From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mailserv2.iuinc.com (IDENT:qmailr@mailserv2.iuinc.com [206.245.164.55]) by puffin.external.hp.com (8.9.3/8.9.3) with SMTP id JAA04027 for ; Wed, 31 May 2000 09:40:46 -0600 Received: from ppp-2-148.cvx3.telinco.net (HELO rhirst.linuxcare.com) (@212.1.145.148) by mailserv2.iuinc.com with SMTP; 31 May 2000 15:41:13 -0000 Received: by rhirst.linuxcare.com (Postfix, from userid 501) id 3FDBCB009; Wed, 31 May 2000 16:41:02 +0100 (BST) Date: Wed, 31 May 2000 16:41:02 +0100 From: Richard Hirst To: parisc-linux@thepuffingroup.com Message-ID: <20000531164102.L2253@linuxcare.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Subject: [parisc-linux] ttyS00 reported twice List-ID: There was some discussion on irc yesterday about bootup messages reporting something like: Found serial #0 at 0xf0823000, IRQ 538 ttyS00 at iomem 0xf0823800 (irq = 538) is a 16550A Found serial #1 at 0xf0822000, IRQ 537 ttyS00 at iomem 0xf0822800 (irq = 537) is a 16550A i.e. both ports reported as ttyS00. This seems to be a consequence of us calling register_serial() before rs_init() is called. In addition there is a chunk of initialisation in rs_init() that is #ifndef CONFIG_GSC; that would need to be enabled so that rs_table[].line is initialised, before we would report ttyS01. The solution might be to make register_serial() aware of whether rs_init() has yet been called. If it hasn't, register_serial() just queues the struct_serial that was passed in and returns success. At the end of rs_init() we call register_serial() again for each item in the queue. This is similar to what happens with the net devices. Richard