Pausing durring boot/compile requiring keystroke to continue

[tamusroyce]

!FIXED... https://forums.gentoo.org/viewtopic-t-572613-highlight-speakup.html...


Hello...

This is my first post, so hopefully it is helpful to others as much as me... And that it isn't too noobie-ish, though I am one...

But my problem is that the startup pauses randomly, and the only way I can continue to get to the login is to press a key. It pauses anywhere between once to three times (I think the same places everytime, too).

It also pauses requently durring compiles (also requiring a key being pressed to continue). I have overcome this problem in the past (another computer) by being careful on what I checkmark while I compile the kernel, but this lost many features I wanted to use. It would require as many as 15 compiles to determine what I have checked in this current kernel which causes this.

Also, this is the third computer (both bootstrapped, such as this one, and non-bootstrapped) that I have came across this problem.

1. So basically, all I need is a suggestion or so on what to uncheckmart to stop the pausing from occuring.

2. I hope it isn't a conflict between two or more options which is causing this & that it is easy to diagnose.

3. I would rather not bother someone with step by step examination, so if you know of a thread or website
which would better help me, I would be most greatful.

It never seems to pause though before stage 3 durring the boot, if that is helpful. And disabling my vixie-cron and nf-syslog from rc-update didn't help. I am on the last page of the x86 gentoo handbook, so no X is installed, or other programs not in handbook.

------------------------------------------------------------------------------------------------------------------------------------------
Kernel is v2.6.21-gentoo-r4
gcc is 4.1.2

Computer info is at bottom... Look at bold section for where error is!
------------------------------------------------------------------------------------------------------------------------------------------
Code maturity level options --->[*] Prompt for development and/or incomplete code/drivers

General Setup --->() Local version - append version information to the version string
[*] Automatically append version information to the version string
[*] Support for paging of anonymous memory (swap)
[*] System V IPC
[ ] IPC Namespaces
[*] POSIX Message Queues
[ ] BSD Processing Accounting
[ ] Export task/process statistics through netlink (EXPERIMENTAL)
[ ] UTS Namespaces
[ ] Auditing support
<*> Kernel .config support
[*] Enable access to .config through /proc/config.gz
[*] Kernel->user space relay support (formerly relayfs)
[ ] Initial RAM filesystem and RAM disk (initramfs/initrd) support
[*] Optimize for size (...)
[ ] Configure standard kernel features (for small systems) --->

Loadable Module Support --->[*] Enable loadable module support
[*] Module unloading
[*] Forced Module Unloading
[ ] Module versioning support
[ ] Source checksum for all modules
[ ] Automatic kernel module loading

Block Layer --->[*] Support for Large Block Devices
[ ] Support for tracing block io actions
[ ] Support for Large Single Files
IO Schedulers ---> <*> Anticipatory I/O scheduler
<*> Deadline I/O scheduler
<*> CFQ I/O scheduler
default I/O scheduler (Anticipatory) ---> (X) Anticipatory

Processor type and features ---> [ ] Tickless System (Dynamic Ticks)
[ ] High Resolution Timer Support
[ ] Symmetric multi-processing support
Subarchitecture Type (PC-Compatable) ---> ...
[*] Paravertualization support (EXPERIMENTAL) # had working on other computer
Processor family (K6/K6-II/K6-III) ---> ...
[*] Generic x86 support
[*] HPET Timer Support
Preemtive Model (Voluntary Kernel Preemption (Desktop)) --> ...
[ ] Local APIC support on uniprocessors # would like enabled if possible?
[*] Machine Check Exception
<*> Check for non-fatal errors on AMD Athlon/Duron / Intel Pentium
< > Toshiba Laptop support
< > Dell laptop support
[ ] Enable X86 board specific fixups for reboot
<*> /dev/cpu/microcode - Intel IA32 CPU microcode support
<*> /dev/cpu/*/msr - Model-specific register support
<*> /dev/cpu/*/cpuid - CPU information support
Firmware Drivers ---> < > Bios Enhanced Disk...
< > Bios update support for DELL systems via sysfs
< > Dell System Management Base Driver
High Memory Support (4GB) ---> ...
Memory model (Flag memory) ---> ...
#too much to type, so from here on, just check-marked and selected
[*] MTRR (Memory Type Range Register) support
[*] Enable seccomp to safely compute untrusted bytecode
Timer frequency (250 HZ) ---> ...
[*] kexec system call
[*] Compat VDSO support

Power management options (ACPI, APM) ---> [*] Power Management support #I doubt it's power managment
[*] Driver model /sys/devices/.../power/state files (DEPRECATED)
ACPI (...) Support ---> [*] ACPI Support
...defaults...
APM (...) Support ---> [*] APM (...) BIOS support
[*] Enable console blanking using APM
[*] Use real mode APM BIOS call to power off
CPU Frequency scaling ---> [ ] CPU Frequency scaling
Bus Options (...) ---> [*] PCI support
PCI access mode (Any) --> ...
[*] ISA support
[*] EISA support
[*] Generic PCI/EISA bridge
[*] EISA device name database

Executable file formats --->[*] Kernel support for ELF binaries

Networking --- > Networking Options ---> <*> Packet socket
<*> Unix domain sockets
[*] TCP/IP networking
[*] IP: multicasting
[*] IP: kernel level autoconfiguration
[*] IP: DHCP support
<*> IP: IPsec transport mode
<*> IP: IPsec tunnel mode
<*> INET: socket monitoring interface
[*] TCP: advanced congestion control --->[*] CUBIC TCP
<*> The IPv6 Protocall
<*> IPv6: IPsec transport mode
<*> IPv6: Ipsec tunnel mode
<*> IPv6: IPv6-in-IPv4 tunnel (SIT driver)
QoS and/or fair queuing ---> Stochastic Fairness Queueing

#This must be where my problem is occuring... But too many options
Device Drivers ---> Generic Driver Options ---> [*] Select only drivers that don't need compile-time external firmware
[*] Prevent firmware from being built
Parellel port support ---> [*] PC-style hardware
<*> AX88796 Parallel Port
[*] IEEE 1284 transfer modes
Block Devices ---> <*> Normal floppy disk support
<*> Loopback device support
<*> Ram Disk support
<*> Packet writing on CD/DVD media
ATA/ATAPI/MFM/RLL Support ---> <*> ATA/ATAPI/MFM/RLL support
<*> Enhanced IDE/MFM/RLL disk/cdrom/tape/floppy support
<*> Include IDE/ATA-2 DISK support
[*] Use multi-mode my default
<*> Include IDE/ATAPI CDROM support
<*> SCSI emulation support
[*] IDE ACPI support
<*> generic/default IDE chipset support
<*> PCI IDE chipset support
[*] Generic PCI bus-master DMA support
[*] ALI M15x3 chipset support
[*] SiS5513 chipset support
SCSI Support ---><*> SCSI disk support
<*>SCSI generic support
Multi-device support (RAID and LVM) --->[*] Multiple devices driver support (RAID and LVM)
<*> Device mapper support
IEEE 1394 (FireWire) support ---><*> IEEE 1394 (FireWire) support
<*> OHCI-1394 support
<*> Raw IEEE1394 I/O support
I20 device support ---> <*> I20 support
[*] Enable LCT notifications
[*] Enable Adaptec extensions
Network device support ---> [*] Netowrk device support
<*> Universal TUN/TAP device driver support
Ethernet (10 or 100Mbit) ---> [*] Ethernet (10 or 100Mbit)
[*] EISA, VLB, PCI and on board control
[*] PCI NE2000 and clones
[*] RealTek RTL-8139 c+ PCI
[*] RealTec RTL-8129/8130/8139 PCI
[*] SiS 900/7016 PCI
Tulip ---><*> DECchip Tulip PCI support
[*] PLIP (parallel port) support
[*] PPP (Point-to-Point protocall) support
[*] PPP Deflate compression
[*] SLIP (serial line) support
[*] CSLIP compressed headers
[*] Keepalive and linefill
Telephony support ---> <*> Linux telephony support
<*> QuckNet Internet LineJack/PhoneJack support
Input device support ---> [*] Provide legacy /dev/psaux device
<*>Joystick interface
<*> Event interface
[*] Mouse ---> PS/2 mouse
[*] Joystick ---> <M> all of them
<*> Gameport data dumper
Hardware I/O ports ---> <M> ct82c710 Aux port controller
<*> Raw access to serio ports
<M> Classic ISA and PnP gameport support
<M> SB Live and Audigy gameport support
<M> ForteMedia FM801 gameport support
Character devices --->[*] Legacy (BSD) PTY support
<*> Parallel printer support
<M> Texas Instruments parallel link cable support
<*> Enhanced REal Time Clock Support
<*> Double Talk PC internal speech card support
<*> /dev/agpgart (AGP Support)
<*> ALI chipset support
<*> SiS chipset support
<*> RAW driver (/dev/raw/rawN) (OBSOLETE)
[*] HPET - High Precesion Event Timer
[*] Allow mmap of HPET
<*> Hangcheck timer
Graphics support --->Console display driver support --->[*] Enable \
\ Scrollback Buffer in System Ram
[*] Video mode \
\ selection support

#Currently I have no speech card, but am interested in getting one.
Speakup console speech ---> <*> Build speakup console speech
<M> All others except
<*> DoubleTalk PC, dtlk
<*> DoubleTalk LT or LiteTalk, ltlk
<*> Software synth, sftsyn # Must be module
<*> Speak Out, spkout
<*> Transport, txprt
(sftsyn) Choose Default synth for Speakup #cannot be default
Sound --> <*> Sound card support
Advanced Linux Sound Architecture ---> [*] Adv. Linux Snd Arch
<*> OSS Mixer API
[/b] <*> OSS PCM API
[*] OSS PCM API - plugin
[*] Support old ALSA
[*] Verbose procfs
Open Sound System ---> <*> Open Sound System (DEPRICIAT)
[*] Obsolete OSS drivers
[M] OSS sound modules
[M] 100% Sound Blaster Compatables
USB support ---> <*> Support for Host-side USB
[*] USB device filesystem
<*> EHCI HCD (USB 2.0) support
<*> OHCI HCD support
<*> UHCI HCD (...) support
<*> USB Printer support
<*> USB Mass Storage support
<*> USB Human Interface Devide (full HID)
[*] /dev/hiddev raw HID device support
# yes, I know it's a toy...
<*> KB Gear JamStudio tablet support
<*> X-Box gamepad support
# for monitoring usb, or is a usb monitor?
<*> USB Monitor
MMC/SD Card support ---> <*> MMC support
<*> MMC block device driver
<*> TI Flash Media MMC/SD Inter. (EXP
Virtualization ---> <*> Kernel-based Virtual Machine
#doubt it has it, but worth a shot.
<*> KVM for AMD processors support
File systems ---> # I assume this isn't needed. If so, please reply.

#I don't have any instrum., and on other computer still had prob w/out anything selected here...
Instrumentation Support --->[*] Profiling support (EXP)
[*] OProfile system profiling (EXP)
[*] Kprobes (EXP)

Kernel Hacking ---> # Default, without change, from setup. Reply if needed.

Library routines ---> # All checked, but seemed to work on other computer w/ checked
# after I fixed it (which I don't know what I unchecked...)

------------------------------------------------------------------------------------------------------------------------------------------
Seems Non-Critical Below
------------------------------------------------------------------------------------------------------------------------------------------
# Last other computer was amd athlon-xp 2Ghz w/ 1 gig of ram and nforce motherboard. 5200 fx nVidia graphics card.
# CFLAGS="march=athlon-xp -O2 -fweb -finline-functions"

# amd K6-2 400Mhz processor w/ 64 MB of ram. ALI Motherboard. SiS pci video card.
# Currently I have no speech card, but am interested in getting one. Also getting Nvidia Quatro 2 agp card in future.
# I would like this computer setup for the deaf and blind, since it will be used for my local lions club.

# My current make.conf compiler setting
CFLAGS="-02 -march=k6-2 -pipe -fomit-frame-pointer -finline-functions -fgcse-after-reload -fpeel-loops -fweb -frename-registers -ftracer"
CXXFLAGS="${CFLAGS}"
CHOST="i586-pc-linux-gnu"
#Use profile is .../linux-default/2007.0/desktop/
USE=""
------------------------------------------------------------------------------------------------------------------------------------------

[mattsteven]

I'd be surprised if this was a kernel problem at all- does it get to INIT without requiring a keypress? If so, you should rule out the kernel.

Your CFLAGS are a little wild and I'd be leaning toward either hardware failure (requiring an interrupt to get the cpu unlocked?!) or something in these CFLAGS giving you this condition on the little bit of info I have been given.

Try rebuilding the entire system using conservative/default CFLAGS, and leave it up to the program author whether they should inline functions, peel loops, etc, they probably know what is best for their programs more than we do.
_________________
Matthew Steven
Linux-only desktop since 1998
Graying hair since 2006

[tamusroyce]

Thanks for your reply mattsteven.

Now that I have rebooted a few times, it does pause before INIT. I think it was a fluke or I didn't catch that it did that my first post.

I know it's a problem with my kernel because when I have configured my kernel with the default settings (without checking any features except default ones), it doesn't have this problem. Also, the genkernel works fine without this happening with the current compiler flags (-O2 plus a select amount that are pulled in from -O3).

I know on this build my compile flags are quite strange, but really they are between -O2 and -O3 stage. I didn't want --fun-loop-unrolling since this increases compiler code more than I wish, with maybe results on increasing or even reducing execution speeds. And since I have had this happen on a previous computer (the amd athlon xp 2800, nforce chipset w/ basic -O2 setting for programming), I didn't think it was needed to recompile my entire system & kernel with these setting.

If it was hardware, I know it wouldn't repeat itself with three computers (well, since 2005). Also, the pause points would be different each time it boots, I would assume, if it was hardware. It always happens in the same places, though it is kinda random to how it's distrubuted.

Thank you very much for your advice. I will see if it is a hardware or excessive compiler flag error, and re-compile my kernel until I solve the problem (with basic -O2 options, until successful) if possible. I will post my solution upon doing so.

p.s. If anyone has info about any kernel flags which have caused them problems in the past, I would again be greatful to hear about them...


I am greatful that the programmers do not add these wild compiler optimizations to their programs defaultly. If they had done so, I would not have been able to setup my other computer for programming, since these compiler options make debugging code impossible. But since this computer is older and not being setup for a programmer, it will hopefully benifits from a few of the -O3 optimizations.