[PATCH] PCI: Fix regression in powerpc MSI-X

View: New views
4 Messages — Rating Filter:   Alert me  

[PATCH] PCI: Fix regression in powerpc MSI-X

by André Detsch :: Rate this Message:

Reply to Author | View Threaded | Show Only this Message

Patch f598282f5145036312d90875d0ed5c14b49fd8a7 exposed a problem in
powerpc MSI-X functionality, making network interfaces such as ixgbe
and cxgb3 stop to work when MSI-X is enabled. RX interrupts were not
being generated.

The problem was caused because MSI irq was not being effectively
unmasked after device initialization.

Signed-off-by: Andre Detsch <adetsch@...>
Signed-off-by: Michael Ellerman <michael@...>

Index: linux-2.6/arch/powerpc/platforms/pseries/msi.c
===================================================================
--- linux-2.6.orig/arch/powerpc/platforms/pseries/msi.c 2009-11-04
06:35:39.000000000 -0700
+++ linux-2.6/arch/powerpc/platforms/pseries/msi.c 2009-11-04
07:23:27.000000000 -0700
@@ -432,8 +432,6 @@ static int rtas_setup_msi_irqs(struct pc
  /* Read config space back so we can restore after reset */
  read_msi_msg(virq, &msg);
  entry->msg = msg;
-
- unmask_msi_irq(virq);
  }
 
  return 0;
Index: linux-2.6/arch/powerpc/platforms/pseries/xics.c
===================================================================
--- linux-2.6.orig/arch/powerpc/platforms/pseries/xics.c 2009-11-04
06:35:39.000000000 -0700
+++ linux-2.6/arch/powerpc/platforms/pseries/xics.c 2009-11-04
07:23:27.000000000 -0700
@@ -18,6 +18,7 @@
 #include <linux/init.h>
 #include <linux/radix-tree.h>
 #include <linux/cpu.h>
+#include <linux/msi.h>
 #include <linux/of.h>
 
 #include <asm/firmware.h>
@@ -219,6 +220,14 @@ static void xics_unmask_irq(unsigned int
 
 static unsigned int xics_startup(unsigned int virq)
 {
+ /*
+ * The generic MSI code returns with the interrupt disabled on the
+ * card, using the MSI mask bits. Firmware doesn't appear to unmask
+ * at that level, so we do it here by hand.
+ */
+ if (irq_to_desc(virq)->msi_desc)
+ unmask_msi_irq(virq);
+
  /* unmask it */
  xics_unmask_irq(virq);
  return 0;
_______________________________________________
Linuxppc-dev mailing list
Linuxppc-dev@...
https://lists.ozlabs.org/listinfo/linuxppc-dev

Re: [PATCH] PCI: Fix regression in powerpc MSI-X

by Michael Ellerman :: Rate this Message:

Reply to Author | View Threaded | Show Only this Message

On Wed, 2009-11-04 at 13:03 -0200, Andre Detsch wrote:
> Patch f598282f5145036312d90875d0ed5c14b49fd8a7 exposed a problem in
> powerpc MSI-X functionality, making network interfaces such as ixgbe
> and cxgb3 stop to work when MSI-X is enabled. RX interrupts were not
> being generated.

Which means we'd like it in 32 please Ben.

Also Andre, once Ben has merged this you should forward it to
stable@... for the 31 stable series.

cheers


_______________________________________________
Linuxppc-dev mailing list
Linuxppc-dev@...
https://lists.ozlabs.org/listinfo/linuxppc-dev

signature.asc (204 bytes) Download Attachment

Re: [PATCH] PCI: Fix regression in powerpc MSI-X

by Benjamin Herrenschmidt :: Rate this Message:

Reply to Author | View Threaded | Show Only this Message

On Wed, 2009-11-04 at 13:03 -0200, Andre Detsch wrote:
> Patch f598282f5145036312d90875d0ed5c14b49fd8a7 exposed a problem in
> powerpc MSI-X functionality, making network interfaces such as ixgbe
> and cxgb3 stop to work when MSI-X is enabled. RX interrupts were not
> being generated.
>
> The problem was caused because MSI irq was not being effectively
> unmasked after device initialization.

Patch is wrapped...

Ben.

> Signed-off-by: Andre Detsch <adetsch@...>
> Signed-off-by: Michael Ellerman <michael@...>
>
> Index: linux-2.6/arch/powerpc/platforms/pseries/msi.c
> ===================================================================
> --- linux-2.6.orig/arch/powerpc/platforms/pseries/msi.c 2009-11-04
> 06:35:39.000000000 -0700
> +++ linux-2.6/arch/powerpc/platforms/pseries/msi.c 2009-11-04
> 07:23:27.000000000 -0700
> @@ -432,8 +432,6 @@ static int rtas_setup_msi_irqs(struct pc
>   /* Read config space back so we can restore after reset */
>   read_msi_msg(virq, &msg);
>   entry->msg = msg;
> -
> - unmask_msi_irq(virq);
>   }
>  
>   return 0;
> Index: linux-2.6/arch/powerpc/platforms/pseries/xics.c
> ===================================================================
> --- linux-2.6.orig/arch/powerpc/platforms/pseries/xics.c 2009-11-04
> 06:35:39.000000000 -0700
> +++ linux-2.6/arch/powerpc/platforms/pseries/xics.c 2009-11-04
> 07:23:27.000000000 -0700
> @@ -18,6 +18,7 @@
>  #include <linux/init.h>
>  #include <linux/radix-tree.h>
>  #include <linux/cpu.h>
> +#include <linux/msi.h>
>  #include <linux/of.h>
>  
>  #include <asm/firmware.h>
> @@ -219,6 +220,14 @@ static void xics_unmask_irq(unsigned int
>  
>  static unsigned int xics_startup(unsigned int virq)
>  {
> + /*
> + * The generic MSI code returns with the interrupt disabled on the
> + * card, using the MSI mask bits. Firmware doesn't appear to unmask
> + * at that level, so we do it here by hand.
> + */
> + if (irq_to_desc(virq)->msi_desc)
> + unmask_msi_irq(virq);
> +
>   /* unmask it */
>   xics_unmask_irq(virq);
>   return 0;
> _______________________________________________
> Linuxppc-dev mailing list
> Linuxppc-dev@...
> https://lists.ozlabs.org/listinfo/linuxppc-dev


_______________________________________________
Linuxppc-dev mailing list
Linuxppc-dev@...
https://lists.ozlabs.org/listinfo/linuxppc-dev

[PATCH] PCI: Fix regression in powerpc MSI-X

by André Detsch :: Rate this Message:

Reply to Author | View Threaded | Show Only this Message

Patch f598282f5145036312d90875d0ed5c14b49fd8a7 exposed a problem in
powerpc MSI-X functionality, making network interfaces such as ixgbe
and cxgb3 stop to work when MSI-X is enabled. RX interrupts were not
being generated.

The problem was caused because MSI irq was not being effectively
unmasked after device initialization.

Signed-off-by: Andre Detsch <adetsch@...>
Signed-off-by: Michael Ellerman <michael@...>

Index: linux-2.6/arch/powerpc/platforms/pseries/msi.c
===================================================================
--- linux-2.6.orig/arch/powerpc/platforms/pseries/msi.c 2009-11-04 06:35:39.000000000 -0700
+++ linux-2.6/arch/powerpc/platforms/pseries/msi.c 2009-11-04 07:23:27.000000000 -0700
@@ -432,8 +432,6 @@ static int rtas_setup_msi_irqs(struct pc
  /* Read config space back so we can restore after reset */
  read_msi_msg(virq, &msg);
  entry->msg = msg;
-
- unmask_msi_irq(virq);
  }
 
  return 0;
Index: linux-2.6/arch/powerpc/platforms/pseries/xics.c
===================================================================
--- linux-2.6.orig/arch/powerpc/platforms/pseries/xics.c 2009-11-04 06:35:39.000000000 -0700
+++ linux-2.6/arch/powerpc/platforms/pseries/xics.c 2009-11-04 07:23:27.000000000 -0700
@@ -18,6 +18,7 @@
 #include <linux/init.h>
 #include <linux/radix-tree.h>
 #include <linux/cpu.h>
+#include <linux/msi.h>
 #include <linux/of.h>
 
 #include <asm/firmware.h>
@@ -219,6 +220,14 @@ static void xics_unmask_irq(unsigned int
 
 static unsigned int xics_startup(unsigned int virq)
 {
+ /*
+ * The generic MSI code returns with the interrupt disabled on the
+ * card, using the MSI mask bits. Firmware doesn't appear to unmask
+ * at that level, so we do it here by hand.
+ */
+ if (irq_to_desc(virq)->msi_desc)
+ unmask_msi_irq(virq);
+
  /* unmask it */
  xics_unmask_irq(virq);
  return 0;
_______________________________________________
Linuxppc-dev mailing list
Linuxppc-dev@...
https://lists.ozlabs.org/listinfo/linuxppc-dev