On Sat, 5 Jan 2008, David Miller wrote:

> From: Herbert Xu <[EMAIL PROTECTED]>
> Date: Sun, 06 Jan 2008 11:29:35 +1100
> 
> > We should never use inline except when it's on the fast path and this
> > is definitely not a fast path.  If a function ends up being called
> > just once the compiler will most likely inline it anyway, making the
> > use of the keyword inline redundant.
> 
> Similarly I question just about any inline usage at all in *.c files
> these days.
> 
> I even would discourage it's use for fast-path cases as well.

There still seems to be good candidates for inline in *.c files, in worst 
case I had +172 due to inline removed and ~60 are on +10-+90 range with 
my gcc, later gccs might do better but I definately would just blindly 
remove them all. Here's the other end of the list:

+35 static inline hci_encrypt_change_evt        net/bluetooth/hci_event.c
+36 static __inline__ tcp_in_window     net/ipv4/tcp_minisocks.c
+38 static inline hci_conn_complete_evt net/bluetooth/hci_event.c
+38 static inline hci_conn_request_evt  net/bluetooth/hci_event.c
+42 static inline gred_wred_mode        net/sched/sch_gred.c
+45 static inline secpath_has_nontransport      net/xfrm/xfrm_policy.c
+52 static inline bool port_match       net/netfilter/xt_tcpudp.c
+67 static inline dn_check_idf  net/decnet/dn_nsp_in.c
+90 static inline __ieee80211_queue_stopped     net/mac80211/tx.c
+172 static inline sctp_chunk_length_valid      net/sctp/sm_statefuns.c


-- 
 i.
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Reply via email to