arch/powerpc/platforms/powernv/pci-ioda.c:1164:3: error: implic

1164 if (pci_pcie_type(bridge) = PCI_EXP_TYPE_PCI_BRIDGE) {

Hi Gavin,

FYI, kernel build failed on

tree:   git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci.git pci/gavin-window-alignment
head:   4814ef2ec1b8ebaabfc54bb50578313c2f94119a
commit: 4814ef2ec1b8ebaabfc54bb50578313c2f94119a [5/5] powerpc/powernv: I/O and memory alignment for P2P bridges
config: powerpc-allmodconfig

All related error/warning messages:

arch/powerpc/platforms/powernv/pci-ioda.c: In function 'pnv_pci_window_alignment':
arch/powerpc/platforms/powernv/pci-ioda.c:1164:3: error: implicit declaration of function 'pci_pcie_type' [-Werror=implicit-function-declaration]
cc1: some warnings being treated as errors

vim +1164 arch/powerpc/platforms/powernv/pci-ioda.c
1159 struct pnv_phb *phb = hose->private_data;
1160 int num_pci_bridges = 0;
1161
1162 bridge = bus->self;
1163 while (bridge) {

Quoted 1 line

1164 if (pci_pcie_type(bridge) = PCI_EXP_TYPE_PCI_BRIDGE) {

1165 num_pci_bridges++;
1166 if (num_pci_bridges >= 2)
1167 return 1;
1168 }
1169

---
0-DAY kernel build testing backend Open Source Technology Centre
Fengguang Wu <[email protected]> Intel Corporation

So I did that rebase and pushed both pci/gavin-window-alignment and next again. I think the result is sensible now.

Sorry, this is my fault. I fixed this earlier, but I did it the wrong way.

What I did before was just update the patch in the pci/gavin-window-alignment branch. But what I should have done was to rebase that branch on top of pci/jiang-pcie-cap, which the branch it actually depends on (that's where pci_pcie_type() was added).

So I did that rebase and pushed both pci/gavin-window-alignment and next again. I think the result is sensible now.

Quoted 10 lines · Bjorn Helgaas

Sorry, this is my fault. I fixed this earlier, but I did it the wrong way.

What I did before was just update the patch in the pci/gavin-window-alignment branch. But what I should have done was to rebase that branch on top of pci/jiang-pcie-cap, which the branch it actually depends on (that's where pci_pcie_type() was added).

So I did that rebase and pushed both pci/gavin-window-alignment and next again. I think the result is sensible now.

The result has content conflicts against a couple of drivers. If I fix that up locally that means that things might end up being a bit more complicated than expected when reconciling everything with Linus and that worries me.

Any chance you can pull in a recent Linus tree & fixup the merges before I pull that in myself ?

I merged v3.6-rc5 into my "next" branch, fixed the conflicts in

drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c drivers/rapidio/devices/tsi721.c

and pushed the result. Let me know if this is not what you expected.