linux-stable/kernel/irq/msi.c

1626 lines
44 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0
/*
* Copyright (C) 2014 Intel Corp.
* Author: Jiang Liu <jiang.liu@linux.intel.com>
*
* This file is licensed under GPLv2.
*
* This file contains common code to support Message Signaled Interrupts for
* PCI compatible and non PCI compatible devices.
*/
#include <linux/types.h>
#include <linux/device.h>
#include <linux/irq.h>
#include <linux/irqdomain.h>
#include <linux/msi.h>
#include <linux/slab.h>
#include <linux/sysfs.h>
#include <linux/pci.h>
#include "internals.h"
/**
* struct msi_ctrl - MSI internal management control structure
* @domid: ID of the domain on which management operations should be done
* @first: First (hardware) slot index to operate on
* @last: Last (hardware) slot index to operate on
* @nirqs: The number of Linux interrupts to allocate. Can be larger
* than the range due to PCI/multi-MSI.
*/
struct msi_ctrl {
unsigned int domid;
unsigned int first;
unsigned int last;
unsigned int nirqs;
};
/* Invalid Xarray index which is outside of any searchable range */
#define MSI_XA_MAX_INDEX (ULONG_MAX - 1)
/* The maximum domain size */
#define MSI_XA_DOMAIN_SIZE (MSI_MAX_INDEX + 1)
static void msi_domain_free_locked(struct device *dev, struct msi_ctrl *ctrl);
static unsigned int msi_domain_get_hwsize(struct device *dev, unsigned int domid);
static inline int msi_sysfs_create_group(struct device *dev);
/**
* msi_alloc_desc - Allocate an initialized msi_desc
* @dev: Pointer to the device for which this is allocated
* @nvec: The number of vectors used in this entry
* @affinity: Optional pointer to an affinity mask array size of @nvec
*
genirq: Fix kernel-doc warnings in pm.c, msi.c and ipi.c Fix all kernel-doc warnings in these 3 files and do some simple editing (capitalize acronyms, capitalize Linux). kernel/irq/pm.c:235: warning: expecting prototype for irq_pm_syscore_ops(). Prototype was for irq_pm_syscore_resume() instead kernel/irq/msi.c:530: warning: expecting prototype for __msi_domain_free_irqs(). Prototype was for msi_domain_free_irqs() instead kernel/irq/msi.c:31: warning: No description found for return value of 'alloc_msi_entry' kernel/irq/msi.c:103: warning: No description found for return value of 'msi_domain_set_affinity' kernel/irq/msi.c:288: warning: No description found for return value of 'msi_create_irq_domain' kernel/irq/msi.c:499: warning: No description found for return value of 'msi_domain_alloc_irqs' kernel/irq/msi.c:545: warning: No description found for return value of 'msi_get_domain_info' kernel/irq/ipi.c:264: warning: expecting prototype for ipi_send_mask(). Prototype was for __ipi_send_mask() instead kernel/irq/ipi.c:25: warning: No description found for return value of 'irq_reserve_ipi' kernel/irq/ipi.c:116: warning: No description found for return value of 'irq_destroy_ipi' kernel/irq/ipi.c:163: warning: No description found for return value of 'ipi_get_hwirq' kernel/irq/ipi.c:222: warning: No description found for return value of '__ipi_send_single' kernel/irq/ipi.c:308: warning: No description found for return value of 'ipi_send_single' kernel/irq/ipi.c:329: warning: No description found for return value of 'ipi_send_mask' Signed-off-by: Randy Dunlap <rdunlap@infradead.org> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Link: https://lore.kernel.org/r/20210810234835.12547-1-rdunlap@infradead.org
2021-08-10 23:48:35 +00:00
* If @affinity is not %NULL then an affinity array[@nvec] is allocated
genirq/core: Introduce struct irq_affinity_desc The interrupt affinity management uses straight cpumask pointers to convey the automatically assigned affinity masks for managed interrupts. The core interrupt descriptor allocation also decides based on the pointer being non NULL whether an interrupt is managed or not. Devices which use managed interrupts usually have two classes of interrupts: - Interrupts for multiple device queues - Interrupts for general device management Currently both classes are treated the same way, i.e. as managed interrupts. The general interrupts get the default affinity mask assigned while the device queue interrupts are spread out over the possible CPUs. Treating the general interrupts as managed is both a limitation and under certain circumstances a bug. Assume the following situation: default_irq_affinity = 4..7 So if CPUs 4-7 are offlined, then the core code will shut down the device management interrupts because the last CPU in their affinity mask went offline. It's also a limitation because it's desired to allow manual placement of the general device interrupts for various reasons. If they are marked managed then the interrupt affinity setting from both user and kernel space is disabled. To remedy that situation it's required to convey more information than the cpumasks through various interfaces related to interrupt descriptor allocation. Instead of adding yet another argument, create a new data structure 'irq_affinity_desc' which for now just contains the cpumask. This struct can be expanded to convey auxilliary information in the next step. No functional change, just preparatory work. [ tglx: Simplified logic and clarified changelog ] Suggested-by: Thomas Gleixner <tglx@linutronix.de> Suggested-by: Bjorn Helgaas <bhelgaas@google.com> Signed-off-by: Dou Liyang <douliyangs@gmail.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Cc: linux-pci@vger.kernel.org Cc: kashyap.desai@broadcom.com Cc: shivasharan.srikanteshwara@broadcom.com Cc: sumit.saxena@broadcom.com Cc: ming.lei@redhat.com Cc: hch@lst.de Cc: douliyang1@huawei.com Link: https://lkml.kernel.org/r/20181204155122.6327-2-douliyangs@gmail.com
2018-12-04 15:51:20 +00:00
* and the affinity masks and flags from @affinity are copied.
genirq: Fix kernel-doc warnings in pm.c, msi.c and ipi.c Fix all kernel-doc warnings in these 3 files and do some simple editing (capitalize acronyms, capitalize Linux). kernel/irq/pm.c:235: warning: expecting prototype for irq_pm_syscore_ops(). Prototype was for irq_pm_syscore_resume() instead kernel/irq/msi.c:530: warning: expecting prototype for __msi_domain_free_irqs(). Prototype was for msi_domain_free_irqs() instead kernel/irq/msi.c:31: warning: No description found for return value of 'alloc_msi_entry' kernel/irq/msi.c:103: warning: No description found for return value of 'msi_domain_set_affinity' kernel/irq/msi.c:288: warning: No description found for return value of 'msi_create_irq_domain' kernel/irq/msi.c:499: warning: No description found for return value of 'msi_domain_alloc_irqs' kernel/irq/msi.c:545: warning: No description found for return value of 'msi_get_domain_info' kernel/irq/ipi.c:264: warning: expecting prototype for ipi_send_mask(). Prototype was for __ipi_send_mask() instead kernel/irq/ipi.c:25: warning: No description found for return value of 'irq_reserve_ipi' kernel/irq/ipi.c:116: warning: No description found for return value of 'irq_destroy_ipi' kernel/irq/ipi.c:163: warning: No description found for return value of 'ipi_get_hwirq' kernel/irq/ipi.c:222: warning: No description found for return value of '__ipi_send_single' kernel/irq/ipi.c:308: warning: No description found for return value of 'ipi_send_single' kernel/irq/ipi.c:329: warning: No description found for return value of 'ipi_send_mask' Signed-off-by: Randy Dunlap <rdunlap@infradead.org> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Link: https://lore.kernel.org/r/20210810234835.12547-1-rdunlap@infradead.org
2021-08-10 23:48:35 +00:00
*
* Return: pointer to allocated &msi_desc on success or %NULL on failure
*/
static struct msi_desc *msi_alloc_desc(struct device *dev, int nvec,
const struct irq_affinity_desc *affinity)
{
struct msi_desc *desc = kzalloc(sizeof(*desc), GFP_KERNEL);
if (!desc)
return NULL;
desc->dev = dev;
desc->nvec_used = nvec;
if (affinity) {
desc->affinity = kmemdup(affinity, nvec * sizeof(*desc->affinity), GFP_KERNEL);
if (!desc->affinity) {
kfree(desc);
return NULL;
}
}
return desc;
}
static void msi_free_desc(struct msi_desc *desc)
{
kfree(desc->affinity);
kfree(desc);
}
static int msi_insert_desc(struct device *dev, struct msi_desc *desc,
unsigned int domid, unsigned int index)
{
struct msi_device_data *md = dev->msi.data;
struct xarray *xa = &md->__domains[domid].store;
unsigned int hwsize;
int ret;
hwsize = msi_domain_get_hwsize(dev, domid);
genirq/msi: Provide msi_domain_alloc_irq_at() For supporting post MSI-X enable allocations and for the upcoming PCI/IMS support a separate interface is required which allows not only the allocation of a specific index, but also the allocation of any, i.e. the next free index. The latter is especially required for IMS because IMS completely does away with index to functionality mappings which are often found in MSI/MSI-X implementation. But even with MSI-X there are devices where only the first few indices have a fixed functionality and the rest is freely assignable by software, e.g. to queues. msi_domain_alloc_irq_at() is also different from the range based interfaces as it always enforces that the MSI descriptor is allocated by the core code and not preallocated by the caller like the PCI/MSI[-X] enable code path does. msi_domain_alloc_irq_at() can be invoked with the index argument set to MSI_ANY_INDEX which makes the core code pick the next free index. The irq domain can provide a prepare_desc() operation callback in it's msi_domain_ops to do domain specific post allocation initialization before the actual Linux interrupt and the associated interrupt descriptor and hierarchy alloccations are conducted. The function also takes an optional @icookie argument which is of type union msi_instance_cookie. This cookie is not used by the core code and is stored in the allocated msi_desc::data::icookie. The meaning of the cookie is completely implementation defined. In case of IMS this might be a PASID or a pointer to a device queue, but for the MSI core it's opaque and not used in any way. The function returns a struct msi_map which on success contains the allocated index number and the Linux interrupt number so the caller can spare the index to Linux interrupt number lookup. On failure map::index contains the error code and map::virq is 0. Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Reviewed-by: Kevin Tian <kevin.tian@intel.com> Acked-by: Marc Zyngier <maz@kernel.org> Link: https://lore.kernel.org/r/20221124232326.501359457@linutronix.de
2022-11-24 23:26:18 +00:00
if (index == MSI_ANY_INDEX) {
struct xa_limit limit = { .min = 0, .max = hwsize - 1 };
unsigned int index;
genirq/msi: Provide msi_domain_alloc_irq_at() For supporting post MSI-X enable allocations and for the upcoming PCI/IMS support a separate interface is required which allows not only the allocation of a specific index, but also the allocation of any, i.e. the next free index. The latter is especially required for IMS because IMS completely does away with index to functionality mappings which are often found in MSI/MSI-X implementation. But even with MSI-X there are devices where only the first few indices have a fixed functionality and the rest is freely assignable by software, e.g. to queues. msi_domain_alloc_irq_at() is also different from the range based interfaces as it always enforces that the MSI descriptor is allocated by the core code and not preallocated by the caller like the PCI/MSI[-X] enable code path does. msi_domain_alloc_irq_at() can be invoked with the index argument set to MSI_ANY_INDEX which makes the core code pick the next free index. The irq domain can provide a prepare_desc() operation callback in it's msi_domain_ops to do domain specific post allocation initialization before the actual Linux interrupt and the associated interrupt descriptor and hierarchy alloccations are conducted. The function also takes an optional @icookie argument which is of type union msi_instance_cookie. This cookie is not used by the core code and is stored in the allocated msi_desc::data::icookie. The meaning of the cookie is completely implementation defined. In case of IMS this might be a PASID or a pointer to a device queue, but for the MSI core it's opaque and not used in any way. The function returns a struct msi_map which on success contains the allocated index number and the Linux interrupt number so the caller can spare the index to Linux interrupt number lookup. On failure map::index contains the error code and map::virq is 0. Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Reviewed-by: Kevin Tian <kevin.tian@intel.com> Acked-by: Marc Zyngier <maz@kernel.org> Link: https://lore.kernel.org/r/20221124232326.501359457@linutronix.de
2022-11-24 23:26:18 +00:00
/* Let the xarray allocate a free index within the limit */
ret = xa_alloc(xa, &index, desc, limit, GFP_KERNEL);
if (ret)
goto fail;
desc->msi_index = index;
return 0;
} else {
if (index >= hwsize) {
ret = -ERANGE;
goto fail;
}
desc->msi_index = index;
ret = xa_insert(xa, index, desc, GFP_KERNEL);
if (ret)
goto fail;
return 0;
}
fail:
msi_free_desc(desc);
return ret;
}
/**
* msi_domain_insert_msi_desc - Allocate and initialize a MSI descriptor and
* insert it at @init_desc->msi_index
*
* @dev: Pointer to the device for which the descriptor is allocated
* @domid: The id of the interrupt domain to which the desriptor is added
* @init_desc: Pointer to an MSI descriptor to initialize the new descriptor
*
* Return: 0 on success or an appropriate failure code.
*/
int msi_domain_insert_msi_desc(struct device *dev, unsigned int domid,
struct msi_desc *init_desc)
{
struct msi_desc *desc;
lockdep_assert_held(&dev->msi.data->mutex);
desc = msi_alloc_desc(dev, init_desc->nvec_used, init_desc->affinity);
if (!desc)
return -ENOMEM;
/* Copy type specific data to the new descriptor. */
desc->pci = init_desc->pci;
return msi_insert_desc(dev, desc, domid, init_desc->msi_index);
}
static bool msi_desc_match(struct msi_desc *desc, enum msi_desc_filter filter)
{
switch (filter) {
case MSI_DESC_ALL:
return true;
case MSI_DESC_NOTASSOCIATED:
return !desc->irq;
case MSI_DESC_ASSOCIATED:
return !!desc->irq;
}
WARN_ON_ONCE(1);
return false;
}
static bool msi_ctrl_valid(struct device *dev, struct msi_ctrl *ctrl)
{
unsigned int hwsize;
if (WARN_ON_ONCE(ctrl->domid >= MSI_MAX_DEVICE_IRQDOMAINS ||
(dev->msi.domain &&
!dev->msi.data->__domains[ctrl->domid].domain)))
return false;
hwsize = msi_domain_get_hwsize(dev, ctrl->domid);
if (WARN_ON_ONCE(ctrl->first > ctrl->last ||
ctrl->first >= hwsize ||
ctrl->last >= hwsize))
return false;
return true;
}
static void msi_domain_free_descs(struct device *dev, struct msi_ctrl *ctrl)
{
struct msi_desc *desc;
struct xarray *xa;
unsigned long idx;
lockdep_assert_held(&dev->msi.data->mutex);
if (!msi_ctrl_valid(dev, ctrl))
return;
xa = &dev->msi.data->__domains[ctrl->domid].store;
xa_for_each_range(xa, idx, desc, ctrl->first, ctrl->last) {
xa_erase(xa, idx);
/* Leak the descriptor when it is still referenced */
if (WARN_ON_ONCE(msi_desc_match(desc, MSI_DESC_ASSOCIATED)))
continue;
msi_free_desc(desc);
}
}
/**
* msi_domain_free_msi_descs_range - Free a range of MSI descriptors of a device in an irqdomain
* @dev: Device for which to free the descriptors
* @domid: Id of the domain to operate on
* @first: Index to start freeing from (inclusive)
* @last: Last index to be freed (inclusive)
*/
void msi_domain_free_msi_descs_range(struct device *dev, unsigned int domid,
unsigned int first, unsigned int last)
{
struct msi_ctrl ctrl = {
.domid = domid,
.first = first,
.last = last,
};
msi_domain_free_descs(dev, &ctrl);
}
/**
* msi_domain_add_simple_msi_descs - Allocate and initialize MSI descriptors
* @dev: Pointer to the device for which the descriptors are allocated
* @ctrl: Allocation control struct
*
* Return: 0 on success or an appropriate failure code.
*/
static int msi_domain_add_simple_msi_descs(struct device *dev, struct msi_ctrl *ctrl)
{
struct msi_desc *desc;
unsigned int idx;
int ret;
lockdep_assert_held(&dev->msi.data->mutex);
if (!msi_ctrl_valid(dev, ctrl))
return -EINVAL;
for (idx = ctrl->first; idx <= ctrl->last; idx++) {
desc = msi_alloc_desc(dev, 1, NULL);
if (!desc)
goto fail_mem;
ret = msi_insert_desc(dev, desc, ctrl->domid, idx);
if (ret)
goto fail;
}
return 0;
fail_mem:
ret = -ENOMEM;
fail:
msi_domain_free_descs(dev, ctrl);
return ret;
}
void __get_cached_msi_msg(struct msi_desc *entry, struct msi_msg *msg)
{
*msg = entry->msg;
}
void get_cached_msi_msg(unsigned int irq, struct msi_msg *msg)
{
struct msi_desc *entry = irq_get_msi_desc(irq);
__get_cached_msi_msg(entry, msg);
}
EXPORT_SYMBOL_GPL(get_cached_msi_msg);
static void msi_device_data_release(struct device *dev, void *res)
{
struct msi_device_data *md = res;
int i;
for (i = 0; i < MSI_MAX_DEVICE_IRQDOMAINS; i++) {
2022-11-24 23:25:56 +00:00
msi_remove_device_irq_domain(dev, i);
WARN_ON_ONCE(!xa_empty(&md->__domains[i].store));
xa_destroy(&md->__domains[i].store);
}
dev->msi.data = NULL;
}
/**
* msi_setup_device_data - Setup MSI device data
* @dev: Device for which MSI device data should be set up
*
* Return: 0 on success, appropriate error code otherwise
*
* This can be called more than once for @dev. If the MSI device data is
* already allocated the call succeeds. The allocated memory is
* automatically released when the device is destroyed.
*/
int msi_setup_device_data(struct device *dev)
{
struct msi_device_data *md;
int ret, i;
if (dev->msi.data)
return 0;
md = devres_alloc(msi_device_data_release, sizeof(*md), GFP_KERNEL);
if (!md)
return -ENOMEM;
ret = msi_sysfs_create_group(dev);
if (ret) {
devres_free(md);
return ret;
}
for (i = 0; i < MSI_MAX_DEVICE_IRQDOMAINS; i++)
genirq/msi: Provide msi_domain_alloc_irq_at() For supporting post MSI-X enable allocations and for the upcoming PCI/IMS support a separate interface is required which allows not only the allocation of a specific index, but also the allocation of any, i.e. the next free index. The latter is especially required for IMS because IMS completely does away with index to functionality mappings which are often found in MSI/MSI-X implementation. But even with MSI-X there are devices where only the first few indices have a fixed functionality and the rest is freely assignable by software, e.g. to queues. msi_domain_alloc_irq_at() is also different from the range based interfaces as it always enforces that the MSI descriptor is allocated by the core code and not preallocated by the caller like the PCI/MSI[-X] enable code path does. msi_domain_alloc_irq_at() can be invoked with the index argument set to MSI_ANY_INDEX which makes the core code pick the next free index. The irq domain can provide a prepare_desc() operation callback in it's msi_domain_ops to do domain specific post allocation initialization before the actual Linux interrupt and the associated interrupt descriptor and hierarchy alloccations are conducted. The function also takes an optional @icookie argument which is of type union msi_instance_cookie. This cookie is not used by the core code and is stored in the allocated msi_desc::data::icookie. The meaning of the cookie is completely implementation defined. In case of IMS this might be a PASID or a pointer to a device queue, but for the MSI core it's opaque and not used in any way. The function returns a struct msi_map which on success contains the allocated index number and the Linux interrupt number so the caller can spare the index to Linux interrupt number lookup. On failure map::index contains the error code and map::virq is 0. Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Reviewed-by: Kevin Tian <kevin.tian@intel.com> Acked-by: Marc Zyngier <maz@kernel.org> Link: https://lore.kernel.org/r/20221124232326.501359457@linutronix.de
2022-11-24 23:26:18 +00:00
xa_init_flags(&md->__domains[i].store, XA_FLAGS_ALLOC);
genirq/msi: Add pointers for per device irq domains With the upcoming per device MSI interrupt domain support it is necessary to store the domain pointers per device. Instead of delegating that storage to device drivers or subsystems add a domain pointer to the msi_dev_domain array in struct msi_device_data. This pointer is also used to take care of tearing down the irq domains when msi_device_data is cleaned up via devres. The interfaces into the MSI core will be changed from irqdomain pointer based interfaces to domain id based interfaces to support multiple MSI domains on a single device (e.g. PCI/MSI[-X] and PCI/IMS. Once the per device domain support is complete the irq domain pointer in struct device::msi.domain will not longer contain a pointer to the "global" MSI domain. It will contain a pointer to the MSI parent domain instead. It would be a horrible maze of conditionals to evaluate all over the place which domain pointer should be used, i.e. the "global" one in device::msi::domain or one from the internal pointer array. To avoid this evaluate in msi_setup_device_data() whether the irq domain which is associated to a device is a "global" or a parent MSI domain. If it is global then copy the pointer into the first entry of the msi_dev_domain array. This allows to convert interfaces and implementation to domain ids while keeping everything existing working. Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Reviewed-by: Kevin Tian <kevin.tian@intel.com> Acked-by: Marc Zyngier <maz@kernel.org> Link: https://lore.kernel.org/r/20221124230313.923860399@linutronix.de
2022-11-24 23:24:22 +00:00
/*
* If @dev::msi::domain is set and is a global MSI domain, copy the
* pointer into the domain array so all code can operate on domain
* ids. The NULL pointer check is required to keep the legacy
* architecture specific PCI/MSI support working.
*/
if (dev->msi.domain && !irq_domain_is_msi_parent(dev->msi.domain))
md->__domains[MSI_DEFAULT_DOMAIN].domain = dev->msi.domain;
mutex_init(&md->mutex);
dev->msi.data = md;
devres_add(dev, md);
return 0;
}
/**
* msi_lock_descs - Lock the MSI descriptor storage of a device
* @dev: Device to operate on
*/
void msi_lock_descs(struct device *dev)
{
mutex_lock(&dev->msi.data->mutex);
}
EXPORT_SYMBOL_GPL(msi_lock_descs);
/**
* msi_unlock_descs - Unlock the MSI descriptor storage of a device
* @dev: Device to operate on
*/
void msi_unlock_descs(struct device *dev)
{
/* Invalidate the index which was cached by the iterator */
dev->msi.data->__iter_idx = MSI_XA_MAX_INDEX;
mutex_unlock(&dev->msi.data->mutex);
}
EXPORT_SYMBOL_GPL(msi_unlock_descs);
static struct msi_desc *msi_find_desc(struct msi_device_data *md, unsigned int domid,
enum msi_desc_filter filter)
{
struct xarray *xa = &md->__domains[domid].store;
struct msi_desc *desc;
xa_for_each_start(xa, md->__iter_idx, desc, md->__iter_idx) {
if (msi_desc_match(desc, filter))
return desc;
}
md->__iter_idx = MSI_XA_MAX_INDEX;
return NULL;
}
/**
* msi_domain_first_desc - Get the first MSI descriptor of an irqdomain associated to a device
* @dev: Device to operate on
* @domid: The id of the interrupt domain which should be walked.
* @filter: Descriptor state filter
*
* Must be called with the MSI descriptor mutex held, i.e. msi_lock_descs()
* must be invoked before the call.
*
* Return: Pointer to the first MSI descriptor matching the search
* criteria, NULL if none found.
*/
struct msi_desc *msi_domain_first_desc(struct device *dev, unsigned int domid,
enum msi_desc_filter filter)
{
struct msi_device_data *md = dev->msi.data;
if (WARN_ON_ONCE(!md || domid >= MSI_MAX_DEVICE_IRQDOMAINS))
return NULL;
lockdep_assert_held(&md->mutex);
md->__iter_idx = 0;
return msi_find_desc(md, domid, filter);
}
EXPORT_SYMBOL_GPL(msi_domain_first_desc);
/**
* msi_next_desc - Get the next MSI descriptor of a device
* @dev: Device to operate on
* @domid: The id of the interrupt domain which should be walked.
* @filter: Descriptor state filter
*
* The first invocation of msi_next_desc() has to be preceeded by a
* successful invocation of __msi_first_desc(). Consecutive invocations are
* only valid if the previous one was successful. All these operations have
* to be done within the same MSI mutex held region.
*
* Return: Pointer to the next MSI descriptor matching the search
* criteria, NULL if none found.
*/
struct msi_desc *msi_next_desc(struct device *dev, unsigned int domid,
enum msi_desc_filter filter)
{
struct msi_device_data *md = dev->msi.data;
if (WARN_ON_ONCE(!md || domid >= MSI_MAX_DEVICE_IRQDOMAINS))
return NULL;
lockdep_assert_held(&md->mutex);
if (md->__iter_idx >= (unsigned long)MSI_MAX_INDEX)
return NULL;
md->__iter_idx++;
return msi_find_desc(md, domid, filter);
}
EXPORT_SYMBOL_GPL(msi_next_desc);
/**
* msi_domain_get_virq - Lookup the Linux interrupt number for a MSI index on a interrupt domain
* @dev: Device to operate on
* @domid: Domain ID of the interrupt domain associated to the device
* @index: MSI interrupt index to look for (0-based)
*
* Return: The Linux interrupt number on success (> 0), 0 if not found
*/
unsigned int msi_domain_get_virq(struct device *dev, unsigned int domid, unsigned int index)
{
struct msi_desc *desc;
unsigned int ret = 0;
bool pcimsi = false;
struct xarray *xa;
if (!dev->msi.data)
return 0;
if (WARN_ON_ONCE(index > MSI_MAX_INDEX || domid >= MSI_MAX_DEVICE_IRQDOMAINS))
return 0;
/* This check is only valid for the PCI default MSI domain */
if (dev_is_pci(dev) && domid == MSI_DEFAULT_DOMAIN)
pcimsi = to_pci_dev(dev)->msi_enabled;
msi_lock_descs(dev);
xa = &dev->msi.data->__domains[domid].store;
desc = xa_load(xa, pcimsi ? 0 : index);
if (desc && desc->irq) {
/*
* PCI-MSI has only one descriptor for multiple interrupts.
* PCI-MSIX and platform MSI use a descriptor per
* interrupt.
*/
if (pcimsi) {
if (index < desc->nvec_used)
ret = desc->irq + index;
} else {
ret = desc->irq;
}
}
msi_unlock_descs(dev);
return ret;
}
EXPORT_SYMBOL_GPL(msi_domain_get_virq);
#ifdef CONFIG_SYSFS
static struct attribute *msi_dev_attrs[] = {
NULL
};
static const struct attribute_group msi_irqs_group = {
.name = "msi_irqs",
.attrs = msi_dev_attrs,
};
static inline int msi_sysfs_create_group(struct device *dev)
{
return devm_device_add_group(dev, &msi_irqs_group);
}
static ssize_t msi_mode_show(struct device *dev, struct device_attribute *attr,
char *buf)
{
/* MSI vs. MSIX is per device not per interrupt */
bool is_msix = dev_is_pci(dev) ? to_pci_dev(dev)->msix_enabled : false;
return sysfs_emit(buf, "%s\n", is_msix ? "msix" : "msi");
}
static void msi_sysfs_remove_desc(struct device *dev, struct msi_desc *desc)
{
struct device_attribute *attrs = desc->sysfs_attrs;
int i;
if (!attrs)
return;
desc->sysfs_attrs = NULL;
for (i = 0; i < desc->nvec_used; i++) {
if (attrs[i].show)
sysfs_remove_file_from_group(&dev->kobj, &attrs[i].attr, msi_irqs_group.name);
kfree(attrs[i].attr.name);
}
kfree(attrs);
}
static int msi_sysfs_populate_desc(struct device *dev, struct msi_desc *desc)
{
struct device_attribute *attrs;
int ret, i;
attrs = kcalloc(desc->nvec_used, sizeof(*attrs), GFP_KERNEL);
if (!attrs)
return -ENOMEM;
desc->sysfs_attrs = attrs;
for (i = 0; i < desc->nvec_used; i++) {
sysfs_attr_init(&attrs[i].attr);
attrs[i].attr.name = kasprintf(GFP_KERNEL, "%d", desc->irq + i);
if (!attrs[i].attr.name) {
ret = -ENOMEM;
goto fail;
}
attrs[i].attr.mode = 0444;
attrs[i].show = msi_mode_show;
ret = sysfs_add_file_to_group(&dev->kobj, &attrs[i].attr, msi_irqs_group.name);
if (ret) {
attrs[i].show = NULL;
goto fail;
}
}
return 0;
fail:
msi_sysfs_remove_desc(dev, desc);
return ret;
}
#ifdef CONFIG_PCI_MSI_ARCH_FALLBACKS
/**
* msi_device_populate_sysfs - Populate msi_irqs sysfs entries for a device
* @dev: The device (PCI, platform etc) which will get sysfs entries
*/
int msi_device_populate_sysfs(struct device *dev)
{
struct msi_desc *desc;
int ret;
msi_for_each_desc(desc, dev, MSI_DESC_ASSOCIATED) {
if (desc->sysfs_attrs)
continue;
ret = msi_sysfs_populate_desc(dev, desc);
if (ret)
return ret;
}
return 0;
}
/**
* msi_device_destroy_sysfs - Destroy msi_irqs sysfs entries for a device
* @dev: The device (PCI, platform etc) for which to remove
* sysfs entries
*/
void msi_device_destroy_sysfs(struct device *dev)
{
struct msi_desc *desc;
msi_for_each_desc(desc, dev, MSI_DESC_ALL)
msi_sysfs_remove_desc(dev, desc);
}
#endif /* CONFIG_PCI_MSI_ARCH_FALLBACK */
#else /* CONFIG_SYSFS */
static inline int msi_sysfs_create_group(struct device *dev) { return 0; }
static inline int msi_sysfs_populate_desc(struct device *dev, struct msi_desc *desc) { return 0; }
static inline void msi_sysfs_remove_desc(struct device *dev, struct msi_desc *desc) { }
#endif /* !CONFIG_SYSFS */
static struct irq_domain *msi_get_device_domain(struct device *dev, unsigned int domid)
{
struct irq_domain *domain;
lockdep_assert_held(&dev->msi.data->mutex);
if (WARN_ON_ONCE(domid >= MSI_MAX_DEVICE_IRQDOMAINS))
return NULL;
domain = dev->msi.data->__domains[domid].domain;
if (!domain)
return NULL;
if (WARN_ON_ONCE(irq_domain_is_msi_parent(domain)))
return NULL;
return domain;
}
static unsigned int msi_domain_get_hwsize(struct device *dev, unsigned int domid)
{
struct msi_domain_info *info;
struct irq_domain *domain;
domain = msi_get_device_domain(dev, domid);
if (domain) {
info = domain->host_data;
return info->hwsize;
}
/* No domain, default to MSI_XA_DOMAIN_SIZE */
return MSI_XA_DOMAIN_SIZE;
}
static inline void irq_chip_write_msi_msg(struct irq_data *data,
struct msi_msg *msg)
{
data->chip->irq_write_msi_msg(data, msg);
}
static void msi_check_level(struct irq_domain *domain, struct msi_msg *msg)
{
struct msi_domain_info *info = domain->host_data;
/*
* If the MSI provider has messed with the second message and
* not advertized that it is level-capable, signal the breakage.
*/
WARN_ON(!((info->flags & MSI_FLAG_LEVEL_CAPABLE) &&
(info->chip->flags & IRQCHIP_SUPPORTS_LEVEL_MSI)) &&
(msg[1].address_lo || msg[1].address_hi || msg[1].data));
}
/**
* msi_domain_set_affinity - Generic affinity setter function for MSI domains
* @irq_data: The irq data associated to the interrupt
* @mask: The affinity mask to set
* @force: Flag to enforce setting (disable online checks)
*
* Intended to be used by MSI interrupt controllers which are
* implemented with hierarchical domains.
genirq: Fix kernel-doc warnings in pm.c, msi.c and ipi.c Fix all kernel-doc warnings in these 3 files and do some simple editing (capitalize acronyms, capitalize Linux). kernel/irq/pm.c:235: warning: expecting prototype for irq_pm_syscore_ops(). Prototype was for irq_pm_syscore_resume() instead kernel/irq/msi.c:530: warning: expecting prototype for __msi_domain_free_irqs(). Prototype was for msi_domain_free_irqs() instead kernel/irq/msi.c:31: warning: No description found for return value of 'alloc_msi_entry' kernel/irq/msi.c:103: warning: No description found for return value of 'msi_domain_set_affinity' kernel/irq/msi.c:288: warning: No description found for return value of 'msi_create_irq_domain' kernel/irq/msi.c:499: warning: No description found for return value of 'msi_domain_alloc_irqs' kernel/irq/msi.c:545: warning: No description found for return value of 'msi_get_domain_info' kernel/irq/ipi.c:264: warning: expecting prototype for ipi_send_mask(). Prototype was for __ipi_send_mask() instead kernel/irq/ipi.c:25: warning: No description found for return value of 'irq_reserve_ipi' kernel/irq/ipi.c:116: warning: No description found for return value of 'irq_destroy_ipi' kernel/irq/ipi.c:163: warning: No description found for return value of 'ipi_get_hwirq' kernel/irq/ipi.c:222: warning: No description found for return value of '__ipi_send_single' kernel/irq/ipi.c:308: warning: No description found for return value of 'ipi_send_single' kernel/irq/ipi.c:329: warning: No description found for return value of 'ipi_send_mask' Signed-off-by: Randy Dunlap <rdunlap@infradead.org> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Link: https://lore.kernel.org/r/20210810234835.12547-1-rdunlap@infradead.org
2021-08-10 23:48:35 +00:00
*
* Return: IRQ_SET_MASK_* result code
*/
int msi_domain_set_affinity(struct irq_data *irq_data,
const struct cpumask *mask, bool force)
{
struct irq_data *parent = irq_data->parent_data;
struct msi_msg msg[2] = { [1] = { }, };
int ret;
ret = parent->chip->irq_set_affinity(parent, mask, force);
if (ret >= 0 && ret != IRQ_SET_MASK_OK_DONE) {
BUG_ON(irq_chip_compose_msi_msg(irq_data, msg));
msi_check_level(irq_data->domain, msg);
irq_chip_write_msi_msg(irq_data, msg);
}
return ret;
}
static int msi_domain_activate(struct irq_domain *domain,
struct irq_data *irq_data, bool early)
{
struct msi_msg msg[2] = { [1] = { }, };
BUG_ON(irq_chip_compose_msi_msg(irq_data, msg));
msi_check_level(irq_data->domain, msg);
irq_chip_write_msi_msg(irq_data, msg);
return 0;
}
static void msi_domain_deactivate(struct irq_domain *domain,
struct irq_data *irq_data)
{
struct msi_msg msg[2];
memset(msg, 0, sizeof(msg));
irq_chip_write_msi_msg(irq_data, msg);
}
static int msi_domain_alloc(struct irq_domain *domain, unsigned int virq,
unsigned int nr_irqs, void *arg)
{
struct msi_domain_info *info = domain->host_data;
struct msi_domain_ops *ops = info->ops;
irq_hw_number_t hwirq = ops->get_hwirq(info, arg);
int i, ret;
if (irq_find_mapping(domain, hwirq) > 0)
return -EEXIST;
if (domain->parent) {
ret = irq_domain_alloc_irqs_parent(domain, virq, nr_irqs, arg);
if (ret < 0)
return ret;
}
for (i = 0; i < nr_irqs; i++) {
ret = ops->msi_init(domain, info, virq + i, hwirq + i, arg);
if (ret < 0) {
if (ops->msi_free) {
for (i--; i > 0; i--)
ops->msi_free(domain, info, virq + i);
}
irq_domain_free_irqs_top(domain, virq, nr_irqs);
return ret;
}
}
return 0;
}
static void msi_domain_free(struct irq_domain *domain, unsigned int virq,
unsigned int nr_irqs)
{
struct msi_domain_info *info = domain->host_data;
int i;
if (info->ops->msi_free) {
for (i = 0; i < nr_irqs; i++)
info->ops->msi_free(domain, info, virq + i);
}
irq_domain_free_irqs_top(domain, virq, nr_irqs);
}
static const struct irq_domain_ops msi_domain_ops = {
.alloc = msi_domain_alloc,
.free = msi_domain_free,
.activate = msi_domain_activate,
.deactivate = msi_domain_deactivate,
};
static irq_hw_number_t msi_domain_ops_get_hwirq(struct msi_domain_info *info,
msi_alloc_info_t *arg)
{
return arg->hwirq;
}
static int msi_domain_ops_prepare(struct irq_domain *domain, struct device *dev,
int nvec, msi_alloc_info_t *arg)
{
memset(arg, 0, sizeof(*arg));
return 0;
}
static void msi_domain_ops_set_desc(msi_alloc_info_t *arg,
struct msi_desc *desc)
{
arg->desc = desc;
}
static int msi_domain_ops_init(struct irq_domain *domain,
struct msi_domain_info *info,
unsigned int virq, irq_hw_number_t hwirq,
msi_alloc_info_t *arg)
{
irq_domain_set_hwirq_and_chip(domain, virq, hwirq, info->chip,
info->chip_data);
if (info->handler && info->handler_name) {
__irq_set_handler(virq, info->handler, 0, info->handler_name);
if (info->handler_data)
irq_set_handler_data(virq, info->handler_data);
}
return 0;
}
static struct msi_domain_ops msi_domain_ops_default = {
.get_hwirq = msi_domain_ops_get_hwirq,
.msi_init = msi_domain_ops_init,
.msi_prepare = msi_domain_ops_prepare,
.set_desc = msi_domain_ops_set_desc,
};
static void msi_domain_update_dom_ops(struct msi_domain_info *info)
{
struct msi_domain_ops *ops = info->ops;
if (ops == NULL) {
info->ops = &msi_domain_ops_default;
return;
}
if (!(info->flags & MSI_FLAG_USE_DEF_DOM_OPS))
return;
if (ops->get_hwirq == NULL)
ops->get_hwirq = msi_domain_ops_default.get_hwirq;
if (ops->msi_init == NULL)
ops->msi_init = msi_domain_ops_default.msi_init;
if (ops->msi_prepare == NULL)
ops->msi_prepare = msi_domain_ops_default.msi_prepare;
if (ops->set_desc == NULL)
ops->set_desc = msi_domain_ops_default.set_desc;
}
static void msi_domain_update_chip_ops(struct msi_domain_info *info)
{
struct irq_chip *chip = info->chip;
BUG_ON(!chip || !chip->irq_mask || !chip->irq_unmask);
if (!chip->irq_set_affinity)
chip->irq_set_affinity = msi_domain_set_affinity;
}
static struct irq_domain *__msi_create_irq_domain(struct fwnode_handle *fwnode,
struct msi_domain_info *info,
unsigned int flags,
struct irq_domain *parent)
{
struct irq_domain *domain;
if (info->hwsize > MSI_XA_DOMAIN_SIZE)
return NULL;
/*
* Hardware size 0 is valid for backwards compatibility and for
* domains which are not backed by a hardware table. Grant the
* maximum index space.
*/
if (!info->hwsize)
info->hwsize = MSI_XA_DOMAIN_SIZE;
msi_domain_update_dom_ops(info);
if (info->flags & MSI_FLAG_USE_DEF_CHIP_OPS)
msi_domain_update_chip_ops(info);
domain = irq_domain_create_hierarchy(parent, flags | IRQ_DOMAIN_FLAG_MSI, 0,
fwnode, &msi_domain_ops, info);
if (domain) {
if (!domain->name && info->chip)
domain->name = info->chip->name;
irq_domain_update_bus_token(domain, info->bus_token);
}
return domain;
}
/**
* msi_create_irq_domain - Create an MSI interrupt domain
* @fwnode: Optional fwnode of the interrupt controller
* @info: MSI domain info
* @parent: Parent irq domain
*
* Return: pointer to the created &struct irq_domain or %NULL on failure
*/
struct irq_domain *msi_create_irq_domain(struct fwnode_handle *fwnode,
struct msi_domain_info *info,
struct irq_domain *parent)
{
return __msi_create_irq_domain(fwnode, info, 0, parent);
}
genirq/msi: Provide struct msi_parent_ops MSI parent domains must have some control over the MSI domains which are built on top. On domain creation they need to fill in e.g. architecture specific chip callbacks or msi domain ops to make the outermost domain parent agnostic which is obviously required for architecture independence etc. The structure contains: 1) A bitfield which exposes the supported functional features. This allows to check for features and is also used in the initialization callback to mask out unsupported features when the actual domain implementation requests a broader range, e.g. on x86 PCI multi-MSI is only supported by remapping domains but not by the underlying vector domain. The PCI/MSI code can then always request multi-MSI support, but the resulting feature set after creation might not have it set. 2) An optional string prefix which is put in front of domain and chip names during creation of the MSI domain. That allows to keep the naming schemes e.g. on x86 where PCI-MSI domains have a IR- prefix when interrupt remapping is enabled. 3) An initialization callback to sanity check the domain info of the to be created MSI domain, to restrict features and to apply changes in MSI ops and interrupt chip callbacks to accomodate to the particular MSI parent implementation and/or the underlying hierarchy. Add a conveniance function to delegate the initialization from the MSI parent domain to an underlying domain in the hierarchy. Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Reviewed-by: Kevin Tian <kevin.tian@intel.com> Acked-by: Marc Zyngier <maz@kernel.org> Link: https://lore.kernel.org/r/20221124232325.382485843@linutronix.de
2022-11-24 23:25:48 +00:00
/**
* msi_parent_init_dev_msi_info - Delegate initialization of device MSI info down
* in the domain hierarchy
* @dev: The device for which the domain should be created
* @domain: The domain in the hierarchy this op is being called on
* @msi_parent_domain: The IRQ_DOMAIN_FLAG_MSI_PARENT domain for the child to
* be created
* @msi_child_info: The MSI domain info of the IRQ_DOMAIN_FLAG_MSI_DEVICE
* domain to be created
*
* Return: true on success, false otherwise
*
* This is the most complex problem of per device MSI domains and the
* underlying interrupt domain hierarchy:
*
* The device domain to be initialized requests the broadest feature set
* possible and the underlying domain hierarchy puts restrictions on it.
*
* That's trivial for a simple parent->child relationship, but it gets
* interesting with an intermediate domain: root->parent->child. The
* intermediate 'parent' can expand the capabilities which the 'root'
* domain is providing. So that creates a classic hen and egg problem:
* Which entity is doing the restrictions/expansions?
*
* One solution is to let the root domain handle the initialization that's
* why there is the @domain and the @msi_parent_domain pointer.
*/
bool msi_parent_init_dev_msi_info(struct device *dev, struct irq_domain *domain,
struct irq_domain *msi_parent_domain,
struct msi_domain_info *msi_child_info)
{
struct irq_domain *parent = domain->parent;
if (WARN_ON_ONCE(!parent || !parent->msi_parent_ops ||
!parent->msi_parent_ops->init_dev_msi_info))
return false;
return parent->msi_parent_ops->init_dev_msi_info(dev, parent, msi_parent_domain,
msi_child_info);
}
2022-11-24 23:25:56 +00:00
/**
* msi_create_device_irq_domain - Create a device MSI interrupt domain
* @dev: Pointer to the device
* @domid: Domain id
* @template: MSI domain info bundle used as template
* @hwsize: Maximum number of MSI table entries (0 if unknown or unlimited)
* @domain_data: Optional pointer to domain specific data which is set in
* msi_domain_info::data
* @chip_data: Optional pointer to chip specific data which is set in
* msi_domain_info::chip_data
*
* Return: True on success, false otherwise
*
* There is no firmware node required for this interface because the per
* device domains are software constructs which are actually closer to the
* hardware reality than any firmware can describe them.
*
* The domain name and the irq chip name for a MSI device domain are
* composed by: "$(PREFIX)$(CHIPNAME)-$(DEVNAME)"
*
* $PREFIX: Optional prefix provided by the underlying MSI parent domain
* via msi_parent_ops::prefix. If that pointer is NULL the prefix
* is empty.
* $CHIPNAME: The name of the irq_chip in @template
* $DEVNAME: The name of the device
*
* This results in understandable chip names and hardware interrupt numbers
* in e.g. /proc/interrupts
*
* PCI-MSI-0000:00:1c.0 0-edge Parent domain has no prefix
* IR-PCI-MSI-0000:00:1c.4 0-edge Same with interrupt remapping prefix 'IR-'
*
* IR-PCI-MSIX-0000:3d:00.0 0-edge Hardware interrupt numbers reflect
* IR-PCI-MSIX-0000:3d:00.0 1-edge the real MSI-X index on that device
* IR-PCI-MSIX-0000:3d:00.0 2-edge
*
* On IMS domains the hardware interrupt number is either a table entry
* index or a purely software managed index but it is guaranteed to be
* unique.
*
* The domain pointer is stored in @dev::msi::data::__irqdomains[]. All
* subsequent operations on the domain depend on the domain id.
*
* The domain is automatically freed when the device is removed via devres
* in the context of @dev::msi::data freeing, but it can also be
* independently removed via @msi_remove_device_irq_domain().
*/
bool msi_create_device_irq_domain(struct device *dev, unsigned int domid,
const struct msi_domain_template *template,
unsigned int hwsize, void *domain_data,
void *chip_data)
{
struct irq_domain *domain, *parent = dev->msi.domain;
const struct msi_parent_ops *pops;
struct msi_domain_template *bundle;
struct fwnode_handle *fwnode;
if (!irq_domain_is_msi_parent(parent))
return false;
if (domid >= MSI_MAX_DEVICE_IRQDOMAINS)
return false;
bundle = kmemdup(template, sizeof(*bundle), GFP_KERNEL);
if (!bundle)
return false;
bundle->info.hwsize = hwsize;
bundle->info.chip = &bundle->chip;
bundle->info.ops = &bundle->ops;
bundle->info.data = domain_data;
bundle->info.chip_data = chip_data;
pops = parent->msi_parent_ops;
snprintf(bundle->name, sizeof(bundle->name), "%s%s-%s",
pops->prefix ? : "", bundle->chip.name, dev_name(dev));
bundle->chip.name = bundle->name;
fwnode = irq_domain_alloc_named_fwnode(bundle->name);
if (!fwnode)
goto free_bundle;
if (msi_setup_device_data(dev))
goto free_fwnode;
msi_lock_descs(dev);
if (WARN_ON_ONCE(msi_get_device_domain(dev, domid)))
goto fail;
if (!pops->init_dev_msi_info(dev, parent, parent, &bundle->info))
goto fail;
domain = __msi_create_irq_domain(fwnode, &bundle->info, IRQ_DOMAIN_FLAG_MSI_DEVICE, parent);
if (!domain)
goto fail;
domain->dev = dev;
dev->msi.data->__domains[domid].domain = domain;
msi_unlock_descs(dev);
return true;
fail:
msi_unlock_descs(dev);
free_fwnode:
kfree(fwnode);
free_bundle:
kfree(bundle);
return false;
}
/**
* msi_remove_device_irq_domain - Free a device MSI interrupt domain
* @dev: Pointer to the device
* @domid: Domain id
*/
void msi_remove_device_irq_domain(struct device *dev, unsigned int domid)
{
struct msi_domain_info *info;
struct irq_domain *domain;
msi_lock_descs(dev);
domain = msi_get_device_domain(dev, domid);
if (!domain || !irq_domain_is_msi_device(domain))
goto unlock;
dev->msi.data->__domains[domid].domain = NULL;
info = domain->host_data;
irq_domain_remove(domain);
kfree(container_of(info, struct msi_domain_template, info));
unlock:
msi_unlock_descs(dev);
}
/**
* msi_match_device_irq_domain - Match a device irq domain against a bus token
* @dev: Pointer to the device
* @domid: Domain id
* @bus_token: Bus token to match against the domain bus token
*
* Return: True if device domain exists and bus tokens match.
*/
bool msi_match_device_irq_domain(struct device *dev, unsigned int domid,
enum irq_domain_bus_token bus_token)
{
struct msi_domain_info *info;
struct irq_domain *domain;
bool ret = false;
msi_lock_descs(dev);
domain = msi_get_device_domain(dev, domid);
if (domain && irq_domain_is_msi_device(domain)) {
info = domain->host_data;
ret = info->bus_token == bus_token;
}
msi_unlock_descs(dev);
return ret;
}
int msi_domain_prepare_irqs(struct irq_domain *domain, struct device *dev,
int nvec, msi_alloc_info_t *arg)
{
struct msi_domain_info *info = domain->host_data;
struct msi_domain_ops *ops = info->ops;
return ops->msi_prepare(domain, dev, nvec, arg);
}
int msi_domain_populate_irqs(struct irq_domain *domain, struct device *dev,
int virq_base, int nvec, msi_alloc_info_t *arg)
{
struct msi_domain_info *info = domain->host_data;
struct msi_domain_ops *ops = info->ops;
struct msi_ctrl ctrl = {
.domid = MSI_DEFAULT_DOMAIN,
.first = virq_base,
.last = virq_base + nvec - 1,
};
struct msi_desc *desc;
struct xarray *xa;
int ret, virq;
if (!msi_ctrl_valid(dev, &ctrl))
return -EINVAL;
msi_lock_descs(dev);
ret = msi_domain_add_simple_msi_descs(dev, &ctrl);
if (ret)
goto unlock;
xa = &dev->msi.data->__domains[ctrl.domid].store;
for (virq = virq_base; virq < virq_base + nvec; virq++) {
desc = xa_load(xa, virq);
desc->irq = virq;
ops->set_desc(arg, desc);
ret = irq_domain_alloc_irqs_hierarchy(domain, virq, 1, arg);
if (ret)
goto fail;
irq_set_msi_desc(virq, desc);
}
msi_unlock_descs(dev);
return 0;
fail:
for (--virq; virq >= virq_base; virq--)
irq_domain_free_irqs_common(domain, virq, 1);
msi_domain_free_descs(dev, &ctrl);
unlock:
msi_unlock_descs(dev);
return ret;
}
genirq/msi, x86/vector: Prevent reservation mode for non maskable MSI The new reservation mode for interrupts assigns a dummy vector when the interrupt is allocated and assigns a real vector when the interrupt is requested. The reservation mode prevents vector pressure when devices with a large amount of queues/interrupts are initialized, but only a minimal subset of those queues/interrupts is actually used. This mode has an issue with MSI interrupts which cannot be masked. If the driver is not careful or the hardware emits an interrupt before the device irq is requestd by the driver then the interrupt ends up on the dummy vector as a spurious interrupt which can cause malfunction of the device or in the worst case a lockup of the machine. Change the logic for the reservation mode so that the early activation of MSI interrupts checks whether: - the device is a PCI/MSI device - the reservation mode of the underlying irqdomain is activated - PCI/MSI masking is globally enabled - the PCI/MSI device uses either MSI-X, which supports masking, or MSI with the maskbit supported. If one of those conditions is false, then clear the reservation mode flag in the irq data of the interrupt and invoke irq_domain_activate_irq() with the reserve argument cleared. In the x86 vector code, clear the can_reserve flag in the vector allocation data so a subsequent free_irq() won't create the same situation again. The interrupt stays assigned to a real vector until pci_disable_msi() is invoked and all allocations are undone. Fixes: 4900be83602b ("x86/vector/msi: Switch to global reservation mode") Reported-by: Alexandru Chirvasitu <achirvasub@gmail.com> Reported-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Tested-by: Alexandru Chirvasitu <achirvasub@gmail.com> Tested-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Cc: Dou Liyang <douly.fnst@cn.fujitsu.com> Cc: Pavel Machek <pavel@ucw.cz> Cc: Maciej W. Rozycki <macro@linux-mips.org> Cc: Mikael Pettersson <mikpelinux@gmail.com> Cc: Josh Poulson <jopoulso@microsoft.com> Cc: Mihai Costache <v-micos@microsoft.com> Cc: Stephen Hemminger <sthemmin@microsoft.com> Cc: Marc Zyngier <marc.zyngier@arm.com> Cc: linux-pci@vger.kernel.org Cc: Haiyang Zhang <haiyangz@microsoft.com> Cc: Dexuan Cui <decui@microsoft.com> Cc: Simon Xiao <sixiao@microsoft.com> Cc: Saeed Mahameed <saeedm@mellanox.com> Cc: Jork Loeser <Jork.Loeser@microsoft.com> Cc: Bjorn Helgaas <bhelgaas@google.com> Cc: devel@linuxdriverproject.org Cc: KY Srinivasan <kys@microsoft.com> Cc: Alan Cox <alan@linux.intel.com> Cc: Sakari Ailus <sakari.ailus@intel.com>, Cc: linux-media@vger.kernel.org Link: https://lkml.kernel.org/r/alpine.DEB.2.20.1712291406420.1899@nanos Link: https://lkml.kernel.org/r/alpine.DEB.2.20.1712291409460.1899@nanos
2017-12-29 09:47:22 +00:00
/*
* Carefully check whether the device can use reservation mode. If
* reservation mode is enabled then the early activation will assign a
* dummy vector to the device. If the PCI/MSI device does not support
* masking of the entry then this can result in spurious interrupts when
* the device driver is not absolutely careful. But even then a malfunction
* of the hardware could result in a spurious interrupt on the dummy vector
* and render the device unusable. If the entry can be masked then the core
* logic will prevent the spurious interrupt and reservation mode can be
* used. For now reservation mode is restricted to PCI/MSI.
*/
static bool msi_check_reservation_mode(struct irq_domain *domain,
struct msi_domain_info *info,
struct device *dev)
genirq/msi: Handle reactivation only on success When analyzing the fallout of the x86 vector allocation rework it turned out that the error handling in msi_domain_alloc_irqs() is broken. If MSI_FLAG_MUST_REACTIVATE is set for a MSI domain then it clears the activation flag for a successfully initialized msi descriptor. If a subsequent initialization fails then the error handling code path does not deactivate the interrupt because the activation flag got cleared. Move the clearing of the activation flag outside of the initialization loop so that an eventual failure can be cleaned up correctly. Fixes: 22d0b12f3560 ("genirq/irqdomain: Add force reactivation flag to irq domains") Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Tested-by: Alexandru Chirvasitu <achirvasub@gmail.com> Tested-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Cc: Dou Liyang <douly.fnst@cn.fujitsu.com> Cc: Pavel Machek <pavel@ucw.cz> Cc: Maciej W. Rozycki <macro@linux-mips.org> Cc: Mikael Pettersson <mikpelinux@gmail.com> Cc: Josh Poulson <jopoulso@microsoft.com> Cc: Mihai Costache <v-micos@microsoft.com> Cc: Stephen Hemminger <sthemmin@microsoft.com> Cc: Marc Zyngier <marc.zyngier@arm.com> Cc: linux-pci@vger.kernel.org Cc: Haiyang Zhang <haiyangz@microsoft.com> Cc: Dexuan Cui <decui@microsoft.com> Cc: Simon Xiao <sixiao@microsoft.com> Cc: Saeed Mahameed <saeedm@mellanox.com> Cc: Jork Loeser <Jork.Loeser@microsoft.com> Cc: Bjorn Helgaas <bhelgaas@google.com> Cc: devel@linuxdriverproject.org Cc: KY Srinivasan <kys@microsoft.com> Cc: Alan Cox <alan@linux.intel.com> Cc: Sakari Ailus <sakari.ailus@intel.com>, Cc: linux-media@vger.kernel.org
2017-12-29 09:42:10 +00:00
{
genirq/msi, x86/vector: Prevent reservation mode for non maskable MSI The new reservation mode for interrupts assigns a dummy vector when the interrupt is allocated and assigns a real vector when the interrupt is requested. The reservation mode prevents vector pressure when devices with a large amount of queues/interrupts are initialized, but only a minimal subset of those queues/interrupts is actually used. This mode has an issue with MSI interrupts which cannot be masked. If the driver is not careful or the hardware emits an interrupt before the device irq is requestd by the driver then the interrupt ends up on the dummy vector as a spurious interrupt which can cause malfunction of the device or in the worst case a lockup of the machine. Change the logic for the reservation mode so that the early activation of MSI interrupts checks whether: - the device is a PCI/MSI device - the reservation mode of the underlying irqdomain is activated - PCI/MSI masking is globally enabled - the PCI/MSI device uses either MSI-X, which supports masking, or MSI with the maskbit supported. If one of those conditions is false, then clear the reservation mode flag in the irq data of the interrupt and invoke irq_domain_activate_irq() with the reserve argument cleared. In the x86 vector code, clear the can_reserve flag in the vector allocation data so a subsequent free_irq() won't create the same situation again. The interrupt stays assigned to a real vector until pci_disable_msi() is invoked and all allocations are undone. Fixes: 4900be83602b ("x86/vector/msi: Switch to global reservation mode") Reported-by: Alexandru Chirvasitu <achirvasub@gmail.com> Reported-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Tested-by: Alexandru Chirvasitu <achirvasub@gmail.com> Tested-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Cc: Dou Liyang <douly.fnst@cn.fujitsu.com> Cc: Pavel Machek <pavel@ucw.cz> Cc: Maciej W. Rozycki <macro@linux-mips.org> Cc: Mikael Pettersson <mikpelinux@gmail.com> Cc: Josh Poulson <jopoulso@microsoft.com> Cc: Mihai Costache <v-micos@microsoft.com> Cc: Stephen Hemminger <sthemmin@microsoft.com> Cc: Marc Zyngier <marc.zyngier@arm.com> Cc: linux-pci@vger.kernel.org Cc: Haiyang Zhang <haiyangz@microsoft.com> Cc: Dexuan Cui <decui@microsoft.com> Cc: Simon Xiao <sixiao@microsoft.com> Cc: Saeed Mahameed <saeedm@mellanox.com> Cc: Jork Loeser <Jork.Loeser@microsoft.com> Cc: Bjorn Helgaas <bhelgaas@google.com> Cc: devel@linuxdriverproject.org Cc: KY Srinivasan <kys@microsoft.com> Cc: Alan Cox <alan@linux.intel.com> Cc: Sakari Ailus <sakari.ailus@intel.com>, Cc: linux-media@vger.kernel.org Link: https://lkml.kernel.org/r/alpine.DEB.2.20.1712291406420.1899@nanos Link: https://lkml.kernel.org/r/alpine.DEB.2.20.1712291409460.1899@nanos
2017-12-29 09:47:22 +00:00
struct msi_desc *desc;
switch(domain->bus_token) {
case DOMAIN_BUS_PCI_MSI:
case DOMAIN_BUS_PCI_DEVICE_MSI:
case DOMAIN_BUS_PCI_DEVICE_MSIX:
case DOMAIN_BUS_VMD_MSI:
break;
default:
genirq/msi, x86/vector: Prevent reservation mode for non maskable MSI The new reservation mode for interrupts assigns a dummy vector when the interrupt is allocated and assigns a real vector when the interrupt is requested. The reservation mode prevents vector pressure when devices with a large amount of queues/interrupts are initialized, but only a minimal subset of those queues/interrupts is actually used. This mode has an issue with MSI interrupts which cannot be masked. If the driver is not careful or the hardware emits an interrupt before the device irq is requestd by the driver then the interrupt ends up on the dummy vector as a spurious interrupt which can cause malfunction of the device or in the worst case a lockup of the machine. Change the logic for the reservation mode so that the early activation of MSI interrupts checks whether: - the device is a PCI/MSI device - the reservation mode of the underlying irqdomain is activated - PCI/MSI masking is globally enabled - the PCI/MSI device uses either MSI-X, which supports masking, or MSI with the maskbit supported. If one of those conditions is false, then clear the reservation mode flag in the irq data of the interrupt and invoke irq_domain_activate_irq() with the reserve argument cleared. In the x86 vector code, clear the can_reserve flag in the vector allocation data so a subsequent free_irq() won't create the same situation again. The interrupt stays assigned to a real vector until pci_disable_msi() is invoked and all allocations are undone. Fixes: 4900be83602b ("x86/vector/msi: Switch to global reservation mode") Reported-by: Alexandru Chirvasitu <achirvasub@gmail.com> Reported-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Tested-by: Alexandru Chirvasitu <achirvasub@gmail.com> Tested-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Cc: Dou Liyang <douly.fnst@cn.fujitsu.com> Cc: Pavel Machek <pavel@ucw.cz> Cc: Maciej W. Rozycki <macro@linux-mips.org> Cc: Mikael Pettersson <mikpelinux@gmail.com> Cc: Josh Poulson <jopoulso@microsoft.com> Cc: Mihai Costache <v-micos@microsoft.com> Cc: Stephen Hemminger <sthemmin@microsoft.com> Cc: Marc Zyngier <marc.zyngier@arm.com> Cc: linux-pci@vger.kernel.org Cc: Haiyang Zhang <haiyangz@microsoft.com> Cc: Dexuan Cui <decui@microsoft.com> Cc: Simon Xiao <sixiao@microsoft.com> Cc: Saeed Mahameed <saeedm@mellanox.com> Cc: Jork Loeser <Jork.Loeser@microsoft.com> Cc: Bjorn Helgaas <bhelgaas@google.com> Cc: devel@linuxdriverproject.org Cc: KY Srinivasan <kys@microsoft.com> Cc: Alan Cox <alan@linux.intel.com> Cc: Sakari Ailus <sakari.ailus@intel.com>, Cc: linux-media@vger.kernel.org Link: https://lkml.kernel.org/r/alpine.DEB.2.20.1712291406420.1899@nanos Link: https://lkml.kernel.org/r/alpine.DEB.2.20.1712291409460.1899@nanos
2017-12-29 09:47:22 +00:00
return false;
}
genirq/msi, x86/vector: Prevent reservation mode for non maskable MSI The new reservation mode for interrupts assigns a dummy vector when the interrupt is allocated and assigns a real vector when the interrupt is requested. The reservation mode prevents vector pressure when devices with a large amount of queues/interrupts are initialized, but only a minimal subset of those queues/interrupts is actually used. This mode has an issue with MSI interrupts which cannot be masked. If the driver is not careful or the hardware emits an interrupt before the device irq is requestd by the driver then the interrupt ends up on the dummy vector as a spurious interrupt which can cause malfunction of the device or in the worst case a lockup of the machine. Change the logic for the reservation mode so that the early activation of MSI interrupts checks whether: - the device is a PCI/MSI device - the reservation mode of the underlying irqdomain is activated - PCI/MSI masking is globally enabled - the PCI/MSI device uses either MSI-X, which supports masking, or MSI with the maskbit supported. If one of those conditions is false, then clear the reservation mode flag in the irq data of the interrupt and invoke irq_domain_activate_irq() with the reserve argument cleared. In the x86 vector code, clear the can_reserve flag in the vector allocation data so a subsequent free_irq() won't create the same situation again. The interrupt stays assigned to a real vector until pci_disable_msi() is invoked and all allocations are undone. Fixes: 4900be83602b ("x86/vector/msi: Switch to global reservation mode") Reported-by: Alexandru Chirvasitu <achirvasub@gmail.com> Reported-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Tested-by: Alexandru Chirvasitu <achirvasub@gmail.com> Tested-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Cc: Dou Liyang <douly.fnst@cn.fujitsu.com> Cc: Pavel Machek <pavel@ucw.cz> Cc: Maciej W. Rozycki <macro@linux-mips.org> Cc: Mikael Pettersson <mikpelinux@gmail.com> Cc: Josh Poulson <jopoulso@microsoft.com> Cc: Mihai Costache <v-micos@microsoft.com> Cc: Stephen Hemminger <sthemmin@microsoft.com> Cc: Marc Zyngier <marc.zyngier@arm.com> Cc: linux-pci@vger.kernel.org Cc: Haiyang Zhang <haiyangz@microsoft.com> Cc: Dexuan Cui <decui@microsoft.com> Cc: Simon Xiao <sixiao@microsoft.com> Cc: Saeed Mahameed <saeedm@mellanox.com> Cc: Jork Loeser <Jork.Loeser@microsoft.com> Cc: Bjorn Helgaas <bhelgaas@google.com> Cc: devel@linuxdriverproject.org Cc: KY Srinivasan <kys@microsoft.com> Cc: Alan Cox <alan@linux.intel.com> Cc: Sakari Ailus <sakari.ailus@intel.com>, Cc: linux-media@vger.kernel.org Link: https://lkml.kernel.org/r/alpine.DEB.2.20.1712291406420.1899@nanos Link: https://lkml.kernel.org/r/alpine.DEB.2.20.1712291409460.1899@nanos
2017-12-29 09:47:22 +00:00
genirq/msi: Handle reactivation only on success When analyzing the fallout of the x86 vector allocation rework it turned out that the error handling in msi_domain_alloc_irqs() is broken. If MSI_FLAG_MUST_REACTIVATE is set for a MSI domain then it clears the activation flag for a successfully initialized msi descriptor. If a subsequent initialization fails then the error handling code path does not deactivate the interrupt because the activation flag got cleared. Move the clearing of the activation flag outside of the initialization loop so that an eventual failure can be cleaned up correctly. Fixes: 22d0b12f3560 ("genirq/irqdomain: Add force reactivation flag to irq domains") Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Tested-by: Alexandru Chirvasitu <achirvasub@gmail.com> Tested-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Cc: Dou Liyang <douly.fnst@cn.fujitsu.com> Cc: Pavel Machek <pavel@ucw.cz> Cc: Maciej W. Rozycki <macro@linux-mips.org> Cc: Mikael Pettersson <mikpelinux@gmail.com> Cc: Josh Poulson <jopoulso@microsoft.com> Cc: Mihai Costache <v-micos@microsoft.com> Cc: Stephen Hemminger <sthemmin@microsoft.com> Cc: Marc Zyngier <marc.zyngier@arm.com> Cc: linux-pci@vger.kernel.org Cc: Haiyang Zhang <haiyangz@microsoft.com> Cc: Dexuan Cui <decui@microsoft.com> Cc: Simon Xiao <sixiao@microsoft.com> Cc: Saeed Mahameed <saeedm@mellanox.com> Cc: Jork Loeser <Jork.Loeser@microsoft.com> Cc: Bjorn Helgaas <bhelgaas@google.com> Cc: devel@linuxdriverproject.org Cc: KY Srinivasan <kys@microsoft.com> Cc: Alan Cox <alan@linux.intel.com> Cc: Sakari Ailus <sakari.ailus@intel.com>, Cc: linux-media@vger.kernel.org
2017-12-29 09:42:10 +00:00
if (!(info->flags & MSI_FLAG_MUST_REACTIVATE))
return false;
genirq/msi, x86/vector: Prevent reservation mode for non maskable MSI The new reservation mode for interrupts assigns a dummy vector when the interrupt is allocated and assigns a real vector when the interrupt is requested. The reservation mode prevents vector pressure when devices with a large amount of queues/interrupts are initialized, but only a minimal subset of those queues/interrupts is actually used. This mode has an issue with MSI interrupts which cannot be masked. If the driver is not careful or the hardware emits an interrupt before the device irq is requestd by the driver then the interrupt ends up on the dummy vector as a spurious interrupt which can cause malfunction of the device or in the worst case a lockup of the machine. Change the logic for the reservation mode so that the early activation of MSI interrupts checks whether: - the device is a PCI/MSI device - the reservation mode of the underlying irqdomain is activated - PCI/MSI masking is globally enabled - the PCI/MSI device uses either MSI-X, which supports masking, or MSI with the maskbit supported. If one of those conditions is false, then clear the reservation mode flag in the irq data of the interrupt and invoke irq_domain_activate_irq() with the reserve argument cleared. In the x86 vector code, clear the can_reserve flag in the vector allocation data so a subsequent free_irq() won't create the same situation again. The interrupt stays assigned to a real vector until pci_disable_msi() is invoked and all allocations are undone. Fixes: 4900be83602b ("x86/vector/msi: Switch to global reservation mode") Reported-by: Alexandru Chirvasitu <achirvasub@gmail.com> Reported-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Tested-by: Alexandru Chirvasitu <achirvasub@gmail.com> Tested-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Cc: Dou Liyang <douly.fnst@cn.fujitsu.com> Cc: Pavel Machek <pavel@ucw.cz> Cc: Maciej W. Rozycki <macro@linux-mips.org> Cc: Mikael Pettersson <mikpelinux@gmail.com> Cc: Josh Poulson <jopoulso@microsoft.com> Cc: Mihai Costache <v-micos@microsoft.com> Cc: Stephen Hemminger <sthemmin@microsoft.com> Cc: Marc Zyngier <marc.zyngier@arm.com> Cc: linux-pci@vger.kernel.org Cc: Haiyang Zhang <haiyangz@microsoft.com> Cc: Dexuan Cui <decui@microsoft.com> Cc: Simon Xiao <sixiao@microsoft.com> Cc: Saeed Mahameed <saeedm@mellanox.com> Cc: Jork Loeser <Jork.Loeser@microsoft.com> Cc: Bjorn Helgaas <bhelgaas@google.com> Cc: devel@linuxdriverproject.org Cc: KY Srinivasan <kys@microsoft.com> Cc: Alan Cox <alan@linux.intel.com> Cc: Sakari Ailus <sakari.ailus@intel.com>, Cc: linux-media@vger.kernel.org Link: https://lkml.kernel.org/r/alpine.DEB.2.20.1712291406420.1899@nanos Link: https://lkml.kernel.org/r/alpine.DEB.2.20.1712291409460.1899@nanos
2017-12-29 09:47:22 +00:00
if (IS_ENABLED(CONFIG_PCI_MSI) && pci_msi_ignore_mask)
return false;
/*
* Checking the first MSI descriptor is sufficient. MSIX supports
* masking and MSI does so when the can_mask attribute is set.
genirq/msi, x86/vector: Prevent reservation mode for non maskable MSI The new reservation mode for interrupts assigns a dummy vector when the interrupt is allocated and assigns a real vector when the interrupt is requested. The reservation mode prevents vector pressure when devices with a large amount of queues/interrupts are initialized, but only a minimal subset of those queues/interrupts is actually used. This mode has an issue with MSI interrupts which cannot be masked. If the driver is not careful or the hardware emits an interrupt before the device irq is requestd by the driver then the interrupt ends up on the dummy vector as a spurious interrupt which can cause malfunction of the device or in the worst case a lockup of the machine. Change the logic for the reservation mode so that the early activation of MSI interrupts checks whether: - the device is a PCI/MSI device - the reservation mode of the underlying irqdomain is activated - PCI/MSI masking is globally enabled - the PCI/MSI device uses either MSI-X, which supports masking, or MSI with the maskbit supported. If one of those conditions is false, then clear the reservation mode flag in the irq data of the interrupt and invoke irq_domain_activate_irq() with the reserve argument cleared. In the x86 vector code, clear the can_reserve flag in the vector allocation data so a subsequent free_irq() won't create the same situation again. The interrupt stays assigned to a real vector until pci_disable_msi() is invoked and all allocations are undone. Fixes: 4900be83602b ("x86/vector/msi: Switch to global reservation mode") Reported-by: Alexandru Chirvasitu <achirvasub@gmail.com> Reported-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Tested-by: Alexandru Chirvasitu <achirvasub@gmail.com> Tested-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Cc: Dou Liyang <douly.fnst@cn.fujitsu.com> Cc: Pavel Machek <pavel@ucw.cz> Cc: Maciej W. Rozycki <macro@linux-mips.org> Cc: Mikael Pettersson <mikpelinux@gmail.com> Cc: Josh Poulson <jopoulso@microsoft.com> Cc: Mihai Costache <v-micos@microsoft.com> Cc: Stephen Hemminger <sthemmin@microsoft.com> Cc: Marc Zyngier <marc.zyngier@arm.com> Cc: linux-pci@vger.kernel.org Cc: Haiyang Zhang <haiyangz@microsoft.com> Cc: Dexuan Cui <decui@microsoft.com> Cc: Simon Xiao <sixiao@microsoft.com> Cc: Saeed Mahameed <saeedm@mellanox.com> Cc: Jork Loeser <Jork.Loeser@microsoft.com> Cc: Bjorn Helgaas <bhelgaas@google.com> Cc: devel@linuxdriverproject.org Cc: KY Srinivasan <kys@microsoft.com> Cc: Alan Cox <alan@linux.intel.com> Cc: Sakari Ailus <sakari.ailus@intel.com>, Cc: linux-media@vger.kernel.org Link: https://lkml.kernel.org/r/alpine.DEB.2.20.1712291406420.1899@nanos Link: https://lkml.kernel.org/r/alpine.DEB.2.20.1712291409460.1899@nanos
2017-12-29 09:47:22 +00:00
*/
desc = msi_first_desc(dev, MSI_DESC_ALL);
return desc->pci.msi_attrib.is_msix || desc->pci.msi_attrib.can_mask;
genirq/msi: Handle reactivation only on success When analyzing the fallout of the x86 vector allocation rework it turned out that the error handling in msi_domain_alloc_irqs() is broken. If MSI_FLAG_MUST_REACTIVATE is set for a MSI domain then it clears the activation flag for a successfully initialized msi descriptor. If a subsequent initialization fails then the error handling code path does not deactivate the interrupt because the activation flag got cleared. Move the clearing of the activation flag outside of the initialization loop so that an eventual failure can be cleaned up correctly. Fixes: 22d0b12f3560 ("genirq/irqdomain: Add force reactivation flag to irq domains") Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Tested-by: Alexandru Chirvasitu <achirvasub@gmail.com> Tested-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Cc: Dou Liyang <douly.fnst@cn.fujitsu.com> Cc: Pavel Machek <pavel@ucw.cz> Cc: Maciej W. Rozycki <macro@linux-mips.org> Cc: Mikael Pettersson <mikpelinux@gmail.com> Cc: Josh Poulson <jopoulso@microsoft.com> Cc: Mihai Costache <v-micos@microsoft.com> Cc: Stephen Hemminger <sthemmin@microsoft.com> Cc: Marc Zyngier <marc.zyngier@arm.com> Cc: linux-pci@vger.kernel.org Cc: Haiyang Zhang <haiyangz@microsoft.com> Cc: Dexuan Cui <decui@microsoft.com> Cc: Simon Xiao <sixiao@microsoft.com> Cc: Saeed Mahameed <saeedm@mellanox.com> Cc: Jork Loeser <Jork.Loeser@microsoft.com> Cc: Bjorn Helgaas <bhelgaas@google.com> Cc: devel@linuxdriverproject.org Cc: KY Srinivasan <kys@microsoft.com> Cc: Alan Cox <alan@linux.intel.com> Cc: Sakari Ailus <sakari.ailus@intel.com>, Cc: linux-media@vger.kernel.org
2017-12-29 09:42:10 +00:00
}
static int msi_handle_pci_fail(struct irq_domain *domain, struct msi_desc *desc,
int allocated)
{
switch(domain->bus_token) {
case DOMAIN_BUS_PCI_MSI:
case DOMAIN_BUS_PCI_DEVICE_MSI:
case DOMAIN_BUS_PCI_DEVICE_MSIX:
case DOMAIN_BUS_VMD_MSI:
if (IS_ENABLED(CONFIG_PCI_MSI))
break;
fallthrough;
default:
return -ENOSPC;
}
/* Let a failed PCI multi MSI allocation retry */
if (desc->nvec_used > 1)
return 1;
/* If there was a successful allocation let the caller know */
return allocated ? allocated : -ENOSPC;
}
#define VIRQ_CAN_RESERVE 0x01
#define VIRQ_ACTIVATE 0x02
#define VIRQ_NOMASK_QUIRK 0x04
static int msi_init_virq(struct irq_domain *domain, int virq, unsigned int vflags)
{
struct irq_data *irqd = irq_domain_get_irq_data(domain, virq);
int ret;
if (!(vflags & VIRQ_CAN_RESERVE)) {
irqd_clr_can_reserve(irqd);
if (vflags & VIRQ_NOMASK_QUIRK)
irqd_set_msi_nomask_quirk(irqd);
/*
* If the interrupt is managed but no CPU is available to
* service it, shut it down until better times. Note that
* we only do this on the !RESERVE path as x86 (the only
* architecture using this flag) deals with this in a
* different way by using a catch-all vector.
*/
if ((vflags & VIRQ_ACTIVATE) &&
irqd_affinity_is_managed(irqd) &&
!cpumask_intersects(irq_data_get_affinity_mask(irqd),
cpu_online_mask)) {
irqd_set_managed_shutdown(irqd);
return 0;
}
}
if (!(vflags & VIRQ_ACTIVATE))
return 0;
ret = irq_domain_activate_irq(irqd, vflags & VIRQ_CAN_RESERVE);
if (ret)
return ret;
/*
* If the interrupt uses reservation mode, clear the activated bit
* so request_irq() will assign the final vector.
*/
if (vflags & VIRQ_CAN_RESERVE)
irqd_clr_activated(irqd);
return 0;
}
static int __msi_domain_alloc_irqs(struct device *dev, struct irq_domain *domain,
struct msi_ctrl *ctrl)
{
struct xarray *xa = &dev->msi.data->__domains[ctrl->domid].store;
struct msi_domain_info *info = domain->host_data;
struct msi_domain_ops *ops = info->ops;
unsigned int vflags = 0, allocated = 0;
msi_alloc_info_t arg = { };
struct msi_desc *desc;
unsigned long idx;
int i, ret, virq;
ret = msi_domain_prepare_irqs(domain, dev, ctrl->nirqs, &arg);
if (ret)
return ret;
/*
* This flag is set by the PCI layer as we need to activate
* the MSI entries before the PCI layer enables MSI in the
* card. Otherwise the card latches a random msi message.
*/
if (info->flags & MSI_FLAG_ACTIVATE_EARLY)
vflags |= VIRQ_ACTIVATE;
/*
* Interrupt can use a reserved vector and will not occupy
* a real device vector until the interrupt is requested.
*/
if (msi_check_reservation_mode(domain, info, dev)) {
vflags |= VIRQ_CAN_RESERVE;
/*
* MSI affinity setting requires a special quirk (X86) when
* reservation mode is active.
*/
if (info->flags & MSI_FLAG_NOMASK_QUIRK)
vflags |= VIRQ_NOMASK_QUIRK;
}
xa_for_each_range(xa, idx, desc, ctrl->first, ctrl->last) {
if (!msi_desc_match(desc, MSI_DESC_NOTASSOCIATED))
continue;
/* This should return -ECONFUSED... */
if (WARN_ON_ONCE(allocated >= ctrl->nirqs))
return -EINVAL;
if (ops->prepare_desc)
ops->prepare_desc(domain, &arg, desc);
ops->set_desc(&arg, desc);
virq = __irq_domain_alloc_irqs(domain, -1, desc->nvec_used,
dev_to_node(dev), &arg, false,
desc->affinity);
if (virq < 0)
return msi_handle_pci_fail(domain, desc, allocated);
for (i = 0; i < desc->nvec_used; i++) {
irq_set_msi_desc_off(virq, i, desc);
irq_debugfs_copy_devname(virq + i, dev);
ret = msi_init_virq(domain, virq + i, vflags);
if (ret)
return ret;
}
if (info->flags & MSI_FLAG_DEV_SYSFS) {
ret = msi_sysfs_populate_desc(dev, desc);
if (ret)
return ret;
}
allocated++;
}
return 0;
}
static int msi_domain_alloc_simple_msi_descs(struct device *dev,
struct msi_domain_info *info,
struct msi_ctrl *ctrl)
{
if (!(info->flags & MSI_FLAG_ALLOC_SIMPLE_MSI_DESCS))
return 0;
return msi_domain_add_simple_msi_descs(dev, ctrl);
}
static int __msi_domain_alloc_locked(struct device *dev, struct msi_ctrl *ctrl)
{
struct msi_domain_info *info;
struct msi_domain_ops *ops;
struct irq_domain *domain;
int ret;
if (!msi_ctrl_valid(dev, ctrl))
return -EINVAL;
domain = msi_get_device_domain(dev, ctrl->domid);
if (!domain)
return -ENODEV;
info = domain->host_data;
ret = msi_domain_alloc_simple_msi_descs(dev, info, ctrl);
if (ret)
return ret;
ops = info->ops;
if (ops->domain_alloc_irqs)
return ops->domain_alloc_irqs(domain, dev, ctrl->nirqs);
return __msi_domain_alloc_irqs(dev, domain, ctrl);
}
static int msi_domain_alloc_locked(struct device *dev, struct msi_ctrl *ctrl)
{
int ret = __msi_domain_alloc_locked(dev, ctrl);
if (ret)
msi_domain_free_locked(dev, ctrl);
return ret;
}
/**
* msi_domain_alloc_irqs_range_locked - Allocate interrupts from a MSI interrupt domain
* @dev: Pointer to device struct of the device for which the interrupts
* are allocated
* @domid: Id of the interrupt domain to operate on
* @first: First index to allocate (inclusive)
* @last: Last index to allocate (inclusive)
*
* Must be invoked from within a msi_lock_descs() / msi_unlock_descs()
* pair. Use this for MSI irqdomains which implement their own descriptor
* allocation/free.
*
* Return: %0 on success or an error code.
*/
int msi_domain_alloc_irqs_range_locked(struct device *dev, unsigned int domid,
unsigned int first, unsigned int last)
{
struct msi_ctrl ctrl = {
.domid = domid,
.first = first,
.last = last,
.nirqs = last + 1 - first,
};
return msi_domain_alloc_locked(dev, &ctrl);
}
/**
* msi_domain_alloc_irqs_range - Allocate interrupts from a MSI interrupt domain
* @dev: Pointer to device struct of the device for which the interrupts
* are allocated
* @domid: Id of the interrupt domain to operate on
* @first: First index to allocate (inclusive)
* @last: Last index to allocate (inclusive)
*
* Return: %0 on success or an error code.
*/
int msi_domain_alloc_irqs_range(struct device *dev, unsigned int domid,
unsigned int first, unsigned int last)
{
int ret;
msi_lock_descs(dev);
ret = msi_domain_alloc_irqs_range_locked(dev, domid, first, last);
msi_unlock_descs(dev);
return ret;
}
/**
* msi_domain_alloc_irqs_all_locked - Allocate all interrupts from a MSI interrupt domain
*
* @dev: Pointer to device struct of the device for which the interrupts
* are allocated
* @domid: Id of the interrupt domain to operate on
* @nirqs: The number of interrupts to allocate
*
* This function scans all MSI descriptors of the MSI domain and allocates interrupts
* for all unassigned ones. That function is to be used for MSI domain usage where
* the descriptor allocation is handled at the call site, e.g. PCI/MSI[X].
*
* Return: %0 on success or an error code.
*/
int msi_domain_alloc_irqs_all_locked(struct device *dev, unsigned int domid, int nirqs)
{
struct msi_ctrl ctrl = {
.domid = domid,
.first = 0,
.last = msi_domain_get_hwsize(dev, domid) - 1,
.nirqs = nirqs,
};
return msi_domain_alloc_locked(dev, &ctrl);
}
genirq/msi: Provide msi_domain_alloc_irq_at() For supporting post MSI-X enable allocations and for the upcoming PCI/IMS support a separate interface is required which allows not only the allocation of a specific index, but also the allocation of any, i.e. the next free index. The latter is especially required for IMS because IMS completely does away with index to functionality mappings which are often found in MSI/MSI-X implementation. But even with MSI-X there are devices where only the first few indices have a fixed functionality and the rest is freely assignable by software, e.g. to queues. msi_domain_alloc_irq_at() is also different from the range based interfaces as it always enforces that the MSI descriptor is allocated by the core code and not preallocated by the caller like the PCI/MSI[-X] enable code path does. msi_domain_alloc_irq_at() can be invoked with the index argument set to MSI_ANY_INDEX which makes the core code pick the next free index. The irq domain can provide a prepare_desc() operation callback in it's msi_domain_ops to do domain specific post allocation initialization before the actual Linux interrupt and the associated interrupt descriptor and hierarchy alloccations are conducted. The function also takes an optional @icookie argument which is of type union msi_instance_cookie. This cookie is not used by the core code and is stored in the allocated msi_desc::data::icookie. The meaning of the cookie is completely implementation defined. In case of IMS this might be a PASID or a pointer to a device queue, but for the MSI core it's opaque and not used in any way. The function returns a struct msi_map which on success contains the allocated index number and the Linux interrupt number so the caller can spare the index to Linux interrupt number lookup. On failure map::index contains the error code and map::virq is 0. Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Reviewed-by: Kevin Tian <kevin.tian@intel.com> Acked-by: Marc Zyngier <maz@kernel.org> Link: https://lore.kernel.org/r/20221124232326.501359457@linutronix.de
2022-11-24 23:26:18 +00:00
/**
* msi_domain_alloc_irq_at - Allocate an interrupt from a MSI interrupt domain at
* a given index - or at the next free index
*
* @dev: Pointer to device struct of the device for which the interrupts
* are allocated
* @domid: Id of the interrupt domain to operate on
* @index: Index for allocation. If @index == %MSI_ANY_INDEX the allocation
* uses the next free index.
* @affdesc: Optional pointer to an interrupt affinity descriptor structure
* @icookie: Optional pointer to a domain specific per instance cookie. If
* non-NULL the content of the cookie is stored in msi_desc::data.
* Must be NULL for MSI-X allocations
*
* This requires a MSI interrupt domain which lets the core code manage the
* MSI descriptors.
*
* Return: struct msi_map
*
* On success msi_map::index contains the allocated index number and
* msi_map::virq the corresponding Linux interrupt number
*
* On failure msi_map::index contains the error code and msi_map::virq
* is %0.
*/
struct msi_map msi_domain_alloc_irq_at(struct device *dev, unsigned int domid, unsigned int index,
const struct irq_affinity_desc *affdesc,
union msi_instance_cookie *icookie)
{
struct msi_ctrl ctrl = { .domid = domid, .nirqs = 1, };
struct irq_domain *domain;
struct msi_map map = { };
struct msi_desc *desc;
int ret;
msi_lock_descs(dev);
domain = msi_get_device_domain(dev, domid);
if (!domain) {
map.index = -ENODEV;
goto unlock;
}
desc = msi_alloc_desc(dev, 1, affdesc);
if (!desc) {
map.index = -ENOMEM;
goto unlock;
}
if (icookie)
desc->data.icookie = *icookie;
ret = msi_insert_desc(dev, desc, domid, index);
if (ret) {
map.index = ret;
goto unlock;
}
ctrl.first = ctrl.last = desc->msi_index;
ret = __msi_domain_alloc_irqs(dev, domain, &ctrl);
if (ret) {
map.index = ret;
msi_domain_free_locked(dev, &ctrl);
} else {
map.index = desc->msi_index;
map.virq = desc->irq;
}
unlock:
msi_unlock_descs(dev);
return map;
}
static void __msi_domain_free_irqs(struct device *dev, struct irq_domain *domain,
struct msi_ctrl *ctrl)
{
struct xarray *xa = &dev->msi.data->__domains[ctrl->domid].store;
struct msi_domain_info *info = domain->host_data;
struct irq_data *irqd;
struct msi_desc *desc;
unsigned long idx;
int i;
xa_for_each_range(xa, idx, desc, ctrl->first, ctrl->last) {
/* Only handle MSI entries which have an interrupt associated */
if (!msi_desc_match(desc, MSI_DESC_ASSOCIATED))
continue;
/* Make sure all interrupts are deactivated */
for (i = 0; i < desc->nvec_used; i++) {
irqd = irq_domain_get_irq_data(domain, desc->irq + i);
if (irqd && irqd_is_activated(irqd))
irq_domain_deactivate_irq(irqd);
genirq: MSI: Fix freeing of unallocated MSI While debugging an unrelated issue with the GICv3 ITS driver, the following trace triggered: WARNING: CPU: 1 PID: 1 at kernel/irq/irqdomain.c:1121 irq_domain_free_irqs+0x160/0x17c() NULL pointer, cannot free irq Modules linked in: CPU: 1 PID: 1 Comm: swapper/0 Tainted: G W 3.19.0-rc6+ #3690 Hardware name: FVP Base (DT) Call trace: [<ffffffc000089398>] dump_backtrace+0x0/0x13c [<ffffffc0000894e4>] show_stack+0x10/0x1c [<ffffffc00066d134>] dump_stack+0x74/0x94 [<ffffffc0000a92f8>] warn_slowpath_common+0x9c/0xd4 [<ffffffc0000a938c>] warn_slowpath_fmt+0x5c/0x80 [<ffffffc0000ee04c>] irq_domain_free_irqs+0x15c/0x17c [<ffffffc0000ef918>] msi_domain_free_irqs+0x58/0x74 [<ffffffc000386f58>] free_msi_irqs+0xb4/0x1c0 // The msi_prepare callback fails here [<ffffffc0003872c0>] pci_enable_msix+0x25c/0x3d4 [<ffffffc00038746c>] pci_enable_msix_range+0x34/0x80 [<ffffffc0003924ac>] vp_try_to_find_vqs+0xec/0x528 [<ffffffc000392954>] vp_find_vqs+0x6c/0xa8 [<ffffffc0003ee2a8>] init_vq+0x120/0x248 [<ffffffc0003eefb0>] virtblk_probe+0xb0/0x6bc [<ffffffc00038fc34>] virtio_dev_probe+0x17c/0x214 [<ffffffc0003d4a04>] driver_probe_device+0x7c/0x23c [<ffffffc0003d4cb0>] __driver_attach+0x98/0xa0 [<ffffffc0003d2c60>] bus_for_each_dev+0x60/0xb4 [<ffffffc0003d455c>] driver_attach+0x1c/0x28 [<ffffffc0003d41b0>] bus_add_driver+0x150/0x208 [<ffffffc0003d54c0>] driver_register+0x64/0x130 [<ffffffc00038f9e8>] register_virtio_driver+0x24/0x68 [<ffffffc00091320c>] init+0x70/0xac [<ffffffc0000828f0>] do_one_initcall+0x94/0x1d0 [<ffffffc0008e9b00>] kernel_init_freeable+0x144/0x1e4 [<ffffffc00066a434>] kernel_init+0xc/0xd8 ---[ end trace f9ee562a77cc7bae ]--- The ITS msi_prepare callback having failed, we end-up trying to free MSIs that have never been allocated. Oddly enough, the kernel is pretty upset about it. It turns out that this behaviour was expected before the MSI domain was introduced (and dealt with in arch_teardown_msi_irqs). The obvious fix is to detect this early enough and bail out. Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Reviewed-by: Jiang Liu <jiang.liu@linux.intel.com> Link: http://lkml.kernel.org/r/1422299419-6051-1-git-send-email-marc.zyngier@arm.com Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
2015-01-26 19:10:19 +00:00
}
irq_domain_free_irqs(desc->irq, desc->nvec_used);
if (info->flags & MSI_FLAG_DEV_SYSFS)
msi_sysfs_remove_desc(dev, desc);
desc->irq = 0;
}
}
static void msi_domain_free_locked(struct device *dev, struct msi_ctrl *ctrl)
{
struct msi_domain_info *info;
struct msi_domain_ops *ops;
struct irq_domain *domain;
if (!msi_ctrl_valid(dev, ctrl))
return;
domain = msi_get_device_domain(dev, ctrl->domid);
if (!domain)
return;
info = domain->host_data;
ops = info->ops;
if (ops->domain_free_irqs)
ops->domain_free_irqs(domain, dev);
else
__msi_domain_free_irqs(dev, domain, ctrl);
if (ops->msi_post_free)
ops->msi_post_free(domain, dev);
if (info->flags & MSI_FLAG_FREE_MSI_DESCS)
msi_domain_free_descs(dev, ctrl);
}
/**
* msi_domain_free_irqs_range_locked - Free a range of interrupts from a MSI interrupt domain
* associated to @dev with msi_lock held
* @dev: Pointer to device struct of the device for which the interrupts
* are freed
* @domid: Id of the interrupt domain to operate on
* @first: First index to free (inclusive)
* @last: Last index to free (inclusive)
*/
void msi_domain_free_irqs_range_locked(struct device *dev, unsigned int domid,
unsigned int first, unsigned int last)
{
struct msi_ctrl ctrl = {
.domid = domid,
.first = first,
.last = last,
};
msi_domain_free_locked(dev, &ctrl);
}
/**
* msi_domain_free_irqs_range - Free a range of interrupts from a MSI interrupt domain
* associated to @dev
* @dev: Pointer to device struct of the device for which the interrupts
* are freed
* @domid: Id of the interrupt domain to operate on
* @first: First index to free (inclusive)
* @last: Last index to free (inclusive)
*/
void msi_domain_free_irqs_range(struct device *dev, unsigned int domid,
unsigned int first, unsigned int last)
{
msi_lock_descs(dev);
msi_domain_free_irqs_range_locked(dev, domid, first, last);
msi_unlock_descs(dev);
}
/**
* msi_domain_free_irqs_all_locked - Free all interrupts from a MSI interrupt domain
* associated to a device
* @dev: Pointer to device struct of the device for which the interrupts
* are freed
* @domid: The id of the domain to operate on
*
* Must be invoked from within a msi_lock_descs() / msi_unlock_descs()
* pair. Use this for MSI irqdomains which implement their own vector
* allocation.
*/
void msi_domain_free_irqs_all_locked(struct device *dev, unsigned int domid)
{
msi_domain_free_irqs_range_locked(dev, domid, 0,
msi_domain_get_hwsize(dev, domid) - 1);
}
/**
* msi_domain_free_irqs_all - Free all interrupts from a MSI interrupt domain
* associated to a device
* @dev: Pointer to device struct of the device for which the interrupts
* are freed
* @domid: The id of the domain to operate on
*/
void msi_domain_free_irqs_all(struct device *dev, unsigned int domid)
{
msi_lock_descs(dev);
msi_domain_free_irqs_all_locked(dev, domid);
msi_unlock_descs(dev);
}
/**
* msi_get_domain_info - Get the MSI interrupt domain info for @domain
* @domain: The interrupt domain to retrieve data from
*
genirq: Fix kernel-doc warnings in pm.c, msi.c and ipi.c Fix all kernel-doc warnings in these 3 files and do some simple editing (capitalize acronyms, capitalize Linux). kernel/irq/pm.c:235: warning: expecting prototype for irq_pm_syscore_ops(). Prototype was for irq_pm_syscore_resume() instead kernel/irq/msi.c:530: warning: expecting prototype for __msi_domain_free_irqs(). Prototype was for msi_domain_free_irqs() instead kernel/irq/msi.c:31: warning: No description found for return value of 'alloc_msi_entry' kernel/irq/msi.c:103: warning: No description found for return value of 'msi_domain_set_affinity' kernel/irq/msi.c:288: warning: No description found for return value of 'msi_create_irq_domain' kernel/irq/msi.c:499: warning: No description found for return value of 'msi_domain_alloc_irqs' kernel/irq/msi.c:545: warning: No description found for return value of 'msi_get_domain_info' kernel/irq/ipi.c:264: warning: expecting prototype for ipi_send_mask(). Prototype was for __ipi_send_mask() instead kernel/irq/ipi.c:25: warning: No description found for return value of 'irq_reserve_ipi' kernel/irq/ipi.c:116: warning: No description found for return value of 'irq_destroy_ipi' kernel/irq/ipi.c:163: warning: No description found for return value of 'ipi_get_hwirq' kernel/irq/ipi.c:222: warning: No description found for return value of '__ipi_send_single' kernel/irq/ipi.c:308: warning: No description found for return value of 'ipi_send_single' kernel/irq/ipi.c:329: warning: No description found for return value of 'ipi_send_mask' Signed-off-by: Randy Dunlap <rdunlap@infradead.org> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Link: https://lore.kernel.org/r/20210810234835.12547-1-rdunlap@infradead.org
2021-08-10 23:48:35 +00:00
* Return: the pointer to the msi_domain_info stored in @domain->host_data.
*/
struct msi_domain_info *msi_get_domain_info(struct irq_domain *domain)
{
return (struct msi_domain_info *)domain->host_data;
}