[Gnuk-users] State of EdDSA in Gnuk / GnuPG

[NIIBE Yutaka]

Thank you for debug output.  It's useful.

On 01/22/2015 09:34 AM, Bertrand Jacquin wrote:
> #2  store_fpr (app=app at entry=0x7fbefc002d40, keynumber=keynumber at entry=2, timestamp=timestamp at entry=1421886581, fpr=fpr at entry=0x7fbf03f24d70
> "\244", card_version=512, key_type=key_type at entry=KEY_TYPE_EDDSA) at /usr/src/debug/app-crypt/gnupg-2.1.1-r1/gnupg-2.1.1/scd/app-openpgp.c:816
>         n = <optimized out>
>         nbits = <optimized out>
>         buffer = 0x7fbefc002f80 "\231"
>         p = <optimized out>
>         tag = <optimized out>
>         tag2 = <optimized out>
>         rc = <optimized out>
>         m = {0x44937a "\t+\006\001\004\001\332G\017\001", 0x7fbefc011a5e
> "@\177!\r\303s\351y\240Q\353\036\263A\307|d\226\356\026N\324th\205\375\070Y\373\373\250lH)(1:d32:\246TWL", 0x7fbefc0008c0 "\006"}
>         mlen = {140458315481098, 33, 11}
                  ^^^^^^^^^^^^^^^
This value is a cause of SEGV.

>         ap = {{
>             gp_offset = 48,
>             fp_offset = 32702,
>             overflow_arg_area = 0x7fbf03f24b90,
>             reg_save_area = 0x7fbf03f24b20
>           }}
>         argc = 2
>         i = 0

The value should be 10.

But it's 0x7fbf0000000a, looks like an address of memory.

I don't know why this bug.  Let me check about variable args.
--