Re: [BUG, workaround] HVR-2200/saa7164 problem with C7 power state
On 02/19/2015 05:22 AM, Steven Toth wrote: I'll post a notice to this list once the driver is ready for download That sounds great. Let me know if/how I can help in any way. David -- To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: [BUG, workaround] HVR-2200/saa7164 problem with C7 power state
On Wed, Feb 18, 2015 at 11:04 PM, David Harty wrote: > On 02/17/2015 04:47 AM, dCrypt wrote: >> >> So the PCI Express change hasn't seemed to help either. Any other ideas? >> David -- > > > My system stops working at least once a day. Does anyone have any > suggestions to try? latest module builds? other firmwares? I notice there > are several up on the http://www.steventoth.net/linux/hvr22xx/ site, would > any of those work better? > > Is there a better card to use? the HVR-2200 and saa7164 are effectively > unusable at this point. I'm planning to release an updated saa7164 driver shortly, with support for the HVR2205 and the HVR2255. Additionally, it contains a fix (which for some people) has significantly reduced the I2C Timeout related errors. (Technically the risc processor crashes on the card, leading to all firmware commands timing out). I'll post a notice to this list once the driver is ready for download. Shortly, I hope. - Steve -- Steven Toth - Kernel Labs http://www.kernellabs.com -- To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: [BUG, workaround] HVR-2200/saa7164 problem with C7 power state
On 02/17/2015 04:47 AM, dCrypt wrote: So the PCI Express change hasn't seemed to help either. Any other ideas? David -- My system stops working at least once a day. Does anyone have any suggestions to try? latest module builds? other firmwares? I notice there are several up on the http://www.steventoth.net/linux/hvr22xx/ site, would any of those work better? Is there a better card to use? the HVR-2200 and saa7164 are effectively unusable at this point. Thanks, David -- To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
RE: [BUG, workaround] HVR-2200/saa7164 problem with C7 power state
Hi, Forcing C-state to C6 doesn't help either. It's just the errors show up in the log later. Last reboot was the 12th of February, and I started getting errors this morning and VDR shutdown (five days working without issues). Find the log attached. I am testing again with the next C-state bios option, C1, in order to check if the problem is related to the C-states or not. Tested in Ubuntu 14.04 with 3.13.0-45-generic kernel, Asrock Q1900-M board. BR > -Mensaje original- > De: linux-media-ow...@vger.kernel.org [mailto:linux-media- > ow...@vger.kernel.org] En nombre de catchall > Enviado el: martes, 17 de febrero de 2015 4:58 > Para: v...@dharty.com; dcr...@telefonica.net; st...@kernellabs.com > CC: linux-media@vger.kernel.org > Asunto: Re: [BUG, workaround] HVR-2200/saa7164 problem with C7 power > state > > On 02/12/2015 03:38 PM, David Harty wrote: > > I hadn't changed the PCI Express Configuration to Gen1 because per > the > > http://whirlpool.net.au/wiki/n54l_all_in_one page it didn't appear to > > help reliably. I've made that change now. I'll report to see if that > > improves anything, perhaps both changes have to be made in > conjunction. > > So the PCI Express change hasn't seemed to help either. Any other > ideas? > > David > > -- > To unsubscribe from this list: send the line "unsubscribe linux-media" > in the body of a message to majord...@vger.kernel.org More majordomo > info at http://vger.kernel.org/majordomo-info.html kern.log Description: Binary data
Re: [BUG, workaround] HVR-2200/saa7164 problem with C7 power state
On 02/12/2015 03:38 PM, David Harty wrote: I hadn't changed the PCI Express Configuration to Gen1 because per the http://whirlpool.net.au/wiki/n54l_all_in_one page it didn't appear to help reliably. I've made that change now. I'll report to see if that improves anything, perhaps both changes have to be made in conjunction. So the PCI Express change hasn't seemed to help either. Any other ideas? David -- To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: [BUG, workaround] HVR-2200/saa7164 problem with C7 power state
On 02/03/2015 05:59 AM, dcr...@telefonica.net wrote: Mensaje original De : st...@kernellabs.com Fecha : 02/02/2015 - 16:39 (GMT) Para : dcr...@telefonica.net CC : linux-media@vger.kernel.org Asunto : Re: [BUG, workaround] HVR-2200/saa7164 problem with C7 power state Basically, it starts working but after a while I get an "Event timed out" message and several i2c errors and VDR shuts down (some hours after reboot). As the web page mentions, I tested downgrading the PCIe bandwith from GEN2 to GEN1 without success. But after playing with different BIOS options, what did the trick was limiting the power-saving C-states. If I select "C7" as the maximum C-state, the card fails as described. After limiting the maximum C-state to "C6", it has been working for a whole weekend. Good feedback on the C7 vs C6 power state, thanks. You are welcome, Steve. Happy to be helpful. I will be at your disposal for testing purposes, if you need. BR -- To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Some additional input: I made similar changes to the bios of my ASRock H87M Pro4. There were multiple settings for CPU C state support. I set the C7 state to disabled and forced selected C6 in the State support dropdown to further force it to C6. Within 3 hours the dmesg was filling up with the no free sequences errors. I hadn't changed the PCI Express Configuration to Gen1 because per the http://whirlpool.net.au/wiki/n54l_all_in_one page it didn't appear to help reliably. I've made that change now. I'll report to see if that improves anything, perhaps both changes have to be made in conjunction. Regards, David -- To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Re: [BUG, workaround] HVR-2200/saa7164 problem with C7 power state
> Mensaje original >De : st...@kernellabs.com >Fecha : 02/02/2015 - 16:39 (GMT) >Para : dcr...@telefonica.net >CC : linux-media@vger.kernel.org >Asunto : Re: [BUG, workaround] HVR-2200/saa7164 problem with C7 power state > >>>Basically, it starts working but after a while I get an "Event timed out" >>>message and several i2c errors and VDR shuts down (some hours after reboot). >>>As the web page mentions, I tested downgrading the PCIe bandwith from GEN2 >>>to GEN1 without success. But after playing with different BIOS options, what >>>did the trick was limiting the power-saving C-states. If I select "C7" as >>>the maximum C-state, the card fails as described. After limiting the maximum >>>C-state to "C6", it has been working for a whole weekend. > >Good feedback on the C7 vs C6 power state, thanks. You are welcome, Steve. Happy to be helpful. I will be at your disposal for testing purposes, if you need. BR -- To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: [BUG, workaround] HVR-2200/saa7164 problem with C7 power state
>>Basically, it starts working but after a while I get an "Event timed out" >>message and several i2c errors and VDR shuts down (some hours after reboot). >>As the web page mentions, I tested downgrading the PCIe bandwith from GEN2 to >>GEN1 without success. But after playing with different BIOS options, what did >>the trick was limiting the power-saving C-states. If I select "C7" as the >>maximum C-state, the card fails as described. After limiting the maximum >>C-state to "C6", it has been working for a whole weekend. Good feedback on the C7 vs C6 power state, thanks. - Steve -- Steven Toth - Kernel Labs http://www.kernellabs.com -- To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: [BUG, workaround] HVR-2200/saa7164 problem with C7 power state
I'm sorry, I resend the message with a descriptive subject > Mensaje original >De : dcr...@telefonica.net >Fecha : 02/02/2015 - 16:04 (GMT) >Para : linux-media@vger.kernel.org >Asunto : > >Hi, > >As I faced problems with my Terratec Cinergy T PCIe Dual and was unable to >solve it (yet), I recently purchased a used Hauppauge HVR-2200 PCIe dual >tuner. I immediately ran into problems with the HVR-2200 as well, perfectly >described here: > >http://whirlpool.net.au/wiki/n54l_all_in_one >(scroll down or search "saa7164_cmd_send() No free sequences") > >Basically, it starts working but after a while I get an "Event timed out" >message and several i2c errors and VDR shuts down (some hours after reboot). >As the web page mentions, I tested downgrading the PCIe bandwith from GEN2 to >GEN1 without success. But after playing with different BIOS options, what did >the trick was limiting the power-saving C-states. If I select "C7" as the >maximum C-state, the card fails as described. After limiting the maximum >C-state to "C6", it has been working for a whole weekend. > >The HVR-2200 was also tested in a GA-H67MA-UD2H-B3 board with an G1610 >Celeron, working without problems (although maximum C-state setup was not >checked). > >Probably the error is present in other saa7164 boards. > >My VDR server is based on BayTrail J1900 (Asrock Q1900M). > >PS: I'm still waiting for advice regarding my Cinergy T/cx23885 problems and >debugging. -- --- dCrypt -- To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html