[PATCH 1/5] samples/bpf: fix test_lru_dist build

[PATCH 1/5] samples/bpf: fix test_lru_dist build

[Matteo Croce]

Fix the following error by removing a duplicate struct definition:

samples/bpf/test_lru_dist.c:39:8: error: redefinition of ‘struct list_head’
   39 | struct list_head {
      |        ^~~~~~~~~
In file included from samples/bpf/test_lru_dist.c:9:
./tools/include/linux/types.h:69:8: note: originally defined here
   69 | struct list_head {
      |        ^~~~~~~~~
make[2]: *** [scripts/Makefile.host:92: samples/bpf/test_lru_dist] Error 1
make[1]: *** [Makefile:1763: samples/bpf/] Error 2

Signed-off-by: Matteo Croce <mcroce@redhat.com>
---
 samples/bpf/test_lru_dist.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/samples/bpf/test_lru_dist.c b/samples/bpf/test_lru_dist.c
index eec3e2509ce8..f532b894654b 100644
--- a/samples/bpf/test_lru_dist.c
+++ b/samples/bpf/test_lru_dist.c
@@ -36,10 +36,6 @@ static int nr_cpus;
 static unsigned long long *dist_keys;
 static unsigned int dist_key_counts;
 
-struct list_head {
-	struct list_head *next, *prev;
-};
-
 static inline void INIT_LIST_HEAD(struct list_head *list)
 {
 	list->next = list;
-- 
2.21.0

[PATCH 2/5] libbpf: add missing typedef

[Matteo Croce]

Sync tools/include/linux/types.h with the UAPI one to fix this build error:

make -C samples/bpf/../../tools/lib/bpf/ RM='rm -rf' LDFLAGS= srctree=samples/bpf/../../ O=
  HOSTCC  samples/bpf/sock_example
In file included from samples/bpf/sock_example.c:27:
/usr/include/linux/ip.h:102:2: error: unknown type name ‘__sum16’
  102 |  __sum16 check;
      |  ^~~~~~~
make[2]: *** [scripts/Makefile.host:92: samples/bpf/sock_example] Error 1
make[1]: *** [Makefile:1763: samples/bpf/] Error 2

Signed-off-by: Matteo Croce <mcroce@redhat.com>
---
 tools/include/linux/types.h | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/tools/include/linux/types.h b/tools/include/linux/types.h
index 154eb4e3ca7c..5266dbfee945 100644
--- a/tools/include/linux/types.h
+++ b/tools/include/linux/types.h
@@ -58,6 +58,9 @@ typedef __u32 __bitwise __be32;
 typedef __u64 __bitwise __le64;
 typedef __u64 __bitwise __be64;
 
+typedef __u16 __bitwise __sum16;
+typedef __u32 __bitwise __wsum;
+
 typedef struct {
 	int counter;
 } atomic_t;
-- 
2.21.0

Re: [PATCH 2/5] libbpf: add missing typedef

[Stanislav Fomichev]

On 05/18, Matteo Croce wrote:
> Sync tools/include/linux/types.h with the UAPI one to fix this build error:
> 
> make -C samples/bpf/../../tools/lib/bpf/ RM='rm -rf' LDFLAGS= srctree=samples/bpf/../../ O=
>   HOSTCC  samples/bpf/sock_example
> In file included from samples/bpf/sock_example.c:27:
> /usr/include/linux/ip.h:102:2: error: unknown type name ‘__sum16’
>   102 |  __sum16 check;
>       |  ^~~~~~~
> make[2]: *** [scripts/Makefile.host:92: samples/bpf/sock_example] Error 1
> make[1]: *** [Makefile:1763: samples/bpf/] Error 2
> 
> Signed-off-by: Matteo Croce <mcroce@redhat.com>
> ---
>  tools/include/linux/types.h | 3 +++
>  1 file changed, 3 insertions(+)
> 
> diff --git a/tools/include/linux/types.h b/tools/include/linux/types.h
> index 154eb4e3ca7c..5266dbfee945 100644
> --- a/tools/include/linux/types.h
> +++ b/tools/include/linux/types.h
> @@ -58,6 +58,9 @@ typedef __u32 __bitwise __be32;
>  typedef __u64 __bitwise __le64;
>  typedef __u64 __bitwise __be64;
>  
> +typedef __u16 __bitwise __sum16;
> +typedef __u32 __bitwise __wsum;
If you do that, you should probably remove 'typedef __u16 __sum16;'
from test_progs.h:

https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf.git/tree/tools/testing/selftests/bpf/test_progs.h#n13

> +
>  typedef struct {
>  	int counter;
>  } atomic_t;
> -- 
> 2.21.0
> 

Re: [PATCH 2/5] libbpf: add missing typedef

[Matteo Croce]

On Mon, May 20, 2019 at 6:53 PM Stanislav Fomichev <sdf@fomichev.me> wrote:
>
> On 05/18, Matteo Croce wrote:
> > Sync tools/include/linux/types.h with the UAPI one to fix this build error:
> >
> > make -C samples/bpf/../../tools/lib/bpf/ RM='rm -rf' LDFLAGS= srctree=samples/bpf/../../ O=
> >   HOSTCC  samples/bpf/sock_example
> > In file included from samples/bpf/sock_example.c:27:
> > /usr/include/linux/ip.h:102:2: error: unknown type name ‘__sum16’
> >   102 |  __sum16 check;
> >       |  ^~~~~~~
> > make[2]: *** [scripts/Makefile.host:92: samples/bpf/sock_example] Error 1
> > make[1]: *** [Makefile:1763: samples/bpf/] Error 2
> >
> > Signed-off-by: Matteo Croce <mcroce@redhat.com>
> > ---
> >  tools/include/linux/types.h | 3 +++
> >  1 file changed, 3 insertions(+)
> >
> > diff --git a/tools/include/linux/types.h b/tools/include/linux/types.h
> > index 154eb4e3ca7c..5266dbfee945 100644
> > --- a/tools/include/linux/types.h
> > +++ b/tools/include/linux/types.h
> > @@ -58,6 +58,9 @@ typedef __u32 __bitwise __be32;
> >  typedef __u64 __bitwise __le64;
> >  typedef __u64 __bitwise __be64;
> >
> > +typedef __u16 __bitwise __sum16;
> > +typedef __u32 __bitwise __wsum;
> If you do that, you should probably remove 'typedef __u16 __sum16;'
> from test_progs.h:
>
> https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf.git/tree/tools/testing/selftests/bpf/test_progs.h#n13
>
> > +
> >  typedef struct {
> >       int counter;
> >  } atomic_t;
> > --
> > 2.21.0
> >

Hi,

I see test_progs.h only included in tools/testing/selftests/bpf/prog_tests/*,
so maybe it's unreladed to my change in samples/bpf/.
Maybe in a different patchset.

Bye,
-- 
Matteo Croce
per aspera ad upstream

Re: [PATCH 2/5] libbpf: add missing typedef

[Stanislav Fomichev]

On 05/20, Matteo Croce wrote:
> On Mon, May 20, 2019 at 6:53 PM Stanislav Fomichev <sdf@fomichev.me> wrote:
> >
> > On 05/18, Matteo Croce wrote:
> > > Sync tools/include/linux/types.h with the UAPI one to fix this build error:
> > >
> > > make -C samples/bpf/../../tools/lib/bpf/ RM='rm -rf' LDFLAGS= srctree=samples/bpf/../../ O=
> > >   HOSTCC  samples/bpf/sock_example
> > > In file included from samples/bpf/sock_example.c:27:
> > > /usr/include/linux/ip.h:102:2: error: unknown type name ‘__sum16’
> > >   102 |  __sum16 check;
> > >       |  ^~~~~~~
> > > make[2]: *** [scripts/Makefile.host:92: samples/bpf/sock_example] Error 1
> > > make[1]: *** [Makefile:1763: samples/bpf/] Error 2
> > >
> > > Signed-off-by: Matteo Croce <mcroce@redhat.com>
> > > ---
> > >  tools/include/linux/types.h | 3 +++
> > >  1 file changed, 3 insertions(+)
> > >
> > > diff --git a/tools/include/linux/types.h b/tools/include/linux/types.h
> > > index 154eb4e3ca7c..5266dbfee945 100644
> > > --- a/tools/include/linux/types.h
> > > +++ b/tools/include/linux/types.h
> > > @@ -58,6 +58,9 @@ typedef __u32 __bitwise __be32;
> > >  typedef __u64 __bitwise __le64;
> > >  typedef __u64 __bitwise __be64;
> > >
> > > +typedef __u16 __bitwise __sum16;
> > > +typedef __u32 __bitwise __wsum;
> > If you do that, you should probably remove 'typedef __u16 __sum16;'
> > from test_progs.h:
> >
> > https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf.git/tree/tools/testing/selftests/bpf/test_progs.h#n13
> >
> > > +
> > >  typedef struct {
> > >       int counter;
> > >  } atomic_t;
> > > --
> > > 2.21.0
> > >
> 
> Hi,
> 
> I see test_progs.h only included in tools/testing/selftests/bpf/prog_tests/*,
> so maybe it's unreladed to my change in samples/bpf/.
> Maybe in a different patchset.
Yes, I'm just saying that now that you have __sum16 defined in
tools/include/linux/types.h you can have another patch to remove
that custom __sum16 typedef from tests_progs.h

[PATCH 3/5] samples/bpf: fix xdpsock_user build error

[Matteo Croce]

Remove duplicate typedef, and use PRIu64 to be both 32 and 64 bit aware.
Fix the following error:

samples/bpf/xdpsock_user.c:52:15: error: conflicting types for ‘u64’
   52 | typedef __u64 u64;
      |               ^~~
In file included from ./tools/include/linux/compiler.h:87,
                 from ./tools/include/asm/barrier.h:2,
                 from samples/bpf/xdpsock_user.c:4:
./tools/include/linux/types.h:30:18: note: previous declaration of ‘u64’ was here
   30 | typedef uint64_t u64;
      |                  ^~~
make[2]: *** [scripts/Makefile.host:109: samples/bpf/xdpsock_user.o] Error 1
make[1]: *** [Makefile:1763: samples/bpf/] Error 2

Signed-off-by: Matteo Croce <mcroce@redhat.com>
---
 samples/bpf/xdpsock_user.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/samples/bpf/xdpsock_user.c b/samples/bpf/xdpsock_user.c
index d08ee1ab7bb4..a4cd42c2f0b0 100644
--- a/samples/bpf/xdpsock_user.c
+++ b/samples/bpf/xdpsock_user.c
@@ -24,6 +24,7 @@
 #include <sys/resource.h>
 #include <sys/socket.h>
 #include <sys/types.h>
+#include <inttypes.h>
 #include <time.h>
 #include <unistd.h>
 
@@ -49,9 +50,6 @@
 #define DEBUG_HEXDUMP 0
 #define MAX_SOCKS 8
 
-typedef __u64 u64;
-typedef __u32 u32;
-
 static unsigned long prev_time;
 
 enum benchmark_type {
@@ -243,7 +241,7 @@ static void hex_dump(void *pkt, size_t length, u64 addr)
 	if (!DEBUG_HEXDUMP)
 		return;
 
-	sprintf(buf, "addr=%llu", addr);
+	sprintf(buf, "addr=%" PRIu64, addr);
 	printf("length = %zu\n", length);
 	printf("%s | ", buf);
 	while (length-- > 0) {
-- 
2.21.0

[PATCH 4/5] samples/bpf: fix tracex5_user build error

[Matteo Croce]

Add missing symbols to tools/include/linux/filter.h to fix a build failure:

make -C samples/bpf/../../tools/lib/bpf/ RM='rm -rf' LDFLAGS= srctree=samples/bpf/../../ O=
  HOSTCC  samples/bpf/tracex5_user.o
samples/bpf/tracex5_user.c: In function ‘install_accept_all_seccomp’:
samples/bpf/tracex5_user.c:17:21: error: array type has incomplete element type ‘struct sock_filter’
   17 |  struct sock_filter filter[] = {
      |                     ^~~~~~
samples/bpf/tracex5_user.c:18:3: warning: implicit declaration of function ‘BPF_STMT’; did you mean ‘BPF_STX’? [-Wimplicit-function-declaration]
   18 |   BPF_STMT(BPF_RET+BPF_K, SECCOMP_RET_ALLOW),
      |   ^~~~~~~~
      |   BPF_STX
samples/bpf/tracex5_user.c:20:9: error: variable ‘prog’ has initializer but incomplete type
   20 |  struct sock_fprog prog = {
      |         ^~~~~~~~~~
samples/bpf/tracex5_user.c:21:4: error: ‘struct sock_fprog’ has no member named ‘len’
   21 |   .len = (unsigned short)(sizeof(filter)/sizeof(filter[0])),
      |    ^~~
samples/bpf/tracex5_user.c:21:10: warning: excess elements in struct initializer
   21 |   .len = (unsigned short)(sizeof(filter)/sizeof(filter[0])),
      |          ^
samples/bpf/tracex5_user.c:21:10: note: (near initialization for ‘prog’)
samples/bpf/tracex5_user.c:22:4: error: ‘struct sock_fprog’ has no member named ‘filter’
   22 |   .filter = filter,
      |    ^~~~~~
samples/bpf/tracex5_user.c:22:13: warning: excess elements in struct initializer
   22 |   .filter = filter,
      |             ^~~~~~
samples/bpf/tracex5_user.c:22:13: note: (near initialization for ‘prog’)
samples/bpf/tracex5_user.c:20:20: error: storage size of ‘prog’ isn’t known
   20 |  struct sock_fprog prog = {
      |                    ^~~~
samples/bpf/tracex5_user.c:20:20: warning: unused variable ‘prog’ [-Wunused-variable]
samples/bpf/tracex5_user.c:17:21: warning: unused variable ‘filter’ [-Wunused-variable]
   17 |  struct sock_filter filter[] = {
      |                     ^~~~~~
make[2]: *** [scripts/Makefile.host:109: samples/bpf/tracex5_user.o] Error 1
make[1]: *** [Makefile:1763: samples/bpf/] Error 2

Signed-off-by: Matteo Croce <mcroce@redhat.com>
---
 tools/include/linux/filter.h | 27 +++++++++++++++++++++++++++
 1 file changed, 27 insertions(+)

diff --git a/tools/include/linux/filter.h b/tools/include/linux/filter.h
index ca28b6ab8db7..6b2ed7eccfa5 100644
--- a/tools/include/linux/filter.h
+++ b/tools/include/linux/filter.h
@@ -7,6 +7,33 @@
 
 #include <linux/bpf.h>
 
+/*
+ *	Try and keep these values and structures similar to BSD, especially
+ *	the BPF code definitions which need to match so you can share filters
+ */
+
+struct sock_filter {	/* Filter block */
+	__u16	code;   /* Actual filter code */
+	__u8	jt;	/* Jump true */
+	__u8	jf;	/* Jump false */
+	__u32	k;      /* Generic multiuse field */
+};
+
+struct sock_fprog {	/* Required for SO_ATTACH_FILTER. */
+	unsigned short		len;	/* Number of filter blocks */
+	struct sock_filter __user *filter;
+};
+
+/*
+ * Macros for filter block array initializers.
+ */
+#ifndef BPF_STMT
+#define BPF_STMT(code, k) { (unsigned short)(code), 0, 0, k }
+#endif
+#ifndef BPF_JUMP
+#define BPF_JUMP(code, k, jt, jf) { (unsigned short)(code), jt, jf, k }
+#endif
+
 /* ArgX, context and stack frame pointer register positions. Note,
  * Arg1, Arg2, Arg3, etc are used as argument mappings of function
  * calls in BPF_CALL instruction.
-- 
2.21.0

Re: [PATCH 4/5] samples/bpf: fix tracex5_user build error

[Daniel Borkmann]

On 05/18/2019 02:46 AM, Matteo Croce wrote:
> Add missing symbols to tools/include/linux/filter.h to fix a build failure:
> 
> make -C samples/bpf/../../tools/lib/bpf/ RM='rm -rf' LDFLAGS= srctree=samples/bpf/../../ O=
>   HOSTCC  samples/bpf/tracex5_user.o
> samples/bpf/tracex5_user.c: In function ‘install_accept_all_seccomp’:
> samples/bpf/tracex5_user.c:17:21: error: array type has incomplete element type ‘struct sock_filter’
>    17 |  struct sock_filter filter[] = {
>       |                     ^~~~~~
> samples/bpf/tracex5_user.c:18:3: warning: implicit declaration of function ‘BPF_STMT’; did you mean ‘BPF_STX’? [-Wimplicit-function-declaration]
>    18 |   BPF_STMT(BPF_RET+BPF_K, SECCOMP_RET_ALLOW),
>       |   ^~~~~~~~
>       |   BPF_STX
> samples/bpf/tracex5_user.c:20:9: error: variable ‘prog’ has initializer but incomplete type
>    20 |  struct sock_fprog prog = {
>       |         ^~~~~~~~~~
> samples/bpf/tracex5_user.c:21:4: error: ‘struct sock_fprog’ has no member named ‘len’
>    21 |   .len = (unsigned short)(sizeof(filter)/sizeof(filter[0])),
>       |    ^~~
> samples/bpf/tracex5_user.c:21:10: warning: excess elements in struct initializer
>    21 |   .len = (unsigned short)(sizeof(filter)/sizeof(filter[0])),
>       |          ^
> samples/bpf/tracex5_user.c:21:10: note: (near initialization for ‘prog’)
> samples/bpf/tracex5_user.c:22:4: error: ‘struct sock_fprog’ has no member named ‘filter’
>    22 |   .filter = filter,
>       |    ^~~~~~
> samples/bpf/tracex5_user.c:22:13: warning: excess elements in struct initializer
>    22 |   .filter = filter,
>       |             ^~~~~~
> samples/bpf/tracex5_user.c:22:13: note: (near initialization for ‘prog’)
> samples/bpf/tracex5_user.c:20:20: error: storage size of ‘prog’ isn’t known
>    20 |  struct sock_fprog prog = {
>       |                    ^~~~
> samples/bpf/tracex5_user.c:20:20: warning: unused variable ‘prog’ [-Wunused-variable]
> samples/bpf/tracex5_user.c:17:21: warning: unused variable ‘filter’ [-Wunused-variable]
>    17 |  struct sock_filter filter[] = {
>       |                     ^~~~~~
> make[2]: *** [scripts/Makefile.host:109: samples/bpf/tracex5_user.o] Error 1
> make[1]: *** [Makefile:1763: samples/bpf/] Error 2
> 
> Signed-off-by: Matteo Croce <mcroce@redhat.com>
> ---
>  tools/include/linux/filter.h | 27 +++++++++++++++++++++++++++
>  1 file changed, 27 insertions(+)
> 
> diff --git a/tools/include/linux/filter.h b/tools/include/linux/filter.h
> index ca28b6ab8db7..6b2ed7eccfa5 100644
> --- a/tools/include/linux/filter.h
> +++ b/tools/include/linux/filter.h
> @@ -7,6 +7,33 @@
>  
>  #include <linux/bpf.h>

This here is also mixing UAPI code below into non-UAPI headers in
tooling infrastructure ..

> +/*
> + *	Try and keep these values and structures similar to BSD, especially
> + *	the BPF code definitions which need to match so you can share filters
> + */
> +
> +struct sock_filter {	/* Filter block */
> +	__u16	code;   /* Actual filter code */
> +	__u8	jt;	/* Jump true */
> +	__u8	jf;	/* Jump false */
> +	__u32	k;      /* Generic multiuse field */
> +};
> +
> +struct sock_fprog {	/* Required for SO_ATTACH_FILTER. */
> +	unsigned short		len;	/* Number of filter blocks */
> +	struct sock_filter __user *filter;
> +};
> +
> +/*
> + * Macros for filter block array initializers.
> + */
> +#ifndef BPF_STMT
> +#define BPF_STMT(code, k) { (unsigned short)(code), 0, 0, k }
> +#endif
> +#ifndef BPF_JUMP
> +#define BPF_JUMP(code, k, jt, jf) { (unsigned short)(code), jt, jf, k }
> +#endif
> +
>  /* ArgX, context and stack frame pointer register positions. Note,
>   * Arg1, Arg2, Arg3, etc are used as argument mappings of function
>   * calls in BPF_CALL instruction.
> 

[PATCH 5/5] samples/bpf: fix hbm build error

[Matteo Croce]

Fix the following build error by declaring bpf_spin_lock in hbm.c.
Including the UAPI header generates tons of redefined symbol errors,
and including it in hbm.h breaks hbm_out_kern.c.

make -C samples/bpf/../../tools/lib/bpf/ RM='rm -rf' LDFLAGS= srctree=samples/bpf/../../ O=
  HOSTCC  samples/bpf/hbm.o
In file included from samples/bpf/hbm.c:49:
samples/bpf/hbm.h:12:23: error: field ‘lock’ has incomplete type
   12 |  struct bpf_spin_lock lock;
      |                       ^~~~
make[2]: *** [scripts/Makefile.host:109: samples/bpf/hbm.o] Error 1
make[1]: *** [Makefile:1763: samples/bpf/] Error 2

Signed-off-by: Matteo Croce <mcroce@redhat.com>
---
 samples/bpf/hbm.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/samples/bpf/hbm.c b/samples/bpf/hbm.c
index a79828ab273f..ca8e567b63c3 100644
--- a/samples/bpf/hbm.c
+++ b/samples/bpf/hbm.c
@@ -40,17 +40,22 @@
 #include <fcntl.h>
 #include <linux/unistd.h>
 
-#include <linux/bpf.h>
 #include <bpf/bpf.h>
+#include <linux/bpf.h>
 
 #include "bpf_load.h"
 #include "bpf_rlimit.h"
 #include "cgroup_helpers.h"
-#include "hbm.h"
 #include "bpf_util.h"
 #include "bpf/bpf.h"
 #include "bpf/libbpf.h"
 
+struct bpf_spin_lock {
+	__u32	val;
+};
+
+#include "hbm.h"
+
 bool outFlag = true;
 int minRate = 1000;		/* cgroup rate limit in Mbps */
 int rate = 1000;		/* can grow if rate conserving is enabled */
-- 
2.21.0

Re: [PATCH 1/5] samples/bpf: fix test_lru_dist build

[Matteo Croce]

On Sat, May 18, 2019 at 2:46 AM Matteo Croce <mcroce@redhat.com> wrote:
>
> Fix the following error by removing a duplicate struct definition:
>

Hi all,

I forget to send a cover letter for this series, but basically what I
wanted to say is that while patches 1-3 are very straightforward,
patches 4-5 are a bit rough and I accept suggstions to make a cleaner
work.

Regards,
-- 
Matteo Croce
per aspera ad upstream

Re: [PATCH 1/5] samples/bpf: fix test_lru_dist build

[Jakub Kicinski]

On Mon, 20 May 2019 19:46:27 +0200, Matteo Croce wrote:
> On Sat, May 18, 2019 at 2:46 AM Matteo Croce <mcroce@redhat.com> wrote:
> >
> > Fix the following error by removing a duplicate struct definition:
> >  
> 
> Hi all,
> 
> I forget to send a cover letter for this series, but basically what I
> wanted to say is that while patches 1-3 are very straightforward,
> patches 4-5 are a bit rough and I accept suggstions to make a cleaner
> work.

samples depend on headers being locally installed:

make headers_install

Are you intending to change that?

Re: [PATCH 1/5] samples/bpf: fix test_lru_dist build

[Daniel Borkmann]

On 05/20/2019 10:38 PM, Jakub Kicinski wrote:
> On Mon, 20 May 2019 19:46:27 +0200, Matteo Croce wrote:
>> On Sat, May 18, 2019 at 2:46 AM Matteo Croce <mcroce@redhat.com> wrote:
>>>
>>> Fix the following error by removing a duplicate struct definition:
>>
>> Hi all,
>>
>> I forget to send a cover letter for this series, but basically what I
>> wanted to say is that while patches 1-3 are very straightforward,
>> patches 4-5 are a bit rough and I accept suggstions to make a cleaner
>> work.
> 
> samples depend on headers being locally installed:
> 
> make headers_install
> 
> Are you intending to change that?

+1, Matteo, could you elaborate?

On latest bpf tree, everything compiles just fine:

[root@linux bpf]# make headers_install
[root@linux bpf]# make -C samples/bpf/
make: Entering directory '/home/darkstar/trees/bpf/samples/bpf'
make -C ../../ /home/darkstar/trees/bpf/samples/bpf/ BPF_SAMPLES_PATH=/home/darkstar/trees/bpf/samples/bpf
make[1]: Entering directory '/home/darkstar/trees/bpf'
  CALL    scripts/checksyscalls.sh
  CALL    scripts/atomic/check-atomics.sh
  DESCEND  objtool
make -C /home/darkstar/trees/bpf/samples/bpf/../../tools/lib/bpf/ RM='rm -rf' LDFLAGS= srctree=/home/darkstar/trees/bpf/samples/bpf/../../ O=
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/test_lru_dist
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/sock_example
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/fds_example.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/fds_example
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/sockex1_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/sockex1
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/sockex2_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/sockex2
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/bpf_load.o
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/sockex3_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/sockex3
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/tracex1_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/tracex1
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/tracex2_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/tracex2
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/tracex3_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/tracex3
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/tracex4_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/tracex4
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/tracex5_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/tracex5
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/tracex6_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/tracex6
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/tracex7_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/tracex7
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/test_probe_write_user_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/test_probe_write_user
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/trace_output_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/trace_output
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/lathist_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/lathist
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/offwaketime_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/offwaketime
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/spintest_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/spintest
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/map_perf_test_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/map_perf_test
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/test_overhead_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/test_overhead
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/test_cgrp2_array_pin.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/test_cgrp2_array_pin
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/test_cgrp2_attach.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/test_cgrp2_attach
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/test_cgrp2_attach2.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/test_cgrp2_attach2
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/test_cgrp2_sock.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/test_cgrp2_sock
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/test_cgrp2_sock2.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/test_cgrp2_sock2
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/xdp1_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/xdp1
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/xdp2
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/xdp_router_ipv4_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/xdp_router_ipv4
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/test_current_task_under_cgroup_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/test_current_task_under_cgroup
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/trace_event_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/trace_event
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/sampleip_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/sampleip
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/tc_l2_redirect_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/tc_l2_redirect
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/lwt_len_hist_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/lwt_len_hist
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/xdp_tx_iptunnel_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/xdp_tx_iptunnel
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/test_map_in_map_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/test_map_in_map
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/cookie_uid_helper_example.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/per_socket_stats_example
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/xdp_redirect_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/xdp_redirect
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/xdp_redirect_map_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/xdp_redirect_map
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/xdp_redirect_cpu_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/xdp_redirect_cpu
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/xdp_monitor_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/xdp_monitor
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/xdp_rxq_info_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/xdp_rxq_info
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/syscall_tp_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/syscall_tp
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/cpustat_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/cpustat
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/xdp_adjust_tail_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/xdp_adjust_tail
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/xdpsock_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/xdpsock
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/xdp_fwd_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/xdp_fwd
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/task_fd_query_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/task_fd_query
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/xdp_sample_pkts_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/xdp_sample_pkts
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/ibumad_user.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/ibumad
  HOSTCC  /home/darkstar/trees/bpf/samples/bpf/hbm.o
  HOSTLD  /home/darkstar/trees/bpf/samples/bpf/hbm
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/sockex1_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/sockex1_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/sockex2_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/sockex2_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/sockex3_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/sockex3_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/tracex1_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/tracex1_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/tracex2_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/tracex2_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/tracex3_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/tracex3_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/tracex4_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/tracex4_kern.o
  CC      /home/darkstar/trees/bpf/samples/bpf/syscall_nrs.s
  UPD     /home/darkstar/trees/bpf/samples/bpf/syscall_nrs.h
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/tracex5_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/tracex5_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/tracex6_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/tracex6_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/tracex7_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/tracex7_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/sock_flags_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/sock_flags_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/test_probe_write_user_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/test_probe_write_user_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/trace_output_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/trace_output_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/tcbpf1_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/tcbpf1_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/tc_l2_redirect_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/tc_l2_redirect_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/lathist_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/lathist_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/offwaketime_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/offwaketime_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/spintest_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/spintest_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/map_perf_test_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/map_perf_test_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/test_overhead_tp_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/test_overhead_tp_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/test_overhead_raw_tp_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/test_overhead_raw_tp_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/test_overhead_kprobe_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/test_overhead_kprobe_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/parse_varlen.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/parse_varlen.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/parse_simple.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/parse_simple.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/parse_ldabs.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/parse_ldabs.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/test_cgrp2_tc_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/test_cgrp2_tc_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/xdp1_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/xdp1_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/xdp2_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/xdp2_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/xdp_router_ipv4_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/xdp_router_ipv4_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/test_current_task_under_cgroup_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/test_current_task_under_cgroup_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/trace_event_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/trace_event_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/sampleip_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/sampleip_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/lwt_len_hist_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/lwt_len_hist_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/xdp_tx_iptunnel_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/xdp_tx_iptunnel_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/test_map_in_map_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/test_map_in_map_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/tcp_synrto_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/tcp_synrto_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/tcp_rwnd_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/tcp_rwnd_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/tcp_bufs_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/tcp_bufs_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/tcp_cong_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/tcp_cong_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/tcp_iw_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/tcp_iw_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/tcp_clamp_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/tcp_clamp_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/tcp_basertt_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/tcp_basertt_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/tcp_tos_reflect_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/tcp_tos_reflect_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/xdp_redirect_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/xdp_redirect_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/xdp_redirect_map_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/xdp_redirect_map_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/xdp_redirect_cpu_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/xdp_redirect_cpu_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/xdp_monitor_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/xdp_monitor_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/xdp_rxq_info_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/xdp_rxq_info_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/xdp2skb_meta_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/xdp2skb_meta_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/syscall_tp_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/syscall_tp_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/cpustat_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/cpustat_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/xdp_adjust_tail_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/xdp_adjust_tail_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/xdp_fwd_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/xdp_fwd_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/task_fd_query_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/task_fd_query_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/xdp_sample_pkts_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/xdp_sample_pkts_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/ibumad_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/ibumad_kern.o
  CLANG-bpf  /home/darkstar/trees/bpf/samples/bpf/hbm_out_kern.o
pahole -J /home/darkstar/trees/bpf/samples/bpf/hbm_out_kern.o
make[1]: Leaving directory '/home/darkstar/trees/bpf'
make: Leaving directory '/home/darkstar/trees/bpf/samples/bpf'
[root@linux bpf]#

Re: [PATCH 1/5] samples/bpf: fix test_lru_dist build

[Matteo Croce]

On Tue, May 21, 2019 at 5:21 PM Daniel Borkmann <daniel@iogearbox.net> wrote:
>
> On 05/20/2019 10:38 PM, Jakub Kicinski wrote:
> > On Mon, 20 May 2019 19:46:27 +0200, Matteo Croce wrote:
> >> On Sat, May 18, 2019 at 2:46 AM Matteo Croce <mcroce@redhat.com> wrote:
> >>>
> >>> Fix the following error by removing a duplicate struct definition:
> >>
> >> Hi all,
> >>
> >> I forget to send a cover letter for this series, but basically what I
> >> wanted to say is that while patches 1-3 are very straightforward,
> >> patches 4-5 are a bit rough and I accept suggstions to make a cleaner
> >> work.
> >
> > samples depend on headers being locally installed:
> >
> > make headers_install
> >
> > Are you intending to change that?
>
> +1, Matteo, could you elaborate?
>
> On latest bpf tree, everything compiles just fine:
>
> [root@linux bpf]# make headers_install
> [root@linux bpf]# make -C samples/bpf/
> make: Entering directory '/home/darkstar/trees/bpf/samples/bpf'
> make -C ../../ /home/darkstar/trees/bpf/samples/bpf/ BPF_SAMPLES_PATH=/home/darkstar/trees/bpf/samples/bpf
> make[1]: Entering directory '/home/darkstar/trees/bpf'
>   CALL    scripts/checksyscalls.sh
>   CALL    scripts/atomic/check-atomics.sh
>   DESCEND  objtool
> make -C /home/darkstar/trees/bpf/samples/bpf/../../tools/lib/bpf/ RM='rm -rf' LDFLAGS= srctree=/home/darkstar/trees/bpf/samples/bpf/../../ O=
>   HOSTCC  /home/darkstar/trees/bpf/samples/bpf/test_lru_dist
>   HOSTCC  /home/darkstar/trees/bpf/samples/bpf/sock_example
>

Hi all,

I have kernel-headers installed from master, but yet the samples fail to build:

matteo@turbo:~/src/linux/samples/bpf$ rpm -q kernel-headers
kernel-headers-5.2.0_rc1-38.x86_64

matteo@turbo:~/src/linux/samples/bpf$ git describe HEAD
v5.2-rc1-97-g5bdd9ad875b6

matteo@turbo:~/src/linux/samples/bpf$ make
make -C ../../ /home/matteo/src/linux/samples/bpf/
BPF_SAMPLES_PATH=/home/matteo/src/linux/samples/bpf
make[1]: Entering directory '/home/matteo/src/linux'
  CALL    scripts/checksyscalls.sh
  CALL    scripts/atomic/check-atomics.sh
  DESCEND  objtool
make -C /home/matteo/src/linux/samples/bpf/../../tools/lib/bpf/ RM='rm
-rf' LDFLAGS= srctree=/home/matteo/src/linux/samples/bpf/../../ O=
  HOSTCC  /home/matteo/src/linux/samples/bpf/test_lru_dist
/home/matteo/src/linux/samples/bpf/test_lru_dist.c:39:8: error:
redefinition of ‘struct list_head’
   39 | struct list_head {
      |        ^~~~~~~~~
In file included from /home/matteo/src/linux/samples/bpf/test_lru_dist.c:9:
./tools/include/linux/types.h:69:8: note: originally defined here
   69 | struct list_head {
      |        ^~~~~~~~~
make[2]: *** [scripts/Makefile.host:90:
/home/matteo/src/linux/samples/bpf/test_lru_dist] Error 1
make[1]: *** [Makefile:1762: /home/matteo/src/linux/samples/bpf/] Error 2
make[1]: Leaving directory '/home/matteo/src/linux'
make: *** [Makefile:231: all] Error 2

Am I missing something obvious?


Regards,


--
Matteo Croce
per aspera ad upstream

Re: [PATCH 1/5] samples/bpf: fix test_lru_dist build

[Jakub Kicinski]

On Tue, 21 May 2019 17:36:17 +0200, Matteo Croce wrote:
> On Tue, May 21, 2019 at 5:21 PM Daniel Borkmann <daniel@iogearbox.net> wrote:
> >
> > On 05/20/2019 10:38 PM, Jakub Kicinski wrote:  
> > > On Mon, 20 May 2019 19:46:27 +0200, Matteo Croce wrote:  
> > >> On Sat, May 18, 2019 at 2:46 AM Matteo Croce <mcroce@redhat.com> wrote:  
> > >>>
> > >>> Fix the following error by removing a duplicate struct definition:  
> > >>
> > >> Hi all,
> > >>
> > >> I forget to send a cover letter for this series, but basically what I
> > >> wanted to say is that while patches 1-3 are very straightforward,
> > >> patches 4-5 are a bit rough and I accept suggstions to make a cleaner
> > >> work.  
> > >
> > > samples depend on headers being locally installed:
> > >
> > > make headers_install
> > >
> > > Are you intending to change that?  
> >
> > +1, Matteo, could you elaborate?
> >
> > On latest bpf tree, everything compiles just fine:
> >
> > [root@linux bpf]# make headers_install
> > [root@linux bpf]# make -C samples/bpf/
> > make: Entering directory '/home/darkstar/trees/bpf/samples/bpf'
> > make -C ../../ /home/darkstar/trees/bpf/samples/bpf/ BPF_SAMPLES_PATH=/home/darkstar/trees/bpf/samples/bpf
> > make[1]: Entering directory '/home/darkstar/trees/bpf'
> >   CALL    scripts/checksyscalls.sh
> >   CALL    scripts/atomic/check-atomics.sh
> >   DESCEND  objtool
> > make -C /home/darkstar/trees/bpf/samples/bpf/../../tools/lib/bpf/ RM='rm -rf' LDFLAGS= srctree=/home/darkstar/trees/bpf/samples/bpf/../../ O=
> >   HOSTCC  /home/darkstar/trees/bpf/samples/bpf/test_lru_dist
> >   HOSTCC  /home/darkstar/trees/bpf/samples/bpf/sock_example
> >  
> 
> Hi all,
> 
> I have kernel-headers installed from master, but yet the samples fail to build:
> 
> matteo@turbo:~/src/linux/samples/bpf$ rpm -q kernel-headers
> kernel-headers-5.2.0_rc1-38.x86_64
> 
> matteo@turbo:~/src/linux/samples/bpf$ git describe HEAD
> v5.2-rc1-97-g5bdd9ad875b6
> 
> matteo@turbo:~/src/linux/samples/bpf$ make
> make -C ../../ /home/matteo/src/linux/samples/bpf/
> BPF_SAMPLES_PATH=/home/matteo/src/linux/samples/bpf
> make[1]: Entering directory '/home/matteo/src/linux'
>   CALL    scripts/checksyscalls.sh
>   CALL    scripts/atomic/check-atomics.sh
>   DESCEND  objtool
> make -C /home/matteo/src/linux/samples/bpf/../../tools/lib/bpf/ RM='rm
> -rf' LDFLAGS= srctree=/home/matteo/src/linux/samples/bpf/../../ O=
>   HOSTCC  /home/matteo/src/linux/samples/bpf/test_lru_dist
> /home/matteo/src/linux/samples/bpf/test_lru_dist.c:39:8: error:
> redefinition of ‘struct list_head’
>    39 | struct list_head {
>       |        ^~~~~~~~~
> In file included from /home/matteo/src/linux/samples/bpf/test_lru_dist.c:9:
> ./tools/include/linux/types.h:69:8: note: originally defined here
>    69 | struct list_head {
>       |        ^~~~~~~~~
> make[2]: *** [scripts/Makefile.host:90:
> /home/matteo/src/linux/samples/bpf/test_lru_dist] Error 1
> make[1]: *** [Makefile:1762: /home/matteo/src/linux/samples/bpf/] Error 2
> make[1]: Leaving directory '/home/matteo/src/linux'
> make: *** [Makefile:231: all] Error 2
> 
> Am I missing something obvious?

Yes ;)  Samples use a local installation of headers in $objtree/usr (I
think, maybe $srctree/usr).  So you need to do make headers_install in
your kernel source tree, otherwise the include path from tools/ takes
priority over your global /usr/include and causes these issues.  I had
this path in my tree for some time, but I don't like enough to post it:

commit 35fb614049e93d46af708c0eaae6601df54017b3
Author: Jakub Kicinski <jakub.kicinski@netronome.com>
Date:   Mon Dec 3 15:00:24 2018 -0800

    bpf: maybe warn ppl about hrds_install
    
    Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>

diff --git a/samples/bpf/Makefile b/samples/bpf/Makefile
index 4f0a1cdbfe7c..f79a4ed2f9f7 100644
--- a/samples/bpf/Makefile
+++ b/samples/bpf/Makefile
@@ -208,6 +208,15 @@ HOSTCC = $(CROSS_COMPILE)gcc
 CLANG_ARCH_ARGS = -target $(ARCH)
 endif
 
+HDR_PROBE := $(shell echo "\#include <linux/types.h>\n struct list_head { int a; }; int main() { return 0; }" | \
+       gcc $(KBUILD_HOSTCFLAGS) -x c - -o /dev/null 2>/dev/null && \
+       echo okay)
+
+ifeq ($(HDR_PROBE),)
+$(warning Detected possible issues with include path.)
+$(warning Please install kernel headers locally (make headers_install))
+endif
+
 BTF_LLC_PROBE := $(shell $(LLC) -march=bpf -mattr=help 2>&1 | grep dwarfris)
 BTF_PAHOLE_PROBE := $(shell $(BTF_PAHOLE) --help 2>&1 | grep BTF)
 BTF_OBJCOPY_PROBE := $(shell $(LLVM_OBJCOPY) --help 2>&1 | grep -i 'usage.*llvm')

Re: [PATCH 1/5] samples/bpf: fix test_lru_dist build

[Matteo Croce]

On Tue, May 21, 2019 at 7:07 PM Jakub Kicinski
<jakub.kicinski@netronome.com> wrote:
>
> On Tue, 21 May 2019 17:36:17 +0200, Matteo Croce wrote:
> > On Tue, May 21, 2019 at 5:21 PM Daniel Borkmann <daniel@iogearbox.net> wrote:
> > >
> > > On 05/20/2019 10:38 PM, Jakub Kicinski wrote:
> > > > On Mon, 20 May 2019 19:46:27 +0200, Matteo Croce wrote:
> > > >> On Sat, May 18, 2019 at 2:46 AM Matteo Croce <mcroce@redhat.com> wrote:
> > > >>>
> > > >>> Fix the following error by removing a duplicate struct definition:
> > > >>
> > > >> Hi all,
> > > >>
> > > >> I forget to send a cover letter for this series, but basically what I
> > > >> wanted to say is that while patches 1-3 are very straightforward,
> > > >> patches 4-5 are a bit rough and I accept suggstions to make a cleaner
> > > >> work.
> > > >
> > > > samples depend on headers being locally installed:
> > > >
> > > > make headers_install
> > > >
> > > > Are you intending to change that?
> > >
> > > +1, Matteo, could you elaborate?
> > >
> > > On latest bpf tree, everything compiles just fine:
> > >
> > > [root@linux bpf]# make headers_install
> > > [root@linux bpf]# make -C samples/bpf/
> > > make: Entering directory '/home/darkstar/trees/bpf/samples/bpf'
> > > make -C ../../ /home/darkstar/trees/bpf/samples/bpf/ BPF_SAMPLES_PATH=/home/darkstar/trees/bpf/samples/bpf
> > > make[1]: Entering directory '/home/darkstar/trees/bpf'
> > >   CALL    scripts/checksyscalls.sh
> > >   CALL    scripts/atomic/check-atomics.sh
> > >   DESCEND  objtool
> > > make -C /home/darkstar/trees/bpf/samples/bpf/../../tools/lib/bpf/ RM='rm -rf' LDFLAGS= srctree=/home/darkstar/trees/bpf/samples/bpf/../../ O=
> > >   HOSTCC  /home/darkstar/trees/bpf/samples/bpf/test_lru_dist
> > >   HOSTCC  /home/darkstar/trees/bpf/samples/bpf/sock_example
> > >
> >
> > Hi all,
> >
> > I have kernel-headers installed from master, but yet the samples fail to build:
> >
> > matteo@turbo:~/src/linux/samples/bpf$ rpm -q kernel-headers
> > kernel-headers-5.2.0_rc1-38.x86_64
> >
> > matteo@turbo:~/src/linux/samples/bpf$ git describe HEAD
> > v5.2-rc1-97-g5bdd9ad875b6
> >
> > matteo@turbo:~/src/linux/samples/bpf$ make
> > make -C ../../ /home/matteo/src/linux/samples/bpf/
> > BPF_SAMPLES_PATH=/home/matteo/src/linux/samples/bpf
> > make[1]: Entering directory '/home/matteo/src/linux'
> >   CALL    scripts/checksyscalls.sh
> >   CALL    scripts/atomic/check-atomics.sh
> >   DESCEND  objtool
> > make -C /home/matteo/src/linux/samples/bpf/../../tools/lib/bpf/ RM='rm
> > -rf' LDFLAGS= srctree=/home/matteo/src/linux/samples/bpf/../../ O=
> >   HOSTCC  /home/matteo/src/linux/samples/bpf/test_lru_dist
> > /home/matteo/src/linux/samples/bpf/test_lru_dist.c:39:8: error:
> > redefinition of ‘struct list_head’
> >    39 | struct list_head {
> >       |        ^~~~~~~~~
> > In file included from /home/matteo/src/linux/samples/bpf/test_lru_dist.c:9:
> > ./tools/include/linux/types.h:69:8: note: originally defined here
> >    69 | struct list_head {
> >       |        ^~~~~~~~~
> > make[2]: *** [scripts/Makefile.host:90:
> > /home/matteo/src/linux/samples/bpf/test_lru_dist] Error 1
> > make[1]: *** [Makefile:1762: /home/matteo/src/linux/samples/bpf/] Error 2
> > make[1]: Leaving directory '/home/matteo/src/linux'
> > make: *** [Makefile:231: all] Error 2
> >
> > Am I missing something obvious?
>
> Yes ;)  Samples use a local installation of headers in $objtree/usr (I
> think, maybe $srctree/usr).  So you need to do make headers_install in
> your kernel source tree, otherwise the include path from tools/ takes
> priority over your global /usr/include and causes these issues.  I had
> this path in my tree for some time, but I don't like enough to post it:
>
> commit 35fb614049e93d46af708c0eaae6601df54017b3
> Author: Jakub Kicinski <jakub.kicinski@netronome.com>
> Date:   Mon Dec 3 15:00:24 2018 -0800
>
>     bpf: maybe warn ppl about hrds_install
>
>     Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
>
> diff --git a/samples/bpf/Makefile b/samples/bpf/Makefile
> index 4f0a1cdbfe7c..f79a4ed2f9f7 100644
> --- a/samples/bpf/Makefile
> +++ b/samples/bpf/Makefile
> @@ -208,6 +208,15 @@ HOSTCC = $(CROSS_COMPILE)gcc
>  CLANG_ARCH_ARGS = -target $(ARCH)
>  endif
>
> +HDR_PROBE := $(shell echo "\#include <linux/types.h>\n struct list_head { int a; }; int main() { return 0; }" | \
> +       gcc $(KBUILD_HOSTCFLAGS) -x c - -o /dev/null 2>/dev/null && \
> +       echo okay)
> +
> +ifeq ($(HDR_PROBE),)
> +$(warning Detected possible issues with include path.)
> +$(warning Please install kernel headers locally (make headers_install))
> +endif
> +
>  BTF_LLC_PROBE := $(shell $(LLC) -march=bpf -mattr=help 2>&1 | grep dwarfris)
>  BTF_PAHOLE_PROBE := $(shell $(BTF_PAHOLE) --help 2>&1 | grep BTF)
>  BTF_OBJCOPY_PROBE := $(shell $(LLVM_OBJCOPY) --help 2>&1 | grep -i 'usage.*llvm')

Hi Jakub,

I see now, It worked, thanks. This is a bit error prone IMHO, if you
ever think about sending this patch, consider it ACKed by me.

Thanks,
-- 
Matteo Croce
per aspera ad upstream