Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Wed, 19 Oct 2022 09:34:16 -0400
From:   Steven Rostedt <rostedt@goodmis.org>
To:     kernel test robot <lkp@intel.com>
Cc:     kbuild-all@lists.01.org, linux-kernel@vger.kernel.org,
        Tom Zanussi <zanussi@kernel.org>
Subject: Re: kernel/trace/trace_events_synth.c:436:68: sparse: sparse:
 incorrect type in argument 2 (different address spaces)
Message-ID: <20221019093416.1155d790@gandalf.local.home>
In-Reply-To: <202210191111.XTgMRb3E-lkp@intel.com>
References: <202210191111.XTgMRb3E-lkp@intel.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Precedence: bulk

On Wed, 19 Oct 2022 11:43:54 +0800
kernel test robot <lkp@intel.com> wrote:

> sparse warnings: (new ones prefixed by >>)
> >> kernel/trace/trace_events_synth.c:436:68: sparse: sparse: incorrect type in argument 2 (different address spaces) @@     expected void const [noderef] __user *unsafe_addr @@     got char *str_val @@  
>    kernel/trace/trace_events_synth.c:436:68: sparse:     expected void const [noderef] __user *unsafe_addr
>    kernel/trace/trace_events_synth.c:436:68: sparse:     got char *str_val
> 
> vim +436 kernel/trace/trace_events_synth.c
> 
>    404	
>    405	static unsigned int trace_string(struct synth_trace_event *entry,
>    406					 struct synth_event *event,
>    407					 char *str_val,
>    408					 bool is_dynamic,
>    409					 unsigned int data_size,
>    410					 unsigned int *n_u64)
>    411	{
>    412		unsigned int len = 0;
>    413		char *str_field;
>    414		int ret;
>    415	
>    416		if (is_dynamic) {
>    417			u32 data_offset;
>    418	
>    419			data_offset = offsetof(typeof(*entry), fields);
>    420			data_offset += event->n_u64 * sizeof(u64);
>    421			data_offset += data_size;
>    422	
>    423			len = kern_fetch_store_strlen((unsigned long)str_val);
>    424	
>    425			data_offset |= len << 16;
>    426			*(u32 *)&entry->fields[*n_u64] = data_offset;
>    427	
>    428			ret = kern_fetch_store_string((unsigned long)str_val, &entry->fields[*n_u64], entry);
>    429	
>    430			(*n_u64)++;
>    431		} else {
>    432			str_field = (char *)&entry->fields[*n_u64];
>    433	
>    434	#ifdef CONFIG_ARCH_HAS_NON_OVERLAPPING_ADDRESS_SPACE
>    435			if ((unsigned long)str_val < TASK_SIZE)
>  > 436				ret = strncpy_from_user_nofault(str_field, str_val, STR_VAR_LEN_MAX);  
>    437			else
>    438	#endif
>    439				ret = strncpy_from_kernel_nofault(str_field, str_val, STR_VAR_LEN_MAX);
>    440	


Does this fix it?

-- Steve

diff --git a/kernel/trace/trace_events_synth.c b/kernel/trace/trace_events_synth.c
index e310052dc83c..2562d7522999 100644
--- a/kernel/trace/trace_events_synth.c
+++ b/kernel/trace/trace_events_synth.c
@@ -433,7 +433,8 @@ static unsigned int trace_string(struct synth_trace_event *entry,
 
 #ifdef CONFIG_ARCH_HAS_NON_OVERLAPPING_ADDRESS_SPACE
 		if ((unsigned long)str_val < TASK_SIZE)
-			ret = strncpy_from_user_nofault(str_field, str_val, STR_VAR_LEN_MAX);
+			ret = strncpy_from_user_nofault(str_field,
+				(__force const void __user *)str_val, STR_VAR_LEN_MAX);
 		else
 #endif
 			ret = strncpy_from_kernel_nofault(str_field, str_val, STR_VAR_LEN_MAX);