Jump to content
XPEnology Community

DSM 6.2 Loader


jun

Recommended Posts

18 minutes ago, cyberphoy said:

Trying to migrate from 6.1 to 6.2. It seemed to go ok but like others after 5 min or so the network craps out. Most have been able to fix this by deleting the 

.xpenoboot folder in the root drive, however when I ssh in I don't see that folder? Is it hidden or something or am I doing something wrong that i can't see it when using putty and logging in as admin?

 

 

try this to see if the folder exists:

ls -la /

have you installed 6.2 or 6.2.1?

Edited by ideasman69
  • Thanks 1
Link to comment
Share on other sites

45 minutes ago, cyberphoy said:

Trying to migrate from 6.1 to 6.2. It seemed to go ok but like others after 5 min or so the network craps out. Most have been able to fix this by deleting the 

.xpenoboot folder in the root drive, however when I ssh in I don't see that folder? Is it hidden or something or am I doing something wrong that i can't see it when using putty and logging in as admin?

 

I recommend to pick an intel desktop GB adapter. It's native supported and much more reliable then realtek crap and its costs as  little as 5 beers.. 

Link to comment
Share on other sites

On 11/9/2018 at 3:26 PM, ideasman69 said:

Back to the Asrock boards:

 

J3455B-ITX: works

J3455-M: works

J3455-ITX: i915 driver fails which explains why /dev/dri is missing:


===== trigger device plug event =====
[  219.079271] BUG: unable to handle kernel NULL pointer dereference at           (null)
[  219.087187] IP: [<          (null)>]           (null)
[  219.092288] PGD 1783da067 PUD 1783db067 PMD 0
[  219.096825] Oops: 0010 [#1] PREEMPT SMP
[  219.100846] Modules linked in: i915(E+) drm_kms_helper(E) syscopyarea(E) sysfillrect(E) sysimgblt(E) fb_sys_fops(E) cfbfillrect(E) cfbcopyarea(E) cfbimgblt(E) drm(E) fb(E) fbdev intel_agp(E) intel_gtt(E) agpgart(E) video backlight button synoacl_vfs(PO) hfsplus md4 hmac r8168(OE) ixgbe(E) igb(E) i2c_algo_bit e1000e(E) vxlan ip6_udp_tunnel udp_tunnel fuse vfat fat crc32c_intel aesni_intel glue_helper lrw gf128mul ablk_helper arc4 cryptd ecryptfs sha256_generic ecb aes_x86_64 authenc des_generic ansi_cprng cts md5 cbc cpufreq_powersave cpufreq_performance acpi_cpufreq processor cpufreq_stats dm_snapshot dm_bufio crc_itu_t crc_ccitt quota_v2 quota_tree psnap p8022 llc sit tunnel4 ip_tunnel ipv6 zram sg etxhci_hcd rtc_cmos(E) mdio(E) mpt3sas(E) raid_class(E) scsi_transport_sas(E) megaraid_sas(E) megaraid(E) mptctl(E) mptspi(E) mptscsih(E) mptbase(E) scsi_transport_spi(E) megaraid_mbox(E) megaraid_mm(E) vmw_pvscsi(E) BusLogic(E) usb_storage xhci_pci xhci_hcd usbcore usb_common imwz(OE) [last unloaded: apollolake_synobios]
[  219.193595] CPU: 3 PID: 6837 Comm: insmod Tainted: P           OE   4.4.59+ #23824
[  219.201196] Hardware name: To Be Filled By O.E.M. To Be Filled By O.E.M./J3455-ITX, BIOS P1.80 05/21/2018
[  219.210792] task: ffff880178b0e240 ti: ffff880072450000 task.ti: ffff880072450000
[  219.218296] RIP: 0010:[<0000000000000000>]  [<          (null)>]           (null)
[  219.225833] RSP: 0018:ffff880072453950  EFLAGS: 00010202
[  219.231162] RAX: 0000000000000246 RBX: ffff8800727341c0 RCX: 0000000000000010
[  219.238322] RDX: 0000000000000002 RSI: 0000000000000002 RDI: ffff8800727341c0
[  219.245482] RBP: ffff880072453970 R08: 0000000000000000 R09: 0000000000000000
[  219.252641] R10: 0000000000bc5624 R11: 0000000000000001 R12: ffff880072453990
[  219.259803] R13: 0000000000000002 R14: 0000000000000002 R15: ffff880072734ba0
[  219.266962] FS:  00007fb83f4f3700(0000) GS:ffff88017fd80000(0000) knlGS:0000000000000000
[  219.275077] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  219.280847] CR2: 0000000000000000 CR3: 0000000174ffd000 CR4: 00000000003406f0
[  219.288006] Stack:
[  219.290031]  ffffffff8140c409 ffff8800727341c0 ffff880072734000 ffff8800727341c0
[  219.297556]  ffff8800724539b8 ffffffffa059da68 ffff8800724539e7 0000000000000000
[  219.305082]  0000000100000040 ffff88007245398c ffff001000010040 ffff8800724539d0
[  219.312605] Call Trace:
[  219.315076]  [<ffffffff8140c409>] ? i2c_transfer+0x79/0x90
[  219.320594]  [<ffffffffa059da68>] drm_dp_dual_mode_read+0x58/0x80 [drm_kms_helper]
[  219.328194]  [<ffffffffa059dcb2>] drm_dp_dual_mode_detect+0x32/0x150 [drm_kms_helper]
[  219.336121]  [<ffffffffa0683adc>] lspcon_init+0x6c/0x1a0 [i915]
[  219.342121]  [<ffffffffa066b480>] intel_ddi_init+0x2d0/0x4f0 [i915]
[  219.348474]  [<ffffffffa0649537>] intel_modeset_init+0x15e7/0x1a70 [i915]
[  219.355342]  [<ffffffffa0683507>] ? intel_setup_gmbus+0x2e7/0x310 [i915]
[  219.362118]  [<ffffffffa05beccf>] i915_driver_load+0xa0f/0xe00 [i915]
[  219.368637]  [<ffffffffa05c9797>] i915_pci_probe+0x27/0x40 [i915]
[  219.374761]  [<ffffffff812fc85c>] pci_device_probe+0x8c/0x100
[  219.380537]  [<ffffffff813842d1>] driver_probe_device+0x1f1/0x310
[  219.386654]  [<ffffffff81384472>] __driver_attach+0x82/0x90
[  219.392254]  [<ffffffff813843f0>] ? driver_probe_device+0x310/0x310
[  219.398544]  [<ffffffff81382361>] bus_for_each_dev+0x61/0xa0
[  219.404226]  [<ffffffff81383d69>] driver_attach+0x19/0x20
[  219.409643]  [<ffffffff81383993>] bus_add_driver+0x1b3/0x230
[  219.415324]  [<ffffffffa06fa000>] ? 0xffffffffa06fa000
[  219.420483]  [<ffffffff81384c7b>] driver_register+0x5b/0xe0
[  219.426079]  [<ffffffff812fb337>] __pci_register_driver+0x47/0x50
[  219.432243]  [<ffffffffa06fa03e>] i915_init+0x3e/0x45 [i915]
[  219.437926]  [<ffffffff810003b6>] do_one_initcall+0x86/0x1b0
[  219.443608]  [<ffffffff810dfdd8>] do_init_module+0x56/0x1be
[  219.449204]  [<ffffffff810b61ad>] load_module+0x1ded/0x2070
[  219.454798]  [<ffffffff810b3510>] ? __symbol_put+0x50/0x50
[  219.460307]  [<ffffffff810b65b9>] SYSC_finit_module+0x79/0x80
[  219.466082]  [<ffffffff810b65d9>] SyS_finit_module+0x9/0x10
[  219.471679]  [<ffffffff81567444>] entry_SYSCALL_64_fastpath+0x18/0x8c
[  219.478146] Code:  Bad RIP value.
[  219.481540] RIP  [<          (null)>]           (null)
[  219.486723]  RSP <ffff880072453950>
[  219.490238] CR2: 0000000000000000
[  219.493572] ---[ end trace 21ef5019a03d0eeb ]---
[  219.498323] ------------[ cut here ]------------

I've tried:

  • all official versions of the J3455-ITX BIOS including the newly released 1.80
  • flashing the J3455B-ITX and J3455-M BIOSs to the board using an external flashing device
    • the J3455B-ITX BIOS booted but for whatever reason said it was a J3355B-ITX and wouldn't get past post saying the CPU didn't match the BIOS.
    • the J3455-M BIOS also booted but the model string was incorrect
  • all combinations of BIOS settings
  • installing 6.2 using 3615 bootloader, then upgrading to 6.2.1 and the 918+ bootloader
  • changing the GPU priority to PCIe and sticking in a PCIe video card


All of the above failed :(

 

It looks like the original release of the 918+ boot loader works on this board so maybe something to do with the newer i915 driver @jun put in?

 

Hi, I've spent several days on this HW transcoding problem with asrock j3455-itx motherboard. I've read all the post of this thread and some other related threads in this forum. Gotta say this might be a general problem with this motherboard. Basically, it's the i915 driver's problem and it can be searched(/dev/dri missing, i915 driver problem) from other linux forums. Finally I kind of found a way to make /dev/dri appear when install DS918+ DSM 6.2 using j3455-itx. Here is my method(But to be knoest, the performance is poor.)

 

1. My method only work on 1.03a2 + extra.lzma_v0.5(I'm not sure whether the extra.lzma is necessary but I included it when making my usb boot)

    I tested that my method won't work in 1.04b, maybe because the updated i915 driver as @jun stated.

2. When you modify the grub.cfg, changing the vid/pid/SN/MAC, also add a new arg 'i915.alpha_support=1' in the line 'set comon_args_918 = xxx', so it looks like:

 

set common_args_918='syno_hdd_powerup_seq=1 HddHotplug=0 syno_hw_version=DS918+ vender_format_version=2 console=ttyS0,115200n8 withefi elevator=elevator quiet syno_hdd_detect=0 syno_port_thaw=1 i915.alpha_support=1'

 

Install the system(I never tried migratiion) and then the /dev/dri appears and HW transcoding works fine in video station if you also have real SN/MAC. (But the performance is quite poor. I've a 35 min 4K h264 movie. For offling hw transcoding to medium quality, it says about 3 hours to finish. But SW transcoding saied 1hr to finish....sign...)

 

Another thing: DSM 6.2 seems quite slow than DSM 6.1. When I tried in install DSM 6.2, after the screen said 'find.synology.com', it took 5min for DS assistant to find my nas, and 10 min to finish install the DSM. Then the machine reboot, another 6 min before DS assistant finds the ip of my nas.  While for installing DSM 6.1, normally it only took 2 min to finish each step.

 

Besides that, in 6.2, my package center seems different from the others. Because I only got the category 'all' in the left panel,  while from the picture I browsed from the internet, there should be categories like 'recommended', 'back up', 'utility'....

 

Last but not least, shutdown is not working for my 6.2. I have to use the shutdown command in command line to turn off my nas.

 

At first I thought this motherborad would be perfect for dsm 6.2 and ds918+ since its specification is quite close to the genuine one. But now I stayed at DS916+ and DS6.1.7.
 

 

Edited by fenguoerbian
add more detail
  • Like 4
Link to comment
Share on other sites

6 hours ago, fenguoerbian said:

Hi, I've spent several days ...

 

Yes we too ... lol but thanks for your effort. I'm confident that sooner or later it will be added. You forgot to mention what the performance is while streaming 4K content to a mobile device. That makes more sense using your nas for this usage compared to offline convert HEVC / 4K / h265 content to FullHD or so. For that usecase I have a i7 8700K running macOS (hackintosh) with Davinci Resolve and RX Vega64. Nothing beats that speed. I was really interested in the streaming usage. I have some 4K content but in most cases I can't view it because my wifi is not capable to stream that speed sustained to a mobile device. If the Nas can transcode 4K to FullHD I don't have to convert movies anymore. According to the Synology documentation the 916+ and the 918+ are capable doing this.  

Edited by hannibal1969
Link to comment
Share on other sites

12 hours ago, fenguoerbian said:

Hi, I've spent several days on this HW transcoding problem with asrock j3455-itx motherboard.

 

 

Hey. I have a Asrock J4205-itx and with the 1.03a2 Loader the /dev/dri folder was there. without adding the alpha support. But the performance was also bad and it worked only after a reboot. A few minutes after the reboot the hardware transcoding stucked. I will make some test with the alpha support.

 

Edit: I added the alpha Support in the 1.04b loader but the /dev/dir folder still missing. So no luck on the J4205. But it wasn't a fresh install. Just saying🤷‍♂️

Edited by Profiler64
Link to comment
Share on other sites

8 hours ago, hannibal1969 said:

 

Yes we too ... lol but thanks for your effort. I'm confident that sooner or later it will be added. You forgot to mention what the performance is while streaming 4K content to a mobile device. That makes more sense using your nas for this usage compared to offline convert HEVC / 4K / h265 content to FullHD or so. For that usecase I have a i7 8700K running macOS (hackintosh) with Davinci Resolve and RX Vega64. Nothing beats that speed. I was really interested in the streaming usage. I have some 4K content but in most cases I can't view it because my wifi is not capable to stream that speed sustained to a mobile device. If the Nas can transcode 4K to FullHD I don't have to convert movies anymore. According to the Synology documentation the 916+ and the 918+ are capable doing this.  

Quality set to auto, on my ios in the same local network, same 4K video I tested in my previous post play 2-3s, freeze 2-3s, play 2-3s, freeze 2-3s.... can't be compared to the performance of SW transcoding

  • Thanks 1
Link to comment
Share on other sites

1 hour ago, Hostilian said:

That's crying out for LACP!  :)

Yes and LACP is crying out with 10Gbit, read please ... JBOD with 3 old disks... so we have a bottleneck here. And although its a little offtopic as is it concerns to LACP, as far as I know, its MAC address / session based. So you can hook up 2 NICs and bond them together with LACP but what about the other side ... the one PC you are working on is connected with 1Gbit? Even when the PC is hooked up with LACP too I don't think it wil utilise the 2Gbit bandwidth whit that nas. Only initiating two transfers simultaneously maybe. Maybe its much improved but that's what I understood in the past.  

  • Like 1
Link to comment
Share on other sites

18 minutes ago, fenguoerbian said:

Quality set to auto, on my ios in the same local network, same 4K video I tested in my previous post play 2-3s, freeze 2-3s, play 2-3s, freeze 2-3s.... can't be compared to the performance of SW transcoding

 

I had same experience. I'm interested if someone knows how this works on a real 918+ anyway.

Link to comment
Share on other sites

18 hours ago, hannibal1969 said:

Yes and LACP is crying out with 10Gbit, read please ... JBOD with 3 old disks... so we have a bottleneck here. And although its a little offtopic as is it concerns to LACP, as far as I know, its MAC address / session based. So you can hook up 2 NICs and bond them together with LACP but what about the other side ... the one PC you are working on is connected with 1Gbit? Even when the PC is hooked up with LACP too I don't think it wil utilise the 2Gbit bandwidth whit that nas. Only initiating two transfers simultaneously maybe. Maybe its much improved but that's what I understood in the past.  

Why so seriousssssssss. Jeez, it was said in jest..

 

Ok.. Being serious then.. Your disks CAN obviously pump out more - if they're bouncing around at the limits of 1GB Ethernet.. So the bottleneck is obviously the NIC - not the disks.

Depends on what you use it for though..
I have LACP bonds between my various NAS boxes  and whilst it might not bring it much above 113MB/s on an individual transfer, having LACP definitely helps when you have other things going to the box (camera streams) or other people using it - which I do..

Having camera streams (even only a few MB/s) cuts down MY peak transfer rate by about 40% (to about 70MB/s). With LACP it still hits 113MB/s..

Link to comment
Share on other sites

1 hour ago, Hostilian said:

Why so seriousssssssss. Jeez, it was said in jest..

 

Ok.. Being serious then.. Your disks CAN obviously pump out more - if they're bouncing around at the limits of 1GB Ethernet.. So the bottleneck is obviously the NIC - not the disks.

Depends on what you use it for though..
I have LACP bonds between my various NAS boxes  and whilst it might not bring it much above 113MB/s on an individual transfer, having LACP definitely helps when you have other things going to the box (camera streams) or other people using it - which I do..

Having camera streams (even only a few MB/s) cuts down MY peak transfer rate by about 40% (to about 70MB/s). With LACP it still hits 113MB/s..

Ah so my setup with jbod and a camera stream without lacp with a sustained throughput of 113MB is actually a very good result? It is more reasonable to think that it can be tuned with the intel NIC settings such as tcp offloading etc. 

 

My reply maybe to serious, was made because I don’t see the real benefit to go for lacp in a private setup. Simultaneous writes from different processes are killing for spinning disks anyway. The max. IOPs per disk is 100 so you will reach this quite soon. I think if you make two pools with 2 disks and you write to two different pools you will sustain the NIC. If you have one pool with 4 disks in jbod you won’t. Lacp won’t help you on this I think. 

Link to comment
Share on other sites

if you have one of the Asrock J3455 boards - check out these disk performance benchmark results comparing the 3615 and 918+ boot loaders using the same hardware and same 4 drive RAID5 volume:

 

1.0.4b 918+ / DSM 6.2.1:

On 11/10/2018 at 11:15 PM, ideasman69 said:

with 4 drives in RAID5 on the J3455B-ITX:

1073741824 bytes (1.1 GB) copied, 7.00778 s, 153 MB/s

 

with 4 drives in RAID5 on the J3455M:

1073741824 bytes (1.1 GB) copied, 7.98305 s, 135 MB/s

 

1.0.2b 3615 / DSM 6.1.7:

Just now, ideasman69 said:

with the same hardware and same 4 drive RAID5 volume but using the 1.02b 3615 bootloader and DSM 6.1.7:

1073741824 bytes (1.1 GB) copied, 1.17239 s, 916 MB/s

 

why is it so slow? who knows! maybe these boards just suck.

 

Link to comment
Share on other sites

12 hours ago, hannibal1969 said:

My reply maybe to serious, was made because I don’t see the real benefit to go for lacp in a private setup. Simultaneous writes from different processes are killing for spinning disks anyway. The max. IOPs per disk is 100 so you will reach this quite soon. I think if you make two pools with 2 disks and you write to two different pools you will sustain the NIC. If you have one pool with 4 disks in jbod you won’t. Lacp won’t help you on this I think. 

All of my disks (on this machine) are single disks; I don't need any sort of fault tolerance here..

Pretty sure your camera (I have three) wasn't streaming when you were getting 113MBs**..

.....But then, you could have just said it wouldn't help you for your setup - and left it at that.. Instead, you decided to be an arse. 😉

 

**Unless you have some sort of magic NAS..

Edited by Hostilian
Link to comment
Share on other sites

16 hours ago, ideasman69 said:

if you have one of the Asrock J3455 boards - check out these disk performance benchmark results ... s just suck.

 

 

Ideasman,

 

Is this just internal read and copy from the raid5 array? Maybe it has to do with the way memory caching is handled. Maybe write cache is enabled in the below version.

 

 

 

Link to comment
Share on other sites

What about this one? https://tweakers.net/pricewatch/704203/gigabyte-ga-h270n-wifi/specificaties/

6 sata ports on an itx board. What would be the best CPU candidate with the dev/dri support?

I think it will beat the performance of the j3455 on all points (6 disks, 2x intelNICs) with indeed a little more power consumption. But who cares with solar panels ;-)

 

 

 

Link to comment
Share on other sites

8 hours ago, simoareva said:

Are there any news on progress about the modded loader for 3615xs / 3617xs? My Gen8 is getting sad :(

 

First of all, I'd like to thank Jun for your awesome hard work and yes I would like to know that as well. 

Again thank you all for your hard work.

 

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...