tech-userlevel archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: Request for review: librefuse overhaul for supporting more filesystems

On 1/13/22 12:18 PM, Joerg Sonnenberger wrote:
> On Thu, Jan 13, 2022 at 10:20:53AM +0900, PHO wrote:
>> +/* The documentation for FUSE is not clear as to what "struct
>> + * fuse_session" is, why it exists, or how it's different from "struct
>> + * fuse". For now we define it as an empty struct and treat "struct
>> + * fuse_session *" as being identical to "struct fuse *". */
>> +struct fuse_session {};
> Specific remark here: try to avoid empty structures. They can create a
> real hassle when interacting between C and C++ as that's explicitly
> invalid for the latter.

Thank you for pointing that out. In this case the actual size or layout of the struct does not matter, so I'll just leave undefined and let compilers treat it as an incomplete type. Do you think this is reasonable?

Home | Main Index | Thread Index | Old Index