| Message ID | 1413269253-8990-2-git-send-email-weiyang@linux.vnet.ibm.com (mailing list archive) |
|---|---|
| State | New, archived |
| Delegated to: | Bjorn Helgaas |
| Headers | show |
[+cc Yinghai] On Tue, Oct 14, 2014 at 02:47:30PM +0800, Wei Yang wrote: > In some quirks, it tries to search a pci cap and use a ttl value to avoid > infinite loop. While the value is hard coded to 48, which is the same as marco > PCI_FIND_CAP_TTL. > > This patch moves the definition of PCI_FIND_CAP_TTL to pci.h and replace the > hard coded value with it. This seems reasonable (though I added Yinghai in case he knows of any reason why HT capabilities should be different from plain PCI capabilities in this respect). But I'd prefer to have the definition in drivers/pci/pci.h rather than in include/linux/pci.h, because both users already include drivers/pci/pci.h, and it is less visible. Bjorn > > Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> > --- > drivers/pci/pci.c | 1 - > drivers/pci/quirks.c | 8 ++++---- > include/linux/pci.h | 2 ++ > 3 files changed, 6 insertions(+), 5 deletions(-) > > diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c > index 2c9ac70..76b002b1 100644 > --- a/drivers/pci/pci.c > +++ b/drivers/pci/pci.c > @@ -137,7 +137,6 @@ void __iomem *pci_ioremap_bar(struct pci_dev *pdev, int bar) > EXPORT_SYMBOL_GPL(pci_ioremap_bar); > #endif > > -#define PCI_FIND_CAP_TTL 48 > > static int __pci_find_next_cap_ttl(struct pci_bus *bus, unsigned int devfn, > u8 pos, int cap, int *ttl) > diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c > index 80c2d01..a5f46b8 100644 > --- a/drivers/pci/quirks.c > +++ b/drivers/pci/quirks.c > @@ -2176,7 +2176,7 @@ DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_AMD, 0x9601, quirk_amd_780_apc_msi); > * return 1 if a HT MSI capability is found and enabled */ > static int msi_ht_cap_enabled(struct pci_dev *dev) > { > - int pos, ttl = 48; > + int pos, ttl = PCI_FIND_CAP_TTL; > > pos = pci_find_ht_capability(dev, HT_CAPTYPE_MSI_MAPPING); > while (pos && ttl--) { > @@ -2235,7 +2235,7 @@ DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_CK804_PCIE, > /* Force enable MSI mapping capability on HT bridges */ > static void ht_enable_msi_mapping(struct pci_dev *dev) > { > - int pos, ttl = 48; > + int pos, ttl = PCI_FIND_CAP_TTL; > > pos = pci_find_ht_capability(dev, HT_CAPTYPE_MSI_MAPPING); > while (pos && ttl--) { > @@ -2314,7 +2314,7 @@ DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_NVIDIA, > > static int ht_check_msi_mapping(struct pci_dev *dev) > { > - int pos, ttl = 48; > + int pos, ttl = PCI_FIND_CAP_TTL; > int found = 0; > > /* check if there is HT MSI cap or enabled on this device */ > @@ -2439,7 +2439,7 @@ out: > > static void ht_disable_msi_mapping(struct pci_dev *dev) > { > - int pos, ttl = 48; > + int pos, ttl = PCI_FIND_CAP_TTL; > > pos = pci_find_ht_capability(dev, HT_CAPTYPE_MSI_MAPPING); > while (pos && ttl--) { > diff --git a/include/linux/pci.h b/include/linux/pci.h > index 96453f9..b27b79e 100644 > --- a/include/linux/pci.h > +++ b/include/linux/pci.h > @@ -33,6 +33,8 @@ > > #include <linux/pci_ids.h> > > +#define PCI_FIND_CAP_TTL 48 > + > /* > * The PCI interface treats multi-function devices as independent > * devices. The slot/function address of each device is encoded > -- > 1.7.9.5 > -- To unsubscribe from this list: send the line "unsubscribe linux-pci" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Wed, Oct 22, 2014 at 05:18:30PM -0600, Bjorn Helgaas wrote: >[+cc Yinghai] > >On Tue, Oct 14, 2014 at 02:47:30PM +0800, Wei Yang wrote: >> In some quirks, it tries to search a pci cap and use a ttl value to avoid >> infinite loop. While the value is hard coded to 48, which is the same as marco >> PCI_FIND_CAP_TTL. >> >> This patch moves the definition of PCI_FIND_CAP_TTL to pci.h and replace the >> hard coded value with it. > >This seems reasonable (though I added Yinghai in case he knows of any >reason why HT capabilities should be different from plain PCI capabilities >in this respect). > >But I'd prefer to have the definition in drivers/pci/pci.h rather than in >include/linux/pci.h, because both users already include drivers/pci/pci.h, >and it is less visible. Got it. > >Bjorn > >> >> Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> >> --- >> drivers/pci/pci.c | 1 - >> drivers/pci/quirks.c | 8 ++++---- >> include/linux/pci.h | 2 ++ >> 3 files changed, 6 insertions(+), 5 deletions(-) >> >> diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c >> index 2c9ac70..76b002b1 100644 >> --- a/drivers/pci/pci.c >> +++ b/drivers/pci/pci.c >> @@ -137,7 +137,6 @@ void __iomem *pci_ioremap_bar(struct pci_dev *pdev, int bar) >> EXPORT_SYMBOL_GPL(pci_ioremap_bar); >> #endif >> >> -#define PCI_FIND_CAP_TTL 48 >> >> static int __pci_find_next_cap_ttl(struct pci_bus *bus, unsigned int devfn, >> u8 pos, int cap, int *ttl) >> diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c >> index 80c2d01..a5f46b8 100644 >> --- a/drivers/pci/quirks.c >> +++ b/drivers/pci/quirks.c >> @@ -2176,7 +2176,7 @@ DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_AMD, 0x9601, quirk_amd_780_apc_msi); >> * return 1 if a HT MSI capability is found and enabled */ >> static int msi_ht_cap_enabled(struct pci_dev *dev) >> { >> - int pos, ttl = 48; >> + int pos, ttl = PCI_FIND_CAP_TTL; >> >> pos = pci_find_ht_capability(dev, HT_CAPTYPE_MSI_MAPPING); >> while (pos && ttl--) { >> @@ -2235,7 +2235,7 @@ DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_CK804_PCIE, >> /* Force enable MSI mapping capability on HT bridges */ >> static void ht_enable_msi_mapping(struct pci_dev *dev) >> { >> - int pos, ttl = 48; >> + int pos, ttl = PCI_FIND_CAP_TTL; >> >> pos = pci_find_ht_capability(dev, HT_CAPTYPE_MSI_MAPPING); >> while (pos && ttl--) { >> @@ -2314,7 +2314,7 @@ DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_NVIDIA, >> >> static int ht_check_msi_mapping(struct pci_dev *dev) >> { >> - int pos, ttl = 48; >> + int pos, ttl = PCI_FIND_CAP_TTL; >> int found = 0; >> >> /* check if there is HT MSI cap or enabled on this device */ >> @@ -2439,7 +2439,7 @@ out: >> >> static void ht_disable_msi_mapping(struct pci_dev *dev) >> { >> - int pos, ttl = 48; >> + int pos, ttl = PCI_FIND_CAP_TTL; >> >> pos = pci_find_ht_capability(dev, HT_CAPTYPE_MSI_MAPPING); >> while (pos && ttl--) { >> diff --git a/include/linux/pci.h b/include/linux/pci.h >> index 96453f9..b27b79e 100644 >> --- a/include/linux/pci.h >> +++ b/include/linux/pci.h >> @@ -33,6 +33,8 @@ >> >> #include <linux/pci_ids.h> >> >> +#define PCI_FIND_CAP_TTL 48 >> + >> /* >> * The PCI interface treats multi-function devices as independent >> * devices. The slot/function address of each device is encoded >> -- >> 1.7.9.5 >>
On Wed, Oct 22, 2014 at 05:18:30PM -0600, Bjorn Helgaas wrote: >[+cc Yinghai] > >On Tue, Oct 14, 2014 at 02:47:30PM +0800, Wei Yang wrote: >> In some quirks, it tries to search a pci cap and use a ttl value to avoid >> infinite loop. While the value is hard coded to 48, which is the same as marco >> PCI_FIND_CAP_TTL. >> >> This patch moves the definition of PCI_FIND_CAP_TTL to pci.h and replace the >> hard coded value with it. > >This seems reasonable (though I added Yinghai in case he knows of any >reason why HT capabilities should be different from plain PCI capabilities >in this respect). Hi, Yinghai, Do you have some historical story on the definition? > >But I'd prefer to have the definition in drivers/pci/pci.h rather than in >include/linux/pci.h, because both users already include drivers/pci/pci.h, >and it is less visible. > >Bjorn > >> >> Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> >> --- >> drivers/pci/pci.c | 1 - >> drivers/pci/quirks.c | 8 ++++---- >> include/linux/pci.h | 2 ++ >> 3 files changed, 6 insertions(+), 5 deletions(-) >> >> diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c >> index 2c9ac70..76b002b1 100644 >> --- a/drivers/pci/pci.c >> +++ b/drivers/pci/pci.c >> @@ -137,7 +137,6 @@ void __iomem *pci_ioremap_bar(struct pci_dev *pdev, int bar) >> EXPORT_SYMBOL_GPL(pci_ioremap_bar); >> #endif >> >> -#define PCI_FIND_CAP_TTL 48 >> >> static int __pci_find_next_cap_ttl(struct pci_bus *bus, unsigned int devfn, >> u8 pos, int cap, int *ttl) >> diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c >> index 80c2d01..a5f46b8 100644 >> --- a/drivers/pci/quirks.c >> +++ b/drivers/pci/quirks.c >> @@ -2176,7 +2176,7 @@ DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_AMD, 0x9601, quirk_amd_780_apc_msi); >> * return 1 if a HT MSI capability is found and enabled */ >> static int msi_ht_cap_enabled(struct pci_dev *dev) >> { >> - int pos, ttl = 48; >> + int pos, ttl = PCI_FIND_CAP_TTL; >> >> pos = pci_find_ht_capability(dev, HT_CAPTYPE_MSI_MAPPING); >> while (pos && ttl--) { >> @@ -2235,7 +2235,7 @@ DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_CK804_PCIE, >> /* Force enable MSI mapping capability on HT bridges */ >> static void ht_enable_msi_mapping(struct pci_dev *dev) >> { >> - int pos, ttl = 48; >> + int pos, ttl = PCI_FIND_CAP_TTL; >> >> pos = pci_find_ht_capability(dev, HT_CAPTYPE_MSI_MAPPING); >> while (pos && ttl--) { >> @@ -2314,7 +2314,7 @@ DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_NVIDIA, >> >> static int ht_check_msi_mapping(struct pci_dev *dev) >> { >> - int pos, ttl = 48; >> + int pos, ttl = PCI_FIND_CAP_TTL; >> int found = 0; >> >> /* check if there is HT MSI cap or enabled on this device */ >> @@ -2439,7 +2439,7 @@ out: >> >> static void ht_disable_msi_mapping(struct pci_dev *dev) >> { >> - int pos, ttl = 48; >> + int pos, ttl = PCI_FIND_CAP_TTL; >> >> pos = pci_find_ht_capability(dev, HT_CAPTYPE_MSI_MAPPING); >> while (pos && ttl--) { >> diff --git a/include/linux/pci.h b/include/linux/pci.h >> index 96453f9..b27b79e 100644 >> --- a/include/linux/pci.h >> +++ b/include/linux/pci.h >> @@ -33,6 +33,8 @@ >> >> #include <linux/pci_ids.h> >> >> +#define PCI_FIND_CAP_TTL 48 >> + >> /* >> * The PCI interface treats multi-function devices as independent >> * devices. The slot/function address of each device is encoded >> -- >> 1.7.9.5 >>
On Thu, Nov 06, 2014 at 11:12:02AM +0800, Wei Yang wrote: >On Wed, Oct 22, 2014 at 05:18:30PM -0600, Bjorn Helgaas wrote: >>[+cc Yinghai] >> >>On Tue, Oct 14, 2014 at 02:47:30PM +0800, Wei Yang wrote: >>> In some quirks, it tries to search a pci cap and use a ttl value to avoid >>> infinite loop. While the value is hard coded to 48, which is the same as marco >>> PCI_FIND_CAP_TTL. >>> >>> This patch moves the definition of PCI_FIND_CAP_TTL to pci.h and replace the >>> hard coded value with it. >> >>This seems reasonable (though I added Yinghai in case he knows of any >>reason why HT capabilities should be different from plain PCI capabilities >>in this respect). > >Hi, Yinghai, > >Do you have some historical story on the definition? > Hi, I am not sure if I missed some mail. This is the last mail I sent, but no reply I got yet. Is is ok to send an updated version? >> >>But I'd prefer to have the definition in drivers/pci/pci.h rather than in >>include/linux/pci.h, because both users already include drivers/pci/pci.h, >>and it is less visible. >> >>Bjorn >> >>> >>> Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> >>> --- >>> drivers/pci/pci.c | 1 - >>> drivers/pci/quirks.c | 8 ++++---- >>> include/linux/pci.h | 2 ++ >>> 3 files changed, 6 insertions(+), 5 deletions(-) >>> >>> diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c >>> index 2c9ac70..76b002b1 100644 >>> --- a/drivers/pci/pci.c >>> +++ b/drivers/pci/pci.c >>> @@ -137,7 +137,6 @@ void __iomem *pci_ioremap_bar(struct pci_dev *pdev, int bar) >>> EXPORT_SYMBOL_GPL(pci_ioremap_bar); >>> #endif >>> >>> -#define PCI_FIND_CAP_TTL 48 >>> >>> static int __pci_find_next_cap_ttl(struct pci_bus *bus, unsigned int devfn, >>> u8 pos, int cap, int *ttl) >>> diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c >>> index 80c2d01..a5f46b8 100644 >>> --- a/drivers/pci/quirks.c >>> +++ b/drivers/pci/quirks.c >>> @@ -2176,7 +2176,7 @@ DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_AMD, 0x9601, quirk_amd_780_apc_msi); >>> * return 1 if a HT MSI capability is found and enabled */ >>> static int msi_ht_cap_enabled(struct pci_dev *dev) >>> { >>> - int pos, ttl = 48; >>> + int pos, ttl = PCI_FIND_CAP_TTL; >>> >>> pos = pci_find_ht_capability(dev, HT_CAPTYPE_MSI_MAPPING); >>> while (pos && ttl--) { >>> @@ -2235,7 +2235,7 @@ DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_CK804_PCIE, >>> /* Force enable MSI mapping capability on HT bridges */ >>> static void ht_enable_msi_mapping(struct pci_dev *dev) >>> { >>> - int pos, ttl = 48; >>> + int pos, ttl = PCI_FIND_CAP_TTL; >>> >>> pos = pci_find_ht_capability(dev, HT_CAPTYPE_MSI_MAPPING); >>> while (pos && ttl--) { >>> @@ -2314,7 +2314,7 @@ DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_NVIDIA, >>> >>> static int ht_check_msi_mapping(struct pci_dev *dev) >>> { >>> - int pos, ttl = 48; >>> + int pos, ttl = PCI_FIND_CAP_TTL; >>> int found = 0; >>> >>> /* check if there is HT MSI cap or enabled on this device */ >>> @@ -2439,7 +2439,7 @@ out: >>> >>> static void ht_disable_msi_mapping(struct pci_dev *dev) >>> { >>> - int pos, ttl = 48; >>> + int pos, ttl = PCI_FIND_CAP_TTL; >>> >>> pos = pci_find_ht_capability(dev, HT_CAPTYPE_MSI_MAPPING); >>> while (pos && ttl--) { >>> diff --git a/include/linux/pci.h b/include/linux/pci.h >>> index 96453f9..b27b79e 100644 >>> --- a/include/linux/pci.h >>> +++ b/include/linux/pci.h >>> @@ -33,6 +33,8 @@ >>> >>> #include <linux/pci_ids.h> >>> >>> +#define PCI_FIND_CAP_TTL 48 >>> + >>> /* >>> * The PCI interface treats multi-function devices as independent >>> * devices. The slot/function address of each device is encoded >>> -- >>> 1.7.9.5 >>> > >-- >Richard Yang >Help you, Help me
diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index 2c9ac70..76b002b1 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -137,7 +137,6 @@ void __iomem *pci_ioremap_bar(struct pci_dev *pdev, int bar) EXPORT_SYMBOL_GPL(pci_ioremap_bar); #endif -#define PCI_FIND_CAP_TTL 48 static int __pci_find_next_cap_ttl(struct pci_bus *bus, unsigned int devfn, u8 pos, int cap, int *ttl) diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index 80c2d01..a5f46b8 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -2176,7 +2176,7 @@ DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_AMD, 0x9601, quirk_amd_780_apc_msi); * return 1 if a HT MSI capability is found and enabled */ static int msi_ht_cap_enabled(struct pci_dev *dev) { - int pos, ttl = 48; + int pos, ttl = PCI_FIND_CAP_TTL; pos = pci_find_ht_capability(dev, HT_CAPTYPE_MSI_MAPPING); while (pos && ttl--) { @@ -2235,7 +2235,7 @@ DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_CK804_PCIE, /* Force enable MSI mapping capability on HT bridges */ static void ht_enable_msi_mapping(struct pci_dev *dev) { - int pos, ttl = 48; + int pos, ttl = PCI_FIND_CAP_TTL; pos = pci_find_ht_capability(dev, HT_CAPTYPE_MSI_MAPPING); while (pos && ttl--) { @@ -2314,7 +2314,7 @@ DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_NVIDIA, static int ht_check_msi_mapping(struct pci_dev *dev) { - int pos, ttl = 48; + int pos, ttl = PCI_FIND_CAP_TTL; int found = 0; /* check if there is HT MSI cap or enabled on this device */ @@ -2439,7 +2439,7 @@ out: static void ht_disable_msi_mapping(struct pci_dev *dev) { - int pos, ttl = 48; + int pos, ttl = PCI_FIND_CAP_TTL; pos = pci_find_ht_capability(dev, HT_CAPTYPE_MSI_MAPPING); while (pos && ttl--) { diff --git a/include/linux/pci.h b/include/linux/pci.h index 96453f9..b27b79e 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -33,6 +33,8 @@ #include <linux/pci_ids.h> +#define PCI_FIND_CAP_TTL 48 + /* * The PCI interface treats multi-function devices as independent * devices. The slot/function address of each device is encoded
In some quirks, it tries to search a pci cap and use a ttl value to avoid infinite loop. While the value is hard coded to 48, which is the same as marco PCI_FIND_CAP_TTL. This patch moves the definition of PCI_FIND_CAP_TTL to pci.h and replace the hard coded value with it. Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> --- drivers/pci/pci.c | 1 - drivers/pci/quirks.c | 8 ++++---- include/linux/pci.h | 2 ++ 3 files changed, 6 insertions(+), 5 deletions(-)