CTPCI and Realtek 8139D

[JoeAtari]

@JoeAtari: lspci (and other "MinT executables") are intended to run in a UNIX-like
environment, which is not part part of FreeMinT snapshot. E.g. lspci should be started
from a shell (sh or bash) and exepts the executable "gzip" in path /bin/ or /usr/bin/ for
extracting the database file "pci_ids.gz". Without this file (or if "gzip" is not found),
lspci only shows numerical values instead of the names and parameters found in database.
I don't know your current MinT installation, so perhaps i explained things you already know...

Latz

I have the full mint install with the UNIX-like environment (I think). I just don't know what I am doing
Here is what I tried:

Code: Select all

root@easymint:/home>exec /home/lspci -vv -i /home/pci.gz
bash: /home/lspci: Permission denied
bash: exec: /home/lspci: cannot execute: Success
root@easymint:/home>

[Latz]

Hi,

I have the full mint install with the UNIX-like environment (I think). I just don't know what I am doing

No problem, my first MinT experiences weren´t much better...

Here is what I tried:

Code: Select all

root@easymint:/home>exec /home/lspci -vv -i /home/pci.gz
bash: /home/lspci: Permission denied
bash: exec: /home/lspci: cannot execute: Success
root@easymint:/home>

Some issues:
- Place lspci in folder /bin/ (or in one of the folders(paths) which are defined in mint.cnf/environment variable PATH). All UNIX
executables are there, according to the users access rights (root or user).
- Check that lspci is set to "executable".
- In bash, you don´t need to type "exec", the first command calls the executable in $PATH.
- You are already in folder /home (root@easymint:/home>), so you don´t need to enter /home once again. So enter:

Code: Select all

lspci -vv -i pci_ids.gz

-vv defines the output level (v, vv or vvv)
-i defines path to database file
- output is displayed in bash

With

Code: Select all

lspci -vv -i pci_ids.gz > pcilist.txt

the output is written as file /home/pcilist.txt

More: With

Code: Select all

man <command>

in bash you get documentation for most executables e.g. "man cp" or "man ping".

I can´t attach the full "pciutils.tar.gz" archive here in forum...

Latz

[wongck]

I have the full mint install with the UNIX-like environment (I think). I just don't know what I am doing

It's a Unix-like environment.... no one knows what they are doing.
Unless the GUI is like blocks of building as in Jurassic Park movie.... "it's Unix I know this"... LOL

[JoeAtari]

I had actually tried all of this with the exception of making lspci executable. I'm learning!

Code: Select all

00:00.0 Bridge: PLX Technology, Inc. PCI9054 32-bit 33MHz PCI <-> IOBus Bridge (rev 0c)
   Subsystem: PLX Technology, Inc. PCI9054 32-bit 33MHz PCI <-> IOBus Bridge
   Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV+ VGASnoop- ParErr+ Stepping- SERR+ FastB2B- DisINTx-
   Status: Cap+ 66MHz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
   Latency: 8 (250ns min, 8000ns max), Cache Line Size: 32 bytes
   Interrupt: pin ? routed to IRQ 255
   Region 0: Memory at ffffff00 (32-bit, non-prefetchable) [size=256]
   Region 2: Memory at 10d80000 (32-bit, non-prefetchable) [size=512K]
   Expansion ROM at <unassigned> [disabled] [size=64K]
   Capabilities: [40] #00 [0000]

00:01.0 VGA compatible controller: ATI Technologies Inc RV280 [Radeon 9200 PRO] (rev 01) (prog-if 00 [VGA controller])
   Subsystem: Hightech Information System Ltd. Excalibur Radeon 9250
   Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR+ FastB2B- DisINTx-
   Status: Cap+ 66MHz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
   Latency: 0 (2000ns min), Cache Line Size: 32 bytes
   Interrupt: pin A routed to IRQ 255
   Region 0: Memory at 40000000 (32-bit, prefetchable) [size=128M]
   Region 1: I/O ports at <unassigned> [size=256]
   Region 2: Memory at 60000000 (32-bit, non-prefetchable) [size=64K]
   Expansion ROM at 60020000 [size=128K]
   Capabilities: [50] Power Management version 2
      Flags: PMEClk- DSI- D1+ D2+ AuxCurrent=0mA PME(D0-,D1-,D2-,D3hot-,D3cold-)
      Status: D0 NoSoftRst- PME-Enable- DSel=0 DScale=0 PME-

00:01.1 Display controller: ATI Technologies Inc RV280 [Radeon 9200 PRO] (Secondary) (rev 01)
   Subsystem: Hightech Information System Ltd. Excalibur Radeon 9250 (Secondary)
   Control: I/O- Mem- BusMaster- SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx-
   Status: Cap+ 66MHz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
   Interrupt: pin ? routed to IRQ 255
   Region 0: Memory at <unassigned> (32-bit, prefetchable) [disabled] [size=128M]
   Region 1: Memory at <unassigned> (32-bit, non-prefetchable) [disabled] [size=64K]
   Capabilities: [50] Power Management version 2
      Flags: PMEClk- DSI- D1+ D2+ AuxCurrent=0mA PME(D0-,D1-,D2-,D3hot-,D3cold-)
      Status: D0 NoSoftRst- PME-Enable- DSel=0 DScale=0 PME-

00:02.0 USB Controller: NEC Corporation USB (rev 43) (prog-if 10 [OHCI])
   Subsystem: Belkin Root Hub
   Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV+ VGASnoop- ParErr+ Stepping- SERR+ FastB2B- DisINTx-
   Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=medium >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
   Latency: 8 (250ns min, 10500ns max), Cache Line Size: 32 bytes
   Interrupt: pin A routed to IRQ 255
   Region 0: Memory at 60040000 (32-bit, non-prefetchable) [size=4K]
   Capabilities: [40] Power Management version 2
      Flags: PMEClk- DSI- D1+ D2+ AuxCurrent=0mA PME(D0+,D1+,D2+,D3hot+,D3cold-)
      Status: D0 NoSoftRst- PME-Enable- DSel=0 DScale=0 PME-

00:02.1 USB Controller: NEC Corporation USB (rev 43) (prog-if 10 [OHCI])
   Subsystem: Belkin Root Hub
   Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV+ VGASnoop- ParErr+ Stepping- SERR+ FastB2B- DisINTx-
   Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=medium >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
   Latency: 8 (250ns min, 10500ns max), Cache Line Size: 32 bytes
   Interrupt: pin B routed to IRQ 255
   Region 0: Memory at 60041000 (32-bit, non-prefetchable) [size=4K]
   Capabilities: [40] Power Management version 2
      Flags: PMEClk- DSI- D1+ D2+ AuxCurrent=0mA PME(D0+,D1+,D2+,D3hot+,D3cold-)
      Status: D0 NoSoftRst- PME-Enable- DSel=0 DScale=0 PME-

00:02.2 USB Controller: NEC Corporation USB 2.0 (rev 04) (prog-if 20 [EHCI])
   Subsystem: Belkin Root Hub
   Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV+ VGASnoop- ParErr+ Stepping- SERR+ FastB2B- DisINTx-
   Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=medium >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
   Latency: 128 (4000ns min, 8500ns max), Cache Line Size: 32 bytes
   Interrupt: pin C routed to IRQ 255
   Region 0: Memory at 60042000 (32-bit, non-prefetchable) [size=256]
   Capabilities: [40] Power Management version 2
      Flags: PMEClk- DSI- D1+ D2+ AuxCurrent=0mA PME(D0+,D1+,D2+,D3hot+,D3cold-)
      Status: D0 NoSoftRst- PME-Enable- DSel=0 DScale=0 PME-

00:03.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL-8139/8139C/8139C+ (rev 10)
   Subsystem: Allied Telesyn International Device c10f
   Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr+ Stepping- SERR+ FastB2B- DisINTx-
   Status: Cap+ 66MHz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort- <TAbort- <MAbort- >SERR+ <PERR+ INTx-
   Latency: 0 (8000ns min, 16000ns max)
   Interrupt: pin A routed to IRQ 255
   Region 0: I/O ports at 0100 [size=256]
   Region 1: Memory at 60042100 (32-bit, non-prefetchable) [size=256]
   Capabilities: [50] Power Management version 2
      Flags: PMEClk- DSI- D1+ D2+ AuxCurrent=0mA PME(D0-,D1+,D2+,D3hot+,D3cold-)
      Status: D0 NoSoftRst- PME-Enable- DSel=0 DScale=0 PME-



[piku]

Is there any source code for this anywhere? I ordered an RTL8139 and was sent an RTL8169. Maybe it could work.

EDIT nevermind I see there is source included with the download I have

[wongck]

EDIT nevermind I see there is source included with the download I have

Was that from sometime in 2018?
IIRC, Mikro did another XIF that was working for us. Not sure if he changed any code.

[piku]

Maybe. I do see someone else had an RTL8169 as well. I wonder what changed with the inclusion of gigabit ethernet. I've asked the seller to send me the proper card I ordered. We'll see if it happens

[mikro]

piku: the "fixed" (big quotes here) version is actually based on my work here: https://github.com/mikrosk/freemint/tree/rtl8139

I just shaved a few unnecessary bits but to this day I don't really understand how it could improve anything. ;-)

That reminds me that I should really include this driver into upstream. IIRC I wanted to include most of the calls in the freemint kernel interface but Alan was against it for some reason.

EDIT: Found it: https://mikro.naprvyraz.sk/mint/freemin ... 01206.html

IMHO, we shouldn't rely on some hypothetical futuristic plans and just implement the kernel interface.

[mikro]

That reminds me that I should really include this driver into upstream. IIRC I wanted to include most of the calls in the freemint kernel interface but Alan was against it for some reason.

Nah, what the hell. I have merged it into upstream as is -- clearly it works for somebody better than the original driver and I can always improve it later. So you should find the driver in freemint 020+ snapshots from 1-19-848 onward. Feel free to give it a try, thanks to gcc 7.4.0 I found a couple of minor bugs, too.

[arf]

I have the full mint install with the UNIX-like environment (I think). I just don't know what I am doing

It's a Unix-like environment.... no one knows what they are doing.
Unless the GUI is like blocks of building as in Jurassic Park movie.... "it's Unix I know this"... LOL

That one was hilarious!

[wongck]

So you should find the driver in freemint 020+ snapshots from 1-19-848 onward. Feel free to give it a try, thanks to gcc 7.4.0 I found a couple of minor bugs, too.

Thanks Mikro.
Will try over weekend if I have the time.
One OT question about mint-848.prg... must I have it by this name or can I rename it ?
( i.e. does it uses that 848 to find the mint system folder ?)
I uses the old-school installation method.

[mikro]

One OT question about mint-848.prg... must I have it by this name or can I rename it ?
( i.e. does it uses that 848 to find the mint system folder ?)
I uses the old-school installation method.

Yes, it's a kind of protection against messing your working system up. :) But that applies only to mint.prg, you can copy any other part of the distro wherever you want.

But I do remember about fixing the old school builds. In fact, I'm about 70% done but it's such a boring task I can force myself to work on it only a few hours per week. :-P

[piku]

I still have to read through this thread but it's not working for me. The device is found in the PCI list, the driver loads and the rtk0 device is there. However, when attempting to connect to something after configuring it says that it autonegotiated 100mbps full duplex but otherwise doesn't work and the out-errors counter goes up.

[wongck]

Finally got some time to update my Mint to 848. The rtl8139 .xif works great.
Tested with iPerf on the old 2018 build and new 2019 build, no difference in throughput.
Thanks Mikro

Yes, it's a kind of protection against messing your working system up. But that applies only to mint.prg, you can copy any other part of the distro wherever you want.

This new way of release is actually easier to update, than the old school method, I do not have to rename MINT.PRG in the AUTO. I used to encode versions & architecture in the name so that I can still select the version & 060/030 depending. I don't really know why I kept the 030 version as I don't even remember the last time I used Mint on 030 mode.
I wonder why I had issues with this new method previously.... may be not enough sleep back then.