summaryrefslogtreecommitdiff
path: root/lib/select.c
Commit message (Collapse)AuthorAgeFilesLines
...
* Matt Witherspoon fixed a problem case when the CPU load went to 100% when aDaniel Stenberg2006-12-051-1/+1
| | | | | | | | | | | | | HTTP upload was disconnected: "What appears to be happening is that my system (Linux 2.6.17 and 2.6.13) is setting *only* POLLHUP on poll() when the conditions in my previous mail occur. As you can see, select.c:Curl_select() does not check for POLLHUP. So basically what was happening, is poll() was returning immediately (with POLLHUP set), but when Curl_select() looked at the bits, neither POLLERR or POLLOUT was set. This still caused Curl_readwrite() to be called, which quickly returned. Then the transfer() loop kept continuing at full speed forever."
* Check for USE_WINSOCK instead of WIN32 where the check was doneYang Tse2006-10-181-2/+2
| | | | to verify winsock API availability.
* Cygwin 1.5.21 needs this hack to pass test 160.Yang Tse2006-10-091-2/+9
| | | | In this way 304 tests out of 304 reported OK.
* Cory Nelson made libcurl use the WSAPoll() function if built for WindowsDaniel Stenberg2006-09-241-2/+8
| | | | Vista (_WIN32_WINNT >= 0x0600)
* additional renames of Curl_ourerrno => Curl_sockerrnoDaniel Stenberg2006-05-051-2/+2
|
* Fixed signed/unsigned convertion errors in Salford-C.Gisle Vanem2006-04-261-2/+2
| | | | #ifdef around WSAEDISCON in strerror.c.
* First commit of David McCreedy's EBCDIC and TPF changes.Daniel Stenberg2006-04-071-2/+22
|
* Thanks to this nice summary of poll() implementations:Daniel Stenberg2005-11-131-1/+1
| | | | | | | http://www.greenend.org.uk/rjk/2001/06/poll.html and further tests by Eugene Kotlyarov, we now know that cygwin's poll returns only POLLHUP on remote connection closure so we check for that case (too) and re-enable poll for cygwin builds.
* Cory Nelson's work on nuking compiler warnings when building on x64 withDaniel Stenberg2005-04-261-2/+2
| | | | VS2005.
* Updated the copyright year since changes have been this year.Daniel Stenberg2005-03-311-1/+1
|
* Fixed typo.Dan Fandrich2005-03-221-1/+1
|
* Modified the VALID_SOCK() macro to become VERIFY_SOCK() instead. It is slighlyDaniel Stenberg2005-03-211-10/+10
| | | | | more involved, but should hopefully not generate any compiler warnings on win32 systems (that can't check the socket based on the numeric).
* errrno can by freak accident become EINTR on DOS orGisle Vanem2005-01-151-4/+8
| | | | | Windows (unrelated to select). select() can never set errno to EINTR on Windows.
* Inspired by Martijn Koster's patch and example source atDaniel Stenberg2005-01-131-7/+16
| | | | | | | | | | | http://www.greenhills.co.uk/mak/gentoo/curl-eintr-bug.c, I now made the select() and poll() calls properly loop if they return -1 and errno is EINTR. glibc docs for this is found here: http://www.gnu.org/software/libc/manual/html_node/Interrupted-Primitives.html This last link says BSD doesn't have this "effect". Will there be a problem if we do this unconditionally? S: ----------------------------------------------------------------------
* Marcin Konicki provided two configure fixes and a source fix to make curlDaniel Stenberg2004-12-221-0/+5
| | | | build out-of-the-box on BeOS.
* include sys/types.h before sys/select.hDaniel Stenberg2004-12-211-4/+3
|
* Dan Fandrich fix to compile with libc5Daniel Stenberg2004-11-201-0/+7
|
* Suppress signed vs. unsigned warnings on Win32Gisle Vanem2004-11-191-4/+6
|
* Curl_select() now uses curl_socket_t on socket argumentsDaniel Stenberg2004-11-191-2/+2
|
* Winsock sockets are not in range 0..FD_SETSIZE.Gisle Vanem2004-11-191-2/+8
| | | | Shouldn't Curl_select() use curl_socket_t ?
* David Phillips' FD_SETSIZE fixDaniel Stenberg2004-11-191-0/+231