Re: [PATCH v2 3/8] tests/9pfs: add local Tlcreate test

[Christian Schoenebeck]

On Donnerstag, 22. Oktober 2020 10:51:46 CEST Greg Kurz wrote:
> On Wed, 21 Oct 2020 14:25:33 +0200
> 
> Christian Schoenebeck <qemu_oss@crudebyte.com> wrote:
> > This test case uses a Tlcreate 9p request to create a regular file inside
> > host's test directory.
> > 
> > Signed-off-by: Christian Schoenebeck <qemu_oss@crudebyte.com>
> > ---
> 
> Just one remark, see below.
> 
> Reviewed-by: Greg Kurz <groug@kaod.org>
> 
> >  tests/qtest/virtio-9p-test.c | 77 ++++++++++++++++++++++++++++++++++++
> >  1 file changed, 77 insertions(+)
> > 
> > diff --git a/tests/qtest/virtio-9p-test.c b/tests/qtest/virtio-9p-test.c
> > index abd7e44648..c030bc2a6c 100644
> > --- a/tests/qtest/virtio-9p-test.c
> > +++ b/tests/qtest/virtio-9p-test.c
> > @@ -258,6 +258,7 @@ static const char *rmessage_name(uint8_t id)
> > 
> >          id == P9_RLOPEN ? "RLOPEN" :
> >          id == P9_RWRITE ? "RWRITE" :
> > 
> >          id == P9_RMKDIR ? "RMKDIR" :
> > +        id == P9_RLCREATE ? "RLCREATE" :
> >          id == P9_RUNLINKAT ? "RUNLINKAT" :
> >          id == P9_RFLUSH ? "RFLUSH" :
> > 
> >          id == P9_RREADDIR ? "READDIR" :
> > @@ -694,6 +695,44 @@ static void v9fs_rmkdir(P9Req *req, v9fs_qid *qid)
> > 
> >      v9fs_req_free(req);
> >  
> >  }
> > 
> > +/* size[4] Tlcreate tag[2] fid[4] name[s] flags[4] mode[4] gid[4] */
> > +static P9Req *v9fs_tlcreate(QVirtio9P *v9p, uint32_t fid, const char
> > *name, +                            uint32_t flags, uint32_t mode,
> > uint32_t gid, +                            uint16_t tag)
> > +{
> > +    P9Req *req;
> > +
> > +    uint32_t body_size = 4 + 4 + 4 + 4;
> > +    uint16_t string_size = v9fs_string_size(name);
> > +
> > +    g_assert_cmpint(body_size, <=, UINT32_MAX - string_size);
> > +    body_size += string_size;
> > +
> > +    req = v9fs_req_init(v9p, body_size, P9_TLCREATE, tag);
> > +    v9fs_uint32_write(req, fid);
> > +    v9fs_string_write(req, name);
> > +    v9fs_uint32_write(req, flags);
> > +    v9fs_uint32_write(req, mode);
> > +    v9fs_uint32_write(req, gid);
> > +    v9fs_req_send(req);
> > +    return req;
> > +}
> > +
> > +/* size[4] Rlcreate tag[2] qid[13] iounit[4] */
> > +static void v9fs_rlcreate(P9Req *req, v9fs_qid *qid, uint32_t *iounit)
> > +{
> > +    v9fs_req_recv(req, P9_RLCREATE);
> > +    if (qid) {
> > +        v9fs_memread(req, qid, 13);
> > +    } else {
> > +        v9fs_memskip(req, 13);
> > +    }
> > +    if (iounit) {
> > +        v9fs_uint32_read(req, iounit);
> > +    }
> > +    v9fs_req_free(req);
> > +}
> > +
> > 
> >  /* size[4] Tunlinkat tag[2] dirfd[4] name[s] flags[4] */
> >  static P9Req *v9fs_tunlinkat(QVirtio9P *v9p, uint32_t dirfd, const char
> >  *name,>  
> >                               uint32_t flags, uint16_t tag)
> > 
> > @@ -1032,6 +1071,24 @@ static void do_mkdir(QVirtio9P *v9p, const char
> > *path, const char *cname)> 
> >      g_free(name);
> >  
> >  }
> > 
> > +/* create a regular file with Tlcreate and return file's fid */
> > +static uint32_t do_lcreate(QVirtio9P *v9p, const char *path,
> > +                           const char *cname)
> > +{
> > +    char *const name = g_strdup(cname);
> > +    uint32_t fid;
> > +    P9Req *req;
> > +
> > +    fid = do_walk(v9p, path);
> > +
> > +    req = v9fs_tlcreate(v9p, fid, name, 0, 0750, 0, 0);
> 
> Maybe it could make sense to make the mode a parameter of
> do_lcreate() in case someone wants to write a test case
> around that ? Same remark applies to do_mkdir() actually.
> 
> No need to change this now anyway.

Yeah, I thought about that for a moment, but decided if that's really needed 
one day, it'll be a trivial change, especially under the assumption that there 
will be probably not a lot of code using this function, even on the long-term.

Thanks for looking at these patches!

Best regards,
Christian Schoenebeck