Hello,

now we have Cologne hfc-e1 evaluation board (http://www.colognechip.com/isdn/controllers/frame-hfc-e1.htm) it comes with 32kb on board ram, according to this board we are designing our board with four hfc-e1 chips and would like to put external ram. But before doing this we would like to see if external ram gives more performance. So we attached 512kb external ram to cologne evaluation board.

when I run visdn_configurator and kernel modules are loaded we can read stream from e1 server and kernel.log looks like this:

Mar  2 09:49:57 localhost kernel: [4346991.630000] visdn-netdev: Netdevice gateway loading
Mar  2 09:49:57 localhost kernel: [4346991.649000] visdn-streamport: vISDN streamport module loading
Mar  2 09:49:57 localhost kernel: [4346991.667000] CSLIP: code copyright 1989 Regents of the University of California
Mar  2 09:49:57 localhost kernel: [4346991.672000] PPP generic driver version 2.4.2
Mar  2 09:49:57 localhost kernel: [4346991.688000] visdn-ppp: vISDN ppp gateway module loading
Mar  2 09:49:57 localhost kernel: [ 4346991.718000] vts_timer_open()
Mar  2 09:49:57 localhost kernel: [4346991.722000] visdn-ec-kb:  echo canceller module for vISDN loading
Mar  2 09:49:57 localhost kernel: [4346991.728000] hfc-4s: HFC-4S HFC-8S Driver loading
Mar  2 09:49:57 localhost kernel: [4346991.752000] hfc-pci: HFC-S PCI A Driver loading
Mar  2 09:49:57 localhost kernel: [4346991.764000] hfc-usb: HFC-USB Driver loading
Mar  2 09:49:57 localhost kernel: [4346991.767000 ] usbcore: registered new driver hfc-usb
Mar  2 09:49:57 localhost kernel: [4346991.826000] NET: Registered protocol family 30
Mar  2 09:49:57 localhost kernel: [4346991.839000]
Mar  2 09:49:57 localhost kernel: [ 4346991.840000] lapd: NETDEV_UP visdn3

everything is ok, but by default visdn is configured for 32kb onboard ram. So we have to configure it to use external ram:

[EMAIL PROTECTED]:/sys/bus/pci/devices/0000:03:05.0# echo 512 > ram_size

and then on kernel.log we get this:

Mar  2 09:51:46 localhost kernel: [4347100.524000] scheduling while atomic: bash/0x00000100/9224
Mar  2 09:51:46 localhost kernel: [4347100.524000 ]  [schedule+67/1188] schedule+0x43/0x4a4
Mar  2 09:51:46 localhost kernel: [4347100.524000]  [invalidate_inode_buffers+12/55] invalidate_inode_buffers+0xc/0x37
Mar  2 09:51:46 localhost kernel: [4347100.524000]  [__mod_timer+88/105] __mod_timer+0x58/0x69
Mar  2 09:51:46 localhost kernel: [4347100.524000]  [schedule_timeout+126/157] schedule_timeout+0x7e/0x9d
Mar  2 09:51:46 localhost kernel: [4347100.524000]  [process_timeout+0/9] process_timeout+0x0/0x9
Mar  2 09:51:46 localhost kernel: [ 4347100.524000]  [msleep+46/53] msleep+0x2e/0x35
Mar  2 09:51:46 localhost kernel: [4347100.524000]  [pg0+947952315/1069995008] visdn_chan_unregister+0x14d/0x181 [visdn_core]
Mar  2 09:51:46 localhost kernel: [4347100.524000 ]  [kobject_release+0/10] kobject_release+0x0/0xa
Mar  2 09:51:46 localhost kernel: [4347100.524000]  [pg0+948087093/1069995008] hfc_sys_chan_unregister+0x22/0x3d [visdn_hfc_e1]
Mar  2 09:51:46 localhost kernel: [4347100.524000 ]  [pg0+948079902/1069995008] hfc_sys_port_reconfigure+0x18/0x5a [visdn_hfc_e1]
Mar  2 09:51:46 localhost kernel: [4347100.524000]  [pg0+948065416/1069995008] hfc_store_ram_size+0x96/0xf6 [visdn_hfc_e1]
Mar  2 09:51:46 localhost kernel: [ 4347100.524000]  [dev_attr_store+31/39] dev_attr_store+0x1f/0x27
Mar  2 09:51:46 localhost kernel: [4347100.524000]  [flush_write_buffer+37/42] flush_write_buffer+0x25/0x2a
Mar  2 09:51:46 localhost kernel: [4347100.524000 ]  [sysfs_write_file+56/86] sysfs_write_file+0x38/0x56
Mar  2 09:51:46 localhost kernel: [4347100.524000]  [vfs_write+163/313] vfs_write+0xa3/0x139
Mar  2 09:51:46 localhost kernel: [4347100.524000]  [sys_write+59/99] sys_write+0x3b/0x63
Mar  2 09:51:46 localhost kernel: [4347100.524000]  [sysenter_past_esp+84/117] sysenter_past_esp+0x54/0x75
Mar  2 09:51:46 localhost kernel: [4347100.575000] hfc-e1: pci-0000:03:05.0 resetting

then board leds lights turn off and when trying to read stream from e1 server we get nothing...

Is this a problem with our attached ram? or is this something to do with visdn hfc-e1 drivers? we re-checked twice the connection of the pins to external ram, from our perspective everything should be o.k. but it isn't

Could you Daniele recommend us something?

Thank You,
Tomas Vilda
_______________________________________________
Visdn-hackers mailing list
[email protected]
https://mailman.uli.it/mailman/listinfo/visdn-hackers

Reply via email to