x86, acpi: Remove out-of-date comment of __acpi_map_table

PATCH Superseded Zhang Yanfei 2 participants 3 messages Jul 10, 2013 – Jul 23, 2013

Subsystems (2)

x86 ACPI

lore

Share & tools

lore

Zhang Yanfei

author

2013-07-10 18:23 UTC #0 lore raw

PATCH 1/1 x86, acpi: Remove out-of-date comment of __acpi_map_table

From: Zhang Yanfei <[email protected]>

The implementation of function __acpi_map_table() has been changed long time ago, and now it directly invokes early_ioremap() to setup the temporarily acpi table mappings. So remove its out-of-date comment.

Signed-off-by Zhang Yanfei <[email protected]>

---
 arch/x86/kernel/acpi/boot.c |   13 +------------
 1 files changed, 1 insertions(+), 12 deletions(-)

arch/x86/kernel/acpi/boot.c +1 ~~-12~~

index d81a972..bac8b5b 100644
@@ -140,18 +140,6 @@ static u32 irq_to_gsi(int irq)
 	return gsi;
 }
 
-/*
- * Temporarily use the virtual area starting from FIX_IO_APIC_BASE_END,
- * to map the target physical address. The problem is that set_fixmap()
- * provides a single page, and it is possible that the page is not
- * sufficient.
- * By using this area, we can map up to MAX_IO_APICS pages temporarily,
- * i.e. until the next __va_range() call.
- *
- * Important Safety Note:  The fixed I/O APIC page numbers are *subtracted*
- * from the fixed base.  That's why we start at FIX_IO_APIC_BASE_END and
- * count idx down while incrementing the phys address.
- */
 char *__init __acpi_map_table(unsigned long phys, unsigned long size)
 {

@@ -160,6 +148,7 @@ char *__init __acpi_map_table(unsigned long phys, unsigned long size)
 
 	return early_ioremap(phys, size);
 }
+
 void __init __acpi_unmap_table(char *map, unsigned long size)
 {
 	if (!map || !size)

Signature · 1 line

1.7.1

Ingo Molnar

2013-07-23 09:22 UTC #1 lore raw

Quoted 23 lines · Zhang Yanfei

From: Zhang Yanfei <[email protected]>
The implementation of function __acpi_map_table() has been changed long time ago, and now it directly invokes early_ioremap() to setup the temporarily acpi table mappings. So remove its out-of-date comment.
Signed-off-by Zhang Yanfei <[email protected]>
---
 arch/x86/kernel/acpi/boot.c |   13 +------------
 1 files changed, 1 insertions(+), 12 deletions(-)
arch/x86/kernel/acpi/boot.c +0 -3
index d81a972..bac8b5b 100644
@@ -140,18 +140,6 @@ static u32 irq_to_gsi(int irq)
 	return gsi;
 }
 
-/*
- * Temporarily use the virtual area starting from FIX_IO_APIC_BASE_END,
- * to map the target physical address. The problem is that set_fixmap()

arch/x86/kernel/acpi/boot.c static u32 irq_to_gsi(int irq) · line 143

Quoted 10 lines

arch/x86/kernel/acpi/boot.c +0 -9

- * provides a single page, and it is possible that the page is not
- * sufficient.
- * By using this area, we can map up to MAX_IO_APICS pages temporarily,
- * i.e. until the next __va_range() call.
- *
- * Important Safety Note:  The fixed I/O APIC page numbers are *subtracted*
- * from the fixed base.  That's why we start at FIX_IO_APIC_BASE_END and
- * count idx down while incrementing the phys address.
- */
 char *__init __acpi_map_table(unsigned long phys, unsigned long size)

Good catch. Removing documentation makes me sad, so how about replacing it with a simple comment:

	/*
	 * This is just a simple wrapper around early_ioremap(),
	 * with sanity checks for phys == 0 and size == 0:
	 */

Thanks,

Ingo

Zhang Yanfei

author

2013-07-23 12:50 UTC #2 lore raw

Quoted 40 lines · Ingo Molnar

Quoted 23 lines · Zhang Yanfei
From: Zhang Yanfei <[email protected]>
The implementation of function __acpi_map_table() has been changed long time ago, and now it directly invokes early_ioremap() to setup the temporarily acpi table mappings. So remove its out-of-date comment.
Signed-off-by Zhang Yanfei <[email protected]>
---
 arch/x86/kernel/acpi/boot.c |   13 +------------
 1 files changed, 1 insertions(+), 12 deletions(-)
arch/x86/kernel/acpi/boot.c +0 -3
index d81a972..bac8b5b 100644
@@ -140,18 +140,6 @@ static u32 irq_to_gsi(int irq)
 	return gsi;
 }
 
-/*
- * Temporarily use the virtual area starting from FIX_IO_APIC_BASE_END,
- * to map the target physical address. The problem is that set_fixmap()
arch/x86/kernel/acpi/boot.c static u32 irq_to_gsi(int irq) · line 143
Quoted 10 lines
arch/x86/kernel/acpi/boot.c +0 -9
- * provides a single page, and it is possible that the page is not
- * sufficient.
- * By using this area, we can map up to MAX_IO_APICS pages temporarily,
- * i.e. until the next __va_range() call.
- *
- * Important Safety Note:  The fixed I/O APIC page numbers are *subtracted*
- * from the fixed base.  That's why we start at FIX_IO_APIC_BASE_END and
- * count idx down while incrementing the phys address.
- */
 char *__init __acpi_map_table(unsigned long phys, unsigned long size)
Good catch. Removing documentation makes me sad, so how about replacing it with a simple comment:

Quoted 6 lines

	/*
	 * This is just a simple wrapper around early_ioremap(),
	 * with sanity checks for phys == 0 and size == 0:
	 */

Yeah, agreed. So I will send the v2 patch.

Quoted 4 lines

Thanks,
Ingo

Signature · 2 lines

Thanks.
Zhang Yanfei