From: Chengming Zhou <[email protected]>

[ Upstream commit 8a224ffb3f52b0027f6b7279854c71a31c48fc97 ]

When module loaded and enabled, we will use __ftrace_replace_code
for module if any ftrace_ops referenced it found. But we will get
wrong ftrace_addr for module rec in ftrace_get_addr_new, because
rec->flags has not been setup correctly. It can cause the callback
function of a ftrace_ops has FTRACE_OPS_FL_SAVE_REGS to be called
with pt_regs set to NULL.
So setup correct FTRACE_FL_REGS flags for rec when we call
referenced_filters to find ftrace_ops references it.

Link: 
https://lkml.kernel.org/r/[email protected]

Cc: [email protected]
Fixes: 8c4f3c3fa9681 ("ftrace: Check module functions being traced on reload")
Signed-off-by: Chengming Zhou <[email protected]>
Signed-off-by: Muchun Song <[email protected]>
Signed-off-by: Steven Rostedt (VMware) <[email protected]>
Signed-off-by: Sasha Levin <[email protected]>
---
 kernel/trace/ftrace.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index e4c6f89b6b11f..89ed01911a9a2 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -2823,8 +2823,11 @@ static int referenced_filters(struct dyn_ftrace *rec)
        int cnt = 0;
 
        for (ops = ftrace_ops_list; ops != &ftrace_list_end; ops = ops->next) {
-               if (ops_references_rec(ops, rec))
-                   cnt++;
+               if (ops_references_rec(ops, rec)) {
+                       cnt++;
+                       if (ops->flags & FTRACE_OPS_FL_SAVE_REGS)
+                               rec->flags |= FTRACE_FL_REGS;
+               }
        }
 
        return cnt;
@@ -2874,7 +2877,7 @@ static int ftrace_update_code(struct module *mod, struct 
ftrace_page *new_pgs)
                        p = &pg->records[i];
                        if (test)
                                cnt += referenced_filters(p);
-                       p->flags = cnt;
+                       p->flags += cnt;
 
                        /*
                         * Do the initial record conversion from mcount jump
-- 
2.25.1



Reply via email to