Up until beta2 this machine would install via NFS at least even if the cdrom drove was not probed properly
Now it hangs on bootup when it checks the processor honouring the WP bit. After 15 minutes the screen
blanks and then the keyboard goes dead.
Requires a power off/on to reboot back to RH70. Test ended until NFS tests are run.
Machine ID/name: #4/shonjir
CPU: 486 1
Memory: 16MB 1
Motherboard: N/A 1
Video: Trident 1MB SVGA 1
Storage: WDC AC13200B 3098MB IDE 1
CD-Rom: Panasonic CR-563 5 *
Network: 3c503 Etherlink II-TP ISA 2
22755 sbpcd.o no longer detects at base address 0x230
* Works in all releases from RH7 back.
Additional info: Tried a NFS install and the bootnet.img acts the same way. The diskettes are not defective because I have
used them on two other machines. (both P133's however).
I scared up another 486 with the same results. Machine data as follows:
So this machine also hangs on the WP check.
Machine ID/name: #5/fire
CPU Intel 486 SX/2 - Stepping 0B 1
Memory: 20MB 1
Motherboard: N/A 1
Storage: Quantum Fireball lct10 5 4884MB IDE 1
COMPAQ CR-589 ATAPI CD-Rom Drive 1
Network: 3c509-B ISA 2
Video: VGA 1
No PCI bus
Ignore the machine name, I used another template to generate a new template and forgot to change the machine name.
This is a known problem. The cause has been found to be the fix of an
SMP race in the vmalloc kernel code, which broke the WP bit on some
specific CPU's. (486's, Cyrix and Winchip)
A fix will be included in the next kernel we build.
Please reopen this bug if a kernel > 2.4.1-0.1.13 does not solve this bug.
Can I get a time estimate on this? I have some outstanding beta test bugs that need the use of one of those machines, and
will there be img files that are compatible with the RC1/Wolverine CD's?
Current rawhide should have a sufficiently recent kernel
Marking NEEDINFO pending results of newer kernel
Ran a test with the boot.img and bootnet.img as provided by Arjan van de Ven and both machines got past the WP point.
Since this was a simulated test as the Wolverine CD's have the buggy kernel that means to me it is fixed as far as I can