Sorry i have forgotten the patch :-(
On Sam 26.08.2006 20:04, Alexander Lazic wrote:
>Hi,
>
>On Fre 25.08.2006 09:07, Alexander Lazic wrote:
>>
>>Well then i will go the *only* milliseconds way.
>
>here now my patch, against the todays 'cvs co'.
>
>make test was going thru the cli have give me some vaild times back.
>
>src/curl -Lo /dev/null -skw 'time_total %{time_total} time_connect \
>%{time_connect} time_namelookup %{time_namelookup} time_pretransfer \
>%{time_pretransfer} time_starttransfer %{time_starttransfer} \
>time_redirect %{time_redirect}\n' http://curlm.haxx.se/
>
>time_total 1226 time_connect 392 time_namelookup 9 time_pretransfer 392 \
>time_starttransfer 818 time_redirect 0
>
>I have also change the hugehelp.c because I don't know how to generate
>it :-(
>
>Thoughts, opinions?!
>
>I have also see a curious behaviour in the *original* code, I have add
>into simple.c the following code:
>
>long timesl = 0;
>double times = 0.0;
>
>*after perform*
>
>res = curl_easy_getinfo(curl, CURLINFO_CONNECT_TIME, &times);
>
>if(res != CURLE_OK){
> printf("error getinfo()\n");
> return 0;
>} else
> printf("CURLINFO_CONNECT_TIME %f\n", times);
>
>res = curl_easy_getinfo(curl, CURLINFO_CONNECT_TIME, &timesl);
>if(res != CURLE_OK){
> printf("error getinfo()\n");
> return 0;
>} else
> printf("CURLINFO_CONNECT_TIME %ld\n", timesl);
>
>But i don't get the error message instead a segfault in:
>
>CURLINFO_CONNECT_TIME 0.044362
>CURLINFO_CONNECT_TIME -1235301314
>
>Program received signal SIGSEGV, Segmentation fault.
>Curl_close (data=0x3fa6b69d) at url.c:198
>198 if(data->multi) {
>(gdb) bt
>#0 Curl_close (data=0x3fa6b69d) at url.c:198
>#1 0xa7fc1ae1 in curl_easy_cleanup (curl=0x3fa6b69d) at easy.c:485
>#2 0x080487af in main () at simple.c:47
>(gdb) The program is running. Exit anyway? (y or n) y
>
>could anybody reproduce this behaviour?!
>
>Regards
>
>Alex