[PATCH 0/2] kmsan: fix sparse warnings

[PATCH 0/2] kmsan: fix sparse warnings

[Ilya Leoshkevich]

Hi,

Kernel test robot reported several sparse warnings in the KMSAN code
base [1].  They belong to two broad classes; fix each in a separate
commit.

Best regards,
Ilya

[1] https://lore.kernel.org/linux-mm/202406272033.KejtfLkw-lkp@intel.com/

Ilya Leoshkevich (2):
  kmsan: add missing __user tags
  kmsan: do not pass NULL pointers as 0

 mm/kmsan/core.c            |  4 ++--
 mm/kmsan/hooks.c           | 15 ++++++++-------
 mm/kmsan/instrumentation.c |  4 ++--
 mm/kmsan/kmsan.h           |  6 +++---
 mm/kmsan/report.c          |  2 +-
 5 files changed, 16 insertions(+), 15 deletions(-)

-- 
2.45.2

[PATCH 1/2] kmsan: add missing __user tags

[Ilya Leoshkevich]

sparse complains that __user pointers are being passed to functions
that expect non-__user ones.  In all cases, these functions are in fact
working with user pointers, only the tag is missing. Add it.

Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202406272033.KejtfLkw-lkp@intel.com/
Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
---
 mm/kmsan/core.c   | 4 ++--
 mm/kmsan/kmsan.h  | 6 +++---
 mm/kmsan/report.c | 2 +-
 3 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/mm/kmsan/core.c b/mm/kmsan/core.c
index 81b22220711a..a495debf1436 100644
--- a/mm/kmsan/core.c
+++ b/mm/kmsan/core.c
@@ -249,8 +249,8 @@ struct page *kmsan_vmalloc_to_page_or_null(void *vaddr)
 		return NULL;
 }
 
-void kmsan_internal_check_memory(void *addr, size_t size, const void *user_addr,
-				 int reason)
+void kmsan_internal_check_memory(void *addr, size_t size,
+				 const void __user *user_addr, int reason)
 {
 	depot_stack_handle_t cur_origin = 0, new_origin = 0;
 	unsigned long addr64 = (unsigned long)addr;
diff --git a/mm/kmsan/kmsan.h b/mm/kmsan/kmsan.h
index 91a360a31e85..29555a8bc315 100644
--- a/mm/kmsan/kmsan.h
+++ b/mm/kmsan/kmsan.h
@@ -73,7 +73,7 @@ void kmsan_print_origin(depot_stack_handle_t origin);
  * @off_last corresponding to different @origin values.
  */
 void kmsan_report(depot_stack_handle_t origin, void *address, int size,
-		  int off_first, int off_last, const void *user_addr,
+		  int off_first, int off_last, const void __user *user_addr,
 		  enum kmsan_bug_reason reason);
 
 DECLARE_PER_CPU(struct kmsan_ctx, kmsan_percpu_ctx);
@@ -163,8 +163,8 @@ depot_stack_handle_t kmsan_internal_chain_origin(depot_stack_handle_t id);
 void kmsan_internal_task_create(struct task_struct *task);
 
 bool kmsan_metadata_is_contiguous(void *addr, size_t size);
-void kmsan_internal_check_memory(void *addr, size_t size, const void *user_addr,
-				 int reason);
+void kmsan_internal_check_memory(void *addr, size_t size,
+				 const void __user *user_addr, int reason);
 
 struct page *kmsan_vmalloc_to_page_or_null(void *vaddr);
 void kmsan_setup_meta(struct page *page, struct page *shadow,
diff --git a/mm/kmsan/report.c b/mm/kmsan/report.c
index 92e73ec61435..94a3303fb65e 100644
--- a/mm/kmsan/report.c
+++ b/mm/kmsan/report.c
@@ -148,7 +148,7 @@ void kmsan_print_origin(depot_stack_handle_t origin)
 }
 
 void kmsan_report(depot_stack_handle_t origin, void *address, int size,
-		  int off_first, int off_last, const void *user_addr,
+		  int off_first, int off_last, const void __user *user_addr,
 		  enum kmsan_bug_reason reason)
 {
 	unsigned long stack_entries[KMSAN_STACK_DEPTH];
-- 
2.45.2

[PATCH 2/2] kmsan: do not pass NULL pointers as 0

[Ilya Leoshkevich]

sparse complains about passing NULL pointers as 0.  Fix all instances.

Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202406272033.KejtfLkw-lkp@intel.com/
Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
---
 mm/kmsan/hooks.c           | 15 ++++++++-------
 mm/kmsan/instrumentation.c |  4 ++--
 2 files changed, 10 insertions(+), 9 deletions(-)

diff --git a/mm/kmsan/hooks.c b/mm/kmsan/hooks.c
index 26d86dfdc819..3ea50f09311f 100644
--- a/mm/kmsan/hooks.c
+++ b/mm/kmsan/hooks.c
@@ -303,7 +303,8 @@ void kmsan_handle_urb(const struct urb *urb, bool is_out)
 	if (is_out)
 		kmsan_internal_check_memory(urb->transfer_buffer,
 					    urb->transfer_buffer_length,
-					    /*user_addr*/ 0, REASON_SUBMIT_URB);
+					    /*user_addr*/ NULL,
+					    REASON_SUBMIT_URB);
 	else
 		kmsan_internal_unpoison_memory(urb->transfer_buffer,
 					       urb->transfer_buffer_length,
@@ -316,14 +317,14 @@ static void kmsan_handle_dma_page(const void *addr, size_t size,
 {
 	switch (dir) {
 	case DMA_BIDIRECTIONAL:
-		kmsan_internal_check_memory((void *)addr, size, /*user_addr*/ 0,
-					    REASON_ANY);
+		kmsan_internal_check_memory((void *)addr, size,
+					    /*user_addr*/ NULL, REASON_ANY);
 		kmsan_internal_unpoison_memory((void *)addr, size,
 					       /*checked*/ false);
 		break;
 	case DMA_TO_DEVICE:
-		kmsan_internal_check_memory((void *)addr, size, /*user_addr*/ 0,
-					    REASON_ANY);
+		kmsan_internal_check_memory((void *)addr, size,
+					    /*user_addr*/ NULL, REASON_ANY);
 		break;
 	case DMA_FROM_DEVICE:
 		kmsan_internal_unpoison_memory((void *)addr, size,
@@ -418,8 +419,8 @@ void kmsan_check_memory(const void *addr, size_t size)
 {
 	if (!kmsan_enabled)
 		return;
-	return kmsan_internal_check_memory((void *)addr, size, /*user_addr*/ 0,
-					   REASON_ANY);
+	return kmsan_internal_check_memory((void *)addr, size,
+					   /*user_addr*/ NULL, REASON_ANY);
 }
 EXPORT_SYMBOL(kmsan_check_memory);
 
diff --git a/mm/kmsan/instrumentation.c b/mm/kmsan/instrumentation.c
index 94b49fac9d8b..02a405e55d6c 100644
--- a/mm/kmsan/instrumentation.c
+++ b/mm/kmsan/instrumentation.c
@@ -315,8 +315,8 @@ void __msan_warning(u32 origin)
 	if (!kmsan_enabled || kmsan_in_runtime())
 		return;
 	kmsan_enter_runtime();
-	kmsan_report(origin, /*address*/ 0, /*size*/ 0,
-		     /*off_first*/ 0, /*off_last*/ 0, /*user_addr*/ 0,
+	kmsan_report(origin, /*address*/ NULL, /*size*/ 0,
+		     /*off_first*/ 0, /*off_last*/ 0, /*user_addr*/ NULL,
 		     REASON_ANY);
 	kmsan_leave_runtime();
 }
-- 
2.45.2

Re: [PATCH 1/2] kmsan: add missing __user tags

[Alexander Potapenko]

On Thu, Jun 27, 2024 at 5:14 PM Ilya Leoshkevich <iii@linux.ibm.com> wrote:
>
> sparse complains that __user pointers are being passed to functions
> that expect non-__user ones.  In all cases, these functions are in fact
> working with user pointers, only the tag is missing. Add it.

Thanks!

>
> Reported-by: kernel test robot <lkp@intel.com>
> Closes: https://lore.kernel.org/oe-kbuild-all/202406272033.KejtfLkw-lkp@intel.com/
> Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Reviewed-by: Alexander Potapenko <glider@google.com>

Re: [PATCH 2/2] kmsan: do not pass NULL pointers as 0

[Alexander Potapenko]

On Thu, Jun 27, 2024 at 5:14 PM Ilya Leoshkevich <iii@linux.ibm.com> wrote:
>
> sparse complains about passing NULL pointers as 0.  Fix all instances.

Thanks a lot for fixing this!

> Reported-by: kernel test robot <lkp@intel.com>
> Closes: https://lore.kernel.org/oe-kbuild-all/202406272033.KejtfLkw-lkp@intel.com/

Rant: I noticed recently that checkpatch.pl aggressively demands
having the Closes: tag follow Reported-by:, even when it is not
strictly necessary per
https://www.kernel.org/doc/html/latest/process/submitting-patches.html#using-reported-by-tested-by-reviewed-by-suggested-by-and-fixes.

> Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Reviewed-by: Alexander Potapenko <glider@google.com>