Date: Fri, 20 Apr 2001 00:22:10 +0200 From: "Georg-W. Koltermann" <gwk@sgi.com> To: Warner Losh <imp@village.org> Cc: mobile@freebsd.org Subject: Re: Cardbus status Message-ID: <20010420002209.A911@hunter.munich.sgi.com> In-Reply-To: <200103210328.f2L3StZ48205@billy-club.village.org>; from imp@village.org on Tue, Mar 20, 2001 at 08:28:55PM -0700 References: <20010320121110.B912@hunter.munich.sgi.com> <20010319154450.A10084@hunter.munich.sgi.com> <200010220509.XAA44604@harmony.village.org> <20010319154450.A10084@hunter.munich.sgi.com> <200103200218.f2K2I4Z43387@billy-club.village.org> <20010320121110.B912@hunter.munich.sgi.com> <200103210328.f2L3StZ48205@billy-club.village.org>
next in thread | previous in thread | raw e-mail | index | archive | help
[-- Attachment #1 --]
Warner,
thanks for your help. I have the modem part of my Xircom CBEM56G
working, coming in via this card right now.
In addition to the lines that you indicated I also had to
re-initialize the cfcr register to 8 bit mode, after the initial tests
have passed. If I didn't do that, I got an error from the probe 0
test and the modem would not attach.
I've appended the complete patch to this mail. Please commit if you
like, so others can benefit.
Thanks again,
Georg.
----------------------------------------------------------------------
Who in the world needs 2000 Windows?
On Tue, Mar 20, 2001 at 08:28:55PM -0700, Warner Losh wrote:
> In message <20010320121110.B912@hunter.munich.sgi.com> "Georg-W. Koltermann" writes:
> : OK, so maybe you could hint me at where to find those two lines..
>
> The patch would look something like the following with XXXX replaced
> by the ID of the card. Note, it also has an unrelated patch and
> doesn't do detach, but it would get you going.
>
> Warner
>
>
> Index: sio.c
> ===================================================================
> RCS file: /home/imp/FreeBSD/CVS/src/sys/isa/sio.c,v
> retrieving revision 1.328
> diff -u -r1.328 sio.c
> +++ sio.c 2001/03/21 03:20:55
> @@ -555,6 +555,7 @@
> bus_release_resource(dev, SYS_RES_IOPORT, 0, com->ioportres);
> if (com->tp && (com->tp->t_state & TS_ISOPEN)) {
> device_printf(dev, "still open, forcing close\n");
> + (*linesw[com->tp->t_line].l_close)(com->tp, 0);
> com->tp->t_gen++;
> ttyclose(com->tp);
> ttwakeup(com->tp);
> @@ -577,7 +578,8 @@
>
> static struct pci_ids pci_ids[] = {
> { 0x100812b9, "3COM PCI FaxModem", 0x10 },
> - { 0x048011c1, "ActionTec 56k FAX PCI Modem", 0x14 },
> + { 0x048011c1, "Lucent kermit based PCI Modem", 0x14 },
> + { 0xXXXXXXXX, "Xircom Cardbus modem", 0x10 },
> { 0x00000000, NULL, 0 }
> };
>
> @@ -3352,4 +3354,5 @@
> #endif
> #if NPCI > 0
> DRIVER_MODULE(sio, pci, sio_pci_driver, sio_devclass, 0, 0);
> +DRIVER_MODULE(sio, cardbus, sio_pci_driver, sio_devclass, 0, 0);
> #endif
[-- Attachment #2 --]
--- /home/hunter/gwk/sio.c.1.329 Thu Apr 19 22:47:43 2001
+++ sys/isa/sio.c Thu Apr 19 23:00:22 2001
@@ -30,7 +30,7 @@
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
- * $FreeBSD: /c/ncvs/src/sys/isa/sio.c,v 1.329 2001/03/26 12:40:58 phk Exp $
+ * $FreeBSD: src/sys/isa/sio.c,v 1.329 2001/03/26 12:40:58 phk Exp $
* from: @(#)com.c 7.5 (Berkeley) 5/16/91
* from: i386/isa sio.c,v 1.234
*/
@@ -554,6 +554,7 @@
bus_release_resource(dev, SYS_RES_IOPORT, 0, com->ioportres);
if (com->tp && (com->tp->t_state & TS_ISOPEN)) {
device_printf(dev, "still open, forcing close\n");
+ (*linesw[com->tp->t_line].l_close)(com->tp, 0);
com->tp->t_gen++;
ttyclose(com->tp);
ttwakeup(com->tp);
@@ -576,7 +577,8 @@
static struct pci_ids pci_ids[] = {
{ 0x100812b9, "3COM PCI FaxModem", 0x10 },
- { 0x048011c1, "ActionTec 56k FAX PCI Modem", 0x14 },
+ { 0x048011c1, "Lucent kermit based PCI Modem", 0x14 },
+ { 0x0103115d, "Xircom Cardbus modem", 0x10 },
{ 0x00000000, NULL, 0 }
};
@@ -925,6 +927,13 @@
sio_setreg(com, com_mcr, mcr_image);
/*
+ * It seems my Xircom CBEM56G Cardbus modem wants to be reset
+ * to 8 bits *again*, or else probe test 0 will fail.
+ * gwk@sgi.com, 4/19/2001
+ */
+ sio_setreg(com, com_cfcr, CFCR_8BITS);
+
+ /*
* Some pcmcia cards have the "TXRDY bug", so we check everyone
* for IIR_TXRDY implementation ( Palido 321s, DC-1S... )
*/
@@ -3351,4 +3360,5 @@
#endif
#if NPCI > 0
DRIVER_MODULE(sio, pci, sio_pci_driver, sio_devclass, 0, 0);
+DRIVER_MODULE(sio, cardbus, sio_pci_driver, sio_devclass, 0, 0);
#endif
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20010420002209.A911>