- Based on patch provided by Jacob Moshenko, the transfer logic now properly
authorDaniel Stenberg <daniel@haxx.se>
Tue, 2 Mar 2010 21:20:22 +0000 (21:20 +0000)
committerDaniel Stenberg <daniel@haxx.se>
Tue, 2 Mar 2010 21:20:22 +0000 (21:20 +0000)
commit9b2cce236fa7dac594d1337b9b83bc88debf1871
treed0f4bc03514ae15ebc96007edd1b406b1a34671f
parentcd465e4ada29c947e1694b48abaf910e3c27f35e
- Based on patch provided by Jacob Moshenko, the transfer logic now properly
  makes sure that when using sub-second timeouts, there's no final bad 1000ms
  wait. Previously, a sub-second timeout would often make the elapsed time end
  up the time rounded up to the nearest second (e.g. 1s for 200ms timeout)
CHANGES
RELEASE-NOTES
lib/transfer.c