Skip to contentSkip to conversation

raid6check.c: reduce verbosity

PATCH In review Piergiorgio Sartor 2 participants 4 messages Feb 1, 2014 – Feb 5, 2014 MD RAID

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 raid6check.c: reduce verbosity

This patch removes some printouts, which are not really useful here.
These could be re-added later, in case a verbosity parameter will be provided.

Signed off: [email protected]

raid6check.c +0 -3
diff -uNrp a/raid6check.c b/raid6check.c
@@ -188,8 +188,6 @@ int check_stripes(struct mdinfo *info, i
 	while (length > 0) {
 		int disk[chunk_size >> CHECK_PAGE_BITS];
 
-		printf("pos --> %llu\n", start);
-
 		err = lock_stripe(info, start, chunk_size, data_disks, sig);
 		if(err != 0) {
 			if (err != 2)
@@ -221,7 +219,6 @@ int check_stripes(struct mdinfo *info, i
 			int disk = geo_map(i, start, raid_disks, level, layout);
 			blocks[i] = stripes[disk];
 			block_index_for_slot[disk] = i;
-			printf("%d->%d\n", i, disk);
 		}
 
 		qsyndrome(p, q, (uint8_t**)blocks, data_disks, chunk_size);
Signature · 1 line
piergiorgio
#1 lore raw
Quoted 22 lines · On Sat, 1 Feb 2014 17:03:34 +0100 Piergiorgio Sartor…

This patch removes some printouts, which are not really useful here.
These could be re-added later, in case a verbosity parameter will be provided.

Signed off: [email protected]

diff -uNrp a/raid6check.c b/raid6check.c

raid6check.c +0 -2
@@ -188,8 +188,6 @@ int check_stripes(struct mdinfo *info, i
 	while (length > 0) {
 		int disk[chunk_size >> CHECK_PAGE_BITS];
 
-		printf("pos --> %llu\n", start);
-
 		err = lock_stripe(info, start, chunk_size, data_disks, sig);
 		if(err != 0) {
 			if (err != 2)

raid6check.c int check_stripes(struct mdinfo *info, i · line 224

Quoted 7 lines
raid6check.c +0 -1
@@ -221,7 +219,6 @@ int check_stripes(struct mdinfo *info, i
 			int disk = geo_map(i, start, raid_disks, level, layout);
 			blocks[i] = stripes[disk];
 			block_index_for_slot[disk] = i;
-			printf("%d->%d\n", i, disk);
 		}
 
 		qsyndrome(p, q, (uint8_t**)blocks, data_disks, chunk_size);

This and other 3 all applied - thanks.

NeilBrown

#2 lore raw
Quoted 29 lines · NeilBrown
Quoted 26 lines · On Sat, 1 Feb 2014 17:03:34 +0100 Piergiorgio Sartor…

This patch removes some printouts, which are not really useful here.
These could be re-added later, in case a verbosity parameter will be provided.

Signed off: [email protected]

diff -uNrp a/raid6check.c b/raid6check.c

raid6check.c +0 -3
@@ -188,8 +188,6 @@ int check_stripes(struct mdinfo *info, i
 	while (length > 0) {
 		int disk[chunk_size >> CHECK_PAGE_BITS];
 
-		printf("pos --> %llu\n", start);
-
 		err = lock_stripe(info, start, chunk_size, data_disks, sig);
 		if(err != 0) {
 			if (err != 2)
@@ -221,7 +219,6 @@ int check_stripes(struct mdinfo *info, i
 			int disk = geo_map(i, start, raid_disks, level, layout);
 			blocks[i] = stripes[disk];
 			block_index_for_slot[disk] = i;
-			printf("%d->%d\n", i, disk);
 		}
 
Quoted 7 lines

raid6check.c int check_stripes(struct mdinfo *info, i · line 224

Quoted 3 lines
raid6check.c +0 -0
 		qsyndrome(p, q, (uint8_t**)blocks, data_disks, chunk_size);

This and other 3 all applied - thanks.

NeilBrown

Hi Neil,

thanks for the update.

It seems the patch removing the commented code (the one you complained about), was not added.

Below the patch again.

Thanks,

bye,

--- --- --- --- ---

This patch will remove some legacy code. It is part of the verbosity "cleanup". In any case, if information about the P and Q parity mismatches is required, it should go inside the code handling page size blocks, not full stripe size.

Signed off: [email protected]

raid6check.c 2014-01-23 19:36:23.151837870 +0100 renamed from raid6check.c 2014-01-20 19:48:10.435721613 +0100 +1 -9
diff -uNrp a/raid6check.c b/raid6check.c
@@ -231,15 +231,7 @@ int check_stripes(struct mdinfo *info, i
                block_index_for_slot[diskP] = data_disks;
                blocks[data_disks+1] = stripes[diskQ];
                block_index_for_slot[diskQ] = data_disks+1;
-/* Do we really need the code below? */
-#if 0
-               if (memcmp(p, stripes[diskP], chunk_size) != 0) {
-                       printf("P(%d) wrong at %llu\n", diskP, start);
-               }
-               if (memcmp(q, stripes[diskQ], chunk_size) != 0) {
-                       printf("Q(%d) wrong at %llu\n", diskQ, start);
-               }
-#endif
+
                raid6_collect(chunk_size, p, q, stripes[diskP], stripes[diskQ], results);
                raid6_stats(disk, results, raid_disks, chunk_size);
 
Signature · 1 line
piergiorgio
#3 lore raw
Quoted 74 lines · On Wed, 5 Feb 2014 20:18:45 +0100 Piergiorgio Sartor…
Quoted 29 lines · NeilBrown
Quoted 26 lines · On Sat, 1 Feb 2014 17:03:34 +0100 Piergiorgio Sartor…

This patch removes some printouts, which are not really useful here.
These could be re-added later, in case a verbosity parameter will be provided.

Signed off: [email protected]

diff -uNrp a/raid6check.c b/raid6check.c

raid6check.c +0 -3
@@ -188,8 +188,6 @@ int check_stripes(struct mdinfo *info, i
 	while (length > 0) {
 		int disk[chunk_size >> CHECK_PAGE_BITS];
 
-		printf("pos --> %llu\n", start);
-
 		err = lock_stripe(info, start, chunk_size, data_disks, sig);
 		if(err != 0) {
 			if (err != 2)
@@ -221,7 +219,6 @@ int check_stripes(struct mdinfo *info, i
 			int disk = geo_map(i, start, raid_disks, level, layout);
 			blocks[i] = stripes[disk];
 			block_index_for_slot[disk] = i;
-			printf("%d->%d\n", i, disk);
 		}
 
Quoted 7 lines

raid6check.c int check_stripes(struct mdinfo *info, i · line 224

Quoted 3 lines
raid6check.c +0 -0
 		qsyndrome(p, q, (uint8_t**)blocks, data_disks, chunk_size);

This and other 3 all applied - thanks.

NeilBrown

Hi Neil,

thanks for the update.

It seems the patch removing the commented code (the one you complained about), was not added.

Below the patch again.

Thanks,

bye,

--- --- --- --- ---

This patch will remove some legacy code. It is part of the verbosity "cleanup". In any case, if information about the P and Q parity mismatches is required, it should go inside the code handling page size blocks, not full stripe size.

Signed off: [email protected] diff -uNrp a/raid6check.c b/raid6check.c

raid6check.c 2014-01-23 19:36:23.151837870 +0100 renamed from raid6check.c 2014-01-20 19:48:10.435721613 +0100 +0 -3
@@ -231,15 +231,7 @@ int check_stripes(struct mdinfo *info, i
                block_index_for_slot[diskP] = data_disks;
                blocks[data_disks+1] = stripes[diskQ];
                block_index_for_slot[diskQ] = data_disks+1;
-/* Do we really need the code below? */
-#if 0
-               if (memcmp(p, stripes[diskP], chunk_size) != 0) {

raid6check.c 2014-01-23 19:36:23.151837870 +0100 int check_stripes(struct mdinfo *info, i · line 237

Quoted 10 lines
raid6check.c 2014-01-23 19:36:23.151837870 +0100 renamed from raid6check.c 2014-01-20 19:48:10.435721613 +0100 +1 -6
-                       printf("P(%d) wrong at %llu\n", diskP, start);
-               }
-               if (memcmp(q, stripes[diskQ], chunk_size) != 0) {
-                       printf("Q(%d) wrong at %llu\n", diskQ, start);
-               }
-#endif
+
                raid6_collect(chunk_size, p, q, stripes[diskP], stripes[diskQ], results);
                raid6_stats(disk, results, raid_disks, chunk_size);
 

Thanks for the reminder. This one is now applied.

NeilBrown