Re: PCI: Add stub pci_irq_vector and others

2019-09-20 Thread Herbert Xu
On Fri, Sep 20, 2019 at 02:42:16PM -0500, Bjorn Helgaas wrote: > On Wed, Sep 04, 2019 at 10:26:00PM +1000, Herbert Xu wrote: > > On Wed, Sep 04, 2019 at 05:10:34AM -0700, Ard Biesheuvel wrote: > > > > > > This is the reason we have so many empty static inline functions in > > > header files - it en

Re: PCI: Add stub pci_irq_vector and others

2019-09-20 Thread Bjorn Helgaas
On Wed, Sep 04, 2019 at 10:26:00PM +1000, Herbert Xu wrote: > On Wed, Sep 04, 2019 at 05:10:34AM -0700, Ard Biesheuvel wrote: > > > > This is the reason we have so many empty static inline functions in > > header files - it ensures that the symbols are declared even if the > > only invocations are

Re: PCI: Add stub pci_irq_vector and others

2019-09-05 Thread Herbert Xu
On Thu, Sep 05, 2019 at 04:07:22PM -0500, Bjorn Helgaas wrote: > > > This patch adds stub functions pci_alloc_irq_vectors_affinity and > > pci_irq_vector when CONFIG_PCI is off so that drivers can use them > > without resorting to ifdefs. > > > > It also moves the PCI_IRQ_* macros outside of the i

Re: PCI: Add stub pci_irq_vector and others

2019-09-05 Thread Bjorn Helgaas
On Wed, Sep 04, 2019 at 10:26:00PM +1000, Herbert Xu wrote: > On Wed, Sep 04, 2019 at 05:10:34AM -0700, Ard Biesheuvel wrote: > > > > This is the reason we have so many empty static inline functions in > > header files - it ensures that the symbols are declared even if the > > only invocations are

Re: PCI: Add stub pci_irq_vector and others

2019-09-04 Thread Yuehaibing
On 2019/9/4 20:26, Herbert Xu wrote: > On Wed, Sep 04, 2019 at 05:10:34AM -0700, Ard Biesheuvel wrote: >> >> This is the reason we have so many empty static inline functions in >> header files - it ensures that the symbols are declared even if the >> only invocations are from dead code. > > Does t

PCI: Add stub pci_irq_vector and others

2019-09-04 Thread Herbert Xu
On Wed, Sep 04, 2019 at 05:10:34AM -0700, Ard Biesheuvel wrote: > > This is the reason we have so many empty static inline functions in > header files - it ensures that the symbols are declared even if the > only invocations are from dead code. Does this patch work? ---8<--- This patch adds stub