Jump to content
XPEnology Community

Driver extension jun 1.03b/1.04b for DSM6.2.3 for 918+ / 3615xs / 3617xs


Recommended Posts

 

1 hour ago, mervincm said:

"Clean USB and HDD install. 6.2.3-25426 918+. 104b, v11 extra and extra2, plus 2 files from pat 6.2.3-25426. hw transcoding was working correctly in plex. I swapped the cpu from 8350K to 9600K and now plex will no longer hw transcode." 

 

 

strange

 

root@DS2:~# dmesg | grep drm
root@DS2:~# dmesg | grep i915
root@DS2:~# cat /var/log/dmesg | grep i915
root@DS2:~# cat /var/log/dmesg | grep drm
root@DS2:~# cat /var/log/dmesg | grep DRM
 

no DRM or i915 messages???

 

i have a i3 9100
https://en.wikichip.org/wiki/intel/core_i3/i3-9100

pcie id for the gpu is "3E91"

 

yours

https://en.wikichip.org/wiki/intel/core_i5/i5-9600k

pcie id for the gpu is "3E98"

 

i looked up that number in the driver 915.ko with a editor i only find he UHD 630 id for the 9100, the UHD 630 id of the 9600k is not listed

that might explain why its not found, is not supported in the driver

its the same with jun's i915 driver (slightly newer source), do downgrading would not change anything

 

the 3e98 is also used in other cpu's (that might be avoided

i5-9400
i5-9600k
i7-9700t
i7-9700

 

lucky me, i was thinking about a 9400 instead of 9100 a bought

 

Link to comment
Share on other sites

18 hours ago, deadsquirrel said:

/dev/dri - new driver has /dev/dri/Card0 - 0.10 had /dev/dri/0 interface - same hardware.

i did not change anything in that direction  and its a generic driver, maybe a difference is between the 1st 6.2.3. v25423 and the 2nd 25426?

 

 

18 hours ago, deadsquirrel said:

CPU Frequency governor insmod ./cpufreq_ondemand.ko - FAILS with

insmod ./cpufreq_ondemand.ko
insmod: ERROR: could not insert module ./cpufreq_ondemand.ko: Unknown symbol in module

 

/var/log/messages output.

2020-05-20T09:16:37+10:00 XXXXX kernel: [687147.086168] cpufreq_ondemand: Unknown symbol gov_queue_work (err 0)
2020-05-20T09:16:37+10:00 XXXXX kernel: [687147.086187] cpufreq_ondemand: Unknown symbol cpufreq_governor_dbs (err 0)
2020-05-20T09:16:37+10:00 XXXXX kernel: [687147.086204] cpufreq_ondemand: Unknown symbol dbs_check_cpu (err 0)

 

there might be dependency that is missing or some specific changes in th source by synology that is missing in this module as synology did not used/need it

i will look for a dependency, it its a source problem there is nothing i can do about it

 

 

18 hours ago, deadsquirrel said:

Ironwolf Health Management is also running ok in 0.11 (it did not in 0.10)

lsi sas connected or ahci?

Link to comment
Share on other sites

45 minutes ago, EVOTk said:

My CPU is https://en.wikichip.org/wiki/intel/core_i5/i5-8600k

PCIE ID for my gpu is "3E92"

 

is it included to driver 915.ko with extra 0.11 for DSM 6.2.3 ?

yes, and slight correction, the (new) i915 driver is a "native" driver from synology originally made for geminilake units like the new DS x20+

the 0.11 extra/extra2 makes sure jun's i915 (not compatible with 6.2.3) gets deleted so the native driver from synology is loaded

 

1 hour ago, mervincm said:

Recent plex packages include i965 and iHD drivers.  3E98 is found in both of these (just did a search with a hex editor)

 

i dont think that will help you, without a kernel driver you won't be able to use it

Link to comment
Share on other sites

6 hours ago, yuyuko said:

after I moved 918 to 3617(both are 6.2.3-25426  same version with jun's original extra.lzma),i just found  I can't shutdown in system, don't know why.

thats usually because of drivers crashing an boot, check dmesg so find it

maybe some left overs from 918+ driver set that are still loaded

a safe choice is also to empty /usr/lib/modules/update/ and reboot

the loader will copy all missing files from the extra.lzma on boot back and you get rid of any left overs

 

Link to comment
Share on other sites

12 hours ago, mervincm said:

Recent plex packages include i965 and iHD drivers.  3E98 is found in both of these (just did a search with a hex editor)

that are lib's (.so files are dynamic libraries), independent from the driver and might be newer then the used (backported) driver from synology so they may contain references to newer hardware

i don't think a lib will be working without  a driver and the i915 is the driver for this hardware, its all UHD 630, maybe intel decided on some minor changes and gave it a new number

the name i965 is used since intel graphics gen 4, so its no new standalone kernel driver for any newer hardware

imho differentiating start with x11 and there might be different x11 drivers depending on the intel graphiccs gen (intel, intel 915, intel 965), but the kernel driver is all the same and the base

Link to comment
Share on other sites

On 5/20/2020 at 9:10 PM, IG-88 said:
Quote

2020-05-20T09:16:37+10:00 XXXXX kernel: [687147.086168] cpufreq_ondemand: Unknown symbol gov_queue_work (err 0)
2020-05-20T09:16:37+10:00 XXXXX kernel: [687147.086187] cpufreq_ondemand: Unknown symbol cpufreq_governor_dbs (err 0)
2020-05-20T09:16:37+10:00 XXXXX kernel: [687147.086204] cpufreq_ondemand: Unknown symbol dbs_check_cpu (err 0)

 

there might be dependency that is missing or some specific changes in th source by synology that is missing in this module as synology did not used/need it

i will look for a dependency, it its a source problem there is nothing i can do about it

 

looks like it would need to replace other cpu-frequ driver files  too (coming from synology)

cpufreq_stats.ko

cpufreq_powersave.ko

cpufreq_performance.ko

acpi-cpufreq.ko

that might interfere with other things, so i#m not going to do this in a new extra/extra2 and without any substantial plan what its good for and how to use it im's not goint to invest any more time into this

@deadsquirrel

what exactly is you plan for using this?

i guess it would need more additional code in the system to make use of it

i can send you the files if you want to experiment on your own

Edited by IG-88
Link to comment
Share on other sites

14 hours ago, IG-88 said:

thats usually because of drivers crashing an boot, check dmesg so find it

maybe some left overs from 918+ driver set that are still loaded

a safe choice is also to empty /usr/lib/modules/update/ and reboot

the loader will copy all missing files from the extra.lzma on boot back and you get rid of any left overs

 

thanks.

I did rm -rf /usr/lib/modules/update/ and reboot ,it doesn't work for me.

I also tried let my friends give me all files from modules folder and replace all,doesn't work too.(3617 dsm6.2.3-25426 works on his mechine)

how to use dmesg commands to check errors from logs.

Edited by yuyuko
Link to comment
Share on other sites

8 hours ago, yuyuko said:

0.0,all dmesg log here.

no drivers crashing, some acpi errors i'd need to compare with a test system (i'm testing 315/17 new extra.lzma, so additional work as i'm looking into demsg anyway)

maybe you can try to reduce some acpi / power management features in bios for testing, 3617 has a older kernel then 918+, like using  older or dumped down acpi version

 

maybe disabling HT and using 918+ is a option (when shutdown is working with it)

Edited by IG-88
Link to comment
Share on other sites

59 minutes ago, IG-88 said:

no drivers crashing, some acpi errors i'd need to compare with a test system (i'm testing 315/17 new extra.lzma, so additional work as i'm looking into demsg anyway)

maybe you can try to reduce some acpi / power management features in bios for testing, 3617 has a older kernel then 918+, like using  older or dumped down acpi version

 

maybe disabling HT and using 918+ is a option (when shutdown is working with it)

thx,I'll try with acpi/power bios settings,see if it can fix it.
why i want to use 3617:
1.918 only can detect 4C/8T , 3617 can work with 8C/16T.
2.918 only can active 2 nics there,3617 can active at least 4

XD, hope u have a great weekend.
still looking forward to the day when 3617/15 new extra.lzma  comes out~

Edited by yuyuko
Link to comment
Share on other sites

2 minutes ago, yuyuko said:

2.918 only can active 2 nics there,3617 can active at least 4

thats just a small change in synoinfo.conf that a documented, i use 3 nic's in my 918+ system

i reference to this in the 1st post in the thread about 6.2.2

https://xpenology.com/forum/topic/12679-progress-of-62-loader/page/9/?tab=comments#comment-92682

 

6 minutes ago, yuyuko said:

1.918 only can detect 4C/8T , 3617 can work with 8C/16T.

yes i guessed that from your cpu, but the limit is not about "threads"  ts just cores, the kernel config states 8 cores for 918+ so if you disable HT you can have all 6 cores used in 918+ and would loose only minor performance as HT is far from that much of a gain

Link to comment
Share on other sites

I have a J4105B-ITX board with a dell perc controller flashed to IT mode (LSI 9211-8i). The 918+ 6.2.3 install seems to be using the mpt3sas driver that I thought was only for 12gbs controllers (instead of the mpt2sas). Is this expected?

0000:01:00.0 Serial Attached SCSI controller: Broadcom / LSI SAS2008 PCI-Express Fusion-MPT SAS-2 [Falcon] (rev 03)
        Subsystem: Dell 6Gbps SAS HBA Adapter
        Flags: bus master, fast devsel, latency 0, IRQ 22
        I/O ports at e000 [disabled] [size=256]
        Memory at a1140000 (64-bit, non-prefetchable) [size=64K]
        Memory at a1100000 (64-bit, non-prefetchable) [size=256K]
        Expansion ROM at a1000000 [disabled] [size=1M]
        Capabilities: [50] Power Management version 3
        Capabilities: [68] Express Endpoint, MSI 00
        Capabilities: [d0] Vital Product Data
        Capabilities: [a8] MSI: Enable- Count=1/1 Maskable- 64bit+
        Capabilities: [c0] MSI-X: Enable+ Count=15 Masked-
        Capabilities: [100] Advanced Error Reporting
        Capabilities: [138] Power Budgeting <?>
        Kernel driver in use: mpt3sas

Also, I think I may have found a problem with how the sas controller behaves on 918+. I have a raid-0 volume with 4 drives and a second (basic) volume on a single drive. I can configure HDD hibernation and the drives will enter hibernation fine. However, when I wake them up, the drive with the single volume is fine, but the raid-0 volume ends up with "unable to access system partition" error and a prompt to repair the partition. I think on wake up, the 4 drives making up the raid-0 volume reappear in a different order causing the system partition to fail. I do not have this problem on 6.2.3/3615/1.03b. Any thoughts if the drive order can be fixed?

Edited by richv31
Link to comment
Share on other sites

On 5/21/2020 at 6:27 AM, EVOTk said:

Hello,

My CPU is https://en.wikichip.org/wiki/intel/core_i5/i5-8600k

PCIE ID for my gpu is "3E92"

 

is it included to driver 915.ko with extra 0.11 for DSM 6.2.3 ?

I am currently in DSM 6.2.2-6

Yes, it is - all intel integrated cpu graphics from sandy bridge ix-2xxx to ix-9xxx included, preumable the ix-10xxx as well as they are still UHD620 interfaces. Excluding the 'F' parts, of course (F=disabled GPU on chip)

DeadS

Link to comment
Share on other sites

7 hours ago, deadsquirrel said:

Yes, it is - all intel integrated cpu graphics from sandy bridge ix-2xxx to ix-9xxx included, preumable the ix-10xxx as well as they are still UHD620 interfaces.

read above, to see that you are wrong with that

there are device id's that are not in the driver even when its "only"  UHD630

https://xpenology.com/forum/topic/28321-driver-extension-jun-104b-for-dsm623-for-918/?do=findComment&comment=146656

synology's i915 driver date in dsm 6.2.3: 20171222

you can look up the supported device id in the binary driver with "modinfo" or just use a editor and look for vermagic, that will land you near the place about firmware and devie id's

i added a comment about that and the list of device id's in the 1st post

 

edit: i looked up a new i5-10500 and its device id is "9BC8" and there are no "9xxx" numbers in the driver we use so don't expect any newer gen10 to work even when it "only" has UHD630

https://www.intel.com/content/www/us/en/products/processors/core/i5-processors/i5-10500.html

 

Edited by IG-88
Link to comment
Share on other sites

On 5/22/2020 at 5:19 PM, IG-88 said:

no drivers crashing, some acpi errors i'd need to compare with a test system (i'm testing 315/17 new extra.lzma, so additional work as i'm looking into demsg anyway)

maybe you can try to reduce some acpi / power management features in bios for testing, 3617 has a older kernel then 918+, like using  older or dumped down acpi version

 

maybe disabling HT and using 918+ is a option (when shutdown is working with it)

I've tested with  all ACPI/power settings(with i can see in the pages) on bios,after verious attempts it still can't work T_T 

Link to comment
Share on other sites

0.11 with ASROCK J4105 and 6.2.3 works with HW Transcoding

needed to do the serial number patch first.

 

I could also not access to my old usb stick on win7 and win10 with win32imager, so i needed to create a new one.

Old usb stick is no usable....not that important by may somebody know how to solve it

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...