Message ID | 1440332257-15894-1-git-send-email-bmeng.cn@gmail.com |
---|---|
State | Superseded |
Delegated to: | Simon Glass |
Headers | show |
Hi Bin, On 23 August 2015 at 06:17, Bin Meng <bmeng.cn@gmail.com> wrote: > > If there is no pci device listed in the device tree, > don't bother scanning the device tree. > > Signed-off-by: Bin Meng <bmeng.cn@gmail.com> > > --- > Simon, I see the following v2 patches were applied, but don't see > you replied the patch thread to mention it's been applied. > - x86: fsp: Call fsp_init_phase_pci() in pci_uclass_post_probe() > - fdtdec: Fix possible infinite loop in fdtdec_get_pci_vendev() > - dm: pci: Save devfn without bus number in pci_uclass_child_post_bind() > > I assume these patches look good, so I don't include those patches > in the v3 series. Let me know if there is any issue. They are fine, thanks. > > Changes in v3 respin: > - Fix typo of 'tree' > > Changes in v3: > - Rebase on u-boot-x86/master > - Drop v2 patches which were already applied > > drivers/pci/pci-uclass.c | 7 +++++++ > 1 file changed, 7 insertions(+) > > diff --git a/drivers/pci/pci-uclass.c b/drivers/pci/pci-uclass.c > index 2d12344..b25298f 100644 > --- a/drivers/pci/pci-uclass.c > +++ b/drivers/pci/pci-uclass.c > @@ -632,6 +632,13 @@ error: > static int pci_uclass_post_bind(struct udevice *bus) > { > /* > + * If there is no pci device listed in the device tree, > + * don't bother scanning the device tree. > + */ > + if (bus->of_offset == -1) > + return 0; Since this is a special case for PCI I think it is best that this check goes here instead of dm_scan_fdt_node(). Acked-by: Simon Glass <sjg@chromium.org> > + > + /* > * Scan the device tree for devices. This does not probe the PCI bus, > * as this is not permitted while binding. It just finds devices > * mentioned in the device tree. > -- > 1.8.2.1 > Regards, Simon
diff --git a/drivers/pci/pci-uclass.c b/drivers/pci/pci-uclass.c index 2d12344..b25298f 100644 --- a/drivers/pci/pci-uclass.c +++ b/drivers/pci/pci-uclass.c @@ -632,6 +632,13 @@ error: static int pci_uclass_post_bind(struct udevice *bus) { /* + * If there is no pci device listed in the device tree, + * don't bother scanning the device tree. + */ + if (bus->of_offset == -1) + return 0; + + /* * Scan the device tree for devices. This does not probe the PCI bus, * as this is not permitted while binding. It just finds devices * mentioned in the device tree.
If there is no pci device listed in the device tree, don't bother scanning the device tree. Signed-off-by: Bin Meng <bmeng.cn@gmail.com> --- Simon, I see the following v2 patches were applied, but don't see you replied the patch thread to mention it's been applied. - x86: fsp: Call fsp_init_phase_pci() in pci_uclass_post_probe() - fdtdec: Fix possible infinite loop in fdtdec_get_pci_vendev() - dm: pci: Save devfn without bus number in pci_uclass_child_post_bind() I assume these patches look good, so I don't include those patches in the v3 series. Let me know if there is any issue. Changes in v3 respin: - Fix typo of 'tree' Changes in v3: - Rebase on u-boot-x86/master - Drop v2 patches which were already applied drivers/pci/pci-uclass.c | 7 +++++++ 1 file changed, 7 insertions(+)