Hi together,

I am using arch linux and libcurl 7.53.1-1.
I ran valgrind to check for memory leaks. And to my surprise, I got
thousands of errors, here only the last context:

==18435== 3974 errors in context 1000 of 1000:
==18435== Conditional jump or move depends on uninitialised value(s)
==18435== at 0x4C31CA6: rawmemchr (in
/usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==18435== by 0x5895421: _IO_str_init_static_internal (in
/usr/lib/libc-2.24.so)
==18435== by 0x58845E6: __isoc99_vsscanf (in /usr/lib/libc-2.24.so)
==18435== by 0x5884586: __isoc99_sscanf (in /usr/lib/libc-2.24.so)
==18435== by 0x55BC4F4: ??? (in /usr/lib/libcurl.so.4.4.0)
==18435== by 0x55D4F07: ??? (in /usr/lib/libcurl.so.4.4.0)
==18435== by 0x55E015D: ??? (in /usr/lib/libcurl.so.4.4.0)
==18435== by 0x55E0C50: curl_multi_perform (in /usr/lib/libcurl.so.4.4.0)
==18435== by 0x55D6ABF: curl_easy_perform (in /usr/lib/libcurl.so.4.4.0)
==18435== by 0x404268: main (dcs_data_transfer.c:744)
==18435==
==18435== ERROR SUMMARY: 35049 errors from 1000 contexts (suppressed: 0
from 0)

Here is the code that valgrind complains about:

curl_global_init(CURL_GLOBAL_DEFAULT);
curl = curl_easy_init();
if ( curl ) {
     struct curl_slist *headers=NULL;
     headers = curl_slist_append(headers, "Content-Type: text/xml");
     /* pass our list of custom made headers */
     curl_easy_setopt(curl, CURLOPT_HTTPHEADER, headers);
     /* send all data to this function */
     curl_easy_setopt(curl, CURLOPT_WRITEFUNCTION, WriteMemoryCallback);
     /* we pass our 'chunk' struct to the callback function */
     curl_easy_setopt(curl, CURLOPT_WRITEDATA, (void *)&chunk);
     /* CURLOPT_COOKIEFILE - file name to read cookies from */
     curl_easy_setopt(curl, CURLOPT_COOKIEFILE, "/etc/farming/cookie.txt" );
     /* CURLOPT_COOKIEJAR - file name to store cookies to */
     curl_easy_setopt(curl, CURLOPT_COOKIEJAR, "/etc/farming/cookie.txt" );
     /* Set the CA info */
     curl_easy_setopt(curl, CURLOPT_CAINFO, "/etc/ssl/certs/logicway.ca.cert.pem");
     /* This is just the server URL */
     curl_easy_setopt(curl, CURLOPT_URL, url_string);
     /* tell it to "upload" to the URL */
     curl_easy_setopt(curl, CURLOPT_UPLOAD, 1L);
     /* set where to read from (on Windows you need to use READFUNCTION too) */
     curl_easy_setopt(curl, CURLOPT_READDATA, fd);
     /* and give the size of the upload (optional) */
     curl_easy_setopt(curl, CURLOPT_INFILESIZE_LARGE, (curl_off_t)file_info.st_size);
     /* enable verbose for easier tracing */
     curl_easy_setopt(curl, CURLOPT_VERBOSE, 1L);

     /* Perform the request, res will get the return code */
     res = curl_easy_perform(curl);
     /* Check for errors */
     if(res != CURLE_OK) {
         Log ( 1, "Line: %d curl_easy_perform() failed: %s\n", __LINE__, curl_easy_strerror(res));
     }
     else {
         Log ( 200, "Line: %d chunk.memory: %s\n", __LINE__, chunk.memory );
         if( strstr( chunk.memory, "err:401") ) {
             cookie_okay = false;
         }
     }
     curl_slist_free_all(headers); /* free the header list */
     /* free the allocated memory */
     free(chunk.memory);
     /* always cleanup */
     curl_easy_cleanup(curl);
     /* close open file */
     fclose( fd );
}
/* we're done with libcurl, so clean it up */
curl_global_cleanup();

I think I've done all initializations and also cleanups/frees. Why there
are so many errors? Does anybody got an idea?

Any help would be appreciated.

Regards,

         Ralph
-----------------------------------------------------------
Unsubscribe: https://cool.haxx.se/list/listinfo/curl-users
Etiquette: https://curl.haxx.se/mail/etiquette.html
Received on 2017-03-08