Skip to contentSkip to conversation

x86: nit: wrong page size in init_memory_mapping() printks

RFC PATCH In review Dave Hansen 3 participants 3 messages Feb 10, 2015 – Feb 19, 2015

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

RFC PATCH 1/1 x86: nit: wrong page size in init_memory_mapping() printks

From: Dave Hansen <[email protected]>

With 32-bit non-PAE kernels, we have 2 page sizes available (at most): 4k and 4M.

Enabling PAE replaces that 4M size with a 2M one (which 64-bit systems use too).

But, when booting a 32-bit non-PAE kernel, in one of our early-boot printouts, we say:

[    0.000000] init_memory_mapping: [mem 0x00000000-0x000fffff]
[    0.000000]  [mem 0x00000000-0x000fffff] page 4k
[    0.000000] init_memory_mapping: [mem 0x37000000-0x373fffff]
[    0.000000]  [mem 0x37000000-0x373fffff] page 2M
[    0.000000] init_memory_mapping: [mem 0x00100000-0x36ffffff]
[    0.000000]  [mem 0x00100000-0x003fffff] page 4k
[    0.000000]  [mem 0x00400000-0x36ffffff] page 2M
[    0.000000] init_memory_mapping: [mem 0x37400000-0x377fdfff]
[    0.000000]  [mem 0x37400000-0x377fdfff] page 4k

Which is obviously wrong. There is no 2M page available. This is probably because of a badly-named variable: in the map_range
code: PG_LEVEL_2M.

Instead of renaming all the PG_LEVEL_2M's. This patch just fixes the printout:

[    0.000000] init_memory_mapping: [mem 0x00000000-0x000fffff]
[    0.000000]  [mem 0x00000000-0x000fffff] page 4k
[    0.000000] init_memory_mapping: [mem 0x37000000-0x373fffff]
[    0.000000]  [mem 0x37000000-0x373fffff] page 4M
[    0.000000] init_memory_mapping: [mem 0x00100000-0x36ffffff]
[    0.000000]  [mem 0x00100000-0x003fffff] page 4k
[    0.000000]  [mem 0x00400000-0x36ffffff] page 4M
[    0.000000] init_memory_mapping: [mem 0x37400000-0x377fdfff]
[    0.000000]  [mem 0x37400000-0x377fdfff] page 4k
[    0.000000] BRK [0x03206000, 0x03206fff] PGTABLE

---

 b/arch/x86/mm/init.c |   28 ++++++++++++++++++++++++++--
 1 file changed, 26 insertions(+), 2 deletions(-)
arch/x86/mm/init.c renamed from arch/x86/mm/init.c~spitting-out-bad-page-size +26 -2
diff -puN arch/x86/mm/init.c~spitting-out-bad-page-size arch/x86/mm/init.c
@@ -238,6 +238,31 @@ static void __init_refok adjust_range_pa
 	}
 }
 
+static const char *page_size_string(struct map_range *mr)
+{
+	static const char str_1g[] = "1G";
+	static const char str_2m[] = "2M";
+	static const char str_4m[] = "4M";
+	static const char str_4k[] = "4k";
+
+	if (mr->page_size_mask & (1<<PG_LEVEL_1G))
+		return str_1g;
+	/*
+	 * 32-bit without PAE has a 4M large page size.
+	 * PG_LEVEL_2M is misnamed, but we can at least
+	 * print out the right size in the string.
+	 */
+	if (IS_ENABLED(CONFIG_X86_32) &&
+	    !IS_ENABLED(CONFIG_X86_PAE) &&
+	    mr->page_size_mask & (1<<PG_LEVEL_2M))
+		return str_4m;
+
+	if (mr->page_size_mask & (1<<PG_LEVEL_2M))
+		return str_2m;
+
+	return str_4k;
+}
+
 static int __meminit split_mem_range(struct map_range *mr, int nr_range,
 				     unsigned long start,
 				     unsigned long end)
@@ -333,8 +358,7 @@ static int __meminit split_mem_range(str
 	for (i = 0; i < nr_range; i++)
 		printk(KERN_DEBUG " [mem %#010lx-%#010lx] page %s\n",
 				mr[i].start, mr[i].end - 1,
-			(mr[i].page_size_mask & (1<<PG_LEVEL_1G))?"1G":(
-			 (mr[i].page_size_mask & (1<<PG_LEVEL_2M))?"2M":"4k"));
+				page_size_string(&mr[i]));
 
 	return nr_range;
 }

_

maintainer Acked-by on v1 1/1
#1 lore raw
Quoted 41 lines · Dave Hansen

From: Dave Hansen <[email protected]>

With 32-bit non-PAE kernels, we have 2 page sizes available (at most): 4k and 4M.

Enabling PAE replaces that 4M size with a 2M one (which 64-bit systems use too).

But, when booting a 32-bit non-PAE kernel, in one of our early-boot printouts, we say:

[    0.000000] init_memory_mapping: [mem 0x00000000-0x000fffff]
[    0.000000]  [mem 0x00000000-0x000fffff] page 4k
[    0.000000] init_memory_mapping: [mem 0x37000000-0x373fffff]
[    0.000000]  [mem 0x37000000-0x373fffff] page 2M
[    0.000000] init_memory_mapping: [mem 0x00100000-0x36ffffff]
[    0.000000]  [mem 0x00100000-0x003fffff] page 4k
[    0.000000]  [mem 0x00400000-0x36ffffff] page 2M
[    0.000000] init_memory_mapping: [mem 0x37400000-0x377fdfff]
[    0.000000]  [mem 0x37400000-0x377fdfff] page 4k

Which is obviously wrong. There is no 2M page available. This is probably because of a badly-named variable: in the map_range
code: PG_LEVEL_2M.

Instead of renaming all the PG_LEVEL_2M's. This patch just fixes the printout:

[    0.000000] init_memory_mapping: [mem 0x00000000-0x000fffff]
[    0.000000]  [mem 0x00000000-0x000fffff] page 4k
[    0.000000] init_memory_mapping: [mem 0x37000000-0x373fffff]
[    0.000000]  [mem 0x37000000-0x373fffff] page 4M
[    0.000000] init_memory_mapping: [mem 0x00100000-0x36ffffff]
[    0.000000]  [mem 0x00100000-0x003fffff] page 4k
[    0.000000]  [mem 0x00400000-0x36ffffff] page 4M
[    0.000000] init_memory_mapping: [mem 0x37400000-0x377fdfff]
[    0.000000]  [mem 0x37400000-0x377fdfff] page 4k
[    0.000000] BRK [0x03206000, 0x03206fff] PGTABLE
Quoted 1 line

Looks ok to me considering that we're missing a PG_LEVEL_4M thing completely which would need to be handled depending on X86_PAE and CR4 PSE and PAE bits. Probably not worth the trouble though.

Signature · 5 lines
Regards/Gruss,
    Boris.

ECO tip #101: Trim your mails when you reply.
--
#2 lore raw

[tip:x86/urgent] x86/mm/init: Fix incorrect page size in init_memory_mapping() printks

In reply to Dave Hansen
x86/mm/init: Fix incorrect page size in init_memory_mapping() printks
Commit-ID:  f15e05186c3244e9195378a0a568283a8ccc60b0
Gitweb:     http://git.kernel.org/tip/f15e05186c3244e9195378a0a568283a8ccc60b0
Author:     Dave Hansen <[email protected]>
AuthorDate: Tue, 10 Feb 2015 13:20:30 -0800
Committer:  Borislav Petkov <[email protected]>
CommitDate: Thu, 19 Feb 2015 11:45:27 +0100

x86/mm/init: Fix incorrect page size in init_memory_mapping() printks

With 32-bit non-PAE kernels, we have 2 page sizes available (at most): 4k and 4M.

Enabling PAE replaces that 4M size with a 2M one (which 64-bit systems use too).

But, when booting a 32-bit non-PAE kernel, in one of our early-boot printouts, we say:

init_memory_mapping: [mem 0x00000000-0x000fffff]
[mem 0x00000000-0x000fffff] page 4k
init_memory_mapping: [mem 0x37000000-0x373fffff]
[mem 0x37000000-0x373fffff] page 2M
init_memory_mapping: [mem 0x00100000-0x36ffffff]
[mem 0x00100000-0x003fffff] page 4k [mem 0x00400000-0x36ffffff] page 2M
init_memory_mapping: [mem 0x37400000-0x377fdfff]
[mem 0x37400000-0x377fdfff] page 4k

Which is obviously wrong. There is no 2M page available. This is probably because of a badly-named variable: in the map_range
code: PG_LEVEL_2M.

Instead of renaming all the PG_LEVEL_2M's. This patch just fixes the printout:

init_memory_mapping: [mem 0x00000000-0x000fffff]
[mem 0x00000000-0x000fffff] page 4k
init_memory_mapping: [mem 0x37000000-0x373fffff]
[mem 0x37000000-0x373fffff] page 4M
init_memory_mapping: [mem 0x00100000-0x36ffffff]
[mem 0x00100000-0x003fffff] page 4k [mem 0x00400000-0x36ffffff] page 4M
init_memory_mapping: [mem 0x37400000-0x377fdfff]
[mem 0x37400000-0x377fdfff] page 4k
BRK [0x03206000, 0x03206fff] PGTABLE

---
 arch/x86/mm/init.c | 28 ++++++++++++++++++++++++++--
 1 file changed, 26 insertions(+), 2 deletions(-)
arch/x86/mm/init.c +26 -2
index 079c3b6..7ff2424 100644
@@ -238,6 +238,31 @@ static void __init_refok adjust_range_page_size_mask(struct map_range *mr,
 	}
 }
 
+static const char *page_size_string(struct map_range *mr)
+{
+	static const char str_1g[] = "1G";
+	static const char str_2m[] = "2M";
+	static const char str_4m[] = "4M";
+	static const char str_4k[] = "4k";
+
+	if (mr->page_size_mask & (1<<PG_LEVEL_1G))
+		return str_1g;
+	/*
+	 * 32-bit without PAE has a 4M large page size.
+	 * PG_LEVEL_2M is misnamed, but we can at least
+	 * print out the right size in the string.
+	 */
+	if (IS_ENABLED(CONFIG_X86_32) &&
+	    !IS_ENABLED(CONFIG_X86_PAE) &&
+	    mr->page_size_mask & (1<<PG_LEVEL_2M))
+		return str_4m;
+
+	if (mr->page_size_mask & (1<<PG_LEVEL_2M))
+		return str_2m;
+
+	return str_4k;
+}
+
 static int __meminit split_mem_range(struct map_range *mr, int nr_range,
 				     unsigned long start,
 				     unsigned long end)
@@ -333,8 +358,7 @@ static int __meminit split_mem_range(struct map_range *mr, int nr_range,
 	for (i = 0; i < nr_range; i++)
 		printk(KERN_DEBUG " [mem %#010lx-%#010lx] page %s\n",
 				mr[i].start, mr[i].end - 1,
-			(mr[i].page_size_mask & (1<<PG_LEVEL_1G))?"1G":(
-			 (mr[i].page_size_mask & (1<<PG_LEVEL_2M))?"2M":"4k"));
+				page_size_string(&mr[i]));
 
 	return nr_range;
 }