中国DOS联盟

This is not a case of being wronged, because this problem is actually very strange. If you don't report it, others may also report it. So, by solving your problem, we have accumulated experience. Next time others encounter a similar problem, we will know that the real reason is the use of memdisk. As long as we switch to the built-in map emulation, there will be no problem. Thanks to zjhxqh for the detailed report.

wsfyoo8, should your problem have been solved in the version of 2008-08-14?

Hehe just witnessed the hundred-meter sprinter easily break the world record, and also wish grub stronger!

wsfyoo8, should your issue have been resolved in the version from 2008-08-14?

Tried the 2008-08-14 version, still can't boot, still getting this prompt:
Uncompressing linux...
crc error
---System halted
There's also a reflection of this situation here: http://www.linuxfans.org/bbs/thread-185258-1-1.html

Last edited by wsfyoo8 on 2008-8-17 at 12:48 AM ]

wsfyoo8, from July 28th to early August, the code changes were very minor and did not involve any changes to the file system. So, from a code perspective, there shouldn't be an issue. Either your Linux is problematic, or there's a more hidden issue in grub4dos. Hope you can help determine where the problem might be. Before having more detailed reports, we'll temporarily not consider this issue.

Change log (excerpted from http://bbs.znpc.net/viewthread.php?tid=2431&extra=page%3D1):

2008-08-04 Fixed the error handling problem of operators && and ||; the structure of grldr.mbr has changed, and now it can be placed in boot.ini; several STI instructions were added in dosstart.S and asm.S to try to avoid system crashes due to hardware interrupt issues.

2008-07-28 Added the sti instruction in asm.S to better respond to hardware interrupts or to feed the watchdog. These changes assume that some machines have a watchdog clock. When the system cannot respond to hardware interrupts for a long time, the watchdog is starved, thus triggering a non-maskable interrupt NMI, leading to system deadlock. This change is expected to solve the recent widely reported crash problems. Please step up testing.

Originally posted by Budian at 2008-8-17 10:15 AM:
wsfyoo8, the code changes from July 28th to early August are very minor, and there is no change related to the file system at all. So, there should be no problem from the code perspective. Either your linux is problematic, or gru ...

Thank you, Master Budian, for your reply! :) I'll check where the problem lies again.

By the way, bump this post so that people notice the changes on August 3rd.

Also, someone has found the cause of the Linux boot freeze on boot-land.net. As long as you don't enter the graphical mode, that is, don't use splashimage and fontfile, there will be no problem.

This proves that the new changes have caused a memory conflict in the space occupancy between the graphical mode and the Linux kernel. It should be resolved soon.

Originally posted by Buddy at 2008-9-8 06:42 PM:
By the way, bump this thread so that people can notice the changes on August 3rd.

Also, someone has found the reason for Linux boot freeze on boot-land.net. As long as you don't enter the graphical mode, then ...

Just tried the 2008-08-14 version. After not using splashimage and fontfile, slax really doesn't freeze, but it can't log in automatically anymore. You have to enter the user and password, then execute startx to enter the desktop. It seems that the "autoexec=xconf;telinit~4 changes=/slax/slaxsave.dat" in the menu

title >>>Boot slax6.06
kernel /boot/vmlinuz ramdisk_size=6666 root=/dev/ram0 rw autoexec=xconf;telinit~4 changes=/slax/slaxsave.dat
initrd /boot/initrd.gz

doesn't work