Re: [PATCH v5 1/2] selftests/resctrl: Fix schemata write error check

[Maciej Wieczór-Retman]

Hello!

On 2023-09-29 at 10:04:21 -0700, Reinette Chatre wrote:
>Hi Maciej,
>
>On 9/29/2023 1:21 AM, Maciej Wieczor-Retman wrote:
>...
>
>> diff --git a/tools/testing/selftests/resctrl/resctrlfs.c b/tools/testing/selftests/resctrl/resctrlfs.c
>> index 3a8111362d26..342a3dbcdbb6 100644
>> --- a/tools/testing/selftests/resctrl/resctrlfs.c
>> +++ b/tools/testing/selftests/resctrl/resctrlfs.c
>> @@ -8,6 +8,7 @@
>>   *    Sai Praneeth Prakhya <sai.praneeth.prakhya@xxxxxxxxx>,
>>   *    Fenghua Yu <fenghua.yu@xxxxxxxxx>
>>   */
>> +#include <fcntl.h>
>>  #include <limits.h>
>>  
>>  #include "resctrl.h"
>> @@ -490,9 +491,8 @@ int write_bm_pid_to_resctrl(pid_t bm_pid, char *ctrlgrp, char *mongrp,
>>   */
>>  int write_schemata(char *ctrlgrp, char *schemata, int cpu_no, char *resctrl_val)
>>  {
>> -	char controlgroup[1024], schema[1024], reason[64];
>> -	int resource_id, ret = 0;
>> -	FILE *fp;
>> +	char controlgroup[1024], schema[1024], reason[128];
>> +	int resource_id, fd, schema_len = 0, ret = 0;
>>  
>>  	if (strncmp(resctrl_val, MBA_STR, sizeof(MBA_STR)) &&
>>  	    strncmp(resctrl_val, MBM_STR, sizeof(MBM_STR)) &&
>> @@ -520,27 +520,37 @@ int write_schemata(char *ctrlgrp, char *schemata, int cpu_no, char *resctrl_val)
>>  
>>  	if (!strncmp(resctrl_val, CAT_STR, sizeof(CAT_STR)) ||
>>  	    !strncmp(resctrl_val, CMT_STR, sizeof(CMT_STR)))
>> -		sprintf(schema, "%s%d%c%s", "L3:", resource_id, '=', schemata);
>> +		schema_len = snprintf(schema, sizeof(schema), "%s%d%c%s\n",
>> +				      "L3:", resource_id, '=', schemata);
>>  	if (!strncmp(resctrl_val, MBA_STR, sizeof(MBA_STR)) ||
>>  	    !strncmp(resctrl_val, MBM_STR, sizeof(MBM_STR)))
>> -		sprintf(schema, "%s%d%c%s", "MB:", resource_id, '=', schemata);
>> +		schema_len = snprintf(schema, sizeof(schema), "%s%d%c%s\n",
>> +				      "MB:", resource_id, '=', schemata);
>> +	if (schema_len < 1) {
>
>I find that this complicates the code since this is not the typical snprintf()
>error checking (0 is a valid snprintf() return). I think it will make the code
>easier to understand if it sticks to snprintf() error checking and initialize
>schema_len to match. What I mean with this is something like this:
>
>	int schema_len = -1;
>
>	if (...)
>		schema_len = snprintf(...);
>
>	if (schema_len < 0 || schema_len >= sizeof(schema))
>		/* error handling */

Okay, I'll change it to the more typical error checking then. I suppose
it's not possible to write zero characters to "schema" so my " < 1"
won't be useful.

>> +		snprintf(reason, sizeof(reason),
>> +			 "snprintf() failed with return value : %d", schema_len);
>> +		ret = -1;
>> +		goto out;
>> +	}
>>  
>> -	fp = fopen(controlgroup, "w");
>> -	if (!fp) {
>> -		sprintf(reason, "Failed to open control group");
>> +	fd = open(controlgroup, O_WRONLY);
>> +	if (fd < 0) {
>> +		snprintf(reason, sizeof(reason),
>> +			 "open() failed : %s", strerror(errno));
>>  		ret = -1;
>>  
>>  		goto out;
>>  	}
>> -
>> -	if (fprintf(fp, "%s\n", schema) < 0) {
>> -		sprintf(reason, "Failed to write schemata in control group");
>> -		fclose(fp);
>> +	if (write(fd, schema, schema_len) < 0) {
>> +		snprintf(reason, sizeof(reason),
>> +			 "write() failed : %s", strerror(errno));
>> +		close(fd);
>>  		ret = -1;
>>  
>>  		goto out;
>>  	}
>> -	fclose(fp);
>> +	close(fd);
>> +	schema[schema_len - 1] = 0;
>>  
>>  out:
>>  	ksft_print_msg("Write schema \"%s\" to resctrl FS%s%s\n",
>
>Also please note that _if_ there is an early exit then uninitialized schema
>will be printed. Maybe this also needs a schema[1024] = {} ?
>
>Reinette

Thanks for pointing it out, I'll fix it for the next version.

-- 
Kind regards
Maciej Wieczór-Retman