The quilt patch titled
Subject: uprobes: introduce the global struct vm_special_mapping xol_mapping
has been removed from the -mm tree. Its filename was
uprobes-introduce-the-global-struct-vm_special_mapping-xol_mapping.patch
This patch was dropped because it was merged into the mm-stable branch
of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
------------------------------------------------------
From: Oleg Nesterov <oleg@xxxxxxxxxx>
Subject: uprobes: introduce the global struct vm_special_mapping xol_mapping
Date: Wed, 11 Sep 2024 15:14:07 +0200
Currently each xol_area has its own instance of vm_special_mapping, this
is suboptimal and ugly. Kill xol_area->xol_mapping and add a single
global instance of vm_special_mapping, the ->fault() method can use
area->pages rather than xol_mapping->pages.
As a side effect this fixes the problem introduced by the recent commit
223febc6e557 ("mm: add optional close() to struct vm_special_mapping"), if
special_mapping_close() is called from the __mmput() paths, it will use
vma->vm_private_data = &area->xol_mapping freed by uprobe_clear_state().
Link: https://lkml.kernel.org/r/20240911131407.GB3448@xxxxxxxxxx
Fixes: 223febc6e557 ("mm: add optional close() to struct vm_special_mapping")
Signed-off-by: Oleg Nesterov <oleg@xxxxxxxxxx>
Reported-by: Sven Schnelle <svens@xxxxxxxxxxxxx>
Closes: https://lore.kernel.org/all/yt9dy149vprr.fsf@xxxxxxxxxxxxx/
Cc: Adrian Hunter <adrian.hunter@xxxxxxxxx>
Cc: Alexander Shishkin <alexander.shishkin@xxxxxxxxxxxxxxx>
Cc: Andrii Nakryiko <andrii@xxxxxxxxxx>
Cc: Arnaldo Carvalho de Melo <acme@xxxxxxxxxx>
Cc: Ian Rogers <irogers@xxxxxxxxxx>
Cc: Ingo Molnar <mingo@xxxxxxxxxx>
Cc: Jiri Olsa <jolsa@xxxxxxxxxx>
Cc: Kan Liang <kan.liang@xxxxxxxxxxxxxxx>
Cc: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
Cc: Mark Rutland <mark.rutland@xxxxxxx>
Cc: Masami Hiramatsu <mhiramat@xxxxxxxxxx>
Cc: Michael Ellerman <mpe@xxxxxxxxxxxxxx>
Cc: Namhyung Kim <namhyung@xxxxxxxxxx>
Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
kernel/events/uprobes.c | 21 ++++++++++++++++-----
1 file changed, 16 insertions(+), 5 deletions(-)
--- a/kernel/events/uprobes.c~uprobes-introduce-the-global-struct-vm_special_mapping-xol_mapping
+++ a/kernel/events/uprobes.c
@@ -99,7 +99,6 @@ struct xol_area {
atomic_t slot_count; /* number of in-use slots */
unsigned long *bitmap; /* 0 = free slot */
- struct vm_special_mapping xol_mapping;
struct page *pages[2];
/*
* We keep the vma's vm_start rather than a pointer to the vma
@@ -1433,6 +1432,21 @@ void uprobe_munmap(struct vm_area_struct
set_bit(MMF_RECALC_UPROBES, &vma->vm_mm->flags);
}
+static vm_fault_t xol_fault(const struct vm_special_mapping *sm,
+ struct vm_area_struct *vma, struct vm_fault *vmf)
+{
+ struct xol_area *area = vma->vm_mm->uprobes_state.xol_area;
+
+ vmf->page = area->pages[0];
+ get_page(vmf->page);
+ return 0;
+}
+
+static const struct vm_special_mapping xol_mapping = {
+ .name = "[uprobes]",
+ .fault = xol_fault,
+};
+
/* Slot allocation for XOL */
static int xol_add_vma(struct mm_struct *mm, struct xol_area *area)
{
@@ -1459,7 +1473,7 @@ static int xol_add_vma(struct mm_struct
vma = _install_special_mapping(mm, area->vaddr, PAGE_SIZE,
VM_EXEC|VM_MAYEXEC|VM_DONTCOPY|VM_IO,
- &area->xol_mapping);
+ &xol_mapping);
if (IS_ERR(vma)) {
ret = PTR_ERR(vma);
goto fail;
@@ -1498,9 +1512,6 @@ static struct xol_area *__create_xol_are
if (!area->bitmap)
goto free_area;
- area->xol_mapping.name = "[uprobes]";
- area->xol_mapping.fault = NULL;
- area->xol_mapping.pages = area->pages;
area->pages[0] = alloc_page(GFP_HIGHUSER);
if (!area->pages[0])
goto free_bitmap;
_
Patches currently in -mm which might be from oleg@xxxxxxxxxx are
