On Mon, Jul 25, 2016 at 10:53:49AM -0700, Tyrel Datwyler wrote:

On 07/21/2016 11:36 PM, Gavin Shan wrote:

quoted

On Fri, Jul 22, 2016 at 03:23:36PM +1000, Russell Currey wrote:

quoted

On EEH events the kernel will print a dump of relevant registers.
If EEH is unavailable (i.e. CONFIG_EEH is disabled, a new platform
doesn't have EEH support, etc) this information isn't readily available.

Add a new debugfs handler to trigger a PHB register dump, so that this
information can be made available on demand.

Signed-off-by: Russell Currey <redacted>

Reviewed-by: Gavin Shan <redacted>

quoted

---
arch/powerpc/platforms/powernv/pci-ioda.c | 35 +++++++++++++++++++++++++++++++
1 file changed, 35 insertions(+)

diff --git a/arch/powerpc/platforms/powernv/pci-ioda.c b/arch/powerpc/platforms/powernv/pci-ioda.c
index 891fc4a..ada2f3c 100644
--- a/arch/powerpc/platforms/powernv/pci-ioda.c
+++ b/arch/powerpc/platforms/powernv/pci-ioda.c

@@ -3018,6 +3018,38 @@ static void pnv_ioda_setup_pe_seg(struct pnv_ioda_pe *pe)

	}
}

+#ifdef CONFIG_DEBUG_FS
+static ssize_t pnv_pci_debug_write(struct file *filp,
+				   const char __user *user_buf,
+				   size_t count, loff_t *ppos)
+{
+	struct pci_controller *hose = filp->private_data;
+	struct pnv_phb *phb;
+	int ret = 0;

Needn't initialize @ret in advance. The code might be simpler, but it's
only a personal preference:

I believe its actually preferred that it not be initialized in advance
so that the tooling can warn you about conditional code paths where you
may have forgotten to set a value. Or as Gavin suggests to explicitly
use error values in the return statements.

Yeah, the data type should be int64_t as well.

Thanks,
Gavin