qed (gcc13): use u16 for fid to be big enough

[ Upstream commit 7d84118 ]

gcc 13 correctly reports overflow in qed_grc_dump_addr_range():
In file included from drivers/net/ethernet/qlogic/qed/qed.h:23,
                 from drivers/net/ethernet/qlogic/qed/qed_debug.c:10:
drivers/net/ethernet/qlogic/qed/qed_debug.c: In function 'qed_grc_dump_addr_range':
include/linux/qed/qed_if.h:1217:9: error: overflow in conversion from 'int' to 'u8' {aka 'unsigned char'} changes value from '(int)vf_id << 8 | 128' to '128' [-Werror=overflow]

We do:
  u8 fid;
  ...
  fid = vf_id << 8 | 128;

Since fid is 16bit (and the stored value above too), fid should be u16,
not u8. Fix that.

Cc: Martin Liska <mliska@suse.cz>
Cc: Ariel Elior <aelior@marvell.com>
Cc: Manish Chopra <manishc@marvell.com>
Signed-off-by: Jiri Slaby (SUSE) <jirislaby@kernel.org>
Link: https://lore.kernel.org/r/20221031114354.10398-1-jirislaby@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

drivers/net/ethernet/qlogic/qed/qed_debug.c

@@ -1972,9 +1972,10 @@ static u32 qed_grc_dump_addr_range(struct qed_hwfn *p_hwfn,
 				   u8 split_id)
 {
 	struct dbg_tools_data *dev_data = &p_hwfn->dbg_info;
-	u8 port_id = 0, pf_id = 0, vf_id = 0, fid = 0;
+	u8 port_id = 0, pf_id = 0, vf_id = 0;
 	bool read_using_dmae = false;
 	u32 thresh;
+	u16 fid;
 
 	if (!dump)
 		return len;