Thanks for finding it! I think best is to revert r942 and ask Kurt Jaeger for a different patch. What do you think?
On 2/3/09, Chaskiel Grundman <[email protected]> wrote: > > > On Thu, 29 Jan 2009, Andrey Jivsov wrote: > > > I am attaching the tested patch to the file ifd-ccid.c to add support for > the > > reader. The reader's USB IDs that I tested with are 0b97:7762 and > 0b97:7772. > > Without this patch the ifd-ccid.c code will not work with these readers. > > > > The patch is based on the work done in the pcsc-lite project. The patch is > > made against openct-0.6.15-svn-r1127. I didn't make corresponding updates > to > > text configuration files. > > > The patch is good (if a little conservative). I failed to notice this > badness, despite having one of these devices myself, because the openct > head includes a (broken, now that I look at it*) patch to the usb > descriptor handling that replaces the extra data for an interface with the > extra data for an endpoint, for any endpoint that has extra data (see > revision 942 of usb-descriptors.c). In the event that a new openct > release is prepared from the head, only one of these patches should be > used.... > > > * the usb-descriptors.c patch is broken because it will throw away extra > data that an interface has if an endpoint also has extra data. Also, it > will leak memory if either a) both the interface and an enpoint have extra > data, or b) multiple endpoints have extra data. These scenarios are > unlikely.... > > _______________________________________________ > opensc-devel mailing list > [email protected] > http://www.opensc-project.org/mailman/listinfo/opensc-devel > _______________________________________________ opensc-devel mailing list [email protected] http://www.opensc-project.org/mailman/listinfo/opensc-devel
