Jump to content
XPEnology Community

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


Recommended Posts

7 hours ago, jimmyjin1978 said:

Platform: ASUS strix B250 ITX,   i3 8100T, 16G RAM 

PVE v6.2  , 1.04b boot loader for 6.2.3-25426 virtualized

physical passthrough  TPlink 2.5G NIC  ( RTL8125 chip) 

 

I saw RTL8125 is in supported driver list , but I'd no way to recongize my 2.5G NIC when boot with another NIC (virutalized E1000E  NIC ) 

I  checked /usr/lib/modules/update/*, there is no r8125.ko

Is there a typo for r8152 ?

Since a lot of Intel 10th gen CPU MB will use 2.5G NIC and mainly use rtl8125 chip , could rtl8125 also support in future ?

 

i guess you write about 918+?

the files are in all three extra's but i forgot tot add the entry for 918+ in the rc.modules, so it doe not get loaded and the copy process for /lib/modules/updates only compares about what is in rc.modules, so its not get copied to disk (even if its in extra.lzma) - that only apply's to 918+, in 3615/17 everything looks ok

check the 1st thread i a few minutes, i will upload a new 13.1 for 981+

 

 

Link to comment
Share on other sites

21 hours ago, _ReaL_ said:

I updated my 918+ to DSM 6.2.3-25426 using 1.04b with 0.11 extra.lzma and extra2.lzma.

it does work with jun's extra/extra2 and 6.2.3?

it sounds like you had 6.2.3 and the problems started when using the extra.lzma

the only possible source  of the problems seems to be usb but both drivers, dvb and usb, come from synology and i'm not aware that any of these base usb drivers is loaded in rc.modules of the extra.lzma so i guess everything related to usb is natively handled by  DSM

Link to comment
Share on other sites

2 hours ago, IG-88 said:

it does work with jun's extra/extra2 and 6.2.3?

it sounds like you had 6.2.3 and the problems started when using the extra.lzma

the only possible source  of the problems seems to be usb but both drivers, dvb and usb, come from synology and i'm not aware that any of these base usb drivers is loaded in rc.modules of the extra.lzma so i guess everything related to usb is natively handled by  DSM

 

Do you think I should remove extra.lzma and extra2.lzma? or simply replace with the original 1.04b?

Link to comment
Share on other sites

1 hour ago, _ReaL_ said:

Do you think I should remove extra.lzma and extra2.lzma? or simply replace with the original 1.04b?

replace both with the original, i915 driver will not work (/dev/dri) but even if you use it normally its ok for testing

 

what was the situation under it worked before?

Link to comment
Share on other sites

19 hours ago, IG-88 said:

replace both with the original, i915 driver will not work (/dev/dri) but even if you use it normally its ok for testing

 

what was the situation under it worked before?

 

In 6.2.1 I know it was working fine. In the first 6.2.2 I think it worked fine too, but I'm not sure.

Link to comment
Share on other sites

On 6/5/2020 at 12:50 AM, IG-88 said:

 

i guess you write about 918+?

the files are in all three extra's but i forgot tot add the entry for 918+ in the rc.modules, so it doe not get loaded and the copy process for /lib/modules/updates only compares about what is in rc.modules, so its not get copied to disk (even if its in extra.lzma) - that only apply's to 918+, in 3615/17 everything looks ok

check the 1st thread i a few minutes, i will upload a new 13.1 for 981+

 

 

Thanks for the quick fixing!

However , with the new 13.1 version for DS918+ , I can't  boot  with the RTL8125 NIC only , after adding an E1000 virtual NIC , DSM can boot , but only E1000 NIC get the IP address , while for RTL8125 I can see the rate is 2500M in DSM ,  but no IP address assigned , if I mannually assign the IP address ,  the address won't work with the management page or SMB . Reinstall DSM 6.23 get the same result

Then I changed to DS3617 version 11.1 drive,  same resultI can't boot with RTL8125 standalone , adding E1000 can boot and see the RTL8125 in DSM , but no IP address.

 

Today , I install this RTL8125 on my win10 desktop PC , now I am writing this reply with this NIC .

I checked the TP-link website and there is a linux driver release on 20200307, as attached , I'm not sure but may it is useful for you

r8125-9.003.01.tar.bz2

Edited by jimmyjin1978
Link to comment
Share on other sites

6 hours ago, jimmyjin1978 said:

I checked the TP-link website and there is a linux driver release on 20200307, as attached , I'm not sure but may it is useful for you

i checked the realtek website and they even had a slightly newer version

look in the 1st post for 13.2 for 918+

Link to comment
Share on other sites

5 hours ago, IG-88 said:

i checked the realtek website and they even had a slightly newer version

look in the 1st post for 13.2 for 918+

You are light speed!

The new version works just fine on my PVE VM , SMB speed can sustain on 250+MB/s

 

Thanks for the brilliant work!

 

Will this driver also in 6.22 DS918 and DS3617 ?

Link to comment
Share on other sites

5 hours ago, jimmyjin1978 said:

The new version works just fine on my PVE VM , SMB speed can sustain on 250+MB/s

good, anything above 1GBit networks is really a must have for now days nas (i prefer 10G but even 2.5G would be a noticeable step up for must people

 

5 hours ago, jimmyjin1978 said:

Will this driver also in 6.22 DS918 and DS3617 ?

no, i dont see a reason to put anymore work into this (would be better to spend more time in documentation about the state of the drivers for 6.2.3)

kind of a closed chapter now as synology is back to "normal" with 6.2.3  (and having three different extra version for 918+ was not really that nice, having a "universal" i915 driver make things much easier)

i will do the same realtek drivers for 3615 and 3617 and 6.2.3 this weekend

 

Edited by IG-88
  • Like 1
Link to comment
Share on other sites

Le 01/06/2020 à 11:57, IG-88 a dit :

 

there is more. you can get a cheap reflashed oem version of a lsi 9211-8i or a P420 (or newer) they do support IT/HBA mode

(it has sas connector as P222 and there are also at least two sata 8port controllers with these connectors but both are limited to 1000MB/s as of pcie 2.0 and two lanes)

 

 

I would say a big thank you to IG-88 since he solved my boring issue with the remplacment of my HP P222 with a LSI 9211-8i in IT mode.

 

All my 4 disks are recognized in DSM 6.2.3 25426 and i have been able to configure them fine.

 

I go put my details in "User Reported Compatibility Thread"

Link to comment
Share on other sites

On 6/7/2020 at 4:43 AM, IG-88 said:

good, anything above 1GBit networks is really a must have for now days nas (i prefer 10G but even 2.5G would be a noticeable step up for must people

 

no, i dont see a reason to put anymore work into this (would be better to spend more time in documentation about the state of the drivers for 6.2.3)

kind of a closed chapter now as synology is back to "normal" with 6.2.3  (and having three different extra version for 918+ was not really that nice, having a "universal" i915 driver make things much easier)

i will do the same realtek drivers for 3615 and 3617 and 6.2.3 this weekend

 

I have intel x550-t1 10G NIC which  works perfertly , however 10G NIC's power comsumpiton is a bit high. I use MS04 case with 4 HDD bay  WD Red 3T and RAID5 in DS918+ VM , 10G is kind of overshot while 2.5G is matching on disk speed wise with only PCIe x 1 needed , the whole idle power dropped from 30W to 24W without performance loss.

It seems Intel made some mistake on it's own 2.5G NIC with Gen10 CPU/MB. so most of MB venders provide RTL8125 2.5G NIC instead of intel's on Z490/B460 MB. 

Link to comment
Share on other sites

Platform:  ASUS P9A-I/C2750/SAS/4L/  3615xs

Marvell 88SE948x   Drive not recognized  

Normal identification of the main board's own SATA

NAS:/$ lspci -k | grep 'Kernel driver'
        Kernel driver in use: pcieport
        Kernel driver in use: pcieport
        Kernel driver in use: pcieport
        Kernel driver in use: pcieport
        Kernel driver in use: igb
        Kernel driver in use: igb
        Kernel driver in use: igb
        Kernel driver in use: igb
        Kernel driver in use: ehci-pci
        Kernel driver in use: ahci
        Kernel driver in use: lpc_ich

Array driver not loaded

Is there a solution? Thank you

dmesg

Link to comment
Share on other sites

On 6/11/2020 at 11:35 AM, djkjd said:

Platform:  ASUS P9A-I/C2750/SAS/4L/  3615xs

Marvell 88SE948x   Drive not recognized  

Normal identification of the main board's own SATA

 

thats your device in question (there are two) and in theory mvsas.ko should handle it

[Fri Jun  5 00:56:13 2020] pci 0000:02:00.0: [1b4b:9485] type 00 class 0x010400
[Fri Jun  5 00:56:13 2020] pci 0000:02:00.0: reg 10: [mem 0xde140000-0xde15ffff 64bit]
[Fri Jun  5 00:56:13 2020] pci 0000:02:00.0: reg 18: [mem 0xde100000-0xde13ffff 64bit]
[Fri Jun  5 00:56:13 2020] pci 0000:02:00.0: reg 30: [mem 0xde160000-0xde16ffff pref]
[Fri Jun  5 00:56:13 2020] pci 0000:02:00.0: supports D1
[Fri Jun  5 00:56:13 2020] pci 0000:02:00.0: PME# supported from D0 D1 D3hot

did you use my newer 0.11 extra.lzma? because it should work as the pci i'd is in the driver as supported device

id do remember i made a kernel patch one about a pci id or sub id in the 3.10.x kernel

just checked and it was about device 9485 and i added the subdevice 9485 that was missing in mv_init.c of the 3.10..x source, i redid the patch when starting to use the newer kernel source that came out lately (24922)

https://lists.debian.org/debian-kernel/2013/09/msg00184.html

 

can you check about the subdevice id for this device? maybe there is even a different now

in recent kernel the whole section about the 9480/9485 melted down to this

{ PCI_VDEVICE(MARVELL_EXT, 0x9485), chip_9485 }, /* Marvell 9480/9485 (any vendor/model) */

edit: case closed, the board had a different sub device id for the 9485 and the new single line from above is covering all sub device id's

worked in this case and will be in all three versions an next release

 

 

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

34 minutes ago, root616 said:

extra.lzma/extra2.lzma for loader 1.04b ds918+ DSM 6.2.3 v0.13.1 & v0.13.3 cause Seagate IronWolf HDD didn't display S.M.A.R.T data & Warning INFO: Access Error. Roll back to v0.13 S.M.A.R.T works fine.

 

yes i know, did you read the warning about the possible data loss with the lsi sas driver in the 1st post?

if you have to choose between a broken raid and  s.m.a.r.t. info ...? i prefer to not break a users raid with my drivers and there is a alternative if you need s.m.a.r.t., you still can migrate to 3615/3617, the lsi sas driver in there seem to work without problems

 

Link to comment
Share on other sites

7 hours ago, IG-88 said:

 

yes i know, did you read the warning about the possible data loss with the lsi sas driver in the 1st post?

if you have to choose between a broken raid and  s.m.a.r.t. info ...? i prefer to not break a users raid with my drivers and there is a alternative if you need s.m.a.r.t., you still can migrate to 3615/3617, the lsi sas driver in there seem to work without problems

 

yep, I'm awared the possible data loss,but it looks good.LOL

Link to comment
Share on other sites

13 hours ago, root616 said:

Roll back to v0.13 S.M.A.R.T works fine.

the last version with smart for 918+ should be 0.12 (when it comes to te files i used in the extra/extra2)

 

5 hours ago, root616 said:

yep, I'm awared the possible data loss,but it looks good.LOL

your cpu is a E3-1240L v5, no intel quick sync video / i915 support, as long as you dont use nvme cache there would be not much different when using 3617

Link to comment
Share on other sites

20 minutes ago, IG-88 said:

the last version with smart for 918+ should be 0.12 (when it comes to te files i used in the extra/extra2)

 

your cpu is a E3-1240L v5, no intel quick sync video / i915 support, as long as you dont use nvme cache there would be not much different when using 3617

v0.13 with smart data works on my build, I've tested. but still update to v13.3, mount synologyboot2 replace 2 extra images with SCP easy to test, NVME support needed, so 918+ it is.

Thx for your reply & driver dev work.

Found this Synology Open Source Project https://sourceforge.net/projects/dsgpl/files/ is there any way load driver as insmod modules through bzImage to fix the problem. I'm a noob in linux , don't know if this is possible, i hope it's not offensive.

Link to comment
Share on other sites

1 hour ago, root616 said:

 is there any way load driver as insmod modules through bzImage to fix the problem.

as we dont have the ability to build ouer own kernel and dont know exactly what synology does as we dont have the recent kernel source to build modules (or even the recent config the kernel was made with) its hard to know, i'm not a software developer and have not much ability to find out why that happens and trace it back

beside this, there are limits of time i can invest into this

its easier to choose the way synology gives like using native hardware and drivers, my new build is not using the 9211-8i anymore, switched to JMB585 to have all ahci

ahci is part of dsm as its used in nearly all systems and they might have a hard time to shut other out my modding the kernel source as they do (like no sata_ or pata_ modules since they modded the drivers they need - but i guess this one could be solved with some coding skills, but no one made the effort yet, we do have kernel source 24922 for 3 month new that could be used)

 

1 hour ago, root616 said:

 I'm a noob in linux , don't know if this is possible, i hope it's not offensive.

if you know what bzImage is good for then you are not a noob

 

 

2 hours ago, root616 said:

v0.13 with smart data works on my build,

that would be interesting as this already contains the modules i compiled with 24922 source and all other 13.x version use the same kernel modules

i'd nee to check that, only difference can be in rc.modules i guess

Link to comment
Share on other sites

Hi @IG-88
I've been out last days and this thread and your great drivers change quickly. Now it's easy to get confused among versions.
J4105B with 9211-8i IT here. DSM 6.2.3-25426 918+. 1.04b loader. extra918plus_v0.11.
Smart Info and SHR volumes (BTFRS) seem to work flawlessly. 25 days now. 😀
Just to be sure, is extra918plus_v0.11 unsafe? Should I sacrifice smart info and migrate to extra918plus_v0.13.3?
Thanks a lot for your brilliant work.

Link to comment
Share on other sites

2 hours ago, Eduardo said:

Just to be sure, is extra918plus_v0.11 unsafe? Should I sacrifice smart info and migrate to extra918plus_v0.13.3?
Thanks a lot for your brilliant work.

i was able to reproduce broken raids when hdd hibernation is active, not all disk wake up properly and will result in problems

initially @richv31 discovered the problem, i might not have recognized it as i usually don't wait that long to to see if a disk goes into sleep (and in most tests i dont have 3 or more disks, often one or two and sometimes not even as raid)  and my own new 918+ system does not use the sas controller anymore, i was going for "native" and that's ahci for 918+ (and that's also working for the other two)

with the new driver made from 24922 kernel source or lsi's newer source there is no broken raid, the disk seem not to go inter hibernation either, at least not distinguishable, the log show hibernation and instant wake up within a very short time after hibernation, so the effect might only come from the fact that the disk in real life never really switches off and never reaches the state we see with jun's drivers

 

also there was feedback about 0.13 still showing smart but it already is made with the new drivers - have'nt checked this as i was busy getting nvidia drivers to test so they hopefully can be used independently from intel qsv for transcoding (on all three images)

Link to comment
Share on other sites

11 hours ago, IG-88 said:

i was able to reproduce broken raids when hdd hibernation is active, not all disk wake up properly and will result in problems

initially @richv31 discovered the problem, i might not have recognized it as i usually don't wait that long to to see if a disk goes into sleep (and in most tests i dont have 3 or more disks, often one or two and sometimes not even as raid)  and my own new 918+ system does not use the sas controller anymore, i was going for "native" and that's ahci for 918+ (and that's also working for the other two)

with the new driver made from 24922 kernel source or lsi's newer source there is no broken raid, the disk seem not to go inter hibernation either, at least not distinguishable, the log show hibernation and instant wake up within a very short time after hibernation, so the effect might only come from the fact that the disk in real life never really switches off and never reaches the state we see with jun's drivers

 

also there was feedback about 0.13 still showing smart but it already is made with the new drivers - have'nt checked this as i was busy getting nvidia drivers to test so they hopefully can be used independently from intel qsv for transcoding (on all three images)

Understood.

I don't think my disks go to sleep. It really doesn't worry me from some years ago.

At this point, I'm not sure what to do. I'll wait some other feedback before making a decision.

Edited by Eduardo
Link to comment
Share on other sites

DS918+ 6.2.3 25426

LSI-9207 IT mode with 4  4T Seagate IronWolf HDD(raid5)
issue:

all drives can be detected in dsm system ,but system keeps showing :Volumne 2  Warning:This storage has entered the warning status.Please go to the HDD/SSD page to check the drives with issues.

when I checked in that page ,I just found those 4 HDD don't have S.M.A.R.T values 

IS that what u said in edit2 ?
and except reinstall/move to 3617 ,is there any way to fix it ?

Link to comment
Share on other sites

1 minute ago, yuyuko said:

DS918+ 6.2.3 25426

LSI-9207 IT mode with 4  4T Seagate IronWolf HDD(raid5)
issue:

all drives can be detected in dsm system ,but system keeps showing :Volumne 2  Warning:This storage has entered the warning status.Please go to the HDD/SSD page to check the drives with issues.

when I checked in that page ,I just found those 4 HDD don't have S.M.A.R.T values 

IS that what u said in edit2 ?
and except reinstall/move to 3617 ,is there any way to fix it ?

based on my tests, you can try v0.13 driver package

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...