Merge tag 'trace-v7.3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-tracegrafted

Pull tracing fixes from Steven Rostedt:

 - Fix error output of boot instance creation failure

   Currently if a boot instance creation fails, instead of printing out
   the name of the instance that failed, it prints "(null)". That is
   because it prints "cur_str" that had already been processed by
   strsep(). Print the saved name instead.

   While at it, print the error code of the failure.

 - Fix use-after-free for same named historgrams

   Histograms can be named so that they can be used in multiple events.
   But if the named histogram has a variable attached, the second event
   that uses the named histogram which duplicates it and needs to free
   the original after duplication leaves the old variable in place and
   still visible. If another histogram uses than variable, it will use
   the stale one which will try to reference the freed duplicate
   histogram and crash the kernel.

   Free the duplicate variables along with the duplicated histogram
   data.

 - Check return value of kthread_run() in event self test

   The events self tests uses a kthread for testing but does not check
   if it succeeded in creating a kthread. If the kthread creation were
   to fail, the code will still try to call kthread_stop() on the error
   returned.

 - Fix race between reading trace_pipe and updating subbuffer size

   If a user is reading the trace_pipe file at the same time they update
   the ring buffer sub-buffer size, can cause the trace_pipe read to
   read stale data. Add trace_access_lock() around updating the ring
   buffer sub-buffer size.

 - Fix eventfs_inode on failure path in creation of the events directory

   In the creation of the "events" directory, if after allocating the
   eventfs_inode a failure is detected, it calls cleanup_ei() which
   calls free_ei(). The free_ei() will test if eventfs_inode being freed
   has no children. It is a bug if it does. But on the failure case of
   the creation of the "events" directory, the children lists have not
   yet been initialized and the free will trigger a warning because
   list_empty() on an uninitialized list returns false.

   Move the initialization into init_ei() where it makes more sense and
   makes sure that a created eventfs_inode has its lists initialized
   upon creation.

 - Check return value of kthread_run() in ftrace direct sample code

   The sample code that shows how to use the ftrace direct calls does
   not test the return of kthread_run() to see if it succeeds. Return a
   failure if the kthread_run() doesn't succeed.

 - Clear user events state on fork in case of alloc failure

   On fork, the child gets a pointer to the parent's user events state.
   It makes a copy of it then updates the child's pointer to it. But if
   the allocation fails, the duplication function leaves the child with
   a pointer to its parent's descriptor. When the child cleans up its
   data, it will free the parent's descriptor while the parent is still
   using it.

   In the duplication function, set the child's user_event_mm to NULL
   before testing if the allocation succeeded, and when it exits it will
   not free the parent's descriptor.

 - Fix retry exhaustion in simple ring buffer reader swap

   simple_ring_buffer_swap_reader_page() starts with retry set to 8 and
   post-decrements it only after a failed link replacement. On the final
   attempt, a successful replacement leaves retry at zero, while a
   failed replacement leaves it at -1.

   But the check for success expects the retry value to be non-zero and
   exits with an error on zero. This is the opposite result. Fix it.

 - Fail nicely when the remote swap_reader_page() returns an error

   Currently, if the swap_reader_page() of a remote buffer fails, it
   triggers a WARN_ON_ONCE() and continues normally. Instead, have it
   exit with an error and a pr_warn() print instead of a full WARNING.

* tag 'trace-v7.3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
  ring-buffer: Stop remote reader update when page swap fails
  tracing: Fix retry exhaustion in simple ring buffer reader swap
  tracing/user_events: Clear copied tracing state before fork duplication
  samples/ftrace: Fix kthread_stop() on ERR_PTR in ftrace-direct-multi-modify
  samples/ftrace: Fix kthread_stop() on ERR_PTR in ftrace-direct-modify
  eventfs: Initialize ei->children and ei->list in init_ei()
  tracing: Fix use-after-free in trace_pipe read on sub-buffer order change
  tracing: Fix crash passing ERR_PTR to kthread_stop()
  tracing: Fix use-after-free with same-name named triggers
  tracing: Fix logged instance name on creation failure

1 files changed, 324 insertions, 0 deletions

diff --git a/fs/adfs/dir_f.c b/fs/adfs/dir_f.c
new file mode 100644
index 000000000..05e963402
--- /dev/null
+++ b/fs/adfs/dir_f.c
@@ -0,0 +1,324 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ *  linux/fs/adfs/dir_f.c
+ *
+ * Copyright (C) 1997-1999 Russell King
+ *
+ *  E and F format directory handling
+ */
+#include "adfs.h"
+#include "dir_f.h"
+
+/*
+ * Read an (unaligned) value of length 1..4 bytes
+ */
+static inline unsigned int adfs_readval(unsigned char *p, int len)
+{
+	unsigned int val = 0;
+
+	switch (len) {
+	case 4:		val |= p[3] << 24;
+		fallthrough;
+	case 3:		val |= p[2] << 16;
+		fallthrough;
+	case 2:		val |= p[1] << 8;
+		fallthrough;
+	default:	val |= p[0];
+	}
+	return val;
+}
+
+static inline void adfs_writeval(unsigned char *p, int len, unsigned int val)
+{
+	switch (len) {
+	case 4:		p[3] = val >> 24;
+		fallthrough;
+	case 3:		p[2] = val >> 16;
+		fallthrough;
+	case 2:		p[1] = val >> 8;
+		fallthrough;
+	default:	p[0] = val;
+	}
+}
+
+#define ror13(v) ((v >> 13) | (v << 19))
+
+#define dir_u8(idx)				\
+	({ int _buf = idx >> blocksize_bits;	\
+	   int _off = idx - (_buf << blocksize_bits);\
+	  *(u8 *)(bh[_buf]->b_data + _off);	\
+	})
+
+#define dir_u32(idx)				\
+	({ int _buf = idx >> blocksize_bits;	\
+	   int _off = idx - (_buf << blocksize_bits);\
+	  *(__le32 *)(bh[_buf]->b_data + _off);	\
+	})
+
+#define bufoff(_bh,_idx)			\
+	({ int _buf = _idx >> blocksize_bits;	\
+	   int _off = _idx - (_buf << blocksize_bits);\
+	  (void *)(_bh[_buf]->b_data + _off);	\
+	})
+
+/*
+ * There are some algorithms that are nice in
+ * assembler, but a bitch in C...  This is one
+ * of them.
+ */
+static u8
+adfs_dir_checkbyte(const struct adfs_dir *dir)
+{
+	struct buffer_head * const *bh = dir->bh;
+	const int blocksize_bits = dir->sb->s_blocksize_bits;
+	union { __le32 *ptr32; u8 *ptr8; } ptr, end;
+	u32 dircheck = 0;
+	int last = 5 - 26;
+	int i = 0;
+
+	/*
+	 * Accumulate each word up to the last whole
+	 * word of the last directory entry.  This
+	 * can spread across several buffer heads.
+	 */
+	do {
+		last += 26;
+		do {
+			dircheck = le32_to_cpu(dir_u32(i)) ^ ror13(dircheck);
+
+			i += sizeof(u32);
+		} while (i < (last & ~3));
+	} while (dir_u8(last) != 0);
+
+	/*
+	 * Accumulate the last few bytes.  These
+	 * bytes will be within the same bh.
+	 */
+	if (i != last) {
+		ptr.ptr8 = bufoff(bh, i);
+		end.ptr8 = ptr.ptr8 + last - i;
+
+		do {
+			dircheck = *ptr.ptr8++ ^ ror13(dircheck);
+		} while (ptr.ptr8 < end.ptr8);
+	}
+
+	/*
+	 * The directory tail is in the final bh
+	 * Note that contary to the RISC OS PRMs,
+	 * the first few bytes are NOT included
+	 * in the check.  All bytes are in the
+	 * same bh.
+	 */
+	ptr.ptr8 = bufoff(bh, 2008);
+	end.ptr8 = ptr.ptr8 + 36;
+
+	do {
+		__le32 v = *ptr.ptr32++;
+		dircheck = le32_to_cpu(v) ^ ror13(dircheck);
+	} while (ptr.ptr32 < end.ptr32);
+
+	return (dircheck ^ (dircheck >> 8) ^ (dircheck >> 16) ^ (dircheck >> 24)) & 0xff;
+}
+
+static int adfs_f_validate(struct adfs_dir *dir)
+{
+	struct adfs_dirheader *head = dir->dirhead;
+	struct adfs_newdirtail *tail = dir->newtail;
+
+	if (head->startmasseq != tail->endmasseq ||
+	    tail->dirlastmask || tail->reserved[0] || tail->reserved[1] ||
+	    (memcmp(&head->startname, "Nick", 4) &&
+	     memcmp(&head->startname, "Hugo", 4)) ||
+	    memcmp(&head->startname, &tail->endname, 4) ||
+	    adfs_dir_checkbyte(dir) != tail->dircheckbyte)
+		return -EIO;
+
+	return 0;
+}
+
+/* Read and check that a directory is valid */
+static int adfs_f_read(struct super_block *sb, u32 indaddr, unsigned int size,
+		       struct adfs_dir *dir)
+{
+	const unsigned int blocksize_bits = sb->s_blocksize_bits;
+	int ret;
+
+	if (size && size != ADFS_NEWDIR_SIZE)
+		return -EIO;
+
+	ret = adfs_dir_read_buffers(sb, indaddr, ADFS_NEWDIR_SIZE, dir);
+	if (ret)
+		return ret;
+
+	dir->dirhead = bufoff(dir->bh, 0);
+	dir->newtail = bufoff(dir->bh, 2007);
+
+	if (adfs_f_validate(dir))
+		goto bad_dir;
+
+	dir->parent_id = adfs_readval(dir->newtail->dirparent, 3);
+
+	return 0;
+
+bad_dir:
+	adfs_error(sb, "dir %06x is corrupted", indaddr);
+	adfs_dir_relse(dir);
+
+	return -EIO;
+}
+
+/*
+ * convert a disk-based directory entry to a Linux ADFS directory entry
+ */
+static inline void
+adfs_dir2obj(struct adfs_dir *dir, struct object_info *obj,
+	struct adfs_direntry *de)
+{
+	unsigned int name_len;
+
+	for (name_len = 0; name_len < ADFS_F_NAME_LEN; name_len++) {
+		if (de->dirobname[name_len] < ' ')
+			break;
+
+		obj->name[name_len] = de->dirobname[name_len];
+	}
+
+	obj->name_len =	name_len;
+	obj->indaddr  = adfs_readval(de->dirinddiscadd, 3);
+	obj->loadaddr = adfs_readval(de->dirload, 4);
+	obj->execaddr = adfs_readval(de->direxec, 4);
+	obj->size     = adfs_readval(de->dirlen,  4);
+	obj->attr     = de->newdiratts;
+
+	adfs_object_fixup(dir, obj);
+}
+
+/*
+ * convert a Linux ADFS directory entry to a disk-based directory entry
+ */
+static inline void
+adfs_obj2dir(struct adfs_direntry *de, struct object_info *obj)
+{
+	adfs_writeval(de->dirinddiscadd, 3, obj->indaddr);
+	adfs_writeval(de->dirload, 4, obj->loadaddr);
+	adfs_writeval(de->direxec, 4, obj->execaddr);
+	adfs_writeval(de->dirlen,  4, obj->size);
+	de->newdiratts = obj->attr;
+}
+
+/*
+ * get a directory entry.  Note that the caller is responsible
+ * for holding the relevant locks.
+ */
+static int
+__adfs_dir_get(struct adfs_dir *dir, int pos, struct object_info *obj)
+{
+	struct adfs_direntry de;
+	int ret;
+
+	ret = adfs_dir_copyfrom(&de, dir, pos, 26);
+	if (ret)
+		return ret;
+
+	if (!de.dirobname[0])
+		return -ENOENT;
+
+	adfs_dir2obj(dir, obj, &de);
+
+	return 0;
+}
+
+static int
+adfs_f_setpos(struct adfs_dir *dir, unsigned int fpos)
+{
+	if (fpos >= ADFS_NUM_DIR_ENTRIES)
+		return -ENOENT;
+
+	dir->pos = 5 + fpos * 26;
+	return 0;
+}
+
+static int
+adfs_f_getnext(struct adfs_dir *dir, struct object_info *obj)
+{
+	unsigned int ret;
+
+	ret = __adfs_dir_get(dir, dir->pos, obj);
+	if (ret == 0)
+		dir->pos += 26;
+
+	return ret;
+}
+
+static int adfs_f_iterate(struct adfs_dir *dir, struct dir_context *ctx)
+{
+	struct object_info obj;
+	int pos = 5 + (ctx->pos - 2) * 26;
+
+	while (ctx->pos < 2 + ADFS_NUM_DIR_ENTRIES) {
+		if (__adfs_dir_get(dir, pos, &obj))
+			break;
+		if (!dir_emit(ctx, obj.name, obj.name_len,
+			      obj.indaddr, DT_UNKNOWN))
+			break;
+		pos += 26;
+		ctx->pos++;
+	}
+	return 0;
+}
+
+static int adfs_f_update(struct adfs_dir *dir, struct object_info *obj)
+{
+	struct adfs_direntry de;
+	int offset, ret;
+
+	offset = 5 - (int)sizeof(de);
+
+	do {
+		offset += sizeof(de);
+		ret = adfs_dir_copyfrom(&de, dir, offset, sizeof(de));
+		if (ret) {
+			adfs_error(dir->sb, "error reading directory entry");
+			return -ENOENT;
+		}
+		if (!de.dirobname[0]) {
+			adfs_error(dir->sb, "unable to locate entry to update");
+			return -ENOENT;
+		}
+	} while (adfs_readval(de.dirinddiscadd, 3) != obj->indaddr);
+
+	/* Update the directory entry with the new object state */
+	adfs_obj2dir(&de, obj);
+
+	/* Write the directory entry back to the directory */
+	return adfs_dir_copyto(dir, offset, &de, 26);
+}
+
+static int adfs_f_commit(struct adfs_dir *dir)
+{
+	int ret;
+
+	/* Increment directory sequence number */
+	dir->dirhead->startmasseq += 1;
+	dir->newtail->endmasseq += 1;
+
+	/* Update directory check byte */
+	dir->newtail->dircheckbyte = adfs_dir_checkbyte(dir);
+
+	/* Make sure the directory still validates correctly */
+	ret = adfs_f_validate(dir);
+	if (ret)
+		adfs_msg(dir->sb, KERN_ERR, "error: update broke directory");
+
+	return ret;
+}
+
+const struct adfs_dir_ops adfs_f_dir_ops = {
+	.read		= adfs_f_read,
+	.iterate	= adfs_f_iterate,
+	.setpos		= adfs_f_setpos,
+	.getnext	= adfs_f_getnext,
+	.update		= adfs_f_update,
+	.commit		= adfs_f_commit,
+};