Re: some success - but mostly failure - with Smartlink drivers

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



I retract my previous statement.  It's possible - but I still worry it won't 
be suitable for all systems.  I've made a small patch for slamr that will do 
this - see attached.

Best Regards,
Jeff Trull

On Sunday 05 August 2007 05:57:27 pm Marvin Stodolsky wrote:
> Jeff,
>
> RE: Update: I found a way to get the
> 10b9:5459 (Smartlink SL1800
> to work on my system.  The required magic is:
>     setpci -d 10b9:5459 latency_timer=10
>
> Do you thing a cogent test could be written into either the slmodemd
> or slamr code?
> This is beyond my personal capabilities.
>
> MarvS
>
> On 8/5/07, Jeff Trull <linmodemstudent@xxxxxxxxx> wrote:
> > Update: I found a way to get the SL1800 to work on my system.  The
> > required magic is:
> >
> > setpci -d 10b9:5459 latency_timer=10
> >
> > which anyone with my hardware would need to add to an init script, I
> > suppose. Note that my 1057:3052 card does not require any modification at
> > all to work with the same driver.
> >
> > I have an Asus K8V-X SE motherboard with an Athlon 64 2800+ (with Kubuntu
> > Feisty, but I've verified this problem on kernels as far back as 2.6.8,
> > and Windows 98 as well!).
> >
> > Hopefully this will help someone.
> >
> > Regards,
> > Jeff Trull
> >
> > On Wednesday 18 July 2007 01:57:57 pm Marvin Stodolsky wrote:
> > > Jeff
> > >
> > > 1057:3052 (Motorola) works, miraculously, because the PCI ID 1057:3052
> > > was added into the init recognition code
> > >
> > > 11d4:1805 Analog Devices fails, even though its PCI ID, was added
> > > after the reported success with 1057:3052
> > >
> > > 10b9:5459 (Smartlink SL1800) initializes OK, but hangs system on
> > > dialing This is BAD because slamr was written to support this modem.
> > > Perhaps check for a resource comflict.
> > >
> > > 14e4:4212 (Broadcom) "cannot init card"- There is no hope for
> > > 14e4:4212 cards under 2.6.n kernels. Last support was in mid 2.4.n
> > >
> > > Just get the current scanModem and run:
> > > $ ./scanModem test 1057:3052 ( or orther PCI IDs) to get cogent
> > > gossip/URLs
> > >
> > > MarvS
> > >
> > > On 7/18/07, Jeff Trull <linmodemstudent@xxxxxxxxx> wrote:
> > > > Hi all,
> > > >
> > > > I'm in the unusual position of having access to a lot of used
> > > > winmodems, due to volunteering with some electronics reuse
> > > > organizations, and I've recently started looking at modems that may
> > > > be compatible with the Smartlink driver. I judged which ones might
> > > > work by looking at the slamr code (specifically, amrmo_init.c). 
> > > > There's a nice list in there defined in the amrmo_pci_tbl structure,
> > > > and I decided to get some with matching PCI id's and try them. Here
> > > > are the results:
> > > >
> > > > 1057:3052 (Motorola) works
> > > > 10b9:5459 (Smartlink SL1800) initializes OK, but hangs system on
> > > > dialing 11d4:1805 (Motorola on chip but lspci says Analog Devices)
> > > > "cannot init card" 14e4:4212 (Broadcom) "cannot init card"
> > > >
> > > > The Smartlink brand card system hang occurs on kernels ranging from
> > > > 2.6.8 (Debian Sarge) to 2.6.20 (Ubuntu Feisty) and on more than one
> > > > such modem (I have two).
> > > >
> > > > The two "cannot init card" failures give somewhat different output in
> > > > the system log when I enable debug messages (with the debug= module
> > > > parameter). 11d4:1805 dies pretty quickly, complaining about
> > > > "BaseAddress==0, can not initialize IOSpace".  The Broadcom card gets
> > > > further, detecting a SiL3052 chipset, revision three, and assuming a
> > > > 32.768MHz crystal (which is wrong!) but failing "because DAA not
> > > > ready".
> > > >
> > > > I'd be interested in hearing from people who've tried any of this
> > > > hardware and can report success (and, especially, the debug logs from
> > > > the driver).
> > > >
> > > > Thanks and regards,
> > > > Jeff Trull


diff -Naur slmodem-2.9.11-20070505/drivers/amrmo_init.c slmodem-2.9.11-20070505-new/drivers/amrmo_init.c
--- slmodem-2.9.11-20070505/drivers/amrmo_init.c	2007-05-05 16:56:25.000000000 -0700
+++ slmodem-2.9.11-20070505-new/drivers/amrmo_init.c	2007-08-06 22:22:48.000000000 -0700
@@ -597,6 +597,11 @@
 
 	pci_set_master(pci_dev);
 
+	/* workaround for SL1800 hang on Asus K8V-X SE - reduce latency timer to 16 */
+	if (pci_id->driver_data == SL1800_CARD) {
+	  pci_write_config_byte(pci_dev, PCI_LATENCY_TIMER, 0x10);
+	}
+
 	ret = pci_request_regions(pci_dev,(char*)amrmo->name);
 	if(ret) {
 		printk(KERN_ERR "slamr: failed request regions.\n");

[Index of Archives]     [Linux Media Development]     [Asterisk]     [DCCP]     [Netdev]     [X.org]     [Xfree86]     [Fedora Women]     [Linux USB]

  Powered by Linux