Drivers requests for DSM 5.2


Recommended Posts

I recompiled the kernel with libata-pmp from 4.2 build:

 

http://xpenology.trantor.be/beta/zImage-sil_v2.7z

 

Your DX510 *should* works with this.

 

Also found that syno_pmp_ports_num return 1 when a non synology esata device is connected. I think this why you (kyleday) only get one drive on your esata box.

I will try to patch sources to enable both (synology and others) esata devices to be properly detected.

Link to post
Share on other sites
I recompiled the kernel with libata-pmp from 4.2 build:

 

http://xpenology.trantor.be/beta/zImage-sil_v2.7z

 

Your DX510 *should* works with this.

YES!!! Now it is working properly.

 

So what is already tested? I turned on dx510 with a mix of 5 different hdd and ssd. All of them are now visible in web-gui. Then I builded SHR volume on those 5 hdds plus the sixth one that was used for booting. Next experiment - restart. Nothing suspicious. Volume survived the reboot. System volume on enclosure's part of raid1 is undamaged.

 

But still I do not understand why the original 4.3 code is not working... My ds1511+ died in last November. It had the latest version of DSM. So it should be 3810 I believe. And there were no problems between its onboard sil3132 and dx510...

Link to post
Share on other sites
I recompiled the kernel with libata-pmp from 4.2 build:

 

http://xpenology.trantor.be/beta/zImage-sil_v2.7z

 

Your DX510 *should* works with this.

YES!!! Now it is working properly.

 

So what is already tested? I turned on dx510 with a mix of 5 different hdd and ssd. All of them are now visible in web-gui. Then I builded SHR volume on those 5 hdds plus the sixth one that was used for booting. Next experiment - restart. Nothing suspicious. Volume survived the reboot. System volume on enclosure's part of raid1 is undamaged.

 

But still I do not understand why the original 4.3 code is not working... My ds1511+ died in last November. It had the latest version of DSM. So it should be 3810 I believe. And there were no problems between its onboard sil3132 and dx510...

 

Coool :grin:

 

Synology add some check in 4.3 source code.

Did your DX510 poweron and/or poweroff when xpenology boot/shutdown ? What about performance ?

 

Next step patch source code to properly detect other esata enclosure.

Link to post
Share on other sites

Still no luck, getting the same "failed to configure port multiplier" message during boot. I want to make sure I'm doing the zimage kernel swap correctly. I'm starting with your 4.3 beta 6 DSM installed, then I browse to the usbshare1 on my network and replace the zimage file, and I have DSM reboot. Is that the correct way to do this?

Link to post
Share on other sites
Still no luck, getting the same "failed to configure port multiplier" message during boot. I want to make sure I'm doing the zimage kernel swap correctly. I'm starting with your 4.3 beta 6 DSM installed, then I browse to the usbshare1 on my network and replace the zimage file, and I have DSM reboot. Is that the correct way to do this?

 

Can you post the full error : failed to configure port multiplier (......

Use dmesg command, and post output about port multiplier

 

Yes this is the good way.

Link to post
Share on other sites

This is an updated beta which include several updates/fixes.

 

Before doing anything, backup your data.

No warranty for any damages, loss of data or anything else.

 

Link : http://xpenology.trantor.be/beta/

 

Update DSM to 4.3-3827 build, please update your system to fix security issues in previous version.

 

Two versions of synoboot : standard and hba (with scsi/sas drivers - only use this if you have a storage controller)

 

 

Known issues:

It will take 10 minutes or longer before it writes the configuration files, you can either cancel the installation or wait for it to finish. If you cancel, run assistant again and right click setup.

 

Changelog:

v7 (24/02/2014)

- Update to 4.3-3827 build
- Synology DX510 is properly detected
- Temporarily removed pata_legacy

v6 (09/02/2014)

- Fix grub error 17 (introduced in beta5)
- Add Mellanox Technologies 10Gbit Ethernet support
- Add QLOGIC QLCNIC 1/10Gb Converged Ethernet support
- Add RocketRAID 2640x4 support (rr26xx driver)

v5 (20/01/2014)

- Add VIA CPU support (CENTAUR)
- Update synobios.ko (by k3dt ; fix "buzzer stop button pressed" warning)
- Add Brocade 1010/1020 10Gb support
- iSCSI (block level) works in IET mode (change lio to iet in synoinfo.cfg)
- iSCSI (regular) works in LIO mode
- NFS working
- CIFS mount point working
- Include bootstrap ipkg setup script (/root.profile is already modified) : after setup just log to ssh and type :
sh syno-i686-bootstrap_1.2-7_i686.xsh
reboot, log again to ssh and type
ipkg update


v4 (29/12/2013)

- You can now install DSM on drives connected to a SCSI/SAS controller
- Fix the volume "broken" issue when DSM is installed on a SCSI/SAS controller
- Fix NFS support
- Update megaraid_sas driver to the latest version
- Modify the way XPEnology's modules are loaded in ramdisk
- Include modules from previous beta
- Add PS2 keyboard support

v3 (21/12/2013)

- Add more drivers for KVM / virtio / virtualization support
- Add patched r8169 driver (realtek chips which are not working with the r8168 driver will work with this one)
- Add latest version of sk98lin driver (Marvell Yukon / SysKonnect SK-98xx/SK-95xx support)
- Update Broadcom bnx2 driver to latest version


v2 (05/12/2013)

- AX88179 Gigabit Ethernet USB 3.0
- Latest driver version for RocketRAID 3xxx/4xxx Controller
- Latest (aacraid) driver version for Adaptec
- Support XFS filesystem (no idea if DSM will automount this)


v1

- Update synobios.ko (by k3dt ; fix "buzzer stop button pressed" warning)


SCSI
----

Highpoint Rocketraid 2720SGL
LSI mpt2sas v17
Adaptec aacraid
Intel C600 Series Chipset SAS


NETWORK
-------

3Com 3c590/3c900, 3cr990
Alteon AceNIC/3Com 3C985/NetGear GA620
QLogic QLA3XXX
SiS 900/7016 SiS190/SiS191


USB
---

USB to serial / Prolific 2303 driver


PATA
----

ATI PATA (PATA_ATIIXP)
AMD/NVidia PATA (PATA_AMD)
CMD64x PATA (PATA_CMD64X)
JMicron PATA (PATA_JMICRON)
Marvell PATA (PATA_MARVELL)
Promise PATA 2027x (PATA_PDC2027X)
Intel SCH PATA (PATA_SCH)
CMD / Silicon Image 680 PATA (PATA_SIL680)
SiS PATA (PATA_SIS)
VIA PATA (PATA_VIA)
Winbond SL82C105 PATA (PATA_WINBOND)
Legacy PATA support (PATA_LEGACY)

Link to post
Share on other sites
Did your DX510 poweron and/or poweroff when xpenology boot/shutdown ? What about performance ?

I was away during last two kernels were posted. :smile: So I am now trying beta7. Am I right? Or there were some changes for test in intermediate versions?

 

Anyway now - beta7 with 3827 (hba variant). All 5 hdd-s in dx510 are visible through web-gui. Shutting down dsm from web-gui - dx510 is off automatically. Ok, now I poweron only main unit with dx510 off. During restart it was not turned on automatically. This is really painful: so it makes unusable both WOL and auto restart after power fail.

 

Performance?.. Really I do not expect much from sil3132... :smile: It is pci-e x1 rev1 controller. So it will slow down even on 3 slow hdd-s. Ok, lets check. The best result after 5 runs:

DiskStation> hdparm -t /dev/sdga

/dev/sdga:
Timing buffered disk reads: 326 MB in  3.00 seconds = 108.53 MB/sec

It is for SAMSUNG HD203WI in dx510. Not bad not good - just expected result.

 

Next test: 1 hdd apart of dx510 and 3 hdd-s within dx510. Sorry, all of them of different size and performance. So I build raid0 volume on this set of hdd-s for test. Then I wait till status changes to "Optimizing file system" and finally to "Normal". By the way - can anyone explain the concrete meaning of "Optimizing file system" status?

 

And again - 5 runs of hdparm. The best one:

DiskStation> hdparm -t /dev/md2       

/dev/md2:
Timing buffered disk reads: 502 MB in  3.01 seconds = 168.23 MB/sec

So it worse than expected. But this is unlikely the driver problem. Probably this is an issue of low performance of hdd-s used for testing.

Link to post
Share on other sites

Hi Trantor i have installed beta 6 on my system under ESXi 5.5..

 

I have my Dell Perc 200H set-up through pass-through to a Vm Client. all drives are detected but i am noticing that the internal transfers are slow.

 

copied 48GB from one volume to another and noticed that the transfer was rather slow.

 

I have attached the drive utilization which seems to be max'd out on Volume 1 which is a SHR raid compromised of 2 x 2TB Sata disks and 1 x 3TB Sata disk. write speed of 66MB seems very slow?

 

what do you make of this? is the normal?

 

Thanks

Capture.JPG.ff58ee637fcc59065fd01bbbc473f5e8.JPG

Link to post
Share on other sites

Hello there,

 

i have Intel Atom D525 CPU on my PC. When starting "XPEnology DS3612xs DSM 4.3 build 3810++ (repack v1.0)" i get the message "This kernel requires an x86-64 cpu, but only detected i686 CPU. Unable to boot - please use a Kernel appropriate for your CPU. I wanted to install in VMWare Workstation 10.

 

Is it possible to make it run with driver support in the vm?

 

Regards, Kurt

Link to post
Share on other sites

Hello,

 

I am still a nood when it come to 3rd party network drivers and VMware. I have VMware 5.5 cd and I have been trying to figure out how to inject my new network drivers into it. I just received a ASRock FM2A88X-ITX+ Mobo that has a Qualcomm Atheros AR8171 network card on it. I have been searching the internet and talking with ASRock tech team to find the drivers to inject into the cd to no avail. Would you be able to help me out? :grin:

Link to post
Share on other sites
Two versions of synoboot : standard and hba (with scsi/sas drivers)

 

Stupid question - do the HP NLxx microservers need the SAS or normal version? I think they use at least a SAS cable...

 

 

HBA if you use an additional Host Bus Adaptor (SAS Controller)..otherwise standard

 

which cable you are using doesn't really matter

Link to post
Share on other sites
Two versions of synoboot : standard and hba (with scsi/sas drivers)

 

Stupid question - do the HP NLxx microservers need the SAS or normal version? I think they use at least a SAS cable...

 

 

HBA if you use an additional Host Bus Adaptor (SAS Controller)..otherwise standard

 

which cable you are using doesn't really matter

Hm-m-m... I am using HBA version on two of my four configurations. 'Cause of presence of sas or esata/pmp controllers respectively.

 

But then my questions is - what is the reason for using standard (non-hba) version? What are the advantages of standard version above the hba-version in non-hba environment?

 

I have only two ideas. First - non-hba kernel utilize little less ram space. Second - hba kernel simply less tested till now and so not recommended for those who does not need its special features right now.

Link to post
Share on other sites
Am I correct when reading the changelog that you didn't add support for the dawicontroller yet?

 

No sorry, can you repost me the driver link.

 

did you already had time to checked the missing "/dev/rtc0" ?

 

When I found some time I will look for this.

 

I have only two ideas. First - non-hba kernel utilize little less ram space. Second - hba kernel simply less tested till now and so not recommended for those who does not need its special features right now.

 

Both correct :smile:

 

I think most people just use the SATA ports on the motherboard and don't need SCSI/SAS drivers.

And yes HBA version is less tested and this is not a original synology feature.

Link to post
Share on other sites

Installing beta7 and pleasantly surprised to see that my mac address was correctly picked up. Do we have to mess with the SN too or is that also taken care of?

 

EDIT: I tell a lie. Turns out that it picked it up correctly before installing DSM, but then changed the MAC after installing pat file and first reboot. As such, it's not picking up the diskstation anymore :sad: Any ideas?

Link to post
Share on other sites

Hello there,

 

i have Intel Atom D525 CPU on my PC an i wanted to install in VMWare Workstation 10.. When starting "XPEnology DS3612xs DSM 4.3 build 3810++ (repack v1.0)" i get the message "This kernel requires an x86-64 cpu, but only detected i686 CPU. Unable to boot - please use a Kernel appropriate for your CPU.

 

I think the CPU is good enough but virtualizing is the problem. Can you tell me some expirience or can i do smething now? Maybe another virtualizing software or settings in VMWare?

 

Regards, Kurt

Link to post
Share on other sites

Hey Trantor,

 

Awesome build. I just upgraded from 4.2-3202 to 4.3-3827 and had to restore all apps but that was a given. I upgraded because I was infected with the dreaded lolz and pwned virus. Anyway, I noticed that the thermal status isn't working like a few other people have reported, but it was working with odie's 3202 build. It shows a constant 40 degrees which I know is incorrect.

 

running the following command shows me the correct cpu temperature.

 

cat /sys/bus/platform/devices/coretemp.0/temp1_input

 

Do you know where synology pulls their thermal status from and could it be just an incorrect reference because of the modified installation? Thanks.

 

My hardware is the following

 

Acer AC100 e3-1260l xeon quad core processor

4gb ram

4 x 3tb wd red drives

Link to post
Share on other sites
Symbiot78 wrote:

Am I correct when reading the changelog that you didn't add support for the dawicontroller yet?

 

 

No sorry, can you repost me the driver link.

 

I have a raidcontroller from Dawicontrol.. ds-624e.. it uses a Marvell 9230 chip.

Any chance for support on this?

 

Just found this for your controller:

http://ubuntu.5.x6.nabble.com/3-8-y-z-e ... 50287.html

http://www.spinics.net/lists/linux-ide/msg46859.html

 

I will patch the kernel with their informations in order to handle Marvell 9230 chip.

 

Hi Trantor..

 

All I have is this..

I tried adding the driver with:

 

echo 1b4b 9230 > /sys/bus/pci/drivers/ahci/new_id

 

which worked.. I think...

 

but as you commented.. it's not reboot proof.. :-|

Link to post
Share on other sites
Guest
This topic is now closed to further replies.