Skip to contentSkip to conversation

Fix strncpy() warnings for GCC8 new -Wstringop-truncation

PATCH New Philippe Mathieu-Daudé 1 participant 6 messages Jan 3, 2019 Block layer

lore
Share & tools
lore
#0 lore raw

v5 cover Fix strncpy() warnings for GCC8 new -Wstringop-truncation

GCC 8 new warning prevents builds to success since quite some time. First report on the mailing list is in July 2018:
https://lists.gnu.org/archive/html/qemu-devel/2018-07/msg03723.html

Since v4:
- fixed some checkpatch errors/warnings reported by patchew
- commented checkpatch errors/warnings false positives
- added Reviewed-by

Now this series is fully reviewed.

Since v3:
- patch 1: make sens of description (eblake)
- patch 2: append QEMU_NONSTRING instead of prepending it (mst)
- patch 3: rebased (imammedo), intented
- patch 4: replaced by Marc-André first attempt, improved doc
- patch 5: add assert() and NUL-terminate the buffer (mst)

Various intents has been sent to fix this:
- Incorrectly using g_strlcpy()
https://lists.gnu.org/archive/html/qemu-devel/2018-08/msg03705.html
https://lists.gnu.org/archive/html/qemu-devel/2018-08/msg03706.html
- Using assert() and strpadcpy()
https://lists.gnu.org/archive/html/qemu-devel/2018-11/msg03938.html
This was the approch taken by the previous v2:
https://lists.gnu.org/archive/html/qemu-devel/2018-12/msg04405.html
- Use #pragma GCC diagnostic ignored "-Wstringop-truncation"
https://lists.gnu.org/archive/html/qemu-devel/2018-12/msg04261.html
- adding an inline wrapper with said pragma in there
https://lists.gnu.org/archive/html/qemu-devel/2018-12/msg04261.html
- -Wno-stringop-truncation is the makefile
https://lists.gnu.org/archive/html/qemu-devel/2018-12/msg04261.html
- Use the 'nonstring' attribute
https://lists.gnu.org/archive/html/qemu-devel/2018-12/msg04493.html

This series add the QEMU_NONSTRING definition and use it.

Regards,

Phil.

v4: https://lists.gnu.org/archive/html/qemu-devel/2018-12/msg06175.html
v3: https://lists.gnu.org/archive/html/qemu-devel/2018-12/msg04596.html
v2: https://lists.gnu.org/archive/html/qemu-devel/2018-12/msg04405.html
v1: https://lists.gnu.org/archive/html/qemu-devel/2018-11/msg03938.html

Marc-André Lureau (1):
migration: Fix stringop-truncation warning

Philippe Mathieu-Daudé (4):
qemu/compiler: Define QEMU_NONSTRING
block/sheepdog: Use QEMU_NONSTRING for non NUL-terminated arrays hw/acpi: Use QEMU_NONSTRING for non NUL-terminated arrays
migration: Use strnlen() for fixed-size string

 block/sheepdog.c            |  2 +-
 hw/acpi/core.c              | 12 ++++++++----
 include/hw/acpi/acpi-defs.h | 13 ++++++++-----
 include/qemu/compiler.h     | 15 +++++++++++++++
 migration/global_state.c    | 15 ++++++++++++++-
 5 files changed, 46 insertions(+), 11 deletions(-)
Signature · 1 line
2.17.2
#1 lore raw

PATCH v5 1/5 qemu/compiler: Define QEMU_NONSTRING

GCC 8 introduced the -Wstringop-truncation checker to detect truncation by the strncat and strncpy functions (closely related to -Wstringop-overflow, which detect buffer overflow by string-modifying functions declared in <string.h>).

In tandem of -Wstringop-truncation, the "nonstring" attribute was added:

The nonstring variable attribute specifies that an object or member declaration with type array of char, signed char, or unsigned char, or pointer to such a type is intended to store character arrays that do not necessarily contain a terminating NUL. This is useful in detecting uses of such arrays or pointers with functions that expect NUL-terminated strings, and to avoid warnings when such an array or pointer is used as an argument to a bounded string manipulation function such as strncpy.

From the GCC manual: https://gcc.gnu.org/onlinedocs/gcc/Common-Variable-Attributes.html#index-nonstring-variable-attribute

Add the QEMU_NONSTRING macro which checks if the compiler supports this attribute.

---
v4: reordered the commit description to make sens (eblake)

Note this trigger the following checkpatch warning (patchew) which I'm ignoring because this is consistent with how the rest of this file uses:

WARNING: architecture specific defines should be avoided #50: FILE: include/qemu/compiler.h:163:
+#if __has_attribute(nonstring)

---
 include/qemu/compiler.h | 15 +++++++++++++++
 1 file changed, 15 insertions(+)
include/qemu/compiler.h +15 -0
index 261842beae..2d8f507c73 100644
@@ -151,6 +151,21 @@
 # define QEMU_ERROR(X)
 #endif
 
+/*
+ * The nonstring variable attribute specifies that an object or member
+ * declaration with type array of char or pointer to char is intended
+ * to store character arrays that do not necessarily contain a terminating
+ * NUL character. This is useful in detecting uses of such arrays or pointers
+ * with functions that expect NUL-terminated strings, and to avoid warnings
+ * when such an array or pointer is used as an argument to a bounded string
+ * manipulation function such as strncpy.
+ */
+#if __has_attribute(nonstring)
+# define QEMU_NONSTRING __attribute__((nonstring))
+#else
+# define QEMU_NONSTRING
+#endif
+
 /* Implement C11 _Generic via GCC builtins.  Example:
  *
  *    QEMU_GENERIC(x, (float, sinf), (long double, sinl), sin) (x)
Signature · 1 line
2.17.2
#2 lore raw

PATCH v5 2/5 block/sheepdog: Use QEMU_NONSTRING for non NUL-terminated arrays

In reply to Philippe Mathieu-Daudé

GCC 8 added a -Wstringop-truncation warning:

The -Wstringop-truncation warning added in GCC 8.0 via r254630 for bug 81117 is specifically intended to highlight likely unintended uses of the strncpy function that truncate the terminating NUL character from the source string.

This new warning leads to compilation failures:

CC block/sheepdog.o
qemu/block/sheepdog.c: In function 'find_vdi_name':
qemu/block/sheepdog.c:1239:5: error: 'strncpy' specified bound 256 equals destination size [-Werror=stringop-truncation]

       strncpy(buf + SD_MAX_VDI_LEN, tag, SD_MAX_VDI_TAG_LEN);
       ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

make: *** [qemu/rules.mak:69: block/sheepdog.o] Error 1

As described previous to the strncpy() calls, the use of strncpy() is correct here:

    /* This pair of strncpy calls ensures that the buffer is zero-filled,
     * which is desirable since we'll soon be sending those bytes, and
     * don't want the send_req to read uninitialized data.
     */
    strncpy(buf, filename, SD_MAX_VDI_LEN);
    strncpy(buf + SD_MAX_VDI_LEN, tag, SD_MAX_VDI_TAG_LEN);

Use the QEMU_NONSTRING attribute, since this array is intended to store character arrays that do not necessarily contain a terminating NUL.

---
 block/sheepdog.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
block/sheepdog.c +1 -1
index 0125df9d49..5cd9618432 100644
@@ -1224,7 +1224,7 @@ static int find_vdi_name(BDRVSheepdogState *s, const char *filename,
     SheepdogVdiReq hdr;
     SheepdogVdiRsp *rsp = (SheepdogVdiRsp *)&hdr;
     unsigned int wlen, rlen = 0;
-    char buf[SD_MAX_VDI_LEN + SD_MAX_VDI_TAG_LEN];
+    char buf[SD_MAX_VDI_LEN + SD_MAX_VDI_TAG_LEN] QEMU_NONSTRING;
 
     fd = connect_to_sdog(s, errp);
     if (fd < 0) {
Signature · 1 line
2.17.2
#3 lore raw

PATCH v5 3/5 hw/acpi: Use QEMU_NONSTRING for non NUL-terminated arrays

In reply to Philippe Mathieu-Daudé

GCC 8 added a -Wstringop-truncation warning:

The -Wstringop-truncation warning added in GCC 8.0 via r254630 for bug 81117 is specifically intended to highlight likely unintended uses of the strncpy function that truncate the terminating NUL character from the source string.

This new warning leads to compilation failures:

CC hw/acpi/core.o
In function 'acpi_table_install', inlined from 'acpi_table_add' at qemu/hw/acpi/core.c:296:5:
qemu/hw/acpi/core.c:184:9: error: 'strncpy' specified bound 4 equals destination size [-Werror=stringop-truncation]

           strncpy(ext_hdr->sig, hdrs->sig, sizeof ext_hdr->sig);
           ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

make: *** [qemu/rules.mak:69: hw/acpi/core.o] Error 1

Use the QEMU_NONSTRING attribute, since ACPI tables don't require the strings to be NUL-terminated.

---
v4: rebased
v5: Fix checkpatch error: space prohibited before open square bracket '['

Note this triggers the following checkpatch warning (patchew), which is incorrect because this is a multi-line macro, thus we can not use Linux-kernel-style multiline comments here:

WARNING: Block comments use a leading /* on a separate line #83: FILE: include/hw/acpi/acpi-defs.h:61:
+ QEMU_NONSTRING; /* OEM identification */ \

---
 hw/acpi/core.c              | 12 ++++++++----
 include/hw/acpi/acpi-defs.h | 13 ++++++++-----
 2 files changed, 16 insertions(+), 9 deletions(-)
hw/acpi/core.c +8 -4
index d6f0709691..47877c0ec1 100644
@@ -35,14 +35,18 @@
 struct acpi_table_header {
     uint16_t _length;         /* our length, not actual part of the hdr */
                               /* allows easier parsing for fw_cfg clients */
-    char sig[4];              /* ACPI signature (4 ASCII characters) */
+    char sig[4]
+             QEMU_NONSTRING;  /* ACPI signature (4 ASCII characters) */
     uint32_t length;          /* Length of table, in bytes, including header */
     uint8_t revision;         /* ACPI Specification minor version # */
     uint8_t checksum;         /* To make sum of entire table == 0 */
-    char oem_id[6];           /* OEM identification */
-    char oem_table_id[8];     /* OEM table identification */
+    char oem_id[6]
+             QEMU_NONSTRING;  /* OEM identification */
+    char oem_table_id[8]
+             QEMU_NONSTRING;  /* OEM table identification */
     uint32_t oem_revision;    /* OEM revision number */
-    char asl_compiler_id[4];  /* ASL compiler vendor ID */
+    char asl_compiler_id[4]
+             QEMU_NONSTRING;  /* ASL compiler vendor ID */
     uint32_t asl_compiler_revision; /* ASL compiler revision number */
 } QEMU_PACKED;
 
include/hw/acpi/acpi-defs.h +8 -5
index 5021cb9e79..4ed160afae 100644
@@ -41,8 +41,8 @@ enum {
 };
 
 typedef struct AcpiRsdpData {
-    uint8_t oem_id[6]; /* OEM identification */
-    uint8_t revision;  /* Must be 0 for 1.0, 2 for 2.0 */
+    uint8_t oem_id[6] QEMU_NONSTRING; /* OEM identification */
+    uint8_t revision;                 /* Must be 0 for 1.0, 2 for 2.0 */
 
     unsigned *rsdt_tbl_offset;
     unsigned *xsdt_tbl_offset;
@@ -57,10 +57,13 @@ typedef struct AcpiRsdpData {
     uint32_t length;                 /* Length of table, in bytes, including header */ \
     uint8_t  revision;               /* ACPI Specification minor version # */ \
     uint8_t  checksum;               /* To make sum of entire table == 0 */ \
-    uint8_t  oem_id [6];             /* OEM identification */ \
-    uint8_t  oem_table_id [8];       /* OEM table identification */ \
+    uint8_t  oem_id[6] \
+                 QEMU_NONSTRING;     /* OEM identification */ \
+    uint8_t  oem_table_id[8] \
+                 QEMU_NONSTRING;     /* OEM table identification */ \
     uint32_t oem_revision;           /* OEM revision number */ \
-    uint8_t  asl_compiler_id [4];    /* ASL compiler vendor ID */ \
+    uint8_t  asl_compiler_id[4] \
+                 QEMU_NONSTRING;     /* ASL compiler vendor ID */ \
     uint32_t asl_compiler_revision;  /* ASL compiler revision number */
 
 
Signature · 1 line
2.17.2
#4 lore raw

PATCH v5 4/5 migration: Fix stringop-truncation warning

In reply to Philippe Mathieu-Daudé

From: Marc-André Lureau <[email protected]>

GCC 8 added a -Wstringop-truncation warning:

The -Wstringop-truncation warning added in GCC 8.0 via r254630 for bug 81117 is specifically intended to highlight likely unintended uses of the strncpy function that truncate the terminating NUL character from the source string.

This new warning leads to compilation failures:

CC migration/global_state.o
qemu/migration/global_state.c: In function 'global_state_store_running':
qemu/migration/global_state.c:45:5: error: 'strncpy' specified bound 100 equals destination size [-Werror=stringop-truncation]

       strncpy((char *)global_state.runstate, state, sizeof(global_state.runstate));
       ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

make: *** [qemu/rules.mak:69: migration/global_state.o] Error 1

Adding an assert is enough to silence GCC.

(alternatively, we could hard-code "running")

---
 migration/global_state.c | 1 +
 1 file changed, 1 insertion(+)
migration/global_state.c +1 -0
index 8e8ab5c51e..01805c567a 100644
@@ -42,6 +42,7 @@ int global_state_store(void)
 void global_state_store_running(void)
 {
     const char *state = RunState_str(RUN_STATE_RUNNING);
+    assert(strlen(state) < sizeof(global_state.runstate));
     strncpy((char *)global_state.runstate,
            state, sizeof(global_state.runstate));
 }
Signature · 1 line
2.17.2
#5 lore raw

PATCH v5 5/5 migration: Use strnlen() for fixed-size string

In reply to Philippe Mathieu-Daudé

GCC 8 introduced the -Wstringop-overflow, which detect buffer overflow by string-modifying functions declared in <string.h>, such strncpy(), used in global_state_store_running().

GCC indeed found an incorrect use of strlen(), because this array is loaded by VMSTATE_BUFFER(runstate, GlobalState) then parsed using qapi_enum_parse which does not get the buffer length.

Use strnlen() which returns sizeof(s->runstate) if the array is not NUL-terminated, assert the size is within range, and enforce the array to be NUL-terminated to avoid an overflow in qapi_enum_parse().

This fixes:

CC migration/global_state.o
qemu/migration/global_state.c: In function 'global_state_pre_save':
qemu/migration/global_state.c:109:15: error: 'strlen' argument 1 declared attribute 'nonstring' [-Werror=stringop-overflow=]

       s->size = strlen((char *)s->runstate) + 1;
                 ^~~~~~~~~~~~~~~~~~~~~~~~~~~
  qemu/migration/global_state.c:24:13: note: argument 'runstate' declared here
       uint8_t runstate[100] QEMU_NONSTRING;
               ^~~~~~~~

cc1: all warnings being treated as errors
make: *** [qemu/rules.mak:69: migration/global_state.o] Error 1

---
v5: Use Linux-kernel-style multiline comments

---
 migration/global_state.c | 14 +++++++++++++-
 1 file changed, 13 insertions(+), 1 deletion(-)
migration/global_state.c +13 -1
index 01805c567a..2c8c447239 100644
@@ -89,6 +89,17 @@ static int global_state_post_load(void *opaque, int version_id)
     s->received = true;
     trace_migrate_global_state_post_load(runstate);
 
+    if (strnlen((char *)s->runstate,
+                sizeof(s->runstate)) == sizeof(s->runstate)) {
+        /*
+         * This condition should never happen during migration, because
+         * all runstate names are shorter than 100 bytes (the size of
+         * s->runstate). However, a malicious stream could overflow
+         * the qapi_enum_parse() call, so we force the last character
+         * to a NUL byte.
+         */
+        s->runstate[sizeof(s->runstate) - 1] = '\0';
+    }
     r = qapi_enum_parse(&RunState_lookup, runstate, -1, &local_err);
 
     if (r == -1) {
@@ -107,7 +118,8 @@ static int global_state_pre_save(void *opaque)
     GlobalState *s = opaque;
 
     trace_migrate_global_state_pre_save((char *)s->runstate);
-    s->size = strlen((char *)s->runstate) + 1;
+    s->size = strnlen((char *)s->runstate, sizeof(s->runstate)) + 1;
+    assert(s->size <= sizeof(s->runstate));
 
     return 0;
 }
Signature · 1 line
2.17.2