Hello Guy,
Tuesday, April 19, 2016, 1:20:48 AM, you wrote:
> On Apr 18, 2016, at 4:04 PM, Guy Harris <guy@xxxxxxxxxxxx> wrote:
>> This is a dumpcap bug, and needs to be fixed - "fixed" as in "fixed for 2.0.3".
> Except that (as you already said) it's not present in 2.0.2.
> So...
>> Please file the bug
> ...no need to file the bug.
Hm, I just got confused until I realized you were quoting your own
first answer and replying to it :-)
> Then again, I looked at the code that dumpcap would use to write an
> ISB, and it doesn't seem to have changed since 2.0.x, so *dumpcap*
> should still be writing the ISB correctly.
I just checked - dumpcap does write the ISB correctly. Looks like it's
really Wireshark saving the file with a broken ISB. So I guess I
should file a bug for libwiretap instead.
> It's *libwiretap* that's broken. The new option-handling code
> treats it as a regular 64-bit quantity, which it is *not* - it's two
> 32-bit quantities, a high part and a low part, *always* written in
> that order, with each of the parts written in host byte order.
Cheers,
Jasper
Attachment:
smime.p7s
Description: S/MIME Cryptographic Signature
