Re: powerpc/pci: Use of_irq_parse_and_map_pci() helper

[Rob Herring]

On Fri, Feb 9, 2018 at 4:01 PM, Mathieu Malaterre  wrote:
> Hi Rob,
>
> I can no longer boot my old Mac Mini G4 using a custom config. After a
> tedious bisect it appears that (*). Since Mac Mini does not provide
> any serial port I cannot provide a kernel message dump (wish there was
> some kind of null modem serial cable). All I can see is some crazy
> stuff going on with ATA DMA/66 & DMA/100.
>
> Is there anything suspicious in this commit which may prevent this
> machine to boot ?

Yes, it's broken. You need this patch[1]. Sorry for the breakage.

Rob

[1] https://patchwork.kernel.org/patch/10208267/

powerpc/pci: Use of_irq_parse_and_map_pci() helper

[Mathieu Malaterre]

Hi Rob,

I can no longer boot my old Mac Mini G4 using a custom config. After a
tedious bisect it appears that (*). Since Mac Mini does not provide
any serial port I cannot provide a kernel message dump (wish there was
some kind of null modem serial cable). All I can see is some crazy
stuff going on with ATA DMA/66 & DMA/100.

Is there anything suspicious in this commit which may prevent this
machine to boot ?

Thanks much,

(*)
$ git bisect good
59f47eff03a08cd2d91310f1c15a5343fa0071e5 is the first bad commit
commit 59f47eff03a08cd2d91310f1c15a5343fa0071e5
Author: Rob Herring <r...@kernel.org>
Date:   Thu Jan 4 15:12:14 2018 -0600

    powerpc/pci: Use of_irq_parse_and_map_pci() helper

Instead of calling both of_irq_parse_pci() and irq_create_of_mapping(),
call of_irq_parse_and_map_pci(), which does the same thing. This will allow
making of_irq_parse_pci() a private, static function.

This changes the logic slightly in that the fallback path will also be
taken if irq_create_of_mapping() fails internally.

Signed-off-by: Rob Herring <r...@kernel.org>
[bhelgaas: fold in virq init from Stephen Rothwell <s...@canb.auug.org.au>]
Signed-off-by: Bjorn Helgaas <bhelg...@google.com>
Cc: Benjamin Herrenschmidt <b...@kernel.crashing.org>
Cc: Paul Mackerras <pau...@samba.org>
Cc: Michael Ellerman <m...@ellerman.id.au>

:04 04 b979406deb8957af240d659be1e76ed75fa43288
ec3c127fe30428955e2ad1d716c9be2adbd2623d M arch

[PATCH 2/3] powerpc/pci: use of_irq_parse_and_map_pci helper

[Rob Herring]

Instead of calling both of_irq_parse_pci and irq_create_of_mapping, call
of_irq_parse_and_map_pci instead which does the same thing. This will
allow making of_irq_parse_pci a private, static function.

This changes the logic slightly in that the fallback path will also be
taken if irq_create_of_mapping fails internally.

Cc: Benjamin Herrenschmidt 
Cc: Paul Mackerras 
Cc: Michael Ellerman 
Cc: linuxppc-dev@lists.ozlabs.org
Signed-off-by: Rob Herring 
---
 arch/powerpc/kernel/pci-common.c | 8 +---
 1 file changed, 1 insertion(+), 7 deletions(-)

diff --git a/arch/powerpc/kernel/pci-common.c b/arch/powerpc/kernel/pci-common.c
index 0ac7aa346c69..f65c07dec23e 100644
--- a/arch/powerpc/kernel/pci-common.c
+++ b/arch/powerpc/kernel/pci-common.c
@@ -339,7 +339,6 @@ struct pci_controller* pci_find_hose_for_OF_device(struct 
device_node* node)
  */
 static int pci_read_irq_line(struct pci_dev *pci_dev)
 {
-   struct of_phandle_args oirq;
unsigned int virq;
 
pr_debug("PCI: Try to map irq for %s...\n", pci_name(pci_dev));
@@ -348,7 +347,7 @@ static int pci_read_irq_line(struct pci_dev *pci_dev)
memset(, 0xff, sizeof(oirq));
 #endif
/* Try to get a mapping from the device-tree */
-   if (of_irq_parse_pci(pci_dev, )) {
+   if (!of_irq_parse_and_map_pci(pci_dev, 0, 0)) {
u8 line, pin;
 
/* If that fails, lets fallback to what is in the config
@@ -372,11 +371,6 @@ static int pci_read_irq_line(struct pci_dev *pci_dev)
virq = irq_create_mapping(NULL, line);
if (virq)
irq_set_irq_type(virq, IRQ_TYPE_LEVEL_LOW);
-   } else {
-   pr_debug(" Got one, spec %d cells (0x%08x 0x%08x...) on %pOF\n",
-oirq.args_count, oirq.args[0], oirq.args[1], oirq.np);
-
-   virq = irq_create_of_mapping();
}
 
if (!virq) {
-- 
2.14.1