Thread (106 messages) 106 messages, 16 authors, 2015-03-10

[PATCH v9 01/21] ACPI / table: Use pr_debug() instead of pr_info() for MADT table scanning

From: Grant Likely <hidden>
Date: 2015-03-06 20:17:31
Also in: linux-acpi, lkml

On Wed, 25 Feb 2015 16:39:41 +0800
, Hanjun Guo [off-list ref]
 wrote:
For a normal 8 cpu sockets system, it will up to 240 cpu threads (Xeon E7
v2 family for now), and we need 240 entries for local apic or local x2apic
in MADT table, so it will be much verbose information printed with a slow
uart console when system booted, this will be even worse with large system
with 16/32 cpu sockets.

This patch just use pr_debug() instead of pr_info() for ioapic/iosapic,
local apic/x2apic/sapic structures when scanning the MADT table to remove
those verbose information, but leave other structures unchanged.

CC: Rafael J Wysocki <redacted>
Signed-off-by: Hanjun Guo <redacted>
One nitpick below, but don't respin over this, and don't do a fixup.

Acked-by: Grant Likely <redacted>
quoted hunk ↗ jump to hunk
---
 drivers/acpi/tables.c | 30 ++++++++++++++++--------------
 1 file changed, 16 insertions(+), 14 deletions(-)
diff --git a/drivers/acpi/tables.c b/drivers/acpi/tables.c
index 93b8152..f4e5b88 100644
--- a/drivers/acpi/tables.c
+++ b/drivers/acpi/tables.c
@@ -23,6 +23,8 @@
  *
  */
 
+/* Uncomment next line to get verbose printout */
+/* #define DEBUG */
 #define pr_fmt(fmt) "ACPI: " fmt
 
 #include <linux/init.h>
@@ -61,9 +63,9 @@ void acpi_table_print_madt_entry(struct acpi_subtable_header *header)
 		{
 			struct acpi_madt_local_apic *p =
 			    (struct acpi_madt_local_apic *)header;
-			pr_info("LAPIC (acpi_id[0x%02x] lapic_id[0x%02x] %s)\n",
-				p->processor_id, p->id,
-				(p->lapic_flags & ACPI_MADT_ENABLED) ? "enabled" : "disabled");
+			pr_debug("LAPIC (acpi_id[0x%02x] lapic_id[0x%02x] %s)\n",
+				 p->processor_id, p->id,
+				 (p->lapic_flags & ACPI_MADT_ENABLED) ? "enabled" : "disabled");
The whitespace changes makes each 1 line change into 3 line changes. In
these situations, I would chose to leave the whitespace alone to keep
the diffstat as small as possible. It makes it less likely to conflict
with other patches and easier to find context.
quoted hunk ↗ jump to hunk
 		}
 		break;
 
@@ -71,9 +73,9 @@ void acpi_table_print_madt_entry(struct acpi_subtable_header *header)
 		{
 			struct acpi_madt_local_x2apic *p =
 			    (struct acpi_madt_local_x2apic *)header;
-			pr_info("X2APIC (apic_id[0x%02x] uid[0x%02x] %s)\n",
-				p->local_apic_id, p->uid,
-				(p->lapic_flags & ACPI_MADT_ENABLED) ? "enabled" : "disabled");
+			pr_debug("X2APIC (apic_id[0x%02x] uid[0x%02x] %s)\n",
+				 p->local_apic_id, p->uid,
+				 (p->lapic_flags & ACPI_MADT_ENABLED) ? "enabled" : "disabled");
 		}
 		break;
 
@@ -81,8 +83,8 @@ void acpi_table_print_madt_entry(struct acpi_subtable_header *header)
 		{
 			struct acpi_madt_io_apic *p =
 			    (struct acpi_madt_io_apic *)header;
-			pr_info("IOAPIC (id[0x%02x] address[0x%08x] gsi_base[%d])\n",
-				p->id, p->address, p->global_irq_base);
+			pr_debug("IOAPIC (id[0x%02x] address[0x%08x] gsi_base[%d])\n",
+				 p->id, p->address, p->global_irq_base);
 		}
 		break;
 
@@ -155,9 +157,9 @@ void acpi_table_print_madt_entry(struct acpi_subtable_header *header)
 		{
 			struct acpi_madt_io_sapic *p =
 			    (struct acpi_madt_io_sapic *)header;
-			pr_info("IOSAPIC (id[0x%x] address[%p] gsi_base[%d])\n",
-				p->id, (void *)(unsigned long)p->address,
-				p->global_irq_base);
+			pr_debug("IOSAPIC (id[0x%x] address[%p] gsi_base[%d])\n",
+				 p->id, (void *)(unsigned long)p->address,
+				 p->global_irq_base);
 		}
 		break;
 
@@ -165,9 +167,9 @@ void acpi_table_print_madt_entry(struct acpi_subtable_header *header)
 		{
 			struct acpi_madt_local_sapic *p =
 			    (struct acpi_madt_local_sapic *)header;
-			pr_info("LSAPIC (acpi_id[0x%02x] lsapic_id[0x%02x] lsapic_eid[0x%02x] %s)\n",
-				p->processor_id, p->id, p->eid,
-				(p->lapic_flags & ACPI_MADT_ENABLED) ? "enabled" : "disabled");
+			pr_debug("LSAPIC (acpi_id[0x%02x] lsapic_id[0x%02x] lsapic_eid[0x%02x] %s)\n",
+				 p->processor_id, p->id, p->eid,
+				 (p->lapic_flags & ACPI_MADT_ENABLED) ? "enabled" : "disabled");
 		}
 		break;
 
-- 
1.9.1
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help