diff options
| author | Xie Yuanbin <xieyuanbin1@huawei.com> | 2026-07-28 03:16:42 +0100 |
|---|---|---|
| committer | Russell King <rmk+kernel@armlinux.org.uk> | 2026-08-05 16:15:51 +0100 |
| commit | 1039bffd6ae9c75b42b7d148d6c1106134107b66 (patch) | |
| tree | e8ee4ac80f87e2ad1b6b0c877eccbb62b0915c02 /scripts/coccinelle/null/badzero.cocci | |
| download | linux-stable-1039bffd6ae9c75b42b7d148d6c1106134107b66.tar.gz linux-stable-1039bffd6ae9c75b42b7d148d6c1106134107b66.zip | |
ARM: 9485/1: mm: acquire mmap write lock around show_pte() for user faultsgrafted
When CONFIG_DEBUG_USER=y, and cmdline "user_debug=31" is set,
a user fault may trigger show_pte() without any lock.
If another thread in the same process concurrently calls munmap(),
the page table pages may be freed while show_pte() is still traversing
them, causing a use-after-free in show_pte().
If CONFIG_ARM_LPAE=y, this may cause a kernel panic if the pages table
of PMD are freed when show_pte() is running.
Acquire mmap_write_lock() around show_pte() for user faults to fix the
contention.
For user faults, additionally restrict that show_pte() is called only
when the addr is a user-space address (addr < TASK_SIZE). This is because
the lock of tsk->mm only protects the virtual memory of user address space,
furthermore, dumping the page tables of a kernel-space address for user
faults is unnecessary and may have security implications.
Keep everything unchanged for kernel faults, because the kernel is
already in the "oops" state, acquiring a lock may risk a deadlock.
Co-developed-by: Qi Xi <xiqi2@huawei.com>
Fixes: 6d021b724481 ("ARM: dump pgd, pmd and pte states on unhandled data abort faults")
Link: https://lore.kernel.org/20260716014022.2823-1-xieyuanbin1@huawei.com
Acked-by: Lorenzo Stoakes (ARM) <ljs@kernel.org>
Reviewed-by: Linus Walleij <linusw@kernel.org>
Signed-off-by: Qi Xi <xiqi2@huawei.com>
Signed-off-by: Xie Yuanbin <xieyuanbin1@huawei.com>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Diffstat (limited to 'scripts/coccinelle/null/badzero.cocci')
| -rw-r--r-- | scripts/coccinelle/null/badzero.cocci | 239 |
1 files changed, 239 insertions, 0 deletions
diff --git a/scripts/coccinelle/null/badzero.cocci b/scripts/coccinelle/null/badzero.cocci new file mode 100644 index 000000000..35d443825 --- /dev/null +++ b/scripts/coccinelle/null/badzero.cocci @@ -0,0 +1,239 @@ +// SPDX-License-Identifier: GPL-2.0-only +/// Compare pointer-typed values to NULL rather than 0 +/// +//# This makes an effort to choose between !x and x == NULL. !x is used +//# if it has previously been used with the function used to initialize x. +//# This relies on type information. More type information can be obtained +//# using the option -all_includes and the option -I to specify an +//# include path. +// +// Confidence: High +// Copyright: (C) 2012 Julia Lawall, INRIA/LIP6. +// Copyright: (C) 2012 Gilles Muller, INRIA/LiP6. +// URL: https://coccinelle.gitlabpages.inria.fr/website +// Requires: 1.0.0 +// Options: + +virtual patch +virtual context +virtual org +virtual report + +@initialize:ocaml@ +@@ +let negtable = Hashtbl.create 101 + +@depends on patch@ +expression *E; +identifier f; +@@ + +( + (E = f(...)) == +- 0 ++ NULL +| + (E = f(...)) != +- 0 ++ NULL +| +- 0 ++ NULL + == (E = f(...)) +| +- 0 ++ NULL + != (E = f(...)) +) + + +@t1 depends on !patch@ +expression *E; +identifier f; +position p; +@@ + +( + (E = f(...)) == +* 0@p +| + (E = f(...)) != +* 0@p +| +* 0@p + == (E = f(...)) +| +* 0@p + != (E = f(...)) +) + +@script:python depends on org@ +p << t1.p; +@@ + +coccilib.org.print_todo(p[0], "WARNING comparing pointer to 0") + +@script:python depends on report@ +p << t1.p; +@@ + +coccilib.report.print_report(p[0], "WARNING comparing pointer to 0") + +// Tests of returned values + +@s@ +identifier f; +expression E,E1; +@@ + + E = f(...) + ... when != E = E1 + !E + +@script:ocaml depends on s@ +f << s.f; +@@ + +try let _ = Hashtbl.find negtable f in () +with Not_found -> Hashtbl.add negtable f () + +@ r disable is_zero,isnt_zero exists @ +expression *E; +identifier f; +@@ + +E = f(...) +... +(E == 0 +|E != 0 +|0 == E +|0 != E +) + +@script:ocaml@ +f << r.f; +@@ + +try let _ = Hashtbl.find negtable f in () +with Not_found -> include_match false + +// This rule may lead to inconsistent path problems, if E is defined in two +// places +@ depends on patch disable is_zero,isnt_zero @ +expression *E; +expression E1; +identifier r.f; +@@ + +E = f(...) +<... +( +- E == 0 ++ !E +| +- E != 0 ++ E +| +- 0 == E ++ !E +| +- 0 != E ++ E +) +...> +?E = E1 + +@t2 depends on !patch disable is_zero,isnt_zero @ +expression *E; +expression E1; +identifier r.f; +position p1; +position p2; +@@ + +E = f(...) +<... +( +* E == 0@p1 +| +* E != 0@p2 +| +* 0@p1 == E +| +* 0@p1 != E +) +...> +?E = E1 + +@script:python depends on org@ +p << t2.p1; +@@ + +coccilib.org.print_todo(p[0], "WARNING comparing pointer to 0, suggest !E") + +@script:python depends on org@ +p << t2.p2; +@@ + +coccilib.org.print_todo(p[0], "WARNING comparing pointer to 0") + +@script:python depends on report@ +p << t2.p1; +@@ + +coccilib.report.print_report(p[0], "WARNING comparing pointer to 0, suggest !E") + +@script:python depends on report@ +p << t2.p2; +@@ + +coccilib.report.print_report(p[0], "WARNING comparing pointer to 0") + +@ depends on patch disable is_zero,isnt_zero @ +expression *E; +@@ + +( + E == +- 0 ++ NULL +| + E != +- 0 ++ NULL +| +- 0 ++ NULL + == E +| +- 0 ++ NULL + != E +) + +@ t3 depends on !patch disable is_zero,isnt_zero @ +expression *E; +position p; +@@ + +( +* E == 0@p +| +* E != 0@p +| +* 0@p == E +| +* 0@p != E +) + +@script:python depends on org@ +p << t3.p; +@@ + +coccilib.org.print_todo(p[0], "WARNING comparing pointer to 0") + +@script:python depends on report@ +p << t3.p; +@@ + +coccilib.report.print_report(p[0], "WARNING comparing pointer to 0") |
