Description of problem: pl2303 "type 2" device drops characters on read sometimes. This is the HX version of the chip. The "type 1" device works OK. This is on a thinkpad R32 with an INTEL 82801CA/CAM USB controller. My Other USB devices all work fine. Version-Release number of selected component (if applicable): 2.6.10-1.12_FC2, but I've seen similar behavior in recent kernels. There was a fairly recent patch to pl2303 that doesn't seem to help. How reproducible: Always (as long as you try twice) Steps to Reproduce: I've created a test script that reproduces the problem. There's a little bit of cruft in it to produce the debug log. It also tests the older version of the adapter, which works OK. I've included both so you can see the difference. http://throb.netspace.org/~bperk/serialtest2 It requires this little C helper program. http://throb.netspace.org/~bperk/stest.c This program is designed to work with my sportrak GPS, but it behaves the same if you short pins 2 and three together ( I used a little jumper) Actual results: The first time the test program is run it works. The second time it drops the first character. Additional info: Here are the results of the test with debug=1 turned on. http://throb.netspace.org/~bperk/pl2303-type1b.log (old working adapter) http://throb.netspace.org/~bperk/pl2303-type2b.log (HX version, non working) The problem show up on line 152 or so of pl2303-type2b.log Where it says Mar 22 12:35:17 thump kernel: PL-2303 ttyUSB0: pl2303_read_bulk_callback - length = 1, data = 50 Which should have been preceded by a '$' (0x24) I've talked to the driver maintainer greg, who has pretty much given up. I tried contacting prolific, and they won't respond. I also tried playing the driver, and didn't get very far. I did find that commenting out _all_ of the FISH SOUP macros had no effect, i.e. the driver works, but is still broken as I've described. I tried to compare what the driver is trying to do compared with usbsnoop under windows. I didn't get very far with this either. I spent a lot of time on this, so I thought I'd report it.
Fedora Core 2 has now reached end of life, and no further updates will be provided by Red Hat. The Fedora legacy project will be producing further kernel updates for security problems only. If this bug has not been fixed in the latest Fedora Core 2 update kernel, please try to reproduce it under Fedora Core 3, and reopen if necessary, changing the product version accordingly. Thank you.