Skip to contentSkip to conversation

x86, calgary: use 8M TCE table size by default

PATCH In review WANG Chao 6 participants 9 messages Mar 10, 2014 – Apr 11, 2014

Subsystems (3)
lore
Share & tools
lore

WANG Chao

author
#0 lore raw

PATCH v2 1/1 x86, calgary: use 8M TCE table size by default

New kexec-tools wants to pass kdump kernel needed memmap via E820 directly, instead of memmap=exactmap. This makes saved_max_pfn not be passed down to 2nd kernel. To keep 1st kernel and 2nd kernel using the same TCE table size, Muli suggest to hard code the size to max (8M).

We can't get rid of saved_max_pfn this time, for backward compatibility with old first kernel and new second kernel. However new first kernel and old second kernel can not work unfortunately.

v2->v1:
- retain saved_max_pfn so new 2nd kernel can work with old 1st kernel from Vivek

---
 arch/x86/kernel/pci-calgary_64.c | 31 +++++++++++++++++++------------
 1 file changed, 19 insertions(+), 12 deletions(-)
arch/x86/kernel/pci-calgary_64.c +19 -12
index 299d493..0497f71 100644
@@ -1207,23 +1207,31 @@ error:
 	return ret;
 }
 
-static inline int __init determine_tce_table_size(u64 ram)
+static inline int __init determine_tce_table_size(void)
 {
 	int ret;
 
 	if (specified_table_size != TCE_TABLE_SIZE_UNSPECIFIED)
 		return specified_table_size;
 
-	/*
-	 * Table sizes are from 0 to 7 (TCE_TABLE_SIZE_64K to
-	 * TCE_TABLE_SIZE_8M). Table size 0 has 8K entries and each
-	 * larger table size has twice as many entries, so shift the
-	 * max ram address by 13 to divide by 8K and then look at the
-	 * order of the result to choose between 0-7.
-	 */
-	ret = get_order(ram >> 13);
-	if (ret > TCE_TABLE_SIZE_8M)
+	if (is_kdump_kernel() && saved_max_pfn) {
+		/*
+		 * Table sizes are from 0 to 7 (TCE_TABLE_SIZE_64K to
+		 * TCE_TABLE_SIZE_8M). Table size 0 has 8K entries and each
+		 * larger table size has twice as many entries, so shift the
+		 * max ram address by 13 to divide by 8K and then look at the
+		 * order of the result to choose between 0-7.
+		 */
+		ret = get_order((saved_max_pfn * PAGE_SIZE) >> 13);
+		if (ret > TCE_TABLE_SIZE_8M)
+			ret = TCE_TABLE_SIZE_8M;
+	} else {
+		/*
+		 * Use 8M by default (suggested by Muli) if it's not
+		 * kdump kernel and saved_max_pfn isn't set.
+		 */
 		ret = TCE_TABLE_SIZE_8M;
+	}
 
 	return ret;
 }
@@ -1418,8 +1426,7 @@ int __init detect_calgary(void)
 		return -ENOMEM;
 	}
 
-	specified_table_size = determine_tce_table_size((is_kdump_kernel() ?
-					saved_max_pfn : max_pfn) * PAGE_SIZE);
+	specified_table_size = determine_tce_table_size();
 
 	for (bus = 0; bus < MAX_PHB_BUS_NUM; bus++) {
 		struct calgary_bus_info *info = &bus_info[bus];
Signature · 7 lines
1.8.5.3


_______________________________________________
kexec mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/kexec
Acked-by on v2 1/1
#1 lore raw
Quoted 9 lines · WANG Chao

New kexec-tools wants to pass kdump kernel needed memmap via E820 directly, instead of memmap=exactmap. This makes saved_max_pfn not be passed down to 2nd kernel. To keep 1st kernel and 2nd kernel using the same TCE table size, Muli suggest to hard code the size to max (8M).

We can't get rid of saved_max_pfn this time, for backward compatibility with old first kernel and new second kernel. However new first kernel and old second kernel can not work unfortunately.

Quoted 5 lines

v2->v1:
- retain saved_max_pfn so new 2nd kernel can work with old 1st kernel from Vivek

This patch looks good to me. This will use saved_max_pfn to determine table size and retain backward compatibility with old kernel (if user uses right kexec cmdline option --pass-memmap).

For others (majority of the cases), both first kernel and second kernel will use 8M of calgary table. 8M does sound a lot in terms of memory but assumption here is that there might be only some machines with calgary hardware and only these will pay the penalty of this increased memory usage. This sounds better than exporting table size to user space.

Thanks
Vivek

Quoted 62 lines
---
 arch/x86/kernel/pci-calgary_64.c | 31 +++++++++++++++++++------------
 1 file changed, 19 insertions(+), 12 deletions(-)
arch/x86/kernel/pci-calgary_64.c +19 -12
index 299d493..0497f71 100644
@@ -1207,23 +1207,31 @@ error:
 	return ret;
 }
 
-static inline int __init determine_tce_table_size(u64 ram)
+static inline int __init determine_tce_table_size(void)
 {
 	int ret;
 
 	if (specified_table_size != TCE_TABLE_SIZE_UNSPECIFIED)
 		return specified_table_size;
 
-	/*
-	 * Table sizes are from 0 to 7 (TCE_TABLE_SIZE_64K to
-	 * TCE_TABLE_SIZE_8M). Table size 0 has 8K entries and each
-	 * larger table size has twice as many entries, so shift the
-	 * max ram address by 13 to divide by 8K and then look at the
-	 * order of the result to choose between 0-7.
-	 */
-	ret = get_order(ram >> 13);
-	if (ret > TCE_TABLE_SIZE_8M)
+	if (is_kdump_kernel() && saved_max_pfn) {
+		/*
+		 * Table sizes are from 0 to 7 (TCE_TABLE_SIZE_64K to
+		 * TCE_TABLE_SIZE_8M). Table size 0 has 8K entries and each
+		 * larger table size has twice as many entries, so shift the
+		 * max ram address by 13 to divide by 8K and then look at the
+		 * order of the result to choose between 0-7.
+		 */
+		ret = get_order((saved_max_pfn * PAGE_SIZE) >> 13);
+		if (ret > TCE_TABLE_SIZE_8M)
+			ret = TCE_TABLE_SIZE_8M;
+	} else {
+		/*
+		 * Use 8M by default (suggested by Muli) if it's not
+		 * kdump kernel and saved_max_pfn isn't set.
+		 */
 		ret = TCE_TABLE_SIZE_8M;
+	}
 
 	return ret;
 }
@@ -1418,8 +1426,7 @@ int __init detect_calgary(void)
 		return -ENOMEM;
 	}
 
-	specified_table_size = determine_tce_table_size((is_kdump_kernel() ?
-					saved_max_pfn : max_pfn) * PAGE_SIZE);
+	specified_table_size = determine_tce_table_size();
 
 	for (bus = 0; bus < MAX_PHB_BUS_NUM; bus++) {
 		struct calgary_bus_info *info = &bus_info[bus];
Signature · 1 line
1.8.5.3

_______________________________________________
kexec mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/kexec

#5 lore raw
In reply to WANG Chao

Hi hpa,

I am wondering how to make progress on calgary issue now. We do want to cleanup kexec-tools code and move to passing memory ranges in bootparams.

I see two choices here.

- Take this patch in so that new kernel code will work with modified kexec-tools without any user intervention. (Both first and second kernel need to be new).

- Don't take this patch in and calgary will be broken with new kexec-tools. And that will always require manual user intervention from user to force passing memmap on command line.

I kind of prefer the first option so that even if there are some calgary machines out there, they will work automatically with new kernel and will require manual intervention only for old kernels and new kexec-tool cobination.

But I can live with second option too if that's everybody agrees to and hopefully we don't get lots of reports of breaking kdump with kexec-tools upgrade.

Thanks
Vivek

Quoted 76 lines · WANG Chao

New kexec-tools wants to pass kdump kernel needed memmap via E820 directly, instead of memmap=exactmap. This makes saved_max_pfn not be passed down to 2nd kernel. To keep 1st kernel and 2nd kernel using the same TCE table size, Muli suggest to hard code the size to max (8M).

We can't get rid of saved_max_pfn this time, for backward compatibility with old first kernel and new second kernel. However new first kernel and old second kernel can not work unfortunately.

v2->v1:
- retain saved_max_pfn so new 2nd kernel can work with old 1st kernel from Vivek

---
 arch/x86/kernel/pci-calgary_64.c | 31 +++++++++++++++++++------------
 1 file changed, 19 insertions(+), 12 deletions(-)
arch/x86/kernel/pci-calgary_64.c +19 -12
index 299d493..0497f71 100644
@@ -1207,23 +1207,31 @@ error:
 	return ret;
 }
 
-static inline int __init determine_tce_table_size(u64 ram)
+static inline int __init determine_tce_table_size(void)
 {
 	int ret;
 
 	if (specified_table_size != TCE_TABLE_SIZE_UNSPECIFIED)
 		return specified_table_size;
 
-	/*
-	 * Table sizes are from 0 to 7 (TCE_TABLE_SIZE_64K to
-	 * TCE_TABLE_SIZE_8M). Table size 0 has 8K entries and each
-	 * larger table size has twice as many entries, so shift the
-	 * max ram address by 13 to divide by 8K and then look at the
-	 * order of the result to choose between 0-7.
-	 */
-	ret = get_order(ram >> 13);
-	if (ret > TCE_TABLE_SIZE_8M)
+	if (is_kdump_kernel() && saved_max_pfn) {
+		/*
+		 * Table sizes are from 0 to 7 (TCE_TABLE_SIZE_64K to
+		 * TCE_TABLE_SIZE_8M). Table size 0 has 8K entries and each
+		 * larger table size has twice as many entries, so shift the
+		 * max ram address by 13 to divide by 8K and then look at the
+		 * order of the result to choose between 0-7.
+		 */
+		ret = get_order((saved_max_pfn * PAGE_SIZE) >> 13);
+		if (ret > TCE_TABLE_SIZE_8M)
+			ret = TCE_TABLE_SIZE_8M;
+	} else {
+		/*
+		 * Use 8M by default (suggested by Muli) if it's not
+		 * kdump kernel and saved_max_pfn isn't set.
+		 */
 		ret = TCE_TABLE_SIZE_8M;
+	}
 
 	return ret;
 }
@@ -1418,8 +1426,7 @@ int __init detect_calgary(void)
 		return -ENOMEM;
 	}
 
-	specified_table_size = determine_tce_table_size((is_kdump_kernel() ?
-					saved_max_pfn : max_pfn) * PAGE_SIZE);
+	specified_table_size = determine_tce_table_size();
 
 	for (bus = 0; bus < MAX_PHB_BUS_NUM; bus++) {
 		struct calgary_bus_info *info = &bus_info[bus];
Signature · 1 line
1.8.5.3

_______________________________________________
kexec mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/kexec

#6 lore raw
Quoted 15 lines · Vivek Goyal

Hi hpa,

I am wondering how to make progress on calgary issue now. We do want to cleanup kexec-tools code and move to passing memory ranges in bootparams.

I see two choices here.

- Take this patch in so that new kernel code will work with modified kexec-tools without any user intervention. (Both first and second kernel need to be new).

- Don't take this patch in and calgary will be broken with new kexec-tools. And that will always require manual user intervention from user to force passing memmap on command line.

Quoted 9 lines

I kind of prefer the first option so that even if there are some calgary machines out there, they will work automatically with new kernel and will require manual intervention only for old kernels and new kexec-tool cobination.

But I can live with second option too if that's everybody agrees to and hopefully we don't get lots of reports of breaking kdump with kexec-tools upgrade.

The patch seems simple enough and doesn't require any new command line hacks. It has been widely reviewed and acked, so I'll pick it up.

-hpa

_______________________________________________
kexec mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/kexec

#7 lore raw

[tip:x86/platform] x86, calgary: Use 8M TCE table size by default

In reply to WANG Chao
x86, calgary: Use 8M TCE table size by default
Commit-ID:  936329de1e6bf3dfa8c056074e6fc6b673e7b1d0
Gitweb:     http://git.kernel.org/tip/936329de1e6bf3dfa8c056074e6fc6b673e7b1d0
Author:     WANG Chao <[email protected]>
AuthorDate: Mon, 10 Mar 2014 22:52:00 +0800
Committer:  H. Peter Anvin <[email protected]>
CommitDate: Thu, 20 Mar 2014 14:51:39 -0700

x86, calgary: Use 8M TCE table size by default

New kexec-tools wants to pass kdump kernel needed memmap via E820 directly, instead of memmap=exactmap. This makes saved_max_pfn not be passed down to 2nd kernel. To keep 1st kernel and 2nd kernel using the same TCE table size, Muli suggest to hard code the size to max (8M).

We can't get rid of saved_max_pfn this time, for backward compatibility with old first kernel and new second kernel. However new first kernel and old second kernel can not work unfortunately.

v2->v1:
- retain saved_max_pfn so new 2nd kernel can work with old 1st kernel from Vivek

---
 arch/x86/kernel/pci-calgary_64.c | 31 +++++++++++++++++++------------
 1 file changed, 19 insertions(+), 12 deletions(-)
arch/x86/kernel/pci-calgary_64.c +19 -12
index 299d493..0497f71 100644
@@ -1207,23 +1207,31 @@ error:
 	return ret;
 }
 
-static inline int __init determine_tce_table_size(u64 ram)
+static inline int __init determine_tce_table_size(void)
 {
 	int ret;
 
 	if (specified_table_size != TCE_TABLE_SIZE_UNSPECIFIED)
 		return specified_table_size;
 
-	/*
-	 * Table sizes are from 0 to 7 (TCE_TABLE_SIZE_64K to
-	 * TCE_TABLE_SIZE_8M). Table size 0 has 8K entries and each
-	 * larger table size has twice as many entries, so shift the
-	 * max ram address by 13 to divide by 8K and then look at the
-	 * order of the result to choose between 0-7.
-	 */
-	ret = get_order(ram >> 13);
-	if (ret > TCE_TABLE_SIZE_8M)
+	if (is_kdump_kernel() && saved_max_pfn) {
+		/*
+		 * Table sizes are from 0 to 7 (TCE_TABLE_SIZE_64K to
+		 * TCE_TABLE_SIZE_8M). Table size 0 has 8K entries and each
+		 * larger table size has twice as many entries, so shift the
+		 * max ram address by 13 to divide by 8K and then look at the
+		 * order of the result to choose between 0-7.
+		 */
+		ret = get_order((saved_max_pfn * PAGE_SIZE) >> 13);
+		if (ret > TCE_TABLE_SIZE_8M)
+			ret = TCE_TABLE_SIZE_8M;
+	} else {
+		/*
+		 * Use 8M by default (suggested by Muli) if it's not
+		 * kdump kernel and saved_max_pfn isn't set.
+		 */
 		ret = TCE_TABLE_SIZE_8M;
+	}
 
 	return ret;
 }
@@ -1418,8 +1426,7 @@ int __init detect_calgary(void)
 		return -ENOMEM;
 	}
 
-	specified_table_size = determine_tce_table_size((is_kdump_kernel() ?
-					saved_max_pfn : max_pfn) * PAGE_SIZE);
+	specified_table_size = determine_tce_table_size();
 
 	for (bus = 0; bus < MAX_PHB_BUS_NUM; bus++) {
 		struct calgary_bus_info *info = &bus_info[bus];
#8 lore raw

[tip:x86/platform] x86, calgary: Use 8M TCE table size by default

In reply to WANG Chao
x86, calgary: Use 8M TCE table size by default
Commit-ID:  0534af01cca338193abbfdb53650af91e65dbf10
Gitweb:     http://git.kernel.org/tip/0534af01cca338193abbfdb53650af91e65dbf10
Author:     WANG Chao <[email protected]>
AuthorDate: Mon, 10 Mar 2014 22:52:00 +0800
Committer:  H. Peter Anvin <[email protected]>
CommitDate: Thu, 10 Apr 2014 19:51:32 -0700

x86, calgary: Use 8M TCE table size by default

New kexec-tools wants to pass kdump kernel needed memmap via E820 directly, instead of memmap=exactmap. This makes saved_max_pfn not be passed down to 2nd kernel. To keep 1st kernel and 2nd kernel using the same TCE table size, Muli suggest to hard code the size to max (8M).

We can't get rid of saved_max_pfn this time, for backward compatibility with old first kernel and new second kernel. However new first kernel and old second kernel can not work unfortunately.

v2->v1:
- retain saved_max_pfn so new 2nd kernel can work with old 1st kernel from Vivek

---
 arch/x86/kernel/pci-calgary_64.c | 31 +++++++++++++++++++------------
 1 file changed, 19 insertions(+), 12 deletions(-)
arch/x86/kernel/pci-calgary_64.c +19 -12
index 299d493..0497f71 100644
@@ -1207,23 +1207,31 @@ error:
 	return ret;
 }
 
-static inline int __init determine_tce_table_size(u64 ram)
+static inline int __init determine_tce_table_size(void)
 {
 	int ret;
 
 	if (specified_table_size != TCE_TABLE_SIZE_UNSPECIFIED)
 		return specified_table_size;
 
-	/*
-	 * Table sizes are from 0 to 7 (TCE_TABLE_SIZE_64K to
-	 * TCE_TABLE_SIZE_8M). Table size 0 has 8K entries and each
-	 * larger table size has twice as many entries, so shift the
-	 * max ram address by 13 to divide by 8K and then look at the
-	 * order of the result to choose between 0-7.
-	 */
-	ret = get_order(ram >> 13);
-	if (ret > TCE_TABLE_SIZE_8M)
+	if (is_kdump_kernel() && saved_max_pfn) {
+		/*
+		 * Table sizes are from 0 to 7 (TCE_TABLE_SIZE_64K to
+		 * TCE_TABLE_SIZE_8M). Table size 0 has 8K entries and each
+		 * larger table size has twice as many entries, so shift the
+		 * max ram address by 13 to divide by 8K and then look at the
+		 * order of the result to choose between 0-7.
+		 */
+		ret = get_order((saved_max_pfn * PAGE_SIZE) >> 13);
+		if (ret > TCE_TABLE_SIZE_8M)
+			ret = TCE_TABLE_SIZE_8M;
+	} else {
+		/*
+		 * Use 8M by default (suggested by Muli) if it's not
+		 * kdump kernel and saved_max_pfn isn't set.
+		 */
 		ret = TCE_TABLE_SIZE_8M;
+	}
 
 	return ret;
 }
@@ -1418,8 +1426,7 @@ int __init detect_calgary(void)
 		return -ENOMEM;
 	}
 
-	specified_table_size = determine_tce_table_size((is_kdump_kernel() ?
-					saved_max_pfn : max_pfn) * PAGE_SIZE);
+	specified_table_size = determine_tce_table_size();
 
 	for (bus = 0; bus < MAX_PHB_BUS_NUM; bus++) {
 		struct calgary_bus_info *info = &bus_info[bus];