Change Details

When I upgraded from 3.1.15 to 4.0.0 I noticed that operations involving smartcard access get stuck, i. e. the "call" to scdaemon is never finished and the scdaemon process causes ~15% CPU load. The only way to "fix" this situation is by killing the scdaemon process. Cases where this behavior occurs include: * gpg --card-status * gpg-connect-agent -> SCD SERIALNO * Accessing scdaemon via libassuan By trial and error, I noticed that this is somehow caused by my scdaemon.conf's line: ``` reader-port "Yubico Yubikey NEO OTP+U2F+CCID" ``` As soon as remove the line (which makes the conf empty) and restart the scdaemon process, everything works fine. Maybe this is connected to change https://dev.gnupg.org/T5644. Is "reader-port" not supported anymore or is this a bug with the new heuristic?

When I upgraded from 3.1.15 to 4.0.0 I noticed that operations involving smartcard access get stuck, i. e. the "call" to scdaemon is never finished and the scdaemon process causes ~15% CPU load. The only way to "fix" this situation is by killing the scdaemon process. Cases where this behavior occurs include: * gpg --card-status * gpg-connect-agent -> SCD SERIALNO * Accessing scdaemon via libassuan By trial and error, I noticed that this is somehow caused by my scdaemon.conf's line: ``` reader-port "Yubico Yubikey NEO OTP+U2F+CCID" ``` As soon as remove the line (which makes the conf empty) and restart the scdaemon process, everything works fine. Maybe this is connected to change https://dev.gnupg.org/T5644. Is "reader-port" not supported anymore or is this a bug with the new heuristic? System specs: * WIndows 10 21H1 (19043.1415)

When I upgraded from 3.1.15 to 4.0.0 I noticed that operations involving smartcard access get stuck, i. e. the "call" to scdaemon is never finished and the scdaemon process causes ~15% CPU load. The only way to "fix" this situation is by killing the scdaemon process. Cases where this behavior occurs include: * gpg --card-status * gpg-connect-agent -> SCD SERIALNO * Accessing scdaemon via libassuan By trial and error, I noticed that this is somehow caused by my scdaemon.conf's line: ``` reader-port "Yubico Yubikey NEO OTP+U2F+CCID" ``` As soon as remove the line (which makes the conf empty) and restart the scdaemon process, everything works fine. Maybe this is connected to change https://dev.gnupg.org/T5644. Is "reader-port" not supported anymore or is this a bug with the new heuristic? System specs: * WIndows 10 21H1 (19043.1415)