x86/kprobes: Fix crash when probing CS CALL instructionsgrafted

When using eBPF to probe CS CALL instructions within a function,
a crash can be triggered.

The eBPF tool probes offset 257 of the __hrtimer_run_queues()
function:

<__hrtimer_run_queues+249>:  nopl   0x0(%rax,%rax,1)
<__hrtimer_run_queues+254>:  mov    %r14,%rdi
<__hrtimer_run_queues+257>:  cs call <__x86_indirect_thunk_r12>
<__hrtimer_run_queues+263>:  mov    %eax,%r12d
<__hrtimer_run_queues+266>:  xchg   %ax,%ax
<__hrtimer_run_queues+268>:  mov    %r13,%rdi

Which triggers this crash:

  BUG: unable to handle page fault for address: 00000000000f41c9
  #PF: supervisor write access in kernel mode
  #PF: error_code(0x0002) - not-present page
  PGD 0 P4D 0
  Oops: 0002 [#1] SMP NOPTI
  CPU: 1 PID: 0 Comm: swapper/1 Kdump: loaded Tainted: P
  RIP: 0010:__hrtimer_run_queues+0x106/0x230

Note that __hrtimer_run_queues+0x106 is __hrtimer_run_queues+262, which is
at the 6th byte of the above CS CALL instruction. Since the CS CALL
instruction occupies 6 bytes, the exception occurred in the middle of that
call instruction.

The root cause is that when using eBPF tools to probe in the middle of a
function, a kprobe with INT3 is used as the underlying implementation.

During single-step emulation of the original CALL instruction,
int3_emulate_call() assumes that the probed CALL instruction is 5 bytes
long. However, the actual CS-prefixed CALL instruction occupies 6 bytes,
so it constructs an incorrect exception return address. When the CPU
returns from the kprobe handler, the next instruction to be executed is at
the address of the last byte of that CS CALL instruction. Coincidentally,
starting from that address, the CPU fetches and decodes a completely
different instruction, which ultimately triggers a kernel crash.

Fix the issue by using the actual instruction length obtained from
the instruction decoder when constructing the exception return
address, rather than relying on the hardcoded CALL_INSN_SIZE macro.

[ mingo: Refined the changelog ]

Fixes: 6256e668b7af ("x86/kprobes: Use int3 instead of debug trap for single-step")
Suggested-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Signed-off-by: Jinke Han <jinkehan@didiglobal.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Acked-by: Yafang Shao <laoar.shao@gmail.com>
Acked-by: Borislav Petkov <bp@alien8.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://patch.msgid.link/20260908073742.GA10517@didi-ThinkCentre-M920t-N000

1 files changed, 217 insertions, 0 deletions

diff --git a/fs/erofs/fileio.c b/fs/erofs/fileio.c
new file mode 100644
index 000000000..ebb81a7ff
--- /dev/null
+++ b/fs/erofs/fileio.c
@@ -0,0 +1,217 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (C) 2024, Alibaba Cloud
+ */
+#include "internal.h"
+#include <trace/events/erofs.h>
+
+struct erofs_fileio_rq {
+	struct bio_vec bvecs[16];
+	struct bio bio;
+	struct kiocb iocb;
+	struct super_block *sb;
+	refcount_t ref;
+};
+
+struct erofs_fileio {
+	struct erofs_map_blocks map;
+	struct erofs_map_dev dev;
+	struct erofs_fileio_rq *rq;
+};
+
+static void erofs_fileio_ki_complete(struct kiocb *iocb, long ret)
+{
+	struct erofs_fileio_rq *rq =
+			container_of(iocb, struct erofs_fileio_rq, iocb);
+	struct folio_iter fi;
+
+	if (ret >= 0 && ret != rq->bio.bi_iter.bi_size)
+		ret = -EIO;
+	if (!rq->bio.bi_end_io) {
+		bio_for_each_folio_all(fi, &rq->bio) {
+			DBG_BUGON(folio_test_uptodate(fi.folio));
+			erofs_onlinefolio_end(fi.folio, ret < 0, false);
+		}
+	} else if (ret < 0 && !rq->bio.bi_status) {
+		rq->bio.bi_status = errno_to_blk_status(ret);
+	}
+	bio_endio(&rq->bio);
+	bio_uninit(&rq->bio);
+	if (refcount_dec_and_test(&rq->ref))
+		kfree(rq);
+}
+
+static void erofs_fileio_rq_submit(struct erofs_fileio_rq *rq)
+{
+	struct iov_iter iter;
+	ssize_t ret;
+
+	if (!rq)
+		return;
+	rq->iocb.ki_pos = rq->bio.bi_iter.bi_sector << SECTOR_SHIFT;
+	rq->iocb.ki_ioprio = get_current_ioprio();
+	rq->iocb.ki_complete = erofs_fileio_ki_complete;
+	if (test_opt(&EROFS_SB(rq->sb)->opt, DIRECT_IO) &&
+	    rq->iocb.ki_filp->f_mode & FMODE_CAN_ODIRECT)
+		rq->iocb.ki_flags = IOCB_DIRECT;
+	iov_iter_bvec(&iter, ITER_DEST, rq->bvecs, rq->bio.bi_vcnt,
+		      rq->bio.bi_iter.bi_size);
+	scoped_with_creds(rq->iocb.ki_filp->f_cred)
+		ret = vfs_iocb_iter_read(rq->iocb.ki_filp, &rq->iocb, &iter);
+	if (ret != -EIOCBQUEUED)
+		erofs_fileio_ki_complete(&rq->iocb, ret);
+	if (refcount_dec_and_test(&rq->ref))
+		kfree(rq);
+}
+
+static struct erofs_fileio_rq *erofs_fileio_rq_alloc(struct erofs_map_dev *mdev)
+{
+	struct erofs_fileio_rq *rq = kzalloc_obj(*rq, GFP_KERNEL | __GFP_NOFAIL);
+
+	bio_init(&rq->bio, NULL, rq->bvecs, ARRAY_SIZE(rq->bvecs), REQ_OP_READ);
+	rq->iocb.ki_filp = mdev->m_dif->file;
+	rq->sb = mdev->m_sb;
+	refcount_set(&rq->ref, 2);
+	return rq;
+}
+
+struct bio *erofs_fileio_bio_alloc(struct erofs_map_dev *mdev)
+{
+	return &erofs_fileio_rq_alloc(mdev)->bio;
+}
+
+void erofs_fileio_submit_bio(struct bio *bio)
+{
+	return erofs_fileio_rq_submit(container_of(bio, struct erofs_fileio_rq,
+						   bio));
+}
+
+static int erofs_fileio_scan_folio(struct erofs_fileio *io,
+				   struct inode *inode, struct folio *folio)
+{
+	struct erofs_map_blocks *map = &io->map;
+	unsigned int cur = 0, end = folio_size(folio), len, attached = 0;
+	loff_t pos = folio_pos(folio), ofs;
+	int err = 0;
+
+	erofs_onlinefolio_init(folio);
+	while (cur < end) {
+		if (!in_range(pos + cur, map->m_la, map->m_llen)) {
+			map->m_la = pos + cur;
+			map->m_llen = end - cur;
+			err = erofs_map_blocks(inode, map);
+			if (err)
+				break;
+		}
+
+		ofs = folio_pos(folio) + cur - map->m_la;
+		len = min_t(loff_t, map->m_llen - ofs, end - cur);
+		if (map->m_flags & EROFS_MAP_META) {
+			struct erofs_buf buf = __EROFS_BUF_INITIALIZER;
+			void *src;
+
+			src = erofs_read_metabuf(&buf, inode->i_sb,
+				map->m_pa + ofs, erofs_inode_in_metabox(inode));
+			if (IS_ERR(src)) {
+				err = PTR_ERR(src);
+				break;
+			}
+			memcpy_to_folio(folio, cur, src, len);
+			erofs_put_metabuf(&buf);
+		} else if (!(map->m_flags & EROFS_MAP_MAPPED)) {
+			folio_zero_segment(folio, cur, cur + len);
+			attached = 0;
+		} else {
+			if (io->rq && (map->m_pa + ofs != io->dev.m_pa ||
+				       map->m_deviceid != io->dev.m_deviceid)) {
+io_retry:
+				erofs_fileio_rq_submit(io->rq);
+				io->rq = NULL;
+			}
+
+			if (!io->rq) {
+				io->dev = (struct erofs_map_dev) {
+					.m_pa = io->map.m_pa + ofs,
+					.m_deviceid = io->map.m_deviceid,
+				};
+				err = erofs_map_dev(inode->i_sb, &io->dev);
+				if (err)
+					break;
+				io->rq = erofs_fileio_rq_alloc(&io->dev);
+				io->rq->bio.bi_iter.bi_sector =
+					(io->dev.m_dif->fsoff + io->dev.m_pa) >> 9;
+				attached = 0;
+			}
+			if (!bio_add_folio(&io->rq->bio, folio, len, cur))
+				goto io_retry;
+			if (!attached++)
+				erofs_onlinefolio_split(folio);
+			io->dev.m_pa += len;
+		}
+		cur += len;
+	}
+	erofs_onlinefolio_end(folio, err, false);
+	return err;
+}
+
+static int erofs_fileio_read_folio(struct file *file, struct folio *folio)
+{
+	bool need_iput;
+	struct inode *realinode = erofs_real_inode(folio_inode(folio), &need_iput);
+	struct erofs_fileio io = {};
+	int err;
+
+	trace_erofs_read_folio(realinode, folio, true);
+	err = erofs_fileio_scan_folio(&io, realinode, folio);
+	erofs_fileio_rq_submit(io.rq);
+	if (need_iput)
+		iput(realinode);
+	return err;
+}
+
+static void erofs_fileio_readahead(struct readahead_control *rac)
+{
+	bool need_iput;
+	struct inode *realinode = erofs_real_inode(rac->mapping->host, &need_iput);
+	struct erofs_fileio io = {};
+	struct folio *folio;
+	int err;
+
+	trace_erofs_readahead(realinode, readahead_index(rac),
+			      readahead_count(rac), true);
+	while ((folio = readahead_folio(rac))) {
+		err = erofs_fileio_scan_folio(&io, realinode, folio);
+		if (err && err != -EINTR)
+			erofs_err(realinode->i_sb, "readahead error at folio %lu @ nid %llu",
+				  folio->index, EROFS_I(realinode)->nid);
+	}
+	erofs_fileio_rq_submit(io.rq);
+	if (need_iput)
+		iput(realinode);
+}
+
+const struct address_space_operations erofs_fileio_aops = {
+	.read_folio = erofs_fileio_read_folio,
+	.readahead = erofs_fileio_readahead,
+};
+
+int erofs_read_meta_folio(struct file *file, struct folio *folio)
+{
+	struct erofs_fileio io = {
+		.dev = { .m_pa = folio_pos(folio), },
+	};
+	struct inode *inode = folio_inode(folio);
+	int err;
+
+	err = erofs_map_dev(inode->i_sb, &io.dev);
+	if (err)
+		return err;
+
+	io.rq = erofs_fileio_rq_alloc(&io.dev);
+	io.rq->bio.bi_iter.bi_sector =
+		(io.dev.m_dif->fsoff + io.dev.m_pa) >> 9;
+	erofs_onlinefolio_init(folio);
+	bio_add_folio_nofail(&io.rq->bio, folio, folio_size(folio), 0);
+	erofs_fileio_rq_submit(io.rq);
+	return 0;
+}