TinyCore RedPill Loader (TCRP)

[Orphée]

@Dvalin21 @Kaneske @Amrit86 if you already are on a loader with u2 handled by maintainers working (DS3615xs if i'm not wrong) then you can install u3.

There is no kernel (zImage) update/change between u2 and u3.

 

But if you are on a loader like DS3622xs+ not updated to u2 by maintainers, then you can install u2/u3, as it is not patched to work with updated zImage. You will bootloop.

 

At least this is how I understood it.

[synoxpe]

How does the module identification take place?

I'm on Intel NUC 10 Frost Canyon, the motherboard SATA controller module can't be found. How can I build one?

 

It looks that you will need the following modules : 


Found VGA Controller : pciid 8086d00009bca  Required Extension : 
No matching extension
Found SATA Controller : pciid 8086d000002d3  Required Extension : 
No matching extension

 

tc@box:~$ lspci -tnnvq
-[0000:00]-+-00.0  Intel Corporation Device [8086:9b51]
           +-02.0  Intel Corporation Comet Lake UHD Graphics [8086:9bca]
           +-08.0  Intel Corporation Skylake Gaussian Mixture Model [8086:1911]
           +-12.0  Intel Corporation Comet Lake Thermal Subsytem [8086:02f9]
           +-14.0  Intel Corporation Comet Lake PCH-LP USB 3.1 xHCI Host Controller [8086:02ed]
           +-14.2  Intel Corporation Comet Lake PCH-LP Shared SRAM [8086:02ef]
           +-14.3  Intel Corporation Comet Lake PCH-LP CNVi WiFi [8086:02f0]
           +-15.0  Intel Corporation Serial IO I2C Host Controller [8086:02e8]
           +-15.2  Intel Corporation Comet Lake PCH-LP LPSS: I2C Controller #2 [8086:02ea]
           +-16.0  Intel Corporation Comet Lake Management Engine Interface [8086:02e0]
           +-17.0  Intel Corporation Comet Lake SATA AHCI Controller [8086:02d3]
           +-1c.0-[01-39]----00.0-[02-39]--+-00.0-[03]----00.0  Intel Corporation JHL7540 Thunderbolt 3 NHI [Titan Ridge 2C 2018] [8086:15e8]
           |                               +-01.0-[04-38]--
           |                               \-02.0-[39]----00.0  Intel Corporation JHL7540 Thunderbolt 3 USB Controller [Titan Ridge 2C 2018] [8086:15e9]
           +-1d.0-[3a]----00.0  JMicron Technology Corp. JMB58x AHCI SATA controller [197b:0585]
           +-1d.5-[3b]----00.0  Genesys Logic, Inc GL9755 SD Host Controller [17a0:9755]
           +-1f.0  Intel Corporation Comet Lake PCH-LP LPC Premium Controller/eSPI Controller [8086:0284]
           +-1f.3  Intel Corporation Comet Lake PCH-LP cAVS [8086:02c8]
           +-1f.4  Intel Corporation Comet Lake PCH-LP SMBus Host Controller [8086:02a3]
           +-1f.5  Intel Corporation Comet Lake SPI (flash) Controller [8086:02a4]
           \-1f.6  Intel Corporation Ethernet Connection (10) I219-V [8086:0d4f]

 

tc@box:~$ ./rploader.sh satamap now
HBA: 00:17.0 Disks : 3
HBA: 00:1d.0 Disks : 5
SataPortMap=35
DiskIdxMap=0005

 

SataPortMap should be 15 as the 00:17.0 SATA Controller has only one port. Tagging a few experts as this is kind of fundamental to understand: @pocopico @IG-88 @Orphée @WiteWulf @haydibe

 

 

 

Edited March 4, 2022 by synoxpe

[danellis12561]

2 hours ago, Orphée said:

@Dvalin21 @Kaneske @Amrit86 if you already are on a loader with u2 handled by maintainers working (DS3615xs if i'm not wrong) then you can install u3.

There is no kernel (zImage) update/change between u2 and u3.

 

But if you are on a loader like DS3622xs+ not updated to u2 by maintainers, then you can install u2/u3, as it is not patched to work with updated zImage. You will bootloop.

 

At least this is how I understood it.

I'm currently running ds3622xs+ on 7.0.1.  What would be the process if I wanted to get it updated to u3.  It's a bare metal install and I don't want to lose anything on the box.

[Orphée]

1 minute ago, danellis12561 said:

I'm currently running ds3622xs+ on 7.0.1.  What would be the process if I wanted to get it updated to u3.  It's a bare metal install and I don't want to lose anything on the box.

Create your own patch yourself from source with tools given by TTG.

[Tibag]

13 hours ago, Orphée said:

 

If you don't want to use linux command lines at all, go to ESXi. It is easier to manage disk / img upload and deployment with ESXi GUI.

If you don't fear command lines, ssh, etc... Proxmox VE is free and has a better serial port handling for our Xpenology needs. (Actually serial port from network/telnet access is not available without enterprise license on ESXi.)

 

Proxmox has some usefull features like Let's Encrypt certificates, MFA availability. I did not see it on ESXi.

 

It is my personnal experience with only one week of Proxmox usage.

Thanks! I am quite comfortable with bashing stuff but more experienced with esxi. 

 

You've used esxi right? I couldn't get my VM to boot with the vdmk from Redpill, did you do anything specific with the image?

[Orphée]

9 minutes ago, Tibag said:

Thanks! I am quite comfortable with bashing stuff but more experienced with esxi. 

 

You've used esxi right? I couldn't get my VM to boot with the vdmk from Redpill, did you do anything specific with the image?

 

Yes I have to convert the vmdk with Starwind V2V converter (local file vmdk > to local file vmdk > esxi > preallocated)
You will now have a *.vmdk and a *-flat.vmdk file
But both of them on the ESXi folder.

[LaBoss]

Hello i got a i5-3470S on proxmox machine, i use tinycore and all runs fine until i need instal PAT file, when install start i got a error at 50% that says unable to install

 

Could it be a CPU problem?

[Orphée]

3 minutes ago, LaBoss said:

Hello i got a i5-3470S on proxmox machine, i use tinycore and all runs fine until i need instal PAT file, when install start i got a error at 50% that says unable to install

 

Could it be a CPU problem?

Logs or it didn't happen

[IG-88]

6 hours ago, synoxpe said:

HBA: 00:17.0 Disks : 3

HBA: 00:1d.0 Disks : 5

SataPortMap=35

accourding to intel spec its two sata ports (one on m.2 and one as sata connector) and if its more then two then its 4 or 6, never seen a uneven number on a intel chipset

the 5 disks looks ok as its jmb585 and that has 5 ports

6 hours ago, synoxpe said:

DiskIdxMap=0005

that one is a two digit hex for every controller, that would imply 1st one starts as sda, 2nd as sdf (a=0, b=1, ...

in theory (from the hardware)  it would be

SataPortMap=25

DiskIdxMap=0002

00 and 01 are the two ports of the onboard, 02, 03, 04, 05, 06 are the 5 ports of the jmb585

if you want the jmb to to be the 1st controller in counting it would be

SataPortMap=25

DiskIdxMap=0500

being 00-04 the five ports of the jmb and 05, 06 ...

 

SataPortMap is kind of hardware dependent as it gives the controller layout that is used in DiskIdxMap

SataPortMap might change is pcie slots are swapped between controllers or if onboard sata is disabled in bios (usually that only works if the whol onbard controlleris disabled, only disabling just ports of a controller usually does not work, dsm akak linux kernel still sees the hardware capabilities of the controller)

 

edit: seeing a kernel log whare controller drivers are loaded would be nice and might explain why "satamap now" shows three onboard ports

 

 

Edited March 4, 2022 by IG-88

[AlexFullmoon]

A small question. If I'll add another loader extension (e.g. acpid), would it require DSM reinstall through pat file?

[Dvalin21]

10 hours ago, Amrit86 said:

I didn’t do anything particularly special, just downloaded the update from the Synology website, clicked Manual DSM Update and uploaded the file. It warned me that it may take 20mins+ to install and then came back 30secs later. 
 

I should note that I’m using it on a bare metal install, a Dell OptiPlex 7040 SFF and I didn’t add any extra config, drivers or anything of the sort when I built v7.0.1 with RedPill. It just worked

I'm doing baremetal myself. Poweredge T320. But it always go into recovery mode and then loops. I'll try again after running an update.

[scoobdriver]

9 hours ago, smilenkovski said:

So the story on Plex transcoding hw doesn't give much hope for the moment.

I don’t have any issues on 8th or 9th gen test systems 

8th is on esxi with whole igpu passed through. Plex in docker , mapping /dev/dri ( I do chmod /dev/dri with 666 so the Plex docker user has permissions)

9th is on Proxmox igpu passed as gvtg , same setup as above. 
4k hw transcodes no problem 

 

have you ensured Plex user has correct access to /dev/dri ,  ? 
 

 

I use a scheduled task to run this script on startup. 

[IG-88]

9 hours ago, smilenkovski said:

Cpu Intel J5040  not working. (surprisingly  because is closest to 918+)

from ark intel that 0x3184 ant thats already in the original driver, inmply's a firmware problem (check the dmesg log)

the driver has it as GLK_IDS aka Gemini Lake and thats whats in 920+ and thats the reason synology introduced the new driver in 6.2.3 and 918+ got it along that as it shares the kernel source from 4.4

"glk_dmc_ver1_04.bin" would be the one you are looking for, that one is also part of 920+ *.pat file of DSM (together with "bxt_dmc_ver1_07.bin" for apollo lake)

 

 

9 hours ago, smilenkovski said:

All three platforms transcoding well on OMV.

they are not limited to a driver that is from 12/2017 like the one synology uses, i guess they where looking for a older one with gemini lake support to minimize the backporting efforts to kernel 4.4

 

9 hours ago, smilenkovski said:

On the only working platform (8th gen cpu) results are not good. While Video station was transcoding 4K HDR without problem, plex was struggling with the same file. Transcoding hw visible in dashboard of Plex. So, finally, the implementation of Quicksync

intel qsv was never famous for its picture quality, its fast and thats it, i always used cpu and better piplines for offline transcoding, even if qsv was present in my cpu

imho its meant to be used for lower resolutions like 480p/576p and below when accessing over lower bandwidth

 

9 hours ago, smilenkovski said:

So the story on Plex transcoding hw doesn't give much hope for the moment.

we have nvidia support too, now (a few days) with the dva3221 or with nvidia drivers for 6.2.3 (https://xpenology.com/forum/topic/22272-nvidia-runtime-library/)

plex should be able to use it

[synoxpe]

2 hours ago, IG-88 said:

accourding to intel spec its two sata ports (one on m.2 and one as sata connector) and if its more then two then its 4 or 6, never seen a uneven number on a intel chipset

the 5 disks looks ok as its jmb585 and that has 5 ports

that one is a two digit hex for every controller, that would imply 1st one starts as sda, 2nd as sdf (a=0, b=1, ...

in theory (from the hardware)  it would be

SataPortMap=25

DiskIdxMap=0002

00 and 01 are the two ports of the onboard, 02, 03, 04, 05, 06 are the 5 ports of the jmb585

if you want the jmb to to be the 1st controller in counting it would be

SataPortMap=25

DiskIdxMap=0500

being 00-04 the five ports of the jmb and 05, 06 ...

 

SataPortMap is kind of hardware dependent as it gives the controller layout that is used in DiskIdxMap

SataPortMap might change is pcie slots are swapped between controllers or if onboard sata is disabled in bios (usually that only works if the whol onbard controlleris disabled, only disabling just ports of a controller usually does not work, dsm akak linux kernel still sees the hardware capabilities of the controller)

 

edit: seeing a kernel log whare controller drivers are loaded would be nice and might explain why "satamap now" shows three onboard ports

 

 

 

@IG-88
Yes I’ve plugged in a JMB585 PCIe SATA bridge in the M.2 port which has 5 SATA ports so that’s clear.

My question is the single port onboard AHCI SATA controller whose extension can’t be found and it’s showing up as 3 ports.

Edited March 4, 2022 by synoxpe

[scoobdriver]

On 3/3/2022 at 11:57 AM, ducnhat1309 said:

I have just move from baremental to proxmox, i found the way to passthrough GVTg, but nowhere i can find GVTd. Help me please!

Ok @ducnhat1309

for gvtd i used the following in Proxmox grub (replace ids with those of your gpu)

GRUB_CMDLINE_LINUX_DEFAULT="quiet intel_iommu=on iommu=pt video=efifb:off video=vesafb:off video=simplefb:off vfio-pci.ids=8086:3e91"

Blacklist the following in conf file in /etc/modprobe.d/

blacklist snd_hda_intel
blacklist snd_hda_codec_hdmi
blacklist i915

for gvtg which allows you to split the igpu between 2 vms , do not black list the driver and instead add the following to the grub line 

GRUB_CMDLINE_LINUX_DEFAULT="quiet intel_iommu=on i915.enable_guc=2 i915.enable_gvt=1 kvm.ignore_msrs=1 kvm.report_ignored_msrs=0"

 

to simply pass the igpu in Proxmox you can use the gui , and it works for Plex / video transcoding , however if you want to pass the gpu at pci address 00:02 (required for hw face detection), you need to manually add the args: -set as described in the previous post to your vm.conf file , for whichever method. 
hope that makes sense. 

Edited March 4, 2022 by scoobdriver
Face detection

[scoobdriver]

 

Duplicate

Edited March 4, 2022 by scoobdriver
Duplicate

[scoobdriver]

Duplicate 

Edited March 4, 2022 by scoobdriver
Duplicate

[scoobdriver]

apologies duplicate, iPad not showing posted after pressing button

Edited March 4, 2022 by scoobdriver
Duplicate

[scoobdriver]

Duplicate 

Edited March 4, 2022 by scoobdriver
Duplicate

[gazzac72]

On 3/3/2022 at 12:41 AM, Dvalin21 said:

Here's your nic modules. https://www.intel.com/content/www/us/en/support/articles/000005480/ethernet-products.html

 

Try adding igb module. If it doesn't work, try the e1000 and then e1000e

Hi sorry about this ive tried adding each of the modules from pocopico github then loading up but with each module i still lose internet when adding the sas extension if i dont add it i have net but like i said before all drives messed up and doesnt see the m.2 drives leaving it on sataportmap on 58 it does move first drive to slot 1 but then drive 2 disaperes saying nothing in drive 2 this is what i also found when loading up bromolow-7.0.1-42218 cant load up ds617 as it just goes on a loop saying not installed.

 

Found VGA Controller : pciid 8086d00001912  Required Extension :
No matching extension
Found SATA Controller : pciid 8086d0000a102  Required Extension :
No matching extension
Found Ethernet Interface : pciid 8086d000015b8 Required Extension :
No matching extension
Found SATA Controller : pciid 1b21d00000612  Required Extension :
No matching extension
Found Ethernet Interface : pciid 1969d0000e0a1 Required Extension :
No matching extension

[IG-88]

 

13 minutes ago, synoxpe said:

Yes I’ve plugged in a JMB585 PCIe SATA bridge in the M.2 port which has 5 SATA ports so that’s clear.

 

 

13 minutes ago, synoxpe said:

My question is the single port onboard AHCI SATA controller whose extension can’t be found and it’s showing up as 3 ports.

using the m.2 as pcie slot does not change the sata port also available on the m.2 it is there if you use it or not, its at least two sata ports

and my answer is above, did you try the suggested settings?

without a log its hard to say why 3 ports are detected, in the specs are only two ports

https://ark.intel.com/content/www/us/en/ark/products/188809/intel-nuc-10-performance-kit-nuc10i7fnkp.html

maybe disable onboard sata completely in bios and try the tinycore detection again or try

SataPortMap=5

DiskIdxMap=00

 

[Dvalin21]

10 minutes ago, gazzac72 said:

Hi sorry about this ive tried adding each of the modules from pocopico github then loading up but with each module i still lose internet when adding the sas extension if i dont add it i have net but like i said before all drives messed up and doesnt see the m.2 drives leaving it on sataportmap on 58 it does move first drive to slot 1 but then drive 2 disaperes saying nothing in drive 2 this is what i also found when loading up bromolow-7.0.1-42218 cant load up ds617 as it just goes on a loop saying not installed.

 

Found VGA Controller : pciid 8086d00001912  Required Extension :
No matching extension
Found SATA Controller : pciid 8086d0000a102  Required Extension :
No matching extension
Found Ethernet Interface : pciid 8086d000015b8 Required Extension :
No matching extension
Found SATA Controller : pciid 1b21d00000612  Required Extension :
No matching extension
Found Ethernet Interface : pciid 1969d0000e0a1 Required Extension :
No matching extension

I had the same issue trying to install 3615 and when trying to install 3617, some boot folders were not being created during the install so I would get the error corrupted file. Which modules are you installing? I didn't have these issues when I installed 3622. Maybe @pocopico can bring some further insight? Megaraid and the e1000/e1000e shouldn't be having issues with each other. When you config your user_config.json are you using your Mac or the Mac generator? (It's generated with the serial when you run serialgen)

[pocopico]

16 minutes ago, Dvalin21 said:

I had the same issue trying to install 3615 and when trying to install 3617, some boot folders were not being created during the install so I would get the error corrupted file. Which modules are you installing? I didn't have these issues when I installed 3622. Maybe @pocopico can bring some further insight? Megaraid and the e1000/e1000e shouldn't be having issues with each other. When you config your user_config.json are you using your Mac or the Mac generator? (It's generated with the serial when you run serialgen)

 

The extension loading mechanism was not very solid and it came later with the loader. The extension loading script needed a bit more work from the TTG but they didnt do that and after that they went silent. So, many issues are within the loader itself and has nothing to do with the extensions or TCRP. The extensions loading mechanism, just works as designed.

 

Lets hope someone wil continue working on improving the great work of TTG.

 

Edited March 4, 2022 by pocopico

[Dvalin21]

1 minute ago, pocopico said:

 

The extension loading mechanism was not very solid and it came later with the loader. The extension loading script needed a bit more work from the TTG but they didnt and after that they went silent. So many issues are withing the loader itself and has nothing to do with the extensions or TCRP. 

 

 

Lets hope someone wil continue working on improoving the great work of TTG.

 

Appreciate the insight. I hope someone will pick it up. It's an awesome tool. Thanks again to you with all your work as well.

[gazzac72]

17 minutes ago, Dvalin21 said:

I had the same issue trying to install 3615 and when trying to install 3617, some boot folders were not being created during the install so I would get the error corrupted file. Which modules are you installing? I didn't have these issues when I installed 3622. Maybe @pocopico can bring some further insight? Megaraid and the e1000/e1000e shouldn't be having issues with each other. When you config your user_config.json are you using your Mac or the Mac generator? (It's generated with the serial when you run serialgen)

i do serial gen then i edit file and replace it with my mac addy

Edited March 4, 2022 by gazzac72