On 2017/8/25 18:37, gengdongjiu wrote:

quoted

quoted

+

quoted

quoted

+/* From the ACPI 6.1 spec, "18.3.2.9 Hardware Error Notification" */
+

It's better to refer to the first spec version of this structure and
same with others you define.

 do you mean which spec version? the definition is aligned with the linux kernel.

What I mean here is that it's better to refer to the ACPI spec version
which introduces Hardware Error Notification first time.

quoted

quoted

quoted

quoted

+enum AcpiHestNotifyType {
+    ACPI_HEST_NOTIFY_POLLED = 0,
+    ACPI_HEST_NOTIFY_EXTERNAL = 1,
+    ACPI_HEST_NOTIFY_LOCAL = 2,
+    ACPI_HEST_NOTIFY_SCI = 3,
+    ACPI_HEST_NOTIFY_NMI = 4,
+    ACPI_HEST_NOTIFY_CMCI = 5,  /* ACPI 5.0 */
+    ACPI_HEST_NOTIFY_MCE = 6,   /* ACPI 5.0 */
+    ACPI_HEST_NOTIFY_GPIO = 7,  /* ACPI 6.0 */
+    ACPI_HEST_NOTIFY_SEA = 8,   /* ACPI 6.1 */
+    ACPI_HEST_NOTIFY_SEI = 9,   /* ACPI 6.1 */
+    ACPI_HEST_NOTIFY_GSIV = 10, /* ACPI 6.1 */
+    ACPI_HEST_NOTIFY_RESERVED = 11  /* 11 and greater are reserved */

In ACPI 6.2, 11 is for Software Delegated Exception, is this useful for
your patchset?

  it is usefull, for all the error source, I reserved the space for them.
Because the space is allocated one time, is not dynamically allocated.
so I use the ACPI_HEST_NOTIFY_RESERVED to specify that there is 11 error source.

I mean whether the new type Software Delegated Exception is useful for
RAS. If so, we could add this new type here.

Thanks,
-- 
Shannon