[PATCH bpf-next n2 0/1] selftests/bpf: Fix a few tests for GCC related

[PATCH bpf-next n2 0/1] selftests/bpf: Fix a few tests for GCC related

[Cupertino Miranda]

Hi everyone,

This is version 2 for the proposed patch.
In initial version I rather disabled the warnings from showing for the
particular tests, as I was afraid I would change the test purpose by
changing the code.
This time I corrected the warnings in GCC by adapting the code. 

For jeq_infer_not_null_fail.c, it seems rather more intricate to fix the
warning. In this particular case I left the disabled warning. If you
rather think that it can be fixed I hope you can advise me how to.

Best regards,
Cupertino

Changes from v1:
 - Fixes warnings by actual code changes.

Cc: jose.marchesi@oracle.com
Cc: david.faust@oracle.com
Cc: Yonghong Song <yonghong.song@linux.dev>
Cc: Eduard Zingerman <eddyz87@gmail.com>
Cc: Andrii Nakryiko <andrii.nakryiko@gmail.com>

Cupertino Miranda (1):
  selftests/bpf: Fix a few tests for GCC related warnings.

 .../selftests/bpf/progs/cpumask_failure.c     |  3 --
 .../testing/selftests/bpf/progs/dynptr_fail.c | 12 ++---
 .../bpf/progs/jeq_infer_not_null_fail.c       |  4 ++
 .../selftests/bpf/progs/test_tunnel_kern.c    | 47 +++++++++++--------
 4 files changed, 37 insertions(+), 29 deletions(-)

-- 
2.39.2

[PATCH bpf-next v2 1/1] selftests/bpf: Fix a few tests for GCC related warnings.

[Cupertino Miranda]

This patch corrects a few warnings to allow selftests to compile for
GCC.

-- progs/cpumask_failure.c --

progs/bpf_misc.h:136:22: error: ‘cpumask’ is used uninitialized
[-Werror=uninitialized]
  136 | #define __sink(expr) asm volatile("" : "+g"(expr))
      |                      ^~~
progs/cpumask_failure.c:68:9: note: in expansion of macro ‘__sink’
   68 |         __sink(cpumask);

The macro __sink(cpumask) with the '+' contraint modifier forces the
the compiler to expect a read and write from cpumask. GCC detects
that cpumask is never initialized and reports an error.
This patch removes the spurious non required definitions of cpumask.

-- progs/dynptr_fail.c --

progs/dynptr_fail.c:1444:9: error: ‘ptr1’ may be used uninitialized
[-Werror=maybe-uninitialized]
 1444 |         bpf_dynptr_clone(&ptr1, &ptr2);

Many of the tests in the file are related to the detection of
uninitialized pointers by the verifier. GCC is able to detect possible
uninitialized values, and reports this as an error.
The patch initializes all of the previous uninitialized structs.

-- progs/test_tunnel_kern.c --

progs/test_tunnel_kern.c:590:9: error: array subscript 1 is outside
array bounds of ‘struct geneve_opt[1]’ [-Werror=array-bounds=]
  590 |         *(int *) &gopt.opt_data = bpf_htonl(0xdeadbeef);
      |         ^~~~~~~~~~~~~~~~~~~~~~~
progs/test_tunnel_kern.c:575:27: note: at offset 4 into object ‘gopt’ of
size 4
  575 |         struct geneve_opt gopt;

This tests accesses beyond the defined data for the struct geneve_opt
which contains as last field "u8 opt_data[0]" which clearly does not get
reserved space (in stack) in the function header. This pattern is
repeated in ip6geneve_set_tunnel and geneve_set_tunnel functions.
GCC is able to see this and emits a warning.
The patch introduces a local struct that allocates enough space to
safely allow the write to opt_data field.

-- progs/jeq_infer_not_null_fail.c --

progs/jeq_infer_not_null_fail.c:21:40: error: array subscript ‘struct
bpf_map[0]’ is partly outside array bounds of ‘struct <anonymous>[1]’
[-Werror=array-bounds=]
   21 |         struct bpf_map *inner_map = map->inner_map_meta;
      |                                        ^~
progs/jeq_infer_not_null_fail.c:14:3: note: object ‘m_hash’ of size 32
   14 | } m_hash SEC(".maps");

This example defines m_hash in the context of the compilation unit and
casts it to struct bpf_map which is much smaller than the size of struct
bpf_map. It errors out in GCC when it attempts to access an element that
would be defined in struct bpf_map outsize of the defined limits for
m_hash.
This patch disables the warning through a GCC pragma.

This changes were tested in bpf-next master selftests without any
regressions.

Signed-off-by: Cupertino Miranda <cupertino.miranda@oracle.com>
Cc: jose.marchesi@oracle.com
Cc: david.faust@oracle.com
Cc: Yonghong Song <yonghong.song@linux.dev>
Cc: Eduard Zingerman <eddyz87@gmail.com>
Cc: Andrii Nakryiko <andrii.nakryiko@gmail.com>
---
 .../selftests/bpf/progs/cpumask_failure.c     |  3 --
 .../testing/selftests/bpf/progs/dynptr_fail.c | 12 ++---
 .../bpf/progs/jeq_infer_not_null_fail.c       |  4 ++
 .../selftests/bpf/progs/test_tunnel_kern.c    | 47 +++++++++++--------
 4 files changed, 37 insertions(+), 29 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/cpumask_failure.c b/tools/testing/selftests/bpf/progs/cpumask_failure.c
index a9bf6ea336cf..a988d2823b52 100644
--- a/tools/testing/selftests/bpf/progs/cpumask_failure.c
+++ b/tools/testing/selftests/bpf/progs/cpumask_failure.c
@@ -61,11 +61,8 @@ SEC("tp_btf/task_newtask")
 __failure __msg("bpf_cpumask_set_cpu args#1 expected pointer to STRUCT bpf_cpumask")
 int BPF_PROG(test_mutate_cpumask, struct task_struct *task, u64 clone_flags)
 {
-	struct bpf_cpumask *cpumask;
-
 	/* Can't set the CPU of a non-struct bpf_cpumask. */
 	bpf_cpumask_set_cpu(0, (struct bpf_cpumask *)task->cpus_ptr);
-	__sink(cpumask);
 
 	return 0;
 }
diff --git a/tools/testing/selftests/bpf/progs/dynptr_fail.c b/tools/testing/selftests/bpf/progs/dynptr_fail.c
index 7ce7e827d5f0..66a60bfb5867 100644
--- a/tools/testing/selftests/bpf/progs/dynptr_fail.c
+++ b/tools/testing/selftests/bpf/progs/dynptr_fail.c
@@ -80,7 +80,7 @@ SEC("?raw_tp")
 __failure __msg("Unreleased reference id=2")
 int ringbuf_missing_release1(void *ctx)
 {
-	struct bpf_dynptr ptr;
+	struct bpf_dynptr ptr = {};
 
 	bpf_ringbuf_reserve_dynptr(&ringbuf, val, 0, &ptr);
 
@@ -1385,7 +1385,7 @@ SEC("?raw_tp")
 __failure __msg("Expected an initialized dynptr as arg #1")
 int dynptr_adjust_invalid(void *ctx)
 {
-	struct bpf_dynptr ptr;
+	struct bpf_dynptr ptr = {};
 
 	/* this should fail */
 	bpf_dynptr_adjust(&ptr, 1, 2);
@@ -1398,7 +1398,7 @@ SEC("?raw_tp")
 __failure __msg("Expected an initialized dynptr as arg #1")
 int dynptr_is_null_invalid(void *ctx)
 {
-	struct bpf_dynptr ptr;
+	struct bpf_dynptr ptr = {};
 
 	/* this should fail */
 	bpf_dynptr_is_null(&ptr);
@@ -1411,7 +1411,7 @@ SEC("?raw_tp")
 __failure __msg("Expected an initialized dynptr as arg #1")
 int dynptr_is_rdonly_invalid(void *ctx)
 {
-	struct bpf_dynptr ptr;
+	struct bpf_dynptr ptr = {};
 
 	/* this should fail */
 	bpf_dynptr_is_rdonly(&ptr);
@@ -1424,7 +1424,7 @@ SEC("?raw_tp")
 __failure __msg("Expected an initialized dynptr as arg #1")
 int dynptr_size_invalid(void *ctx)
 {
-	struct bpf_dynptr ptr;
+	struct bpf_dynptr ptr = {};
 
 	/* this should fail */
 	bpf_dynptr_size(&ptr);
@@ -1437,7 +1437,7 @@ SEC("?raw_tp")
 __failure __msg("Expected an initialized dynptr as arg #1")
 int clone_invalid1(void *ctx)
 {
-	struct bpf_dynptr ptr1;
+	struct bpf_dynptr ptr1 = {};
 	struct bpf_dynptr ptr2;
 
 	/* this should fail */
diff --git a/tools/testing/selftests/bpf/progs/jeq_infer_not_null_fail.c b/tools/testing/selftests/bpf/progs/jeq_infer_not_null_fail.c
index f46965053acb..4d619bea9c75 100644
--- a/tools/testing/selftests/bpf/progs/jeq_infer_not_null_fail.c
+++ b/tools/testing/selftests/bpf/progs/jeq_infer_not_null_fail.c
@@ -4,6 +4,10 @@
 #include <bpf/bpf_helpers.h>
 #include "bpf_misc.h"
 
+#ifndef __clang__
+#pragma GCC diagnostic ignored "-Warray-bounds"
+#endif
+
 char _license[] SEC("license") = "GPL";
 
 struct {
diff --git a/tools/testing/selftests/bpf/progs/test_tunnel_kern.c b/tools/testing/selftests/bpf/progs/test_tunnel_kern.c
index 3e436e6f7312..3f5abcf3ff13 100644
--- a/tools/testing/selftests/bpf/progs/test_tunnel_kern.c
+++ b/tools/testing/selftests/bpf/progs/test_tunnel_kern.c
@@ -567,12 +567,18 @@ int ip6vxlan_get_tunnel_src(struct __sk_buff *skb)
 	return TC_ACT_OK;
 }
 
+struct local_geneve_opt {
+	struct geneve_opt gopt;
+	int data;
+};
+
 SEC("tc")
 int geneve_set_tunnel(struct __sk_buff *skb)
 {
 	int ret;
 	struct bpf_tunnel_key key;
-	struct geneve_opt gopt;
+	struct local_geneve_opt local_gopt;
+	struct geneve_opt *gopt = (struct geneve_opt *) &local_gopt;
 
 	__builtin_memset(&key, 0x0, sizeof(key));
 	key.remote_ipv4 = 0xac100164; /* 172.16.1.100 */
@@ -580,14 +586,14 @@ int geneve_set_tunnel(struct __sk_buff *skb)
 	key.tunnel_tos = 0;
 	key.tunnel_ttl = 64;
 
-	__builtin_memset(&gopt, 0x0, sizeof(gopt));
-	gopt.opt_class = bpf_htons(0x102); /* Open Virtual Networking (OVN) */
-	gopt.type = 0x08;
-	gopt.r1 = 0;
-	gopt.r2 = 0;
-	gopt.r3 = 0;
-	gopt.length = 2; /* 4-byte multiple */
-	*(int *) &gopt.opt_data = bpf_htonl(0xdeadbeef);
+	__builtin_memset(gopt, 0x0, sizeof(local_gopt));
+	gopt->opt_class = bpf_htons(0x102); /* Open Virtual Networking (OVN) */
+	gopt->type = 0x08;
+	gopt->r1 = 0;
+	gopt->r2 = 0;
+	gopt->r3 = 0;
+	gopt->length = 2; /* 4-byte multiple */
+	*(int *) &gopt->opt_data = bpf_htonl(0xdeadbeef);
 
 	ret = bpf_skb_set_tunnel_key(skb, &key, sizeof(key),
 				     BPF_F_ZERO_CSUM_TX);
@@ -596,7 +602,7 @@ int geneve_set_tunnel(struct __sk_buff *skb)
 		return TC_ACT_SHOT;
 	}
 
-	ret = bpf_skb_set_tunnel_opt(skb, &gopt, sizeof(gopt));
+	ret = bpf_skb_set_tunnel_opt(skb, gopt, sizeof(local_gopt));
 	if (ret < 0) {
 		log_err(ret);
 		return TC_ACT_SHOT;
@@ -631,7 +637,8 @@ SEC("tc")
 int ip6geneve_set_tunnel(struct __sk_buff *skb)
 {
 	struct bpf_tunnel_key key;
-	struct geneve_opt gopt;
+	struct local_geneve_opt local_gopt;
+	struct geneve_opt *gopt = (struct geneve_opt *) &local_gopt;
 	int ret;
 
 	__builtin_memset(&key, 0x0, sizeof(key));
@@ -647,16 +654,16 @@ int ip6geneve_set_tunnel(struct __sk_buff *skb)
 		return TC_ACT_SHOT;
 	}
 
-	__builtin_memset(&gopt, 0x0, sizeof(gopt));
-	gopt.opt_class = bpf_htons(0x102); /* Open Virtual Networking (OVN) */
-	gopt.type = 0x08;
-	gopt.r1 = 0;
-	gopt.r2 = 0;
-	gopt.r3 = 0;
-	gopt.length = 2; /* 4-byte multiple */
-	*(int *) &gopt.opt_data = bpf_htonl(0xfeedbeef);
+	__builtin_memset(gopt, 0x0, sizeof(local_gopt));
+	gopt->opt_class = bpf_htons(0x102); /* Open Virtual Networking (OVN) */
+	gopt->type = 0x08;
+	gopt->r1 = 0;
+	gopt->r2 = 0;
+	gopt->r3 = 0;
+	gopt->length = 2; /* 4-byte multiple */
+	*(int *) &gopt->opt_data = bpf_htonl(0xfeedbeef);
 
-	ret = bpf_skb_set_tunnel_opt(skb, &gopt, sizeof(gopt));
+	ret = bpf_skb_set_tunnel_opt(skb, gopt, sizeof(gopt));
 	if (ret < 0) {
 		log_err(ret);
 		return TC_ACT_SHOT;
-- 
2.39.2

Re: [PATCH bpf-next n2 0/1] selftests/bpf: Fix a few tests for GCC related

[patchwork-bot+netdevbpf]

Hello:

This patch was applied to bpf/bpf-next.git (master)
by Alexei Starovoitov <ast@kernel.org>:

On Fri, 10 May 2024 19:38:49 +0100 you wrote:
> Hi everyone,
> 
> This is version 2 for the proposed patch.
> In initial version I rather disabled the warnings from showing for the
> particular tests, as I was afraid I would change the test purpose by
> changing the code.
> This time I corrected the warnings in GCC by adapting the code.
> 
> [...]

Here is the summary with links:
  - [bpf-next,v2,1/1] selftests/bpf: Fix a few tests for GCC related warnings.
    https://git.kernel.org/bpf/bpf-next/c/5ddafcc377f9

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html