summaryrefslogtreecommitdiff
path: root/fs
diff options
context:
space:
mode:
Diffstat (limited to 'fs')
-rw-r--r--fs/binfmt_elf.c71
-rw-r--r--fs/btrfs/discard.c17
-rw-r--r--fs/btrfs/fs.h1
-rw-r--r--fs/btrfs/inode.c7
-rw-r--r--fs/btrfs/super.c4
-rw-r--r--fs/buffer.c4
-rw-r--r--fs/eventpoll.c7
-rw-r--r--fs/udf/truncate.c2
-rw-r--r--fs/xattr.c24
9 files changed, 104 insertions, 33 deletions
diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
index 584fa89bc877..4c1ea6b52a53 100644
--- a/fs/binfmt_elf.c
+++ b/fs/binfmt_elf.c
@@ -830,6 +830,7 @@ static int load_elf_binary(struct linux_binprm *bprm)
struct elf_phdr *elf_ppnt, *elf_phdata, *interp_elf_phdata = NULL;
struct elf_phdr *elf_property_phdata = NULL;
unsigned long elf_brk;
+ bool brk_moved = false;
int retval, i;
unsigned long elf_entry;
unsigned long e_entry;
@@ -1097,15 +1098,19 @@ out_free_interp:
/* Calculate any requested alignment. */
alignment = maximum_alignment(elf_phdata, elf_ex->e_phnum);
- /*
- * There are effectively two types of ET_DYN
- * binaries: programs (i.e. PIE: ET_DYN with PT_INTERP)
- * and loaders (ET_DYN without PT_INTERP, since they
- * _are_ the ELF interpreter). The loaders must
- * be loaded away from programs since the program
- * may otherwise collide with the loader (especially
- * for ET_EXEC which does not have a randomized
- * position). For example to handle invocations of
+ /**
+ * DOC: PIE handling
+ *
+ * There are effectively two types of ET_DYN ELF
+ * binaries: programs (i.e. PIE: ET_DYN with
+ * PT_INTERP) and loaders (i.e. static PIE: ET_DYN
+ * without PT_INTERP, usually the ELF interpreter
+ * itself). Loaders must be loaded away from programs
+ * since the program may otherwise collide with the
+ * loader (especially for ET_EXEC which does not have
+ * a randomized position).
+ *
+ * For example, to handle invocations of
* "./ld.so someprog" to test out a new version of
* the loader, the subsequent program that the
* loader loads must avoid the loader itself, so
@@ -1118,6 +1123,9 @@ out_free_interp:
* ELF_ET_DYN_BASE and loaders are loaded into the
* independently randomized mmap region (0 load_bias
* without MAP_FIXED nor MAP_FIXED_NOREPLACE).
+ *
+ * See below for "brk" handling details, which is
+ * also affected by program vs loader and ASLR.
*/
if (interpreter) {
/* On ET_DYN with PT_INTERP, we do the ASLR. */
@@ -1234,8 +1242,6 @@ out_free_interp:
start_data += load_bias;
end_data += load_bias;
- current->mm->start_brk = current->mm->brk = ELF_PAGEALIGN(elf_brk);
-
if (interpreter) {
elf_entry = load_elf_interp(interp_elf_ex,
interpreter,
@@ -1291,27 +1297,44 @@ out_free_interp:
mm->end_data = end_data;
mm->start_stack = bprm->p;
- if ((current->flags & PF_RANDOMIZE) && (snapshot_randomize_va_space > 1)) {
+ /**
+ * DOC: "brk" handling
+ *
+ * For architectures with ELF randomization, when executing a
+ * loader directly (i.e. static PIE: ET_DYN without PT_INTERP),
+ * move the brk area out of the mmap region and into the unused
+ * ELF_ET_DYN_BASE region. Since "brk" grows up it may collide
+ * early with the stack growing down or other regions being put
+ * into the mmap region by the kernel (e.g. vdso).
+ *
+ * In the CONFIG_COMPAT_BRK case, though, everything is turned
+ * off because we're not allowed to move the brk at all.
+ */
+ if (!IS_ENABLED(CONFIG_COMPAT_BRK) &&
+ IS_ENABLED(CONFIG_ARCH_HAS_ELF_RANDOMIZE) &&
+ elf_ex->e_type == ET_DYN && !interpreter) {
+ elf_brk = ELF_ET_DYN_BASE;
+ /* This counts as moving the brk, so let brk(2) know. */
+ brk_moved = true;
+ }
+ mm->start_brk = mm->brk = ELF_PAGEALIGN(elf_brk);
+
+ if ((current->flags & PF_RANDOMIZE) && snapshot_randomize_va_space > 1) {
/*
- * For architectures with ELF randomization, when executing
- * a loader directly (i.e. no interpreter listed in ELF
- * headers), move the brk area out of the mmap region
- * (since it grows up, and may collide early with the stack
- * growing down), and into the unused ELF_ET_DYN_BASE region.
+ * If we didn't move the brk to ELF_ET_DYN_BASE (above),
+ * leave a gap between .bss and brk.
*/
- if (IS_ENABLED(CONFIG_ARCH_HAS_ELF_RANDOMIZE) &&
- elf_ex->e_type == ET_DYN && !interpreter) {
- mm->brk = mm->start_brk = ELF_ET_DYN_BASE;
- } else {
- /* Otherwise leave a gap between .bss and brk. */
+ if (!brk_moved)
mm->brk = mm->start_brk = mm->brk + PAGE_SIZE;
- }
mm->brk = mm->start_brk = arch_randomize_brk(mm);
+ brk_moved = true;
+ }
+
#ifdef compat_brk_randomized
+ if (brk_moved)
current->brk_randomized = 1;
#endif
- }
if (current->personality & MMAP_PAGE_ZERO) {
/* Why this, you ask??? Well SVr4 maps page 0 as read-only,
diff --git a/fs/btrfs/discard.c b/fs/btrfs/discard.c
index d6eef4bd9e9d..de23c4b3515e 100644
--- a/fs/btrfs/discard.c
+++ b/fs/btrfs/discard.c
@@ -94,8 +94,6 @@ static void __add_to_discard_list(struct btrfs_discard_ctl *discard_ctl,
struct btrfs_block_group *block_group)
{
lockdep_assert_held(&discard_ctl->lock);
- if (!btrfs_run_discard_work(discard_ctl))
- return;
if (list_empty(&block_group->discard_list) ||
block_group->discard_index == BTRFS_DISCARD_INDEX_UNUSED) {
@@ -118,6 +116,9 @@ static void add_to_discard_list(struct btrfs_discard_ctl *discard_ctl,
if (!btrfs_is_block_group_data_only(block_group))
return;
+ if (!btrfs_run_discard_work(discard_ctl))
+ return;
+
spin_lock(&discard_ctl->lock);
__add_to_discard_list(discard_ctl, block_group);
spin_unlock(&discard_ctl->lock);
@@ -244,6 +245,18 @@ again:
block_group->used != 0) {
if (btrfs_is_block_group_data_only(block_group)) {
__add_to_discard_list(discard_ctl, block_group);
+ /*
+ * The block group must have been moved to other
+ * discard list even if discard was disabled in
+ * the meantime or a transaction abort happened,
+ * otherwise we can end up in an infinite loop,
+ * always jumping into the 'again' label and
+ * keep getting this block group over and over
+ * in case there are no other block groups in
+ * the discard lists.
+ */
+ ASSERT(block_group->discard_index !=
+ BTRFS_DISCARD_INDEX_UNUSED);
} else {
list_del_init(&block_group->discard_list);
btrfs_put_block_group(block_group);
diff --git a/fs/btrfs/fs.h b/fs/btrfs/fs.h
index bcca43046064..7baa2ed45198 100644
--- a/fs/btrfs/fs.h
+++ b/fs/btrfs/fs.h
@@ -300,6 +300,7 @@ enum {
#define BTRFS_FEATURE_INCOMPAT_SAFE_CLEAR 0ULL
#define BTRFS_DEFAULT_COMMIT_INTERVAL (30)
+#define BTRFS_WARNING_COMMIT_INTERVAL (300)
#define BTRFS_DEFAULT_MAX_INLINE (2048)
struct btrfs_dev_replace {
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index bdafe4d4c4a5..90f5da3c520a 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -1109,6 +1109,7 @@ static void submit_one_async_extent(struct async_chunk *async_chunk,
struct extent_state *cached = NULL;
struct extent_map *em;
int ret = 0;
+ bool free_pages = false;
u64 start = async_extent->start;
u64 end = async_extent->start + async_extent->ram_size - 1;
@@ -1129,7 +1130,10 @@ static void submit_one_async_extent(struct async_chunk *async_chunk,
}
if (async_extent->compress_type == BTRFS_COMPRESS_NONE) {
+ ASSERT(!async_extent->folios);
+ ASSERT(async_extent->nr_folios == 0);
submit_uncompressed_range(inode, async_extent, locked_folio);
+ free_pages = true;
goto done;
}
@@ -1145,6 +1149,7 @@ static void submit_one_async_extent(struct async_chunk *async_chunk,
* fall back to uncompressed.
*/
submit_uncompressed_range(inode, async_extent, locked_folio);
+ free_pages = true;
goto done;
}
@@ -1186,6 +1191,8 @@ static void submit_one_async_extent(struct async_chunk *async_chunk,
done:
if (async_chunk->blkcg_css)
kthread_associate_blkcg(NULL);
+ if (free_pages)
+ free_async_extent_pages(async_extent);
kfree(async_extent);
return;
diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index 7121d8c7a318..7310e2fa8526 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -569,6 +569,10 @@ static int btrfs_parse_param(struct fs_context *fc, struct fs_parameter *param)
break;
case Opt_commit_interval:
ctx->commit_interval = result.uint_32;
+ if (ctx->commit_interval > BTRFS_WARNING_COMMIT_INTERVAL) {
+ btrfs_warn(NULL, "excessive commit interval %u, use with care",
+ ctx->commit_interval);
+ }
if (ctx->commit_interval == 0)
ctx->commit_interval = BTRFS_DEFAULT_COMMIT_INTERVAL;
break;
diff --git a/fs/buffer.c b/fs/buffer.c
index 7be23ff20b27..7ba1807145aa 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -1220,10 +1220,8 @@ void mark_buffer_write_io_error(struct buffer_head *bh)
/* FIXME: do we need to set this in both places? */
if (bh->b_folio && bh->b_folio->mapping)
mapping_set_error(bh->b_folio->mapping, -EIO);
- if (bh->b_assoc_map) {
+ if (bh->b_assoc_map)
mapping_set_error(bh->b_assoc_map, -EIO);
- errseq_set(&bh->b_assoc_map->host->i_sb->s_wb_err, -EIO);
- }
}
EXPORT_SYMBOL(mark_buffer_write_io_error);
diff --git a/fs/eventpoll.c b/fs/eventpoll.c
index 4bc264b854c4..d4dbffdedd08 100644
--- a/fs/eventpoll.c
+++ b/fs/eventpoll.c
@@ -2111,9 +2111,10 @@ static int ep_poll(struct eventpoll *ep, struct epoll_event __user *events,
write_unlock_irq(&ep->lock);
- if (!eavail && ep_schedule_timeout(to))
- timed_out = !schedule_hrtimeout_range(to, slack,
- HRTIMER_MODE_ABS);
+ if (!eavail)
+ timed_out = !ep_schedule_timeout(to) ||
+ !schedule_hrtimeout_range(to, slack,
+ HRTIMER_MODE_ABS);
__set_current_state(TASK_RUNNING);
/*
diff --git a/fs/udf/truncate.c b/fs/udf/truncate.c
index 4f33a4a48886..b4071c9cf8c9 100644
--- a/fs/udf/truncate.c
+++ b/fs/udf/truncate.c
@@ -115,7 +115,7 @@ void udf_truncate_tail_extent(struct inode *inode)
}
/* This inode entry is in-memory only and thus we don't have to mark
* the inode dirty */
- if (ret == 0)
+ if (ret >= 0)
iinfo->i_lenExtents = inode->i_size;
brelse(epos.bh);
}
diff --git a/fs/xattr.c b/fs/xattr.c
index fabb2a04501e..8ec5b0204bfd 100644
--- a/fs/xattr.c
+++ b/fs/xattr.c
@@ -1428,6 +1428,15 @@ static bool xattr_is_trusted(const char *name)
return !strncmp(name, XATTR_TRUSTED_PREFIX, XATTR_TRUSTED_PREFIX_LEN);
}
+static bool xattr_is_maclabel(const char *name)
+{
+ const char *suffix = name + XATTR_SECURITY_PREFIX_LEN;
+
+ return !strncmp(name, XATTR_SECURITY_PREFIX,
+ XATTR_SECURITY_PREFIX_LEN) &&
+ security_ismaclabel(suffix);
+}
+
/**
* simple_xattr_list - list all xattr objects
* @inode: inode from which to get the xattrs
@@ -1460,6 +1469,17 @@ ssize_t simple_xattr_list(struct inode *inode, struct simple_xattrs *xattrs,
if (err)
return err;
+ err = security_inode_listsecurity(inode, buffer, remaining_size);
+ if (err < 0)
+ return err;
+
+ if (buffer) {
+ if (remaining_size < err)
+ return -ERANGE;
+ buffer += err;
+ }
+ remaining_size -= err;
+
read_lock(&xattrs->lock);
for (rbp = rb_first(&xattrs->rb_root); rbp; rbp = rb_next(rbp)) {
xattr = rb_entry(rbp, struct simple_xattr, rb_node);
@@ -1468,6 +1488,10 @@ ssize_t simple_xattr_list(struct inode *inode, struct simple_xattrs *xattrs,
if (!trusted && xattr_is_trusted(xattr->name))
continue;
+ /* skip MAC labels; these are provided by LSM above */
+ if (xattr_is_maclabel(xattr->name))
+ continue;
+
err = xattr_list_one(&buffer, &remaining_size, xattr->name);
if (err)
break;
generated by cgit (git 2.34.1) at 2025-05-15 14:01:17 +0000