btrfs-progs: docs: add extra LLM disclosure following kernel guidelines

Documentation/btrfs-quota.rst

@@ -22,12 +22,23 @@ of a btrfs filesystem. The quota groups (qgroups) are managed by the subcommand
 STABILITY AND PERFORMANCE IMPLICATIONS
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
-The qgroup implementation is considered reasonably stable for daily use and has
-been enabled in various distributions.
+The qgroup mode is considered not recommended for daily usage, unless there is
+no planned new snapshot/subvolume creation and deletion.
 
-When quotas are activated, they affect all extent processing, which takes a
-performance hit. Activation of qgroups is not recommended unless the user
-intends to actually use them.
+The core design of qgroup mode and snapshot are not compatible from day one, and
+are the cause of all kinds of performance problems.
+
+When qgroup mode is activated, it affects all extent processing, which takes a
+performance hit. Operations that modify a whole subvolume/snapshot in one go,
+which include snapshot creation and subvolume deletion, are heavily affected and
+can cause a system hang due to the heavy load.
+
+Although the kernel is taking several workarounds, the problem is not fully resolved
+and has extra costs, e.g. marking qgroup inconsistent, breaking limits and
+requiring extra rescan.
+
+Activation of qgroups is not recommended unless the user intends to actually use them,
+and the usage does not involve new subvolume/snapshot.
 
 .. _man-quota-hierarchical-quota-group-concepts:
 

Documentation/dev/Developer-s-FAQ.rst

@@ -36,6 +36,10 @@ highlight the most frequently used tags and their expected meaning. This only
 briefly mentions the commonly used tags. You're encouraged to read the whole
 document and get familiar with it.
 
+With the recent increased flow of LLM-assisted bug fixes, please also follow the
+kernel documentation at https://docs.kernel.org/process/coding-assistants.html to
+properly disclose LLM usage for every patch.
+
 Signed-off-by:
 ^^^^^^^^^^^^^^
 

cmds/rescue-fix-data-checksum.c

@@ -16,6 +16,7 @@
 
 #include "kerncompat.h"
 #include <ctype.h>
+#include "kernel-lib/bitmap.h"
 #include "kernel-shared/disk-io.h"
 #include "kernel-shared/ctree.h"
 #include "kernel-shared/volumes.h"
@@ -84,14 +85,14 @@ static int add_corrupted_block(struct btrfs_fs_info *fs_info, u64 logical,
 	/* The last entry is the same, just set update the error mirror bitmap. */
 	if (last->logical == logical) {
 		UASSERT(last->error_mirror_bitmap);
-		set_bit(mirror, last->error_mirror_bitmap);
+		set_bit(mirror - 1, last->error_mirror_bitmap);
 		return 0;
 	}
 add:
 	last = calloc(1, sizeof(*last));
 	if (!last)
 		return -ENOMEM;
-	last->error_mirror_bitmap = calloc(1, BITS_TO_LONGS(num_mirrors));
+	last->error_mirror_bitmap = bitmap_zalloc(num_mirrors);
 	if (!last->error_mirror_bitmap) {
 		free(last);
 		return -ENOMEM;
@@ -456,7 +457,7 @@ static void free_corrupted_blocks(void)
 
 		entry = list_entry(corrupted_blocks.next, struct corrupted_block, list);
 		list_del_init(&entry->list);
-		free(entry->error_mirror_bitmap);
+		bitmap_free(entry->error_mirror_bitmap);
 		free(entry);
 	}
 }