kdump: round up the total memory size to 128M for crashkernel reservation

The total memory size we get in kernel is usually slightly less than 2G with a 2G memory module machine. The main reason is bios/firmware reserve some area it will not export all memory as usable to Linux.

2G memory X86 kvm guest test result of the total_mem value:
UEFI boot with ovmf: 0x7ef10000
Legacy boot kvm guest: 0x7ff7cc00

An option is to use dmi/smbios to get physical memory size, but it's not reliable as well. According to Prarit hardware vendors sometimes screw this up. Thus we choose to round up total size to 128M to workaround this problem. This is a best effort workaround, will improve it when we have better way in the future.

Signed-off-by Dave Young <[email protected]>

---
 kernel/crash_core.c |   17 +++++++++++++----
 1 file changed, 13 insertions(+), 4 deletions(-)

kernel/crash_core.c +13 -4

@@ -42,6 +42,15 @@ static int __init parse_crashkernel_mem(
 {
 	char *cur = cmdline, *tmp;
 	bool infinite_end = false;
+	unsigned long long total_mem = system_ram;
+
+	/*
+	 * Firmware usually reserves some memory regions for it's own use.
+	 * so we get less than actual system memory size.
+	 * We workaround this by round up the total size to 128M which is
+	 * enough for most test cases.
+	 */
+	total_mem = roundup(total_mem, 0x8000000);
 
 	/* for each entry of the comma-separated list */
 	do {

@@ -86,13 +95,13 @@ static int __init parse_crashkernel_mem(
 			return -EINVAL;
 		}
 		cur = tmp;
-		if (size >= system_ram) {
+		if (size >= total_mem) {
 			pr_warn("crashkernel: invalid size\n");
 			return -EINVAL;
 		}
 
 		/* match ? */
-		if (system_ram >= start && system_ram < end) {
+		if (total_mem >= start && total_mem < end) {
 			*crash_size = size;
 			if (end == ULLONG_MAX)
 				infinite_end = true;

@@ -126,9 +135,9 @@ static int __init parse_crashkernel_mem(
 				pr_warn("Memory reservation scale order expected after '^'\n");
 				return -EINVAL;
 			}
-			size = (system_ram - *crash_size) >> shift;
+			size = (total_mem - *crash_size) >> shift;
 			size = *crash_size + roundup(size, 1ULL << 20);
-			if (size < system_ram)
+			if (size < total_mem)
 				*crash_size = size;
 			cur = tmp;
 		} else

Hi Dave,

Quoted 8 lines · Dave Young

The total memory size we get in kernel is usually slightly less than 2G with a 2G memory module machine. The main reason is bios/firmware reserve some area it will not export all memory as usable to Linux.

2G memory X86 kvm guest test result of the total_mem value:
UEFI boot with ovmf: 0x7ef10000
Legacy boot kvm guest: 0x7ff7cc00

Quoted 5 lines

An option is to use dmi/smbios to get physical memory size, but it's not reliable as well. According to Prarit hardware vendors sometimes screw this up. Thus we choose to round up total size to 128M to workaround this problem. This is a best effort workaround, will improve it when we have better way in the future.

Thanks for this posting. While I don't get the point of this patch. So firmware take piece of memory, then why we need to count it into the total memory which we want to calculate a crashkernel memory based on.

Not counting that, is there anyting incorrect?

Thanks
Baoquan

Quoted 46 lines

Signed-off-by Dave Young <[email protected]>

---
 kernel/crash_core.c |   17 +++++++++++++----
 1 file changed, 13 insertions(+), 4 deletions(-)

kernel/crash_core.c +12 -3

@@ -42,6 +42,15 @@ static int __init parse_crashkernel_mem(
 {
 	char *cur = cmdline, *tmp;
 	bool infinite_end = false;
+	unsigned long long total_mem = system_ram;
+
+	/*
+	 * Firmware usually reserves some memory regions for it's own use.
+	 * so we get less than actual system memory size.
+	 * We workaround this by round up the total size to 128M which is
+	 * enough for most test cases.
+	 */
+	total_mem = roundup(total_mem, 0x8000000);
 
 	/* for each entry of the comma-separated list */
 	do {

@@ -86,13 +95,13 @@ static int __init parse_crashkernel_mem(
 			return -EINVAL;
 		}
 		cur = tmp;
-		if (size >= system_ram) {
+		if (size >= total_mem) {
 			pr_warn("crashkernel: invalid size\n");
 			return -EINVAL;
 		}
 
 		/* match ? */
-		if (system_ram >= start && system_ram < end) {
+		if (total_mem >= start && total_mem < end) {
 			*crash_size = size;
 			if (end == ULLONG_MAX)
 				infinite_end = true;

@@ -126,9 +135,9 @@ static int __init parse_crashkernel_mem(
 				pr_warn("Memory reservation scale order expected after '^'\n");
 				return -EINVAL;
 			}
-			size = (system_ram - *crash_size) >> shift;
+			size = (total_mem - *crash_size) >> shift;

kernel/crash_core.c static int __init parse_crashkernel_mem( · line 143

Quoted 8 lines

kernel/crash_core.c +1 -1

 			size = *crash_size + roundup(size, 1ULL << 20);
-			if (size < system_ram)
+			if (size < total_mem)
 				*crash_size = size;
 			cur = tmp;
 		} else

Hi Baoquan,

Quoted 21 lines · Baoquan He

Hi Dave,

Quoted 8 lines · Dave Young

The total memory size we get in kernel is usually slightly less than 2G with a 2G memory module machine. The main reason is bios/firmware reserve some area it will not export all memory as usable to Linux.

2G memory X86 kvm guest test result of the total_mem value:
UEFI boot with ovmf: 0x7ef10000
Legacy boot kvm guest: 0x7ff7cc00

Quoted 5 lines

An option is to use dmi/smbios to get physical memory size, but it's not reliable as well. According to Prarit hardware vendors sometimes screw this up. Thus we choose to round up total size to 128M to workaround this problem. This is a best effort workaround, will improve it when we have better way in the future.

Thanks for this posting. While I don't get the point of this patch. So firmware take piece of memory, then why we need to count it into the total memory which we want to calculate a crashkernel memory based on.

Quoted 1 line

Not counting that, is there anyting incorrect?

Yes, considering crashkernel=1G-2G:128M, if we have a 1G memory machine, we get total size 1023M from firmware then it will not fall into 1G-2G thus no memory reserved. User will never know that, it is hard to let user to know the exact total value we get in kernel..

Quoted 47 lines

Quoted 47 lines

Signed-off-by Dave Young <[email protected]>

---
 kernel/crash_core.c |   17 +++++++++++++----
 1 file changed, 13 insertions(+), 4 deletions(-)

kernel/crash_core.c +12 -3

@@ -42,6 +42,15 @@ static int __init parse_crashkernel_mem(
 {
 	char *cur = cmdline, *tmp;
 	bool infinite_end = false;
+	unsigned long long total_mem = system_ram;
+
+	/*
+	 * Firmware usually reserves some memory regions for it's own use.
+	 * so we get less than actual system memory size.
+	 * We workaround this by round up the total size to 128M which is
+	 * enough for most test cases.
+	 */
+	total_mem = roundup(total_mem, 0x8000000);
 
 	/* for each entry of the comma-separated list */
 	do {

@@ -86,13 +95,13 @@ static int __init parse_crashkernel_mem(
 			return -EINVAL;
 		}
 		cur = tmp;
-		if (size >= system_ram) {
+		if (size >= total_mem) {
 			pr_warn("crashkernel: invalid size\n");
 			return -EINVAL;
 		}
 
 		/* match ? */
-		if (system_ram >= start && system_ram < end) {
+		if (total_mem >= start && total_mem < end) {
 			*crash_size = size;
 			if (end == ULLONG_MAX)
 				infinite_end = true;

@@ -126,9 +135,9 @@ static int __init parse_crashkernel_mem(
 				pr_warn("Memory reservation scale order expected after '^'\n");
 				return -EINVAL;
 			}
-			size = (system_ram - *crash_size) >> shift;
+			size = (total_mem - *crash_size) >> shift;

Quoted 8 lines

kernel/crash_core.c static int __init parse_crashkernel_mem( · line 143

Quoted 8 lines

kernel/crash_core.c +1 -1

 			size = *crash_size + roundup(size, 1ULL << 20);
-			if (size < system_ram)
+			if (size < total_mem)
 				*crash_size = size;
 			cur = tmp;
 		} else

Thanks
Dave