Black screen and blinking cursor after vesamenu



  • @Sebastian-Roth said in Black screen and blinking cursor after vesamenu:

    @Luc-Novales Can you tell us more about this old hardware. What make and model is this?

    This model is an IEI AFL-07A-LX : a panel PC used in an operational equipment.

    System is detected as :

    dmidecode: System Information
    dmidecode: 	Manufacturer: ICP / iEi
    dmidecode: 	Product Name: H605
    dmidecode: 	Version: V1.00
    ...
    dmidecode: Base Board Information
    dmidecode: 	Manufacturer: ICP / iEi
    dmidecode: 	Product Name: E089
    dmidecode: 	Version: V1.00
    
    

    This model is AMD Geode™ LX 800 500 MHz Processor
    It is detected under linux as :

    /proc/cpuinfo: processor	: 0
    /proc/cpuinfo: vendor_id	: AuthenticAMD
    /proc/cpuinfo: cpu family	: 5
    /proc/cpuinfo: model		: 10
    /proc/cpuinfo: model name	: Geode(TM) Integrated Processor by AMD PCS
    /proc/cpuinfo: stepping	: 2
    /proc/cpuinfo: cpu MHz		: 498.042
    /proc/cpuinfo: cache size	: 128 KB
    /proc/cpuinfo: physical id	: 0
    /proc/cpuinfo: siblings	: 1
    /proc/cpuinfo: core id		: 0
    /proc/cpuinfo: cpu cores	: 1
    /proc/cpuinfo: apicid		: 0
    /proc/cpuinfo: initial apicid	: 0
    /proc/cpuinfo: fdiv_bug	: no
    /proc/cpuinfo: f00f_bug	: no
    /proc/cpuinfo: coma_bug	: no
    /proc/cpuinfo: fpu		: yes
    /proc/cpuinfo: fpu_exception	: yes
    /proc/cpuinfo: cpuid level	: 1
    /proc/cpuinfo: wp		: yes
    /proc/cpuinfo: flags		: fpu de pse tsc msr cx8 sep pge cmov clflush mmx mmxext 3dnowext 3dnow cpuid 3dnowprefetch vmmcall
    /proc/cpuinfo: bugs		: sysret_ss_attrs spectre_v1 spectre_v2 spec_store_bypass
    /proc/cpuinfo: bogomips	: 996.08
    /proc/cpuinfo: clflush size	: 32
    /proc/cpuinfo: cache_alignment	: 32
    /proc/cpuinfo: address sizes	: 32 bits physical, 32 bits virtual
    /proc/cpuinfo: power management:
    /proc/cpuinfo: 
    
    

    and dmidecode defines CPU as 80486 family

    dmidecode: Processor Information
    dmidecode: 	Socket Designation: Geode LX 
    dmidecode: 	Type: Central Processor
    dmidecode: 	Family: 80486
    dmidecode: 	Manufacturer: AMD
    dmidecode: 	ID: A2 05 00 00 3D A9 88 00
    dmidecode: 	Signature: Type 0, Family 5, Model 10, Stepping 2
    dmidecode: 	Version: Geode LX
    dmidecode: 	Voltage: 1.2 V
    dmidecode: 	External Clock: 64 MHz
    dmidecode: 	Max Speed: 500 MHz
    dmidecode: 	Current Speed: 500 MHz
    dmidecode: 	Status: Populated, Enabled
    dmidecode: 	Upgrade: ZIF Socket
    dmidecode: 	L1 Cache Handle: 0x0004
    dmidecode: 	L2 Cache Handle: 0x0005
    dmidecode: 	L3 Cache Handle: No L3 Cache
    
    

    BIOS information is :

    dmidecode: BIOS Information
    dmidecode: 	Vendor: Phoenix Technologies, LTD
    dmidecode: 	Version: 6.00 PG
    dmidecode: 	Release Date: 05/18/2010
    dmidecode: 	Address: 0xE0000
    dmidecode: 	Runtime Size: 128 kB
    dmidecode: 	ROM Size: 512 kB
    dmidecode: 	Characteristics:
    dmidecode: 		ISA is supported
    dmidecode: 		PCI is supported
    dmidecode: 		PNP is supported
    dmidecode: 		APM is supported
    dmidecode: 		BIOS is upgradeable
    dmidecode: 		BIOS shadowing is allowed
    dmidecode: 		Boot from CD is supported
    dmidecode: 		Selectable boot is supported
    dmidecode: 		BIOS ROM is socketed
    dmidecode: 		EDD is supported
    dmidecode: 		5.25"/360 kB floppy services are supported (int 13h)
    dmidecode: 		5.25"/1.2 MB floppy services are supported (int 13h)
    dmidecode: 		3.5"/720 kB floppy services are supported (int 13h)
    dmidecode: 		3.5"/2.88 MB floppy services are supported (int 13h)
    dmidecode: 		Print screen service is supported (int 5h)
    dmidecode: 		8042 keyboard services are supported (int 9h)
    dmidecode: 		Serial services are supported (int 14h)
    dmidecode: 		Printer services are supported (int 17h)
    dmidecode: 		CGA/mono video services are supported (int 10h)
    dmidecode: 		ACPI is supported
    dmidecode: 		LS-120 boot is supported
    
    

    Others information as memory… in attached hardware-summary.txt.

    I posted in the iPXE forum to see if they can help us debugging this. So far I have only done debugging in the higher level code and don’t know where to start here in the very early assembly code: http://forum.ipxe.org/showthread.php?tid=18855

    Thanks,
    Luc.


  • Developer

    @Luc-Novales Can you tell us more about this old hardware. What make and model is this?

    I posted in the iPXE forum to see if they can help us debugging this. So far I have only done debugging in the higher level code and don’t know where to start here in the very early assembly code: http://forum.ipxe.org/showthread.php?tid=18855



  • Hi Sebastian,

    @Sebastian-Roth said in Black screen and blinking cursor after vesamenu:

    @Luc-Novales Ok, great you posted such detailed information an pictures of the screens. That helps a lot!

    The first thing that comes to my mind is trying different iPXE binaries. Instead of undionly.kpxe try the following realtek.pxe, realtek.kpxe, realtek.kkpxe, undionly.pxe, undionly.kkpxe, ipxe.pxe, ipxe.kpxe, ipxe.kkpxe. It’s interesting we see it hanging even before the “iPXE initialzing devices…” output.

    Using realtek.pxe, realtek.kpxe, realtek.kkpxe, undionly.pxe, undionly.kkpxe, ipxe.pxe, ipxe.kpxe, ipxe.kkpxe, result is exactly the same as previous screenshot (PXE, entry point, UNDI code and data segments). Only free base memory unload change between 562k, 601k and 630k.

    After acknowledgment of last block of retrieved pxe file, client doesn’t reply to ARP requests from FOG server, and nothing is emitted (http request…).

    The other route we can go is compiling iPXE with debugging enabled but let’s keep that for later.

    Can you tell us more about this old hardware. What make and model is this? Please boot into the operating system from disk and get us some more information about the network card. In Windows you open Device Manager -> select network card -> properties -> see if you can find the so called hardware IDs there. In Linux you simply run lspci -nn | grep -i net and post the full output you get here.
    Windows XP embedded does not allow anything. 😉
    The challenge was to solve kernel panic on boot of GRML due to kernel bug #917569.
    Using debian installer recovery mode, detailed hardware detection is attached
    hardware-summary.txt.

    For network part, here a summary :

    00:0b.0 Network controller [0280]: VIA Technologies, Inc. VT6655 WiFi Adapter, 802.11a/b/g [1106:3253]
    00:0c.0 Ethernet controller [0200]: Realtek Semiconductor Co., Ltd. RTL-8100/8101L/8139 PCI Fast Ethernet Adapter [10ec:8139] (rev 10)
    00:0d.0 Ethernet controller [0200]: Realtek Semiconductor Co., Ltd. RTL-8100/8101L/8139 PCI Fast Ethernet Adapter [10ec:8139] (rev 10)
    
    

    Is there something I can do to debug ?

    Thx,
    Luc.


  • Developer

    @Luc-Novales Ok, great you posted such detailed information an pictures of the screens. That helps a lot!

    The first thing that comes to my mind is trying different iPXE binaries. Instead of undionly.kpxe try the following realtek.pxe, realtek.kpxe, realtek.kkpxe, undionly.pxe, undionly.kkpxe, ipxe.pxe, ipxe.kpxe, ipxe.kkpxe. It’s interesting we see it hanging even before the “iPXE initialzing devices…” output.

    The other route we can go is compiling iPXE with debugging enabled but let’s keep that for later.

    Can you tell us more about this old hardware. What make and model is this? Please boot into the operating system from disk and get us some more information about the network card. In Windows you open Device Manager -> select network card -> properties -> see if you can find the so called hardware IDs there. In Linux you simply run lspci -nn | grep -i net and post the full output you get here.



  • Hi sebastian,
    Thanks for this answer.

    @Sebastian-Roth said in Black screen and blinking cursor after vesamenu:

    @Luc-Novales said in Black screen and blinking cursor after vesamenu:

    IPXE didn’t work out of the box, and I applied Chainloading PXE to iPXE using pxelinux.0.

    The wiki article was edited in 2014 last and I don’t think this is up to date information. Would be really cool if we ever find the time to clean up the wiki and remove such old stuff.

    It help to understant, perhaps it should not be removed 😉

    There have been a lot of changes in iPXE as well for newer hardware and I’d suggest we try to find a solution with current iPXE instead of trying to fix the old pxelinux hack.

    The problem is that it is an old hardware that we could clone with an old version of clonezilla, but impossible with newer versions. Now, we decided to use FOG and we need to debug this PXE boot.

    When you say “IPXE didn’t work out of the box”, what exactly does that mean. Please describe in more detail and take a picture of the screen and post here.

    IPXE boot
    Using “stock” configuration with undionly.kpxe, result is conform to this screenshot :
    Booting from IPXE.

    Client gets undionly.kpxe by TFTP and it stops. It doesn’t reply to ARP request send by the server.

    PXE boot
    When applying Chainloading PXE to iPXE using pxelinux.0.
    The client gets pxelinux.0 and the screen is :
    Booting PXElinux.0

    Then it displays the fugitive vesamenu as below :
    Displaying vesamenu

    The ipxe.krn file is downloaded by the client using TFTP and it stops.
    There is a black screen and blinking cursor, no HTTP request or other communication on the LAN from the client.
    No log in apache2 server.
    Only WOL request are emitted by server.

    Luc.


  • Developer

    @Luc-Novales said in Black screen and blinking cursor after vesamenu:

    IPXE didn’t work out of the box, and I applied Chainloading PXE to iPXE using pxelinux.0.

    The wiki article was edited in 2014 last and I don’t think this is up to date information. Would be really cool if we ever find the time to clean up the wiki and remove such old stuff.

    There have been a lot of changes in iPXE as well for newer hardware and I’d suggest we try to find a solution with current iPXE instead of trying to fix the old pxelinux hack.

    When you say “IPXE didn’t work out of the box”, what exactly does that mean. Please describe in more detail and take a picture of the screen and post here.


Log in to reply
 

475
Online

6.6k
Users

14.0k
Topics

132.1k
Posts