2013-05-09 15:57:47

by Zhang Yanfei

[permalink] [raw]
Subject: [PATCH] x86: mm: Add missing comments for initial kernel direct mapping

From: Zhang Yanfei <[email protected]>

For some reason, comments below are lost:
- comments for init_range_memory_mapping()
- comments in init_mem_mapping that is helpful for reminding people
that the pagetable is setup top-down

The comments were written by Yinghai in his patch in
https://lkml.org/lkml/2012/11/28/620

Signed-off-by: Zhang Yanfei <[email protected]>
Signed-off-by: Yinghai Lu <[email protected]>
Cc: "H. Peter Anvin" <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Andrew Morton <[email protected]>
Cc: Yasuaki Ishimatsu <[email protected]>
---
arch/x86/mm/init.c | 19 ++++++++++++++++++-
1 file changed, 18 insertions(+), 1 deletion(-)

diff --git a/arch/x86/mm/init.c b/arch/x86/mm/init.c
index fdc5dca..fed9993 100644
--- a/arch/x86/mm/init.c
+++ b/arch/x86/mm/init.c
@@ -359,7 +359,17 @@ unsigned long __init_refok init_memory_mapping(unsigned long start,
}

/*
- * would have hole in the middle or ends, and only ram parts will be mapped.
+ * We need to iterate through E820 memory map and create direct mappings
+ * for only E820_RAM and E820_KERN_RESERVED regions. We cannot simply
+ * create direct mappings for all pfns from [0 to max_low_pfn) and
+ * [4GB to max_pfn) because of possible memory holes in high addresses
+ * that cannot be marked as UC by fixed/variable range MTRRs.
+ * Depending on the alignment of E820 ranges, this may possibly result
+ * in using smaller size (i.e. 4K instead of 2M or 1G) page tables.
+ *
+ * init_mem_mapping call init_range_memory_mapping with big range.
+ * That range would have hole in the middle or ends, and only ram parts
+ * will be mapped in init_range_memory_mapping.
*/
static unsigned long __init init_range_memory_mapping(
unsigned long r_start,
@@ -419,6 +429,13 @@ void __init init_mem_mapping(void)
max_pfn_mapped = 0; /* will get exact value next */
min_pfn_mapped = real_end >> PAGE_SHIFT;
last_start = start = real_end;
+
+ /*
+ * We start from the top (end of memory) and go to the bottom.
+ * The memblock_find_in_range gets us a block of RAM from the
+ * end of RAM in [min_pfn_mapped, max_pfn_mapped) as new pages
+ * for page table.
+ */
while (last_start > ISA_END_ADDRESS) {
if (last_start > step_size) {
start = round_down(last_start - 1, step_size);
--
1.8.1.2


Subject: [tip:x86/urgent] x86/mm: Add missing comments for initial kernel direct mapping

Commit-ID: cf8b166d5c1c89aad6c436a954fa40fd18a75bfb
Gitweb: http://git.kernel.org/tip/cf8b166d5c1c89aad6c436a954fa40fd18a75bfb
Author: Zhang Yanfei <[email protected]>
AuthorDate: Thu, 9 May 2013 23:57:42 +0800
Committer: Ingo Molnar <[email protected]>
CommitDate: Fri, 10 May 2013 12:00:35 +0200

x86/mm: Add missing comments for initial kernel direct mapping

Two sets of comments were lost during patch-series shuffling:

- comments for init_range_memory_mapping()

- comments in init_mem_mapping that is helpful for reminding people
that the pagetable is setup top-down

The comments were written by Yinghai in his patch in:

https://lkml.org/lkml/2012/11/28/620

This patch reintroduces them.

Originally-From: Yinghai Lu <[email protected]>
Signed-off-by: Zhang Yanfei <[email protected]>
Cc: Yasuaki Ishimatsu <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Andrew Morton <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
[ Tidied it all up a bit. ]
Signed-off-by: Ingo Molnar <[email protected]>
---
arch/x86/mm/init.c | 19 ++++++++++++++++++-
1 file changed, 18 insertions(+), 1 deletion(-)

diff --git a/arch/x86/mm/init.c b/arch/x86/mm/init.c
index fdc5dca..eaac174 100644
--- a/arch/x86/mm/init.c
+++ b/arch/x86/mm/init.c
@@ -359,7 +359,17 @@ unsigned long __init_refok init_memory_mapping(unsigned long start,
}

/*
- * would have hole in the middle or ends, and only ram parts will be mapped.
+ * We need to iterate through the E820 memory map and create direct mappings
+ * for only E820_RAM and E820_KERN_RESERVED regions. We cannot simply
+ * create direct mappings for all pfns from [0 to max_low_pfn) and
+ * [4GB to max_pfn) because of possible memory holes in high addresses
+ * that cannot be marked as UC by fixed/variable range MTRRs.
+ * Depending on the alignment of E820 ranges, this may possibly result
+ * in using smaller size (i.e. 4K instead of 2M or 1G) page tables.
+ *
+ * init_mem_mapping() calls init_range_memory_mapping() with big range.
+ * That range would have hole in the middle or ends, and only ram parts
+ * will be mapped in init_range_memory_mapping().
*/
static unsigned long __init init_range_memory_mapping(
unsigned long r_start,
@@ -419,6 +429,13 @@ void __init init_mem_mapping(void)
max_pfn_mapped = 0; /* will get exact value next */
min_pfn_mapped = real_end >> PAGE_SHIFT;
last_start = start = real_end;
+
+ /*
+ * We start from the top (end of memory) and go to the bottom.
+ * The memblock_find_in_range() gets us a block of RAM from the
+ * end of RAM in [min_pfn_mapped, max_pfn_mapped) used as new pages
+ * for page table.
+ */
while (last_start > ISA_END_ADDRESS) {
if (last_start > step_size) {
start = round_down(last_start - 1, step_size);