OCTEON PCIe Multifunction Device | >> >> >> +---------------------------------------------------------------+ >> >> >> | | | | >> >> >> | | | | >> >> >> +---------------------+ +----------------+ +-----+ +----------------+ >> >> >> | Function 0 | | Function 1 | | ... | | Function 7 | >> >> >> | (Hotplug controller)| | (Hotplug slot) | | | | (Hotplug slot) | >> >> >> +---------------------+ +----------------+ +-----+ +----------------+ >> >> >> | >> >> >> | >> >> >> +-------------------------+ >> >> >> | Controller Firmware | >> >> >> +-------------------------+ >> >> >> >> >> >> The hotplug controller driver enables hotplugging of non-controller >> >> >> functions within the same device. During probing, the driver removes >> >> >> the non-controller functions and registers them as PCI hotplug slots. >> >> >> These slots are added back by the driver, only upon request from the >> >> >> device firmware. >> >> >> >> >> >> The controller uses MSI-X interrupts to notify the host of hotplug >> >> >> events initiated by the OCTEON firmware. Additionally, the driver >> >> >> allows users to enable or disable individual functions via sysfs slot >> >> >> entries, as provided by the PCI hotplug framework. >> >> > >> >> >Can we say something here about what the benefit of this driver is? >> >> >For example, does it save power? >> >> >> >> The driver enables hotplugging of non-controller functions within the >device >> >> without requiring a fully implemented switch, reducing both power >> >consumption >> >> and product cost. >> > >> >Reduced product cost is motivation for the hardware design, not for >> >this hotplug driver. >> > >> >You didn't explicitly say that when function 0 hot-removes another >> >function, it reduces overall power consumption. But I assume that's >> >the case? >> > >> >> Yes, I will explain it in detail below >> >> >> >What causes the function 0 firmware to request a hot-add or >> >> >hot-removal of another function? >> >> >> >> The firmware will enable the required number of non-controller >> >> functions based on runtime demand, allowing control over these >> >> functions. For example, in a vDPA scenario, each function could act >> >> as a different type of device (such as net, crypto, or storage) >> >> depending on the firmware configuration. >> > >> >What is the path for this runtime demand? I assume function 0 >> >provides some interface to request a specific kind of functionality >> >(net, crypo, storage, etc)? >> > >> >> Right now, it done via firmware management console. >> >> >I don't know anything about vDPA, so if that's important here, it >> >needs a little more context. >> > >> >> Hot removal is useful in cases of live firmware updates. >> > >> >So the idea is that function X is hot-removed, which forces the driver >> >to let go of it, the firmware is updated, and X is hot-added again, >> >and the driver binds to it again? >> > >> >> I will explain the process in detail, which should also address the questions >> below. >> >> >And somewhere in there is a reset of function X, and after the reset >> >X is running the new firmware? >> > >> >Who/what initiates this whole path? Some request to function 0, >> >saying "please remove function X"? >> > >> >But I guess maybe it doesn't go through function 0, since octeon_hp >> >claims function 0, and it doesn't provide that functionality. Maybe >> >the individual drivers for *other* functions know how to initiate >> >these things, and those functions internally communicate with function >> >0 to ask it to start a hot-remove/hot-add sequence? >> > >> >That wouldn't explain the power reduction plan, though. A driver for >> >function X could conceivably tell its device "I'm no longer needed" >> >and function X could tell function 0 to remove it. That might enable >> >some power savings. But that doesn't have a path to *re-enable* >> >function X, since function X has been removed and there's no driver to >> >ask for it to be hot-added again. >> > >> >Maybe there's some out-of-band management path that can tell function >> >0 to do things, independent of PCIe? >> > >> >> Our implementation aims to achieve two main objectives: >> >> 1. Enable changing a function's personality at runtime. >> 2. Reduce power consumption. >> >> The OCTEON PCI device has multiple ARM cores running Linux, with its >firmware >> composed of multiple components. For example, the firmware includes >components >> like Virtio-net, NVMe, and Virtio-Crypto, which can be assigned to any >function >> at runtime. The device firmware is accessible via a management console, >allowing >> components to be started or stopped. For each component, an associated >function >> is hot-added on the host to expose its functionality. Initially, after boot, only >> Function 0 and the controller firmware are active. >> >> Here's a breakdown: >> >> At Time 0: >> - Linux boots on the device, starting the controller firmware. >> >> At Time 1: >> - The hotplug driver loads on the host, temporarily removing other functions. >> >> At Time 2: >> - A network device firmware component starts on an ARM core (initiated >through >> a console command). >> - This component sets up the Function 1 configuration space, data, and other >> request handlers for network processing. >> - The firmware issues a hot-add request to Function 0 (hotplug driver) on the >> host to enable Function 1. >> >> At Time 3: >> - The Function 0 hotplug driver on the host receives the hot-add request and >> enables Function 1 on the host. >> - A network driver binds to Function 1 based on device class and ID. >> >> At Time 4: >> - The network device firmware component receives a stop signal. >> - The firmware issues a hot-remove request for Function 1 on the host. >> - The firmware component halts, reducing the device's power consumption. >> >> At Time 5: >> - The Function 0 hotplug driver on the host receives the hot-remove request >and >> disables Function 1 on the host. >> >> At Time 6: >> - A crypto device firmware component starts on an ARM core. >> - This component configures the Function 1 configuration space for crypto >> processing and sets up the required firmware handlers. >> - The firmware issues a hot-add request to enable Function 1 on the host. >> >> At Time 7: >> - The Function 0 hotplug driver on the host receives the hot-add request and >enables Function 1 on the host. >> - A crypto driver binds to Function 1 based on device class and ID. >> >> The firmware component for each function only runs and is hot-added when >> needed. Only Function 0 and the controller firmware remain active >> continuously. This dynamic control reduces power usage by keeping >unnecessary >> components off. Additionally, a single function can adapt its personality based >> on the associated firmware component, enhancing flexibility. >> >> I hope this clarifies the implementation. Let me know if you have any >> questions. > >Thanks very much! I propose adding text like this to the commit log: > > There is an out-of-band management console interface to firmware > running on function 0 whereby an administrator can disable functions > to save power or enable them with one of several personalities > (virtio-net, virtio-crypto, NVMe, etc) for the other functions. > Function 0 initiates hotplug events handled by this driver when the > other functions are enabled or disabled. > >I provisionally applied this to pci/hotplug-octeon, but will be happy >to update the text if necessary. > Thank you. The commit log looks good. I hope no further action is required from my side. Shijith
diff --git a/MAINTAINERS b/MAINTAINERS index 42decde38320..2258f1459440 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -13677,6 +13677,12 @@ R: schalla@marvell.com R: vattunuru@marvell.com F: drivers/vdpa/octeon_ep/ +MARVELL OCTEON HOTPLUG DRIVER +R: Shijith Thotton <sthotton@marvell.com> +R: Vamsi Attunuru <vattunuru@marvell.com> +S: Supported +F: drivers/pci/hotplug/octep_hp.c + MATROX FRAMEBUFFER DRIVER L: linux-fbdev@vger.kernel.org S: Orphan diff --git a/drivers/pci/hotplug/Kconfig b/drivers/pci/hotplug/Kconfig index 1472aef0fb81..123c4c7c2ab5 100644 --- a/drivers/pci/hotplug/Kconfig +++ b/drivers/pci/hotplug/Kconfig @@ -118,6 +118,16 @@ config HOTPLUG_PCI_CPCI_GENERIC When in doubt, say N. +config HOTPLUG_PCI_OCTEONEP + bool "Marvell OCTEON PCI Hotplug driver" + depends on HOTPLUG_PCI + help + Say Y here if you have an OCTEON PCIe device with a hotplug + controller. This driver enables the non-controller functions of the + device to be registered as hotplug slots. + + When in doubt, say N. + config HOTPLUG_PCI_SHPC bool "SHPC PCI Hotplug driver" help diff --git a/drivers/pci/hotplug/Makefile b/drivers/pci/hotplug/Makefile index 240c99517d5e..40aaf31fe338 100644 --- a/drivers/pci/hotplug/Makefile +++ b/drivers/pci/hotplug/Makefile @@ -20,6 +20,7 @@ obj-$(CONFIG_HOTPLUG_PCI_RPA) += rpaphp.o obj-$(CONFIG_HOTPLUG_PCI_RPA_DLPAR) += rpadlpar_io.o obj-$(CONFIG_HOTPLUG_PCI_ACPI) += acpiphp.o obj-$(CONFIG_HOTPLUG_PCI_S390) += s390_pci_hpc.o +obj-$(CONFIG_HOTPLUG_PCI_OCTEONEP) += octep_hp.o # acpiphp_ibm extends acpiphp, so should be linked afterwards. diff --git a/drivers/pci/hotplug/octep_hp.c b/drivers/pci/hotplug/octep_hp.c new file mode 100644 index 000000000000..d2475feb44cc --- /dev/null +++ b/drivers/pci/hotplug/octep_hp.c @@ -0,0 +1,427 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Copyright (C) 2024 Marvell. */ + +#include <linux/cleanup.h> +#include <linux/container_of.h> +#include <linux/delay.h> +#include <linux/dev_printk.h> +#include <linux/init.h> +#include <linux/interrupt.h> +#include <linux/io-64-nonatomic-lo-hi.h> +#include <linux/kernel.h> +#include <linux/list.h> +#include <linux/module.h> +#include <linux/mutex.h> +#include <linux/pci.h> +#include <linux/pci_hotplug.h> +#include <linux/slab.h> +#include <linux/spinlock.h> +#include <linux/workqueue.h> + +#define OCTEP_HP_INTR_OFFSET(x) (0x20400 + ((x) << 4)) +#define OCTEP_HP_INTR_VECTOR(x) (16 + (x)) +#define OCTEP_HP_DRV_NAME "octep_hp" + +/* + * Type of MSI-X interrupts. + * The macros OCTEP_HP_INTR_VECTOR and OCTEP_HP_INTR_OFFSET are used to + * generate the vector and offset for an interrupt type. + */ +enum octep_hp_intr_type { + OCTEP_HP_INTR_INVALID = -1, + OCTEP_HP_INTR_ENA = 0, + OCTEP_HP_INTR_DIS = 1, + OCTEP_HP_INTR_MAX = 2, +}; + +struct octep_hp_cmd { + struct list_head list; + enum octep_hp_intr_type intr_type; + u64 intr_val; +}; + +struct octep_hp_slot { + struct list_head list; + struct hotplug_slot slot; + u16 slot_number; + struct pci_dev *hp_pdev; + unsigned int hp_devfn; + struct octep_hp_controller *ctrl; +}; + +struct octep_hp_intr_info { + enum octep_hp_intr_type type; + int number; + char name[16]; +}; + +struct octep_hp_controller { + void __iomem *base; + struct pci_dev *pdev; + struct octep_hp_intr_info intr[OCTEP_HP_INTR_MAX]; + struct work_struct work; + struct list_head slot_list; + struct mutex slot_lock; /* Protects slot_list */ + struct list_head hp_cmd_list; + spinlock_t hp_cmd_lock; /* Protects hp_cmd_list */ +}; + +static void octep_hp_enable_pdev(struct octep_hp_controller *hp_ctrl, + struct octep_hp_slot *hp_slot) +{ + guard(mutex)(&hp_ctrl->slot_lock); + if (hp_slot->hp_pdev) { + dev_dbg(&hp_slot->hp_pdev->dev, "Slot %s is already enabled\n", + hotplug_slot_name(&hp_slot->slot)); + return; + } + + /* Scan the device and add it to the bus */ + hp_slot->hp_pdev = pci_scan_single_device(hp_ctrl->pdev->bus, + hp_slot->hp_devfn); + pci_bus_assign_resources(hp_ctrl->pdev->bus); + pci_bus_add_device(hp_slot->hp_pdev); + + dev_dbg(&hp_slot->hp_pdev->dev, "Enabled slot %s\n", + hotplug_slot_name(&hp_slot->slot)); +} + +static void octep_hp_disable_pdev(struct octep_hp_controller *hp_ctrl, + struct octep_hp_slot *hp_slot) +{ + guard(mutex)(&hp_ctrl->slot_lock); + if (!hp_slot->hp_pdev) { + dev_dbg(&hp_ctrl->pdev->dev, "Slot %s is already disabled\n", + hotplug_slot_name(&hp_slot->slot)); + return; + } + + dev_dbg(&hp_slot->hp_pdev->dev, "Disabling slot %s\n", + hotplug_slot_name(&hp_slot->slot)); + + /* Remove the device from the bus */ + pci_stop_and_remove_bus_device_locked(hp_slot->hp_pdev); + hp_slot->hp_pdev = NULL; +} + +static int octep_hp_enable_slot(struct hotplug_slot *slot) +{ + struct octep_hp_slot *hp_slot = + container_of(slot, struct octep_hp_slot, slot); + + octep_hp_enable_pdev(hp_slot->ctrl, hp_slot); + return 0; +} + +static int octep_hp_disable_slot(struct hotplug_slot *slot) +{ + struct octep_hp_slot *hp_slot = + container_of(slot, struct octep_hp_slot, slot); + + octep_hp_disable_pdev(hp_slot->ctrl, hp_slot); + return 0; +} + +static struct hotplug_slot_ops octep_hp_slot_ops = { + .enable_slot = octep_hp_enable_slot, + .disable_slot = octep_hp_disable_slot, +}; + +#define SLOT_NAME_SIZE 16 +static struct octep_hp_slot * +octep_hp_register_slot(struct octep_hp_controller *hp_ctrl, + struct pci_dev *pdev, u16 slot_number) +{ + char slot_name[SLOT_NAME_SIZE]; + struct octep_hp_slot *hp_slot; + int ret; + + hp_slot = kzalloc(sizeof(*hp_slot), GFP_KERNEL); + if (!hp_slot) + return ERR_PTR(-ENOMEM); + + hp_slot->ctrl = hp_ctrl; + hp_slot->hp_pdev = pdev; + hp_slot->hp_devfn = pdev->devfn; + hp_slot->slot_number = slot_number; + hp_slot->slot.ops = &octep_hp_slot_ops; + + snprintf(slot_name, sizeof(slot_name), "octep_hp_%u", slot_number); + ret = pci_hp_register(&hp_slot->slot, hp_ctrl->pdev->bus, + PCI_SLOT(pdev->devfn), slot_name); + if (ret) { + kfree(hp_slot); + return ERR_PTR(ret); + } + + dev_info(&pdev->dev, "Registered slot %s for device %s\n", + slot_name, pci_name(pdev)); + + list_add_tail(&hp_slot->list, &hp_ctrl->slot_list); + octep_hp_disable_pdev(hp_ctrl, hp_slot); + + return hp_slot; +} + +static void octep_hp_deregister_slot(void *data) +{ + struct octep_hp_slot *hp_slot = data; + struct octep_hp_controller *hp_ctrl = hp_slot->ctrl; + + pci_hp_deregister(&hp_slot->slot); + octep_hp_enable_pdev(hp_ctrl, hp_slot); + list_del(&hp_slot->list); + kfree(hp_slot); +} + +static const char *octep_hp_cmd_name(enum octep_hp_intr_type type) +{ + switch (type) { + case OCTEP_HP_INTR_ENA: + return "hotplug enable"; + case OCTEP_HP_INTR_DIS: + return "hotplug disable"; + default: + return "invalid"; + } +} + +static void octep_hp_cmd_handler(struct octep_hp_controller *hp_ctrl, + struct octep_hp_cmd *hp_cmd) +{ + struct octep_hp_slot *hp_slot; + + /* + * Enable or disable the slots based on the slot mask. + * intr_val is a bit mask where each bit represents a slot. + */ + list_for_each_entry(hp_slot, &hp_ctrl->slot_list, list) { + if (!(hp_cmd->intr_val & BIT(hp_slot->slot_number))) + continue; + + dev_info(&hp_ctrl->pdev->dev, "Received %s command for slot %s\n", + octep_hp_cmd_name(hp_cmd->intr_type), + hotplug_slot_name(&hp_slot->slot)); + + switch (hp_cmd->intr_type) { + case OCTEP_HP_INTR_ENA: + octep_hp_enable_pdev(hp_ctrl, hp_slot); + break; + case OCTEP_HP_INTR_DIS: + octep_hp_disable_pdev(hp_ctrl, hp_slot); + break; + default: + break; + } + } +} + +static void octep_hp_work_handler(struct work_struct *work) +{ + struct octep_hp_controller *hp_ctrl; + struct octep_hp_cmd *hp_cmd; + unsigned long flags; + + hp_ctrl = container_of(work, struct octep_hp_controller, work); + + /* Process all the hotplug commands */ + spin_lock_irqsave(&hp_ctrl->hp_cmd_lock, flags); + while (!list_empty(&hp_ctrl->hp_cmd_list)) { + hp_cmd = list_first_entry(&hp_ctrl->hp_cmd_list, + struct octep_hp_cmd, list); + list_del(&hp_cmd->list); + spin_unlock_irqrestore(&hp_ctrl->hp_cmd_lock, flags); + + octep_hp_cmd_handler(hp_ctrl, hp_cmd); + kfree(hp_cmd); + + spin_lock_irqsave(&hp_ctrl->hp_cmd_lock, flags); + } + spin_unlock_irqrestore(&hp_ctrl->hp_cmd_lock, flags); +} + +static enum octep_hp_intr_type octep_hp_intr_type(struct octep_hp_intr_info *intr, + int irq) +{ + enum octep_hp_intr_type type; + + for (type = OCTEP_HP_INTR_ENA; type < OCTEP_HP_INTR_MAX; type++) { + if (intr[type].number == irq) + return type; + } + + return OCTEP_HP_INTR_INVALID; +} + +static irqreturn_t octep_hp_intr_handler(int irq, void *data) +{ + struct octep_hp_controller *hp_ctrl = data; + struct pci_dev *pdev = hp_ctrl->pdev; + enum octep_hp_intr_type type; + struct octep_hp_cmd *hp_cmd; + u64 intr_val; + + type = octep_hp_intr_type(hp_ctrl->intr, irq); + if (type == OCTEP_HP_INTR_INVALID) { + dev_err(&pdev->dev, "Invalid interrupt %d\n", irq); + return IRQ_HANDLED; + } + + /* Read and clear the interrupt */ + intr_val = readq(hp_ctrl->base + OCTEP_HP_INTR_OFFSET(type)); + writeq(intr_val, hp_ctrl->base + OCTEP_HP_INTR_OFFSET(type)); + + hp_cmd = kzalloc(sizeof(*hp_cmd), GFP_ATOMIC); + if (!hp_cmd) + return IRQ_HANDLED; + + hp_cmd->intr_val = intr_val; + hp_cmd->intr_type = type; + + /* Add the command to the list and schedule the work */ + spin_lock(&hp_ctrl->hp_cmd_lock); + list_add_tail(&hp_cmd->list, &hp_ctrl->hp_cmd_list); + spin_unlock(&hp_ctrl->hp_cmd_lock); + schedule_work(&hp_ctrl->work); + + return IRQ_HANDLED; +} + +static void octep_hp_irq_cleanup(void *data) +{ + struct octep_hp_controller *hp_ctrl = data; + + pci_free_irq_vectors(hp_ctrl->pdev); + flush_work(&hp_ctrl->work); +} + +static int octep_hp_request_irq(struct octep_hp_controller *hp_ctrl, + enum octep_hp_intr_type type) +{ + struct pci_dev *pdev = hp_ctrl->pdev; + struct octep_hp_intr_info *intr; + int irq; + + irq = pci_irq_vector(pdev, OCTEP_HP_INTR_VECTOR(type)); + if (irq < 0) + return irq; + + intr = &hp_ctrl->intr[type]; + intr->number = irq; + intr->type = type; + snprintf(intr->name, sizeof(intr->name), "octep_hp_%d", type); + + return devm_request_irq(&pdev->dev, irq, octep_hp_intr_handler, + IRQF_SHARED, intr->name, hp_ctrl); +} + +static int octep_hp_controller_setup(struct pci_dev *pdev, + struct octep_hp_controller *hp_ctrl) +{ + struct device *dev = &pdev->dev; + enum octep_hp_intr_type type; + int ret; + + ret = pcim_enable_device(pdev); + if (ret) + return dev_err_probe(dev, ret, "Failed to enable PCI device\n"); + + hp_ctrl->base = pcim_iomap_region(pdev, 0, OCTEP_HP_DRV_NAME); + if (IS_ERR(hp_ctrl->base)) + return dev_err_probe(dev, PTR_ERR(hp_ctrl->base), + "Failed to map PCI device region\n"); + + pci_set_master(pdev); + pci_set_drvdata(pdev, hp_ctrl); + + INIT_LIST_HEAD(&hp_ctrl->slot_list); + INIT_LIST_HEAD(&hp_ctrl->hp_cmd_list); + mutex_init(&hp_ctrl->slot_lock); + spin_lock_init(&hp_ctrl->hp_cmd_lock); + INIT_WORK(&hp_ctrl->work, octep_hp_work_handler); + hp_ctrl->pdev = pdev; + + ret = pci_alloc_irq_vectors(pdev, 1, + OCTEP_HP_INTR_VECTOR(OCTEP_HP_INTR_MAX), + PCI_IRQ_MSIX); + if (ret < 0) + return dev_err_probe(dev, ret, "Failed to alloc MSI-X vectors\n"); + + ret = devm_add_action(&pdev->dev, octep_hp_irq_cleanup, hp_ctrl); + if (ret) + return dev_err_probe(&pdev->dev, ret, "Failed to add IRQ cleanup action\n"); + + for (type = OCTEP_HP_INTR_ENA; type < OCTEP_HP_INTR_MAX; type++) { + ret = octep_hp_request_irq(hp_ctrl, type); + if (ret) + return dev_err_probe(dev, ret, + "Failed to request IRQ for vector %d\n", + OCTEP_HP_INTR_VECTOR(type)); + } + + return 0; +} + +static int octep_hp_pci_probe(struct pci_dev *pdev, + const struct pci_device_id *id) +{ + struct octep_hp_controller *hp_ctrl; + struct pci_dev *tmp_pdev, *next; + struct octep_hp_slot *hp_slot; + u16 slot_number = 0; + int ret; + + hp_ctrl = devm_kzalloc(&pdev->dev, sizeof(*hp_ctrl), GFP_KERNEL); + if (!hp_ctrl) + return -ENOMEM; + + ret = octep_hp_controller_setup(pdev, hp_ctrl); + if (ret) + return ret; + + /* + * Register all hotplug slots. Hotplug controller is the first function + * of the PCI device. The hotplug slots are the remaining functions of + * the PCI device. The hotplug slot functions are logically removed from + * the bus during probing and are re-enabled by the driver when a + * hotplug event is received. + */ + list_for_each_entry_safe(tmp_pdev, next, &pdev->bus->devices, bus_list) { + if (tmp_pdev == pdev) + continue; + + hp_slot = octep_hp_register_slot(hp_ctrl, tmp_pdev, slot_number); + if (IS_ERR(hp_slot)) + return dev_err_probe(&pdev->dev, PTR_ERR(hp_slot), + "Failed to register hotplug slot %u\n", + slot_number); + + ret = devm_add_action(&pdev->dev, octep_hp_deregister_slot, + hp_slot); + if (ret) + return dev_err_probe(&pdev->dev, ret, + "Failed to add action for deregistering slot %u\n", + slot_number); + slot_number++; + } + + return 0; +} + +#define PCI_DEVICE_ID_CAVIUM_OCTEP_HP_CTLR 0xa0e3 +static struct pci_device_id octep_hp_pci_map[] = { + { PCI_DEVICE(PCI_VENDOR_ID_CAVIUM, PCI_DEVICE_ID_CAVIUM_OCTEP_HP_CTLR) }, + { }, +}; + +static struct pci_driver octep_hp = { + .name = OCTEP_HP_DRV_NAME, + .id_table = octep_hp_pci_map, + .probe = octep_hp_pci_probe, +}; + +module_pci_driver(octep_hp); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Marvell"); +MODULE_DESCRIPTION("Marvell OCTEON PCI Hotplug driver");