Allow PFN_EXCLUDED to be tunable via command line option --exclude-threshold

The PFN_EXCLUDED value is used to control at which point a run of zeros in the bitmap (zeros denote excluded pages) is large enough to warrant truncating the current output segment and to create a new output segment (containing non-excluded pages), in an ELF dump.

If the run is smaller than PFN_EXCLUDED, then those excluded pages are still output in the ELF dump, for the current output segment.

By using smaller values of PFN_EXCLUDED, the resulting dump file size can be made smaller by actually removing more excluded pages from the resulting dump file.

This patch adds the command line option --exclude-threshold=<value> to indicate the threshold. The default is 256, the legacy value of PFN_EXCLUDED. The smallest value permitted is 1.

Using an existing vmcore, this was tested by the following:

% makedumpfile -E -d31 --exclude-threshold=256 -x vmlinux vmcore newvmcore256 % makedumpfile -E -d31 --exclude-threshold=4 -x vmlinux vmcore newvmcore4

I utilize -d31 in order to exclude as many page types as possible, resulting in a [significantly] smaller file sizes than the original vmcore.

(quoted diff) +0 -3

-rwxrwx--- 1 edevolde edevolde 4034564096 Jun 27 10:24 vmcore
-rw------- 1 edevolde edevolde 119808156 Jul  6 13:01 newvmcore256
-rw------- 1 edevolde edevolde 100811276 Jul  6 13:08 newvmcore4

The use of smaller value of PFN_EXCLUDED increases the number of output segments (the 'Number of program headers' in the readelf output) in the ELF dump file.

% readelf -h vmcore
ELF Header:
Magic: 7f 45 4c 46 02 01 01 00 00 00 00 00 00 00 00 00

 Class:                             ELF64
 Data:                              2's complement, little endian
 Version:                           1 (current)
 OS/ABI:                            UNIX - System V
 ABI Version:                       0
 Type:                              CORE (Core file)
 Machine:                           Advanced Micro Devices X86-64
 Version:                           0x1
 Entry point address:               0x0
 Start of program headers:          64 (bytes into file)
 Start of section headers:          0 (bytes into file)
 Flags:                             0x0
 Size of this header:               64 (bytes)
 Size of program headers:           56 (bytes)
 Number of program headers:         6
                                    ^^^

Size of section headers: 0 (bytes)
Number of section headers: 0
Section header string table index: 0

% readelf -h newvmcore256
ELF Header:
Magic: 7f 45 4c 46 02 01 01 00 00 00 00 00 00 00 00 00

 Class:                             ELF64
 Data:                              2's complement, little endian
 Version:                           1 (current)
 OS/ABI:                            UNIX - System V
 ABI Version:                       0
 Type:                              CORE (Core file)
 Machine:                           Advanced Micro Devices X86-64
 Version:                           0x1
 Entry point address:               0x0
 Start of program headers:          64 (bytes into file)
 Start of section headers:          0 (bytes into file)
 Flags:                             0x0
 Size of this header:               64 (bytes)
 Size of program headers:           56 (bytes)
 Number of program headers:         18
                                    ^^^

Size of section headers: 0 (bytes)
Number of section headers: 0
Section header string table index: 0

% readelf -h newvmcore4
ELF Header:
Magic: 7f 45 4c 46 02 01 01 00 00 00 00 00 00 00 00 00

 Class:                             ELF64
 Data:                              2's complement, little endian
 Version:                           1 (current)
 OS/ABI:                            UNIX - System V
 ABI Version:                       0
 Type:                              CORE (Core file)
 Machine:                           Advanced Micro Devices X86-64
 Version:                           0x1
 Entry point address:               0x0
 Start of program headers:          64 (bytes into file)
 Start of section headers:          0 (bytes into file)
 Flags:                             0x0
 Size of this header:               64 (bytes)
 Size of program headers:           56 (bytes)
 Number of program headers:         244
                                    ^^^

Size of section headers: 0 (bytes)
Number of section headers: 0
Section header string table index: 0

The newvmcore4 has an even smaller file size than newvmcore256, with the small price being that there are now 244 rather than 18 segments in the dump file.

And with a larger number of segments, loading both vmcore and newvmcore4 into 'crash' resulted in identical outputs when run with the dmesg, ps, files, mount, and net sub-commands.

Signed-off-by Eric DeVolder <[email protected]>

---
v1: Posted 06jul2017 to kexec-tools mailing list
- original
---
makedumpfile.c | 20 +++++++++++++++++---
makedumpfile.h | 4 +++-

2 files changed, 20 insertions(+), 4 deletions(-)

makedumpfile.c +17 -3

index e69b6df..940f64c 100644
@@ -7236,7 +7236,7 @@ get_loads_dumpfile_cyclic(void)
 
 				/*
 				 * If the number of the contiguous pages to be excluded
-				 * is 256 or more, those pages are excluded really.
+				 * is PFN_EXCLUDED or more, those pages are excluded really.
 				 * And a new PT_LOAD segment is created.
 				 */
 				if (num_excluded >= PFN_EXCLUDED) {

@@ -7352,7 +7352,7 @@ write_elf_pages_cyclic(struct cache_data *cd_header, struct cache_data *cd_page)
 					continue;
 					/*
 					 * If the number of the contiguous pages to be excluded
-					 * is 255 or less, those pages are not excluded.
+					 * is less than PFN_EXCLUDED, those pages are not excluded.
 					 */
 				} else if (num_excluded < PFN_EXCLUDED) {
 					if ((pfn == pfn_end - 1) && frac_tail) {

@@ -7370,7 +7370,7 @@ write_elf_pages_cyclic(struct cache_data *cd_header, struct cache_data *cd_page)
 
 				/*
 				 * If the number of the contiguous pages to be excluded
-				 * is 256 or more, those pages are excluded really.
+				 * is PFN_EXCLUDED or more, those pages are excluded really.
 				 * And a new PT_LOAD segment is created.
 				 */
 				load.p_memsz = memsz;

@@ -11007,6 +11007,7 @@ static struct option longopts[] = {
 	{"splitblock-size", required_argument, NULL, OPT_SPLITBLOCK_SIZE},
 	{"work-dir", required_argument, NULL, OPT_WORKING_DIR},
 	{"num-threads", required_argument, NULL, OPT_NUM_THREADS},
+	{"exclude-threshold", required_argument, NULL, OPT_PFN_EXCLUDE_THRESHOLD},
 	{0, 0, 0, 0}
 };
 

@@ -11044,6 +11045,14 @@ main(int argc, char *argv[])
 	 */
 	info->flag_usemmap = MMAP_TRY;
 
+	/*
+	 * A run of zeros in the bitmap (excluded pages) of less than
+	 * pfn_excluded_threshold in length will still be dumped. Runs greater
+	 * than or equal to pfn_excluded_threshold will result in the creation
+	 * of a new output segment, for ELF dumps.
+	 */
+	info->pfn_exclude_threshold = 256;
+
 	info->block_order = DEFAULT_ORDER;
 	message_level = DEFAULT_MSG_LEVEL;
 	while ((opt = getopt_long(argc, argv, "b:cDd:eEFfg:hi:lpRvXx:", longopts,

@@ -11163,6 +11172,11 @@ main(int argc, char *argv[])
 		case OPT_NUM_THREADS:
 			info->num_threads = MAX(atoi(optarg), 0);
 			break;
+		case OPT_PFN_EXCLUDE_THRESHOLD:
+			info->pfn_exclude_threshold = strtoul(optarg, NULL, 0);
+			if (0 == info->pfn_exclude_threshold)
+				info->pfn_exclude_threshold = 1;
+			break;
 		case '?':
 			MSG("Commandline parameter is invalid.\n");
 			MSG("Try `makedumpfile --help' for more information.\n");

makedumpfile.h +3 -2

index e32e567..33d3eb0 100644
@@ -216,7 +216,7 @@ isAnon(unsigned long mapping)
 
 #define BITPERBYTE		(8)
 #define PGMM_CACHED		(512)
-#define PFN_EXCLUDED		(256)
+#define PFN_EXCLUDED		(info->pfn_exclude_threshold)
 #define BUFSIZE			(1024)
 #define BUFSIZE_FGETS		(1500)
 #define BUFSIZE_BITMAP		(4096)

@@ -1139,6 +1139,7 @@ struct DumpInfo {
 	long		page_size;           /* size of page */
 	long		page_shift;
 	mdf_pfn_t	max_mapnr;   /* number of page descriptor */
+	unsigned long	pfn_exclude_threshold;
 	unsigned long   page_offset;
 	unsigned long   section_size_bits;
 	unsigned long   max_physmem_bits;

@@ -2143,6 +2144,7 @@ struct elf_prstatus {
 #define OPT_SPLITBLOCK_SIZE	OPT_START+14
 #define OPT_WORKING_DIR         OPT_START+15
 #define OPT_NUM_THREADS	OPT_START+16
+#define OPT_PFN_EXCLUDE_THRESHOLD	OPT_START+17
 
 /*
 * Function Prototype.
--

2.7.4

Quoted 29 lines

The PFN_EXCLUDED value is used to control at which point a run of zeros in the bitmap (zeros denote excluded pages) is large enough to warrant truncating the current output segment and to create a new output segment (containing non-excluded pages), in an ELF dump.

If the run is smaller than PFN_EXCLUDED, then those excluded pages are still output in the ELF dump, for the current output segment.

By using smaller values of PFN_EXCLUDED, the resulting dump file size can be made smaller by actually removing more excluded pages from the resulting dump file.

This patch adds the command line option --exclude-threshold=<value> to indicate the threshold. The default is 256, the legacy value of PFN_EXCLUDED. The smallest value permitted is 1.

Using an existing vmcore, this was tested by the following:

% makedumpfile -E -d31 --exclude-threshold=256 -x vmlinux vmcore newvmcore256 % makedumpfile -E -d31 --exclude-threshold=4 -x vmlinux vmcore newvmcore4

I utilize -d31 in order to exclude as many page types as possible, resulting in a [significantly] smaller file sizes than the original vmcore.

(quoted diff) +0 -3

-rwxrwx--- 1 edevolde edevolde 4034564096 Jun 27 10:24 vmcore
-rw------- 1 edevolde edevolde 119808156 Jul  6 13:01 newvmcore256
-rw------- 1 edevolde edevolde 100811276 Jul  6 13:08 newvmcore4

Quoted 3 lines

The use of smaller value of PFN_EXCLUDED increases the number of output segments (the 'Number of program headers' in the readelf output) in the ELF dump file.

How will you tune the value ? I'm not sure what is the benefit of the tunable PFN_EXCLUDED. If there is no regression caused by too many PT_LOAD entries, I think we can decide a concrete PFN_EXCLUDED.

The penalty for splitting PT_LOAD is the size of a PT_LOAD header, so the best PFN_EXCLUDED is the minimum number which meets the condition below:

(size of PT_LOAD header) < (PFN_EXCLUDED << PAGE_SIZE)

Quoted 73 lines

% readelf -h vmcore
ELF Header:
Magic: 7f 45 4c 46 02 01 01 00 00 00 00 00 00 00 00 00

 Class:                             ELF64
 Data:                              2's complement, little endian
 Version:                           1 (current)
 OS/ABI:                            UNIX - System V
 ABI Version:                       0
 Type:                              CORE (Core file)
 Machine:                           Advanced Micro Devices X86-64
 Version:                           0x1
 Entry point address:               0x0
 Start of program headers:          64 (bytes into file)
 Start of section headers:          0 (bytes into file)
 Flags:                             0x0
 Size of this header:               64 (bytes)
 Size of program headers:           56 (bytes)
 Number of program headers:         6
                                    ^^^

Size of section headers: 0 (bytes)
Number of section headers: 0
Section header string table index: 0

% readelf -h newvmcore256
ELF Header:
Magic: 7f 45 4c 46 02 01 01 00 00 00 00 00 00 00 00 00

 Class:                             ELF64
 Data:                              2's complement, little endian
 Version:                           1 (current)
 OS/ABI:                            UNIX - System V
 ABI Version:                       0
 Type:                              CORE (Core file)
 Machine:                           Advanced Micro Devices X86-64
 Version:                           0x1
 Entry point address:               0x0
 Start of program headers:          64 (bytes into file)
 Start of section headers:          0 (bytes into file)
 Flags:                             0x0
 Size of this header:               64 (bytes)
 Size of program headers:           56 (bytes)
 Number of program headers:         18
                                    ^^^

Size of section headers: 0 (bytes)
Number of section headers: 0
Section header string table index: 0

% readelf -h newvmcore4
ELF Header:
Magic: 7f 45 4c 46 02 01 01 00 00 00 00 00 00 00 00 00

 Class:                             ELF64
 Data:                              2's complement, little endian
 Version:                           1 (current)
 OS/ABI:                            UNIX - System V
 ABI Version:                       0
 Type:                              CORE (Core file)
 Machine:                           Advanced Micro Devices X86-64
 Version:                           0x1
 Entry point address:               0x0
 Start of program headers:          64 (bytes into file)
 Start of section headers:          0 (bytes into file)
 Flags:                             0x0
 Size of this header:               64 (bytes)
 Size of program headers:           56 (bytes)
 Number of program headers:         244
                                    ^^^

Size of section headers: 0 (bytes)
Number of section headers: 0
Section header string table index: 0

The newvmcore4 has an even smaller file size than newvmcore256, with the small price being that there are now 244 rather than 18 segments in the dump file.

Quoted 3 lines

And with a larger number of segments, loading both vmcore and newvmcore4 into 'crash' resulted in identical outputs when run with the dmesg, ps, files, mount, and net sub-commands.

What about the processing speed of crash, is there no slow down ?

Thanks,
Atsushi Kumagai

Quoted 102 lines

Signed-off-by Eric DeVolder <[email protected]>

---
v1: Posted 06jul2017 to kexec-tools mailing list
- original
---
makedumpfile.c | 20 +++++++++++++++++---
makedumpfile.h | 4 +++-

2 files changed, 20 insertions(+), 4 deletions(-)

makedumpfile.c +17 -3

index e69b6df..940f64c 100644
@@ -7236,7 +7236,7 @@ get_loads_dumpfile_cyclic(void)
 
 				/*
 				 * If the number of the contiguous pages to be excluded
-				 * is 256 or more, those pages are excluded really.
+				 * is PFN_EXCLUDED or more, those pages are excluded really.
 				 * And a new PT_LOAD segment is created.
 				 */
 				if (num_excluded >= PFN_EXCLUDED) {

@@ -7352,7 +7352,7 @@ write_elf_pages_cyclic(struct cache_data *cd_header, struct cache_data *cd_page)
 					continue;
 					/*
 					 * If the number of the contiguous pages to be excluded
-					 * is 255 or less, those pages are not excluded.
+					 * is less than PFN_EXCLUDED, those pages are not excluded.
 					 */
 				} else if (num_excluded < PFN_EXCLUDED) {
 					if ((pfn == pfn_end - 1) && frac_tail) {

@@ -7370,7 +7370,7 @@ write_elf_pages_cyclic(struct cache_data *cd_header, struct cache_data *cd_page)
 
 				/*
 				 * If the number of the contiguous pages to be excluded
-				 * is 256 or more, those pages are excluded really.
+				 * is PFN_EXCLUDED or more, those pages are excluded really.
 				 * And a new PT_LOAD segment is created.
 				 */
 				load.p_memsz = memsz;

@@ -11007,6 +11007,7 @@ static struct option longopts[] = {
 	{"splitblock-size", required_argument, NULL, OPT_SPLITBLOCK_SIZE},
 	{"work-dir", required_argument, NULL, OPT_WORKING_DIR},
 	{"num-threads", required_argument, NULL, OPT_NUM_THREADS},
+	{"exclude-threshold", required_argument, NULL, OPT_PFN_EXCLUDE_THRESHOLD},
 	{0, 0, 0, 0}
 };
 

@@ -11044,6 +11045,14 @@ main(int argc, char *argv[])
 	 */
 	info->flag_usemmap = MMAP_TRY;
 
+	/*
+	 * A run of zeros in the bitmap (excluded pages) of less than
+	 * pfn_excluded_threshold in length will still be dumped. Runs greater
+	 * than or equal to pfn_excluded_threshold will result in the creation
+	 * of a new output segment, for ELF dumps.
+	 */
+	info->pfn_exclude_threshold = 256;
+
 	info->block_order = DEFAULT_ORDER;
 	message_level = DEFAULT_MSG_LEVEL;
 	while ((opt = getopt_long(argc, argv, "b:cDd:eEFfg:hi:lpRvXx:", longopts,

@@ -11163,6 +11172,11 @@ main(int argc, char *argv[])
 		case OPT_NUM_THREADS:
 			info->num_threads = MAX(atoi(optarg), 0);
 			break;
+		case OPT_PFN_EXCLUDE_THRESHOLD:
+			info->pfn_exclude_threshold = strtoul(optarg, NULL, 0);
+			if (0 == info->pfn_exclude_threshold)
+				info->pfn_exclude_threshold = 1;
+			break;
 		case '?':
 			MSG("Commandline parameter is invalid.\n");
 			MSG("Try `makedumpfile --help' for more information.\n");

makedumpfile.h +3 -1

index e32e567..33d3eb0 100644
@@ -216,7 +216,7 @@ isAnon(unsigned long mapping)
 
 #define BITPERBYTE		(8)
 #define PGMM_CACHED		(512)
-#define PFN_EXCLUDED		(256)
+#define PFN_EXCLUDED		(info->pfn_exclude_threshold)
 #define BUFSIZE			(1024)
 #define BUFSIZE_FGETS		(1500)
 #define BUFSIZE_BITMAP		(4096)

@@ -1139,6 +1139,7 @@ struct DumpInfo {
 	long		page_size;           /* size of page */
 	long		page_shift;
 	mdf_pfn_t	max_mapnr;   /* number of page descriptor */
+	unsigned long	pfn_exclude_threshold;
 	unsigned long   page_offset;
 	unsigned long   section_size_bits;
 	unsigned long   max_physmem_bits;

@@ -2143,6 +2144,7 @@ struct elf_prstatus {
 #define OPT_SPLITBLOCK_SIZE	OPT_START+14
 #define OPT_WORKING_DIR         OPT_START+15
 #define OPT_NUM_THREADS	OPT_START+16
+#define OPT_PFN_EXCLUDE_THRESHOLD	OPT_START+17
 

Quoted 4 lines

makedumpfile.h +0 -1

 /*
 * Function Prototype.
--

2.7.4

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

Hi Atsushi,
please see below.
eric

Quoted 31 lines · Atsushi Kumagai

Quoted 31 lines

The PFN_EXCLUDED value is used to control at which point a run of zeros in the bitmap (zeros denote excluded pages) is large enough to warrant truncating the current output segment and to create a new output segment (containing non-excluded pages), in an ELF dump.

If the run is smaller than PFN_EXCLUDED, then those excluded pages are still output in the ELF dump, for the current output segment.

By using smaller values of PFN_EXCLUDED, the resulting dump file size can be made smaller by actually removing more excluded pages from the resulting dump file.

This patch adds the command line option --exclude-threshold=<value> to indicate the threshold. The default is 256, the legacy value of PFN_EXCLUDED. The smallest value permitted is 1.

Using an existing vmcore, this was tested by the following:

% makedumpfile -E -d31 --exclude-threshold=256 -x vmlinux vmcore newvmcore256
% makedumpfile -E -d31 --exclude-threshold=4 -x vmlinux vmcore newvmcore4

I utilize -d31 in order to exclude as many page types as possible, resulting in a [significantly] smaller file sizes than the original vmcore.

(quoted diff) +0 -3

-rwxrwx--- 1 edevolde edevolde 4034564096 Jun 27 10:24 vmcore
-rw------- 1 edevolde edevolde 119808156 Jul  6 13:01 newvmcore256
-rw------- 1 edevolde edevolde 100811276 Jul  6 13:08 newvmcore4

Quoted 8 lines

Quoted 3 lines

The use of smaller value of PFN_EXCLUDED increases the number of output segments (the 'Number of program headers' in the readelf output) in the ELF dump file.

How will you tune the value ? I'm not sure what is the benefit of the tunable PFN_EXCLUDED. If there is no regression caused by too many PT_LOAD
entries, I think we can decide a concrete PFN_EXCLUDED.

Allow me note two things prior to addressing the question.

Note that the value for PFN_EXCLUDED really should be in the range:

1 <= PFN_EXCLUDED <= NUM_PAGES(largest segment)

but that values larger than NUM_PAGES(largest segment) behave the same as NUM_PAGES(largest segment) and simply prevent makedumpfile from ever omitting excluded pages from the dump file.

Also note that the ELF header allows for a 16-bit e_phnum value for the number of segments in the dump file. As it stands today, I doubt that anybody has come close to reaching 65535 segments, but the combination of larger and larger memories as well as the work we (Oracle) are doing to further enhance the capabilities of makedumpfile, I believe we will start to challenge this 65535 number.

The ability to tune PFN_EXCLUDED allows one to minimize file size while still staying within ELF boundaries.

There are two ways in which have PFN_EXCLUDED as a tunable parameter benefits the user.

The first benefit is, when making PFN_EXCLUDED smaller, makedumpfile has more opportunities to NOT write excluded pages to the resulting dump file, thus obtaining a smaller overall dump file size. And since a PT_LOAD header is smaller than a page, this penalty (of more segments) will always result in a smaller file size. (In the example I cite the dump file was 18MB smaller with a PFN_EXCLUDED value of 4 than default 256, in spite of increasing the number of segments from 6 to 244).

The second benefit is, when enabling PFN_EXCLUDED to become larger, it allows makedumpfile to continue to generate valid ELF dump files in the presence of larger and larger memory systems. Generally speaking, the goal is to minimize the size of dump files via the exclusion of uninteresting pages (ie zero, free, etc), especially as the size of memory continues to grow and grow. As the memory increases, there are more and more of these uninteresting pages, and more opportunities for makedumpfile to omit them (even at the current PFN_EXCLUDED value of 256). Furthermore, we are working on additional page exclusion strategies that will drive more and more opportunities for makedumpfile to omit these pages from the dump file. And as makedumpfile omits more and more pages from the dump file, that increases the number of segments needed.

By enabling a user to tune the value of PFN_EXCLUDED, we provide an additional mechanism to balance the size of the ELF dump file with respect to the size of memory.

Quoted 9 lines

The penalty for splitting PT_LOAD is the size of a PT_LOAD header, so the best PFN_EXCLUDED is the minimum number which meets the condition below:

(size of PT_LOAD header) < (PFN_EXCLUDED << PAGE_SIZE)

I admit I don't quite understand, would you mind working through an example or two?

It seems to me that a PT_LOAD header of 56 bytes is always less than a page_size of 4096 bytes, and would always be a win (meaning a new segment is better than dumping the page), especially as the consecutive number of physically contiguous excluded pages is large.

The only caveat here being that we can not exceed e_phnum limitation of 65535 segments.

Quoted 73 lines

Quoted 73 lines

% readelf -h vmcore
ELF Header:
Magic: 7f 45 4c 46 02 01 01 00 00 00 00 00 00 00 00 00

 Class:                             ELF64
 Data:                              2's complement, little endian
 Version:                           1 (current)
 OS/ABI:                            UNIX - System V
 ABI Version:                       0
 Type:                              CORE (Core file)
 Machine:                           Advanced Micro Devices X86-64
 Version:                           0x1
 Entry point address:               0x0
 Start of program headers:          64 (bytes into file)
 Start of section headers:          0 (bytes into file)
 Flags:                             0x0
 Size of this header:               64 (bytes)
 Size of program headers:           56 (bytes)
 Number of program headers:         6
                                    ^^^

Size of section headers: 0 (bytes)
Number of section headers: 0
Section header string table index: 0

% readelf -h newvmcore256
ELF Header:
Magic: 7f 45 4c 46 02 01 01 00 00 00 00 00 00 00 00 00

 Class:                             ELF64
 Data:                              2's complement, little endian
 Version:                           1 (current)
 OS/ABI:                            UNIX - System V
 ABI Version:                       0
 Type:                              CORE (Core file)
 Machine:                           Advanced Micro Devices X86-64
 Version:                           0x1
 Entry point address:               0x0
 Start of program headers:          64 (bytes into file)
 Start of section headers:          0 (bytes into file)
 Flags:                             0x0
 Size of this header:               64 (bytes)
 Size of program headers:           56 (bytes)
 Number of program headers:         18
                                    ^^^

Size of section headers: 0 (bytes)
Number of section headers: 0
Section header string table index: 0

% readelf -h newvmcore4
ELF Header:
Magic: 7f 45 4c 46 02 01 01 00 00 00 00 00 00 00 00 00

 Class:                             ELF64
 Data:                              2's complement, little endian
 Version:                           1 (current)
 OS/ABI:                            UNIX - System V
 ABI Version:                       0
 Type:                              CORE (Core file)
 Machine:                           Advanced Micro Devices X86-64
 Version:                           0x1
 Entry point address:               0x0
 Start of program headers:          64 (bytes into file)
 Start of section headers:          0 (bytes into file)
 Flags:                             0x0
 Size of this header:               64 (bytes)
 Size of program headers:           56 (bytes)
 Number of program headers:         244
                                    ^^^

Size of section headers: 0 (bytes)
Number of section headers: 0
Section header string table index: 0

The newvmcore4 has an even smaller file size than newvmcore256, with the small price being that there are now 244 rather than 18 segments in the dump file.

Quoted 5 lines

Quoted 3 lines

And with a larger number of segments, loading both vmcore and newvmcore4 into 'crash' resulted in identical outputs when run with the dmesg, ps, files, mount, and net sub-commands.

What about the processing speed of crash, is there no slow down ?

I did not observe a noticeable change in processing speed of crash.

Quoted 117 lines

Thanks,
Atsushi Kumagai

Quoted 112 lines

Signed-off-by Eric DeVolder <[email protected]>

---
v1: Posted 06jul2017 to kexec-tools mailing list
- original
---
makedumpfile.c | 20 +++++++++++++++++---
makedumpfile.h | 4 +++-

2 files changed, 20 insertions(+), 4 deletions(-)

makedumpfile.c +1 -1

index e69b6df..940f64c 100644
@@ -7236,7 +7236,7 @@ get_loads_dumpfile_cyclic(void)
 
                /*
                 * If the number of the contiguous pages to be excluded
-                 * is 256 or more, those pages are excluded really.
+                 * is PFN_EXCLUDED or more, those pages are excluded

really.

                 * And a new PT_LOAD segment is created.
                 */
                if (num_excluded >= PFN_EXCLUDED) {

makedumpfile.c +0 -0

@@ -7352,7 +7352,7 @@ write_elf_pages_cyclic(struct cache_data
 *cd_header, struct cache_data *cd_page)
                    continue;
                    /*
                     * If the number of the contiguous pages to be

excluded

makedumpfile.c +1 -1

-                     * is 255 or less, those pages are not excluded.
+                     * is less than PFN_EXCLUDED, those pages are

not excluded.

                     */
                } else if (num_excluded < PFN_EXCLUDED) {
                    if ((pfn == pfn_end - 1) && frac_tail) {

makedumpfile.c +1 -1

@@ -7370,7 +7370,7 @@ write_elf_pages_cyclic(struct cache_data
 *cd_header, struct cache_data *cd_page)
 
                /*
                 * If the number of the contiguous pages to be excluded
-                 * is 256 or more, those pages are excluded really.
+                 * is PFN_EXCLUDED or more, those pages are excluded

really.

                 * And a new PT_LOAD segment is created.
                 */
                load.p_memsz = memsz;

makedumpfile.c +4 -0

@@ -11007,6 +11007,7 @@ static struct option longopts[] = {
    {"splitblock-size", required_argument, NULL, OPT_SPLITBLOCK_SIZE},
    {"work-dir", required_argument, NULL, OPT_WORKING_DIR},
    {"num-threads", required_argument, NULL, OPT_NUM_THREADS},
+    {"exclude-threshold", required_argument, NULL,
 OPT_PFN_EXCLUDE_THRESHOLD},
    {0, 0, 0, 0}
 };

@@ -11044,6 +11045,14 @@ main(int argc, char *argv[])
     */
    info->flag_usemmap = MMAP_TRY;
 
+    /*
+     * A run of zeros in the bitmap (excluded pages) of less than
+     * pfn_excluded_threshold in length will still be dumped. Runs

greater
+ * than or equal to pfn_excluded_threshold will result in the
creation

makedumpfile.c +4 -0

+     * of a new output segment, for ELF dumps.
+     */
+    info->pfn_exclude_threshold = 256;
+
    info->block_order = DEFAULT_ORDER;
    message_level = DEFAULT_MSG_LEVEL;
    while ((opt = getopt_long(argc, argv, "b:cDd:eEFfg:hi:lpRvXx:",

longopts,

makedumpfile.c +5 -0

@@ -11163,6 +11172,11 @@ main(int argc, char *argv[])
        case OPT_NUM_THREADS:
            info->num_threads = MAX(atoi(optarg), 0);
            break;
+        case OPT_PFN_EXCLUDE_THRESHOLD:
+            info->pfn_exclude_threshold = strtoul(optarg, NULL, 0);
+            if (0 == info->pfn_exclude_threshold)
+                info->pfn_exclude_threshold = 1;
+            break;
        case '?':
            MSG("Commandline parameter is invalid.\n");
            MSG("Try `makedumpfile --help' for more information.\n");

makedumpfile.h +3 -1

index e32e567..33d3eb0 100644
@@ -216,7 +216,7 @@ isAnon(unsigned long mapping)
 
 #define BITPERBYTE        (8)
 #define PGMM_CACHED        (512)
-#define PFN_EXCLUDED        (256)
+#define PFN_EXCLUDED        (info->pfn_exclude_threshold)
 #define BUFSIZE            (1024)
 #define BUFSIZE_FGETS        (1500)
 #define BUFSIZE_BITMAP        (4096)

@@ -1139,6 +1139,7 @@ struct DumpInfo {
    long        page_size;           /* size of page */
    long        page_shift;
    mdf_pfn_t    max_mapnr;   /* number of page descriptor */
+    unsigned long    pfn_exclude_threshold;
    unsigned long   page_offset;
    unsigned long   section_size_bits;
    unsigned long   max_physmem_bits;

@@ -2143,6 +2144,7 @@ struct elf_prstatus {
 #define OPT_SPLITBLOCK_SIZE    OPT_START+14
 #define OPT_WORKING_DIR         OPT_START+15
 #define OPT_NUM_THREADS    OPT_START+16
+#define OPT_PFN_EXCLUDE_THRESHOLD    OPT_START+17
 

Quoted 12 lines

Quoted 4 lines

makedumpfile.h +0 -1

 /*
 * Function Prototype.
--

2.7.4

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

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

Quoted 51 lines · Atsushi Kumagai

Quoted 31 lines

Quoted 31 lines

The PFN_EXCLUDED value is used to control at which point a run of zeros in the bitmap (zeros denote excluded pages) is large enough to warrant truncating the current output segment and to create a new output segment (containing non-excluded pages), in an ELF dump.

If the run is smaller than PFN_EXCLUDED, then those excluded pages are still output in the ELF dump, for the current output segment.

By using smaller values of PFN_EXCLUDED, the resulting dump file size can be made smaller by actually removing more excluded pages from the resulting dump file.

This patch adds the command line option --exclude-threshold=<value> to indicate the threshold. The default is 256, the legacy value of PFN_EXCLUDED. The smallest value permitted is 1.

Using an existing vmcore, this was tested by the following:

% makedumpfile -E -d31 --exclude-threshold=256 -x vmlinux vmcore newvmcore256
% makedumpfile -E -d31 --exclude-threshold=4 -x vmlinux vmcore newvmcore4

I utilize -d31 in order to exclude as many page types as possible, resulting in a [significantly] smaller file sizes than the original vmcore.

(quoted diff) +0 -3

-rwxrwx--- 1 edevolde edevolde 4034564096 Jun 27 10:24 vmcore
-rw------- 1 edevolde edevolde 119808156 Jul  6 13:01 newvmcore256
-rw------- 1 edevolde edevolde 100811276 Jul  6 13:08 newvmcore4

Quoted 8 lines

Quoted 3 lines

The use of smaller value of PFN_EXCLUDED increases the number of output segments (the 'Number of program headers' in the readelf output) in the ELF dump file.

How will you tune the value ? I'm not sure what is the benefit of the tunable PFN_EXCLUDED. If there is no regression caused by too many PT_LOAD
entries, I think we can decide a concrete PFN_EXCLUDED.

Allow me note two things prior to addressing the question.

Note that the value for PFN_EXCLUDED really should be in the range:

1 <= PFN_EXCLUDED <= NUM_PAGES(largest segment)

but that values larger than NUM_PAGES(largest segment) behave the same as NUM_PAGES(largest segment) and simply prevent makedumpfile from ever omitting excluded pages from the dump file.

Quoted 6 lines

Also note that the ELF header allows for a 16-bit e_phnum value for the number of segments in the dump file. As it stands today, I doubt that anybody has come close to reaching 65535 segments, but the combination of larger and larger memories as well as the work we (Oracle) are doing to further enhance the capabilities of makedumpfile, I believe we will start to challenge this 65535 number.

Quoted 28 lines

The ability to tune PFN_EXCLUDED allows one to minimize file size while still staying within ELF boundaries.

There are two ways in which have PFN_EXCLUDED as a tunable parameter benefits the user.

The first benefit is, when making PFN_EXCLUDED smaller, makedumpfile has more opportunities to NOT write excluded pages to the resulting dump file, thus obtaining a smaller overall dump file size. And since a PT_LOAD header is smaller than a page, this penalty (of more segments) will always result in a smaller file size. (In the example I cite the dump file was 18MB smaller with a PFN_EXCLUDED value of 4 than default 256, in spite of increasing the number of segments from 6 to 244).

The second benefit is, when enabling PFN_EXCLUDED to become larger, it allows makedumpfile to continue to generate valid ELF dump files in the presence of larger and larger memory systems. Generally speaking, the goal is to minimize the size of dump files via the exclusion of uninteresting pages (ie zero, free, etc), especially as the size of memory continues to grow and grow. As the memory increases, there are more and more of these uninteresting pages, and more opportunities for makedumpfile to omit them (even at the current PFN_EXCLUDED value of 256). Furthermore, we are working on additional page exclusion strategies that will drive more and more opportunities for makedumpfile to omit these pages from the dump file. And as makedumpfile omits more and more pages from the dump file, that increases the number of segments needed.

Quoted 3 lines

By enabling a user to tune the value of PFN_EXCLUDED, we provide an additional mechanism to balance the size of the ELF dump file with respect to the size of memory.

It occurred to me that offering the option "--exclude-threshold=auto" whereby a binary search on the second bitmap in the function get_loads_dumpfile_cyclic() to determine the optimum value of PFN_EXCLUDED (optimum here meaning the smallest possible value while still staying within 65535 segments, which would yield the smallest possible dump file size for the given constraints) would be an excellent feature to have?

Quoted 7 lines

Quoted 5 lines

Quoted 3 lines

And with a larger number of segments, loading both vmcore and newvmcore4 into 'crash' resulted in identical outputs when run with the dmesg, ps, files, mount, and net sub-commands.

What about the processing speed of crash, is there no slow down ?

I did not observe a noticeable change in processing speed of crash.

Quoted 131 lines

Quoted 129 lines

Thanks,
Atsushi Kumagai

Quoted 115 lines

Signed-off-by Eric DeVolder <[email protected]>

---
v1: Posted 06jul2017 to kexec-tools mailing list
- original
---
makedumpfile.c | 20 +++++++++++++++++---
makedumpfile.h | 4 +++-

2 files changed, 20 insertions(+), 4 deletions(-)

makedumpfile.c +1 -1

index e69b6df..940f64c 100644
@@ -7236,7 +7236,7 @@ get_loads_dumpfile_cyclic(void)
 
                /*
                 * If the number of the contiguous pages to be excluded
-                 * is 256 or more, those pages are excluded really.
+                 * is PFN_EXCLUDED or more, those pages are excluded

really.

                 * And a new PT_LOAD segment is created.
                 */
                if (num_excluded >= PFN_EXCLUDED) {

makedumpfile.c +0 -0

@@ -7352,7 +7352,7 @@ write_elf_pages_cyclic(struct cache_data
 *cd_header, struct cache_data *cd_page)
                    continue;
                    /*
                     * If the number of the contiguous pages to be

excluded

makedumpfile.c +1 -1

-                     * is 255 or less, those pages are not excluded.
+                     * is less than PFN_EXCLUDED, those pages are

not excluded.

                     */
                } else if (num_excluded < PFN_EXCLUDED) {
                    if ((pfn == pfn_end - 1) && frac_tail) {

makedumpfile.c +1 -1

@@ -7370,7 +7370,7 @@ write_elf_pages_cyclic(struct cache_data
 *cd_header, struct cache_data *cd_page)
 
                /*
                 * If the number of the contiguous pages to be excluded
-                 * is 256 or more, those pages are excluded really.
+                 * is PFN_EXCLUDED or more, those pages are excluded

really.

                 * And a new PT_LOAD segment is created.
                 */
                load.p_memsz = memsz;

makedumpfile.c +4 -0

@@ -11007,6 +11007,7 @@ static struct option longopts[] = {
    {"splitblock-size", required_argument, NULL, OPT_SPLITBLOCK_SIZE},
    {"work-dir", required_argument, NULL, OPT_WORKING_DIR},
    {"num-threads", required_argument, NULL, OPT_NUM_THREADS},
+    {"exclude-threshold", required_argument, NULL,
 OPT_PFN_EXCLUDE_THRESHOLD},
    {0, 0, 0, 0}
 };

@@ -11044,6 +11045,14 @@ main(int argc, char *argv[])
     */
    info->flag_usemmap = MMAP_TRY;
 
+    /*
+     * A run of zeros in the bitmap (excluded pages) of less than
+     * pfn_excluded_threshold in length will still be dumped. Runs

greater
+ * than or equal to pfn_excluded_threshold will result in the
creation

makedumpfile.c +4 -0

+     * of a new output segment, for ELF dumps.
+     */
+    info->pfn_exclude_threshold = 256;
+
    info->block_order = DEFAULT_ORDER;
    message_level = DEFAULT_MSG_LEVEL;
    while ((opt = getopt_long(argc, argv, "b:cDd:eEFfg:hi:lpRvXx:",

longopts,

makedumpfile.c +5 -0

@@ -11163,6 +11172,11 @@ main(int argc, char *argv[])
        case OPT_NUM_THREADS:
            info->num_threads = MAX(atoi(optarg), 0);
            break;
+        case OPT_PFN_EXCLUDE_THRESHOLD:
+            info->pfn_exclude_threshold = strtoul(optarg, NULL, 0);
+            if (0 == info->pfn_exclude_threshold)
+                info->pfn_exclude_threshold = 1;
+            break;
        case '?':
            MSG("Commandline parameter is invalid.\n");
            MSG("Try `makedumpfile --help' for more information.\n");

makedumpfile.h +3 -2

index e32e567..33d3eb0 100644
@@ -216,7 +216,7 @@ isAnon(unsigned long mapping)
 
 #define BITPERBYTE        (8)
 #define PGMM_CACHED        (512)
-#define PFN_EXCLUDED        (256)
+#define PFN_EXCLUDED        (info->pfn_exclude_threshold)
 #define BUFSIZE            (1024)
 #define BUFSIZE_FGETS        (1500)
 #define BUFSIZE_BITMAP        (4096)

@@ -1139,6 +1139,7 @@ struct DumpInfo {
    long        page_size;           /* size of page */
    long        page_shift;
    mdf_pfn_t    max_mapnr;   /* number of page descriptor */
+    unsigned long    pfn_exclude_threshold;
    unsigned long   page_offset;
    unsigned long   section_size_bits;
    unsigned long   max_physmem_bits;

@@ -2143,6 +2144,7 @@ struct elf_prstatus {
 #define OPT_SPLITBLOCK_SIZE    OPT_START+14
 #define OPT_WORKING_DIR         OPT_START+15
 #define OPT_NUM_THREADS    OPT_START+16
+#define OPT_PFN_EXCLUDE_THRESHOLD    OPT_START+17
 
 /*
 * Function Prototype.
--

Quoted 1 line

2.7.4

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

Quoted 4 lines

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

Hello Eric,

Quoted 51 lines

Quoted 51 lines · Atsushi Kumagai

Quoted 31 lines

Quoted 31 lines

The PFN_EXCLUDED value is used to control at which point a run of zeros in the bitmap (zeros denote excluded pages) is large enough to warrant truncating the current output segment and to create a new output segment (containing non-excluded pages), in an ELF dump.

If the run is smaller than PFN_EXCLUDED, then those excluded pages are still output in the ELF dump, for the current output segment.

By using smaller values of PFN_EXCLUDED, the resulting dump file size can be made smaller by actually removing more excluded pages from the resulting dump file.

This patch adds the command line option --exclude-threshold=<value> to indicate the threshold. The default is 256, the legacy value of PFN_EXCLUDED. The smallest value permitted is 1.

Using an existing vmcore, this was tested by the following:

% makedumpfile -E -d31 --exclude-threshold=256 -x vmlinux vmcore newvmcore256
% makedumpfile -E -d31 --exclude-threshold=4 -x vmlinux vmcore newvmcore4

I utilize -d31 in order to exclude as many page types as possible, resulting in a [significantly] smaller file sizes than the original vmcore.

(quoted diff) +0 -3

-rwxrwx--- 1 edevolde edevolde 4034564096 Jun 27 10:24 vmcore
-rw------- 1 edevolde edevolde 119808156 Jul  6 13:01 newvmcore256
-rw------- 1 edevolde edevolde 100811276 Jul  6 13:08 newvmcore4

Quoted 8 lines

Quoted 3 lines

The use of smaller value of PFN_EXCLUDED increases the number of output segments (the 'Number of program headers' in the readelf output) in the ELF dump file.

How will you tune the value ? I'm not sure what is the benefit of the tunable PFN_EXCLUDED. If there is no regression caused by too many PT_LOAD
entries, I think we can decide a concrete PFN_EXCLUDED.

Allow me note two things prior to addressing the question.

Note that the value for PFN_EXCLUDED really should be in the range:

1 <= PFN_EXCLUDED <= NUM_PAGES(largest segment)

but that values larger than NUM_PAGES(largest segment) behave the same as NUM_PAGES(largest segment) and simply prevent makedumpfile from ever omitting excluded pages from the dump file.

Quoted 6 lines

Quoted 6 lines

Also note that the ELF header allows for a 16-bit e_phnum value for the number of segments in the dump file. As it stands today, I doubt that anybody has come close to reaching 65535 segments, but the combination of larger and larger memories as well as the work we (Oracle) are doing to further enhance the capabilities of makedumpfile, I believe we will start to challenge this 65535 number.

I overlooked the limitation of the number of segments, so I considered only "The first benefit" you said below.

Quoted 32 lines

Quoted 28 lines

The ability to tune PFN_EXCLUDED allows one to minimize file size while still staying within ELF boundaries.

There are two ways in which have PFN_EXCLUDED as a tunable parameter benefits the user.

The first benefit is, when making PFN_EXCLUDED smaller, makedumpfile has more opportunities to NOT write excluded pages to the resulting dump file, thus obtaining a smaller overall dump file size. And since a PT_LOAD header is smaller than a page, this penalty (of more segments) will always result in a smaller file size. (In the example I cite the dump file was 18MB smaller with a PFN_EXCLUDED value of 4 than default 256, in spite of increasing the number of segments from 6 to 244).

The second benefit is, when enabling PFN_EXCLUDED to become larger, it allows makedumpfile to continue to generate valid ELF dump files in the presence of larger and larger memory systems. Generally speaking, the goal is to minimize the size of dump files via the exclusion of uninteresting pages (ie zero, free, etc), especially as the size of memory continues to grow and grow. As the memory increases, there are more and more of these uninteresting pages, and more opportunities for makedumpfile to omit them (even at the current PFN_EXCLUDED value of 256). Furthermore, we are working on additional page exclusion strategies that will drive more and more opportunities for makedumpfile to omit these pages from the dump file. And as makedumpfile omits more and more pages from the dump file, that increases the number of segments needed.

Quoted 3 lines

By enabling a user to tune the value of PFN_EXCLUDED, we provide an additional mechanism to balance the size of the ELF dump file with respect to the size of memory.

Quoted 7 lines

It occurred to me that offering the option "--exclude-threshold=auto" whereby a binary search on the second bitmap in the function get_loads_dumpfile_cyclic() to determine the optimum value of PFN_EXCLUDED (optimum here meaning the smallest possible value while still staying within 65535 segments, which would yield the smallest possible dump file size for the given constraints) would be an excellent feature to have?

I think the "auto" is necessary for --exclude-threshold, the optimum value should be calculated automatically. Otherwise, it imposes trial-and-error on users every time, it doesn't sound practical. IOW, this patch is unacceptable if there is no mechanism to support users.
So now, my only concern for this option is the processing time of the binary search.

[snip]

Quoted 4 lines

Quoted 4 lines

Quoted 4 lines

Quoted 3 lines

And with a larger number of segments, loading both vmcore and newvmcore4 into 'crash' resulted in identical outputs when run with the dmesg, ps, files, mount, and net sub-commands.

Quoted 3 lines

Quoted 3 lines

Quoted 1 line

What about the processing speed of crash, is there no slow down ?

I did not observe a noticeable change in processing speed of crash.

Good, it would be better to be represented by actual measured results.

Thanks,
Atsushi Kumagai