* [PATCH 0/3] trivial changes for MSI
@ 2014-07-26 1:48 wangyijing
2014-07-26 1:48 ` [PATCH 1/3] PCI/MSI: Clean up the kobject in struct msi_desc wangyijing
` (3 more replies)
0 siblings, 4 replies; 6+ messages in thread
From: wangyijing @ 2014-07-26 1:48 UTC (permalink / raw)
To: Bjorn Helgaas; +Cc: linux-pci, Yijing Wang
From: Yijing Wang <wangyijing@huawei.com>
Yijing Wang (3):
PCI/MSI: Clean up the kobject in struct msi_desc
PCI/MSI: Remove msi_desc->msi_attrib.pos
PCI/MSI: Refactor msi_bus to support EP enable/disable MSI
arch/mips/pci/msi-octeon.c | 6 +++---
drivers/pci/host/pcie-designware.c | 2 +-
drivers/pci/msi.c | 13 -------------
drivers/pci/pci-sysfs.c | 12 ++++++------
include/linux/msi.h | 3 ---
5 files changed, 10 insertions(+), 26 deletions(-)
^ permalink raw reply [flat|nested] 6+ messages in thread
* [PATCH 1/3] PCI/MSI: Clean up the kobject in struct msi_desc
2014-07-26 1:48 [PATCH 0/3] trivial changes for MSI wangyijing
@ 2014-07-26 1:48 ` wangyijing
2014-07-27 3:18 ` Greg Kroah-Hartman
2014-07-26 1:48 ` [PATCH 2/3] PCI/MSI: Remove msi_desc->msi_attrib.pos wangyijing
` (2 subsequent siblings)
3 siblings, 1 reply; 6+ messages in thread
From: wangyijing @ 2014-07-26 1:48 UTC (permalink / raw)
To: Bjorn Helgaas; +Cc: linux-pci, Yijing Wang, Greg Kroah-Hartman
From: Yijing Wang <wangyijing@huawei.com>
commit 1c51b50c299 exported MSI mode using attributes
instead of kobject. So clean up the kobject in struct
msi_desc.
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Yijing Wang <wangyijing@huawei.com>
---
drivers/pci/msi.c | 11 -----------
include/linux/msi.h | 2 --
2 files changed, 0 insertions(+), 13 deletions(-)
diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c
index 5a40516..e2aa74e 100644
--- a/drivers/pci/msi.c
+++ b/drivers/pci/msi.c
@@ -384,17 +384,6 @@ static void free_msi_irqs(struct pci_dev *dev)
iounmap(entry->mask_base);
}
- /*
- * Its possible that we get into this path
- * When populate_msi_sysfs fails, which means the entries
- * were not registered with sysfs. In that case don't
- * unregister them.
- */
- if (entry->kobj.parent) {
- kobject_del(&entry->kobj);
- kobject_put(&entry->kobj);
- }
-
list_del(&entry->list);
kfree(entry);
}
diff --git a/include/linux/msi.h b/include/linux/msi.h
index 8103f32..8892d41 100644
--- a/include/linux/msi.h
+++ b/include/linux/msi.h
@@ -47,8 +47,6 @@ struct msi_desc {
/* Last set MSI message */
struct msi_msg msg;
-
- struct kobject kobj;
};
/*
--
1.7.1
^ permalink raw reply related [flat|nested] 6+ messages in thread
* [PATCH 2/3] PCI/MSI: Remove msi_desc->msi_attrib.pos
2014-07-26 1:48 [PATCH 0/3] trivial changes for MSI wangyijing
2014-07-26 1:48 ` [PATCH 1/3] PCI/MSI: Clean up the kobject in struct msi_desc wangyijing
@ 2014-07-26 1:48 ` wangyijing
2014-07-26 1:48 ` [PATCH 3/3] PCI/MSI: Refactor msi_bus to support EP enable/disable MSI wangyijing
2014-08-22 7:46 ` [PATCH 0/3] trivial changes for MSI Yijing Wang
3 siblings, 0 replies; 6+ messages in thread
From: wangyijing @ 2014-07-26 1:48 UTC (permalink / raw)
To: Bjorn Helgaas; +Cc: linux-pci, Yijing Wang
From: Yijing Wang <wangyijing@huawei.com>
Use pci_dev->msi/x_cap instead of msi_desc->msi_attrib.pos,
and remove pos member.
Signed-off-by: Yijing Wang <wangyijing@huawei.com>
---
arch/mips/pci/msi-octeon.c | 6 +++---
drivers/pci/host/pcie-designware.c | 2 +-
drivers/pci/msi.c | 2 --
include/linux/msi.h | 1 -
4 files changed, 4 insertions(+), 7 deletions(-)
diff --git a/arch/mips/pci/msi-octeon.c b/arch/mips/pci/msi-octeon.c
index ab0c5d1..c8814be 100644
--- a/arch/mips/pci/msi-octeon.c
+++ b/arch/mips/pci/msi-octeon.c
@@ -73,8 +73,8 @@ int arch_setup_msi_irq(struct pci_dev *dev, struct msi_desc *desc)
* wants. Most devices only want 1, which will give
* configured_private_bits and request_private_bits equal 0.
*/
- pci_read_config_word(dev, desc->msi_attrib.pos + PCI_MSI_FLAGS,
- &control);
+ pci_read_config_word(dev, dev->msi_cap + PCI_MSI_FLAGS,
+ &control);
/*
* If the number of private bits has been configured then use
@@ -176,7 +176,7 @@ msi_irq_allocated:
/* Update the number of IRQs the device has available to it */
control &= ~PCI_MSI_FLAGS_QSIZE;
control |= request_private_bits << 4;
- pci_write_config_word(dev, desc->msi_attrib.pos + PCI_MSI_FLAGS,
+ pci_write_config_word(dev, dev->msi_cap + PCI_MSI_FLAGS,
control);
irq_set_msi_desc(irq, desc);
diff --git a/drivers/pci/host/pcie-designware.c b/drivers/pci/host/pcie-designware.c
index 52bd3a1..c3706cb 100644
--- a/drivers/pci/host/pcie-designware.c
+++ b/drivers/pci/host/pcie-designware.c
@@ -355,7 +355,7 @@ static int dw_msi_setup_irq(struct msi_chip *chip, struct pci_dev *pdev,
return -EINVAL;
}
- pci_read_config_word(pdev, desc->msi_attrib.pos+PCI_MSI_FLAGS,
+ pci_read_config_word(pdev, pdev->msi_cap + PCI_MSI_FLAGS,
&msg_ctr);
msgvec = (msg_ctr&PCI_MSI_FLAGS_QSIZE) >> 4;
if (msgvec == 0)
diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c
index e2aa74e..0d0f163 100644
--- a/drivers/pci/msi.c
+++ b/drivers/pci/msi.c
@@ -584,7 +584,6 @@ static struct msi_desc *msi_setup_entry(struct pci_dev *dev)
entry->msi_attrib.entry_nr = 0;
entry->msi_attrib.maskbit = !!(control & PCI_MSI_FLAGS_MASKBIT);
entry->msi_attrib.default_irq = dev->irq; /* Save IOAPIC IRQ */
- entry->msi_attrib.pos = dev->msi_cap;
entry->msi_attrib.multi_cap = (control & PCI_MSI_FLAGS_QMASK) >> 1;
if (control & PCI_MSI_FLAGS_64BIT)
@@ -688,7 +687,6 @@ static int msix_setup_entries(struct pci_dev *dev, void __iomem *base,
entry->msi_attrib.is_64 = 1;
entry->msi_attrib.entry_nr = entries[i].entry;
entry->msi_attrib.default_irq = dev->irq;
- entry->msi_attrib.pos = dev->msix_cap;
entry->mask_base = base;
list_add_tail(&entry->list, &dev->msi_list);
diff --git a/include/linux/msi.h b/include/linux/msi.h
index 8892d41..fff7201 100644
--- a/include/linux/msi.h
+++ b/include/linux/msi.h
@@ -29,7 +29,6 @@ struct msi_desc {
__u8 multi_cap : 3; /* log2 num of messages supported */
__u8 maskbit : 1; /* mask-pending bit supported ? */
__u8 is_64 : 1; /* Address size: 0=32bit 1=64bit */
- __u8 pos; /* Location of the msi capability */
__u16 entry_nr; /* specific enabled entry */
unsigned default_irq; /* default pre-assigned irq */
} msi_attrib;
--
1.7.1
^ permalink raw reply related [flat|nested] 6+ messages in thread
* [PATCH 3/3] PCI/MSI: Refactor msi_bus to support EP enable/disable MSI
2014-07-26 1:48 [PATCH 0/3] trivial changes for MSI wangyijing
2014-07-26 1:48 ` [PATCH 1/3] PCI/MSI: Clean up the kobject in struct msi_desc wangyijing
2014-07-26 1:48 ` [PATCH 2/3] PCI/MSI: Remove msi_desc->msi_attrib.pos wangyijing
@ 2014-07-26 1:48 ` wangyijing
2014-08-22 7:46 ` [PATCH 0/3] trivial changes for MSI Yijing Wang
3 siblings, 0 replies; 6+ messages in thread
From: wangyijing @ 2014-07-26 1:48 UTC (permalink / raw)
To: Bjorn Helgaas; +Cc: linux-pci, Yijing Wang
From: Yijing Wang <wangyijing@huawei.com>
Msi_bus attribute is only valid for bridge device.
We can enable or disable MSI capability for a bus,
if we echo 1/0 > /sys/bus/pci/devices/$EP/msi_bus,
the action will be ignored. Sometime we need to
only enable/disable a device MSI, not all devices share
the same bus.
Signed-off-by: Yijing Wang <wangyijing@huawei.com>
---
drivers/pci/pci-sysfs.c | 12 ++++++------
1 files changed, 6 insertions(+), 6 deletions(-)
diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c
index 9ff0a90..b199ad9 100644
--- a/drivers/pci/pci-sysfs.c
+++ b/drivers/pci/pci-sysfs.c
@@ -251,11 +251,9 @@ static ssize_t msi_bus_show(struct device *dev, struct device_attribute *attr,
{
struct pci_dev *pdev = to_pci_dev(dev);
- if (!pdev->subordinate)
- return 0;
-
- return sprintf(buf, "%u\n",
- !(pdev->subordinate->bus_flags & PCI_BUS_FLAGS_NO_MSI));
+ return sprintf(buf, "%u\n", pdev->subordinate ?
+ !(pdev->subordinate->bus_flags & PCI_BUS_FLAGS_NO_MSI)
+ : !pdev->no_msi);
}
static ssize_t msi_bus_store(struct device *dev, struct device_attribute *attr,
@@ -278,8 +276,10 @@ static ssize_t msi_bus_store(struct device *dev, struct device_attribute *attr,
* Maybe devices without subordinate buses shouldn't have this
* attribute in the first place?
*/
- if (!pdev->subordinate)
+ if (!pdev->subordinate) {
+ pdev->no_msi = !val;
return count;
+ }
/* Is the flag going to change, or keep the value it already had? */
if (!(pdev->subordinate->bus_flags & PCI_BUS_FLAGS_NO_MSI) ^
--
1.7.1
^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [PATCH 1/3] PCI/MSI: Clean up the kobject in struct msi_desc
2014-07-26 1:48 ` [PATCH 1/3] PCI/MSI: Clean up the kobject in struct msi_desc wangyijing
@ 2014-07-27 3:18 ` Greg Kroah-Hartman
0 siblings, 0 replies; 6+ messages in thread
From: Greg Kroah-Hartman @ 2014-07-27 3:18 UTC (permalink / raw)
To: wangyijing; +Cc: Bjorn Helgaas, linux-pci
On Sat, Jul 26, 2014 at 09:48:34AM +0800, wangyijing@huawei.com wrote:
> From: Yijing Wang <wangyijing@huawei.com>
>
> commit 1c51b50c299 exported MSI mode using attributes
> instead of kobject. So clean up the kobject in struct
> msi_desc.
>
> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
> Signed-off-by: Yijing Wang <wangyijing@huawei.com>
> ---
> drivers/pci/msi.c | 11 -----------
> include/linux/msi.h | 2 --
> 2 files changed, 0 insertions(+), 13 deletions(-)
>
> diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c
> index 5a40516..e2aa74e 100644
> --- a/drivers/pci/msi.c
> +++ b/drivers/pci/msi.c
> @@ -384,17 +384,6 @@ static void free_msi_irqs(struct pci_dev *dev)
> iounmap(entry->mask_base);
> }
>
> - /*
> - * Its possible that we get into this path
> - * When populate_msi_sysfs fails, which means the entries
> - * were not registered with sysfs. In that case don't
> - * unregister them.
> - */
> - if (entry->kobj.parent) {
> - kobject_del(&entry->kobj);
> - kobject_put(&entry->kobj);
> - }
> -
> list_del(&entry->list);
> kfree(entry);
> }
> diff --git a/include/linux/msi.h b/include/linux/msi.h
> index 8103f32..8892d41 100644
> --- a/include/linux/msi.h
> +++ b/include/linux/msi.h
> @@ -47,8 +47,6 @@ struct msi_desc {
>
> /* Last set MSI message */
> struct msi_msg msg;
> -
> - struct kobject kobj;
> };
>
> /*
> --
> 1.7.1
Oops, missed that one, nice catch.
Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH 0/3] trivial changes for MSI
2014-07-26 1:48 [PATCH 0/3] trivial changes for MSI wangyijing
` (2 preceding siblings ...)
2014-07-26 1:48 ` [PATCH 3/3] PCI/MSI: Refactor msi_bus to support EP enable/disable MSI wangyijing
@ 2014-08-22 7:46 ` Yijing Wang
3 siblings, 0 replies; 6+ messages in thread
From: Yijing Wang @ 2014-08-22 7:46 UTC (permalink / raw)
To: wangyijing, Bjorn Helgaas; +Cc: linux-pci
On 2014/7/26 9:48, wangyijing@huawei.com wrote:
> From: Yijing Wang <wangyijing@huawei.com>
Hi Bjorn, please drop this one, I have sent new version.
Thanks!
Yijing.
>
> Yijing Wang (3):
> PCI/MSI: Clean up the kobject in struct msi_desc
> PCI/MSI: Remove msi_desc->msi_attrib.pos
> PCI/MSI: Refactor msi_bus to support EP enable/disable MSI
>
> arch/mips/pci/msi-octeon.c | 6 +++---
> drivers/pci/host/pcie-designware.c | 2 +-
> drivers/pci/msi.c | 13 -------------
> drivers/pci/pci-sysfs.c | 12 ++++++------
> include/linux/msi.h | 3 ---
> 5 files changed, 10 insertions(+), 26 deletions(-)
>
>
> .
>
--
Thanks!
Yijing
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2014-08-22 7:46 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2014-07-26 1:48 [PATCH 0/3] trivial changes for MSI wangyijing
2014-07-26 1:48 ` [PATCH 1/3] PCI/MSI: Clean up the kobject in struct msi_desc wangyijing
2014-07-27 3:18 ` Greg Kroah-Hartman
2014-07-26 1:48 ` [PATCH 2/3] PCI/MSI: Remove msi_desc->msi_attrib.pos wangyijing
2014-07-26 1:48 ` [PATCH 3/3] PCI/MSI: Refactor msi_bus to support EP enable/disable MSI wangyijing
2014-08-22 7:46 ` [PATCH 0/3] trivial changes for MSI Yijing Wang
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).