Netdev Archive on lore.kernel.org
help / color / mirror / Atom feed
* [PATCH v2 bpf-next] bpf: Fix build without BPF_SYSCALL, but with BPF_JIT.
@ 2020-08-31 15:51 Alexei Starovoitov
  2020-08-31 16:15 ` Paul E. McKenney
  2020-08-31 20:07 ` Daniel Borkmann
  0 siblings, 2 replies; 4+ messages in thread
From: Alexei Starovoitov @ 2020-08-31 15:51 UTC (permalink / raw)
  To: davem; +Cc: daniel, paulmck, netdev, bpf, kernel-team

From: Alexei Starovoitov <ast@kernel.org>

When CONFIG_BPF_SYSCALL is not set, but CONFIG_BPF_JIT=y
the kernel build fails:
In file included from ../kernel/bpf/trampoline.c:11:
../kernel/bpf/trampoline.c: In function ‘bpf_trampoline_update’:
../kernel/bpf/trampoline.c:220:39: error: ‘call_rcu_tasks_trace’ undeclared
../kernel/bpf/trampoline.c: In function ‘__bpf_prog_enter_sleepable’:
../kernel/bpf/trampoline.c:411:2: error: implicit declaration of function ‘rcu_read_lock_trace’
../kernel/bpf/trampoline.c: In function ‘__bpf_prog_exit_sleepable’:
../kernel/bpf/trampoline.c:416:2: error: implicit declaration of function ‘rcu_read_unlock_trace’

This is due to:
obj-$(CONFIG_BPF_JIT) += trampoline.o
obj-$(CONFIG_BPF_JIT) += dispatcher.o
There is a number of functions that arch/x86/net/bpf_jit_comp.c is
using from these two files, but none of them will be used when
only cBPF is on (which is the case for BPF_SYSCALL=n BPF_JIT=y).

Add rcu_trace functions to rcupdate_trace.h. The JITed code won't execute them
and BPF trampoline logic won't be used without BPF_SYSCALL.

Reported-by: kernel test robot <lkp@intel.com>
Fixes: 1e6c62a88215 ("bpf: Introduce sleepable BPF programs")
Acked-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/rcupdate_trace.h | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/include/linux/rcupdate_trace.h b/include/linux/rcupdate_trace.h
index d9015aac78c6..aaaac8ac927c 100644
--- a/include/linux/rcupdate_trace.h
+++ b/include/linux/rcupdate_trace.h
@@ -82,7 +82,14 @@ static inline void rcu_read_unlock_trace(void)
 void call_rcu_tasks_trace(struct rcu_head *rhp, rcu_callback_t func);
 void synchronize_rcu_tasks_trace(void);
 void rcu_barrier_tasks_trace(void);
-
+#else
+/*
+ * The BPF JIT forms these addresses even when it doesn't call these
+ * functions, so provide definitions that result in runtime errors.
+ */
+static inline void call_rcu_tasks_trace(struct rcu_head *rhp, rcu_callback_t func) { BUG(); }
+static inline void rcu_read_lock_trace(void) { BUG(); }
+static inline void rcu_read_unlock_trace(void) { BUG(); }
 #endif /* #ifdef CONFIG_TASKS_TRACE_RCU */
 
 #endif /* __LINUX_RCUPDATE_TRACE_H */
-- 
2.23.0


^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH v2 bpf-next] bpf: Fix build without BPF_SYSCALL, but with BPF_JIT.
  2020-08-31 15:51 [PATCH v2 bpf-next] bpf: Fix build without BPF_SYSCALL, but with BPF_JIT Alexei Starovoitov
@ 2020-08-31 16:15 ` Paul E. McKenney
  2020-08-31 16:19   ` Alexei Starovoitov
  2020-08-31 20:07 ` Daniel Borkmann
  1 sibling, 1 reply; 4+ messages in thread
From: Paul E. McKenney @ 2020-08-31 16:15 UTC (permalink / raw)
  To: Alexei Starovoitov; +Cc: davem, daniel, netdev, bpf, kernel-team

On Mon, Aug 31, 2020 at 08:51:55AM -0700, Alexei Starovoitov wrote:
> From: Alexei Starovoitov <ast@kernel.org>
> 
> When CONFIG_BPF_SYSCALL is not set, but CONFIG_BPF_JIT=y
> the kernel build fails:
> In file included from ../kernel/bpf/trampoline.c:11:
> ../kernel/bpf/trampoline.c: In function ‘bpf_trampoline_update’:
> ../kernel/bpf/trampoline.c:220:39: error: ‘call_rcu_tasks_trace’ undeclared
> ../kernel/bpf/trampoline.c: In function ‘__bpf_prog_enter_sleepable’:
> ../kernel/bpf/trampoline.c:411:2: error: implicit declaration of function ‘rcu_read_lock_trace’
> ../kernel/bpf/trampoline.c: In function ‘__bpf_prog_exit_sleepable’:
> ../kernel/bpf/trampoline.c:416:2: error: implicit declaration of function ‘rcu_read_unlock_trace’
> 
> This is due to:
> obj-$(CONFIG_BPF_JIT) += trampoline.o
> obj-$(CONFIG_BPF_JIT) += dispatcher.o
> There is a number of functions that arch/x86/net/bpf_jit_comp.c is
> using from these two files, but none of them will be used when
> only cBPF is on (which is the case for BPF_SYSCALL=n BPF_JIT=y).
> 
> Add rcu_trace functions to rcupdate_trace.h. The JITed code won't execute them
> and BPF trampoline logic won't be used without BPF_SYSCALL.
> 
> Reported-by: kernel test robot <lkp@intel.com>
> Fixes: 1e6c62a88215 ("bpf: Introduce sleepable BPF programs")
> Acked-by: Paul E. McKenney <paulmck@kernel.org>
> Signed-off-by: Alexei Starovoitov <ast@kernel.org>

Looks good, and unless someone tells me otherwise, I am assuming that
this one goes up the normal BPF patch route.

							Thanx, Paul

> ---
>  include/linux/rcupdate_trace.h | 9 ++++++++-
>  1 file changed, 8 insertions(+), 1 deletion(-)
> 
> diff --git a/include/linux/rcupdate_trace.h b/include/linux/rcupdate_trace.h
> index d9015aac78c6..aaaac8ac927c 100644
> --- a/include/linux/rcupdate_trace.h
> +++ b/include/linux/rcupdate_trace.h
> @@ -82,7 +82,14 @@ static inline void rcu_read_unlock_trace(void)
>  void call_rcu_tasks_trace(struct rcu_head *rhp, rcu_callback_t func);
>  void synchronize_rcu_tasks_trace(void);
>  void rcu_barrier_tasks_trace(void);
> -
> +#else
> +/*
> + * The BPF JIT forms these addresses even when it doesn't call these
> + * functions, so provide definitions that result in runtime errors.
> + */
> +static inline void call_rcu_tasks_trace(struct rcu_head *rhp, rcu_callback_t func) { BUG(); }
> +static inline void rcu_read_lock_trace(void) { BUG(); }
> +static inline void rcu_read_unlock_trace(void) { BUG(); }
>  #endif /* #ifdef CONFIG_TASKS_TRACE_RCU */
>  
>  #endif /* __LINUX_RCUPDATE_TRACE_H */
> -- 
> 2.23.0
> 

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH v2 bpf-next] bpf: Fix build without BPF_SYSCALL, but with BPF_JIT.
  2020-08-31 16:15 ` Paul E. McKenney
@ 2020-08-31 16:19   ` Alexei Starovoitov
  0 siblings, 0 replies; 4+ messages in thread
From: Alexei Starovoitov @ 2020-08-31 16:19 UTC (permalink / raw)
  To: paulmck, Alexei Starovoitov; +Cc: davem, daniel, netdev, bpf, kernel-team

On 8/31/20 9:15 AM, Paul E. McKenney wrote:
> On Mon, Aug 31, 2020 at 08:51:55AM -0700, Alexei Starovoitov wrote:
>> From: Alexei Starovoitov <ast@kernel.org>
>>
>> When CONFIG_BPF_SYSCALL is not set, but CONFIG_BPF_JIT=y
>> the kernel build fails:
>> In file included from ../kernel/bpf/trampoline.c:11:
>> ../kernel/bpf/trampoline.c: In function ‘bpf_trampoline_update’:
>> ../kernel/bpf/trampoline.c:220:39: error: ‘call_rcu_tasks_trace’ undeclared
>> ../kernel/bpf/trampoline.c: In function ‘__bpf_prog_enter_sleepable’:
>> ../kernel/bpf/trampoline.c:411:2: error: implicit declaration of function ‘rcu_read_lock_trace’
>> ../kernel/bpf/trampoline.c: In function ‘__bpf_prog_exit_sleepable’:
>> ../kernel/bpf/trampoline.c:416:2: error: implicit declaration of function ‘rcu_read_unlock_trace’
>>
>> This is due to:
>> obj-$(CONFIG_BPF_JIT) += trampoline.o
>> obj-$(CONFIG_BPF_JIT) += dispatcher.o
>> There is a number of functions that arch/x86/net/bpf_jit_comp.c is
>> using from these two files, but none of them will be used when
>> only cBPF is on (which is the case for BPF_SYSCALL=n BPF_JIT=y).
>>
>> Add rcu_trace functions to rcupdate_trace.h. The JITed code won't execute them
>> and BPF trampoline logic won't be used without BPF_SYSCALL.
>>
>> Reported-by: kernel test robot <lkp@intel.com>
>> Fixes: 1e6c62a88215 ("bpf: Introduce sleepable BPF programs")
>> Acked-by: Paul E. McKenney <paulmck@kernel.org>
>> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
> 
> Looks good, and unless someone tells me otherwise, I am assuming that
> this one goes up the normal BPF patch route.

Yes. It fixes the issue in bpf-next tree.

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH v2 bpf-next] bpf: Fix build without BPF_SYSCALL, but with BPF_JIT.
  2020-08-31 15:51 [PATCH v2 bpf-next] bpf: Fix build without BPF_SYSCALL, but with BPF_JIT Alexei Starovoitov
  2020-08-31 16:15 ` Paul E. McKenney
@ 2020-08-31 20:07 ` Daniel Borkmann
  1 sibling, 0 replies; 4+ messages in thread
From: Daniel Borkmann @ 2020-08-31 20:07 UTC (permalink / raw)
  To: Alexei Starovoitov, davem; +Cc: paulmck, netdev, bpf, kernel-team

On 8/31/20 5:51 PM, Alexei Starovoitov wrote:
> From: Alexei Starovoitov <ast@kernel.org>
> 
> When CONFIG_BPF_SYSCALL is not set, but CONFIG_BPF_JIT=y
> the kernel build fails:
> In file included from ../kernel/bpf/trampoline.c:11:
> ../kernel/bpf/trampoline.c: In function ‘bpf_trampoline_update’:
> ../kernel/bpf/trampoline.c:220:39: error: ‘call_rcu_tasks_trace’ undeclared
> ../kernel/bpf/trampoline.c: In function ‘__bpf_prog_enter_sleepable’:
> ../kernel/bpf/trampoline.c:411:2: error: implicit declaration of function ‘rcu_read_lock_trace’
> ../kernel/bpf/trampoline.c: In function ‘__bpf_prog_exit_sleepable’:
> ../kernel/bpf/trampoline.c:416:2: error: implicit declaration of function ‘rcu_read_unlock_trace’
> 
> This is due to:
> obj-$(CONFIG_BPF_JIT) += trampoline.o
> obj-$(CONFIG_BPF_JIT) += dispatcher.o
> There is a number of functions that arch/x86/net/bpf_jit_comp.c is
> using from these two files, but none of them will be used when
> only cBPF is on (which is the case for BPF_SYSCALL=n BPF_JIT=y).
> 
> Add rcu_trace functions to rcupdate_trace.h. The JITed code won't execute them
> and BPF trampoline logic won't be used without BPF_SYSCALL.
> 
> Reported-by: kernel test robot <lkp@intel.com>
> Fixes: 1e6c62a88215 ("bpf: Introduce sleepable BPF programs")
> Acked-by: Paul E. McKenney <paulmck@kernel.org>
> Signed-off-by: Alexei Starovoitov <ast@kernel.org>

Applied, thanks!

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2020-08-31 20:07 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-08-31 15:51 [PATCH v2 bpf-next] bpf: Fix build without BPF_SYSCALL, but with BPF_JIT Alexei Starovoitov
2020-08-31 16:15 ` Paul E. McKenney
2020-08-31 16:19   ` Alexei Starovoitov
2020-08-31 20:07 ` Daniel Borkmann

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).