(no subject)

[Robert Kaiser]

Hi,

In article <[e-mail]>,
	[e-mail] (Timur Tabi) writes:
> ** Reply to message from Robert Kaiser <[e-mail]> on Tue, 9 Jan 2001
> 23:17:11 +0100
> 
> 
>> temp = mk_pte_phys(__pa(vaddr), PAGE_KERNEL);
>> 	*pte = temp;
>> 
>> where temp is declared "volatile pte_t". I inserted test-prints between the
>> above two lines. Accoding to that, the _first_ line , i.e. the evaluation of the
>> mk_pte_phys() macro is causing the crash!
> 
> In that case, it's either a compiler bug or a race condition.
> 
> Compiling this source file with the -S option will generate an assembly output.
> The output should be the same regardless of whether you use the temp variable or
> not.  That will answer the question as to what the cause is.  If you're lucky,
> it's a compiler bug, because they're easier to fix.
> 
> 
I thought that by declaring temp "volatile" I could force the compiler to not
optimize it away, i.e. make sure that the evaluation of the mk_pte_phys()
macro and the assignment to *pte indeed get seperated. Is this not the case ?

Anyway, the actual code I'm using has debug output in between the lines, so it
looks more like this:

	*((volatile unsigned char*)0xb8000) = '1';
	temp = mk_pte_phys(__pa(vaddr), PAGE_KERNEL);
	*((volatile unsigned char*)0xb8000) = '2';
	*pte = temp;
	*((volatile unsigned char*)0xb8000) = '3';

When the system crashes, just before the BIOS clears the screen, I can clearly
see a '2' in the upper left corner of the screen.

----------------------------------------------------------------
Robert Kaiser                         email: [e-mail]
SYSGO RTS GmbH
Am Pfaffenstein 14                    phone: [phone]
D-55270 Klein-Winternheim / Germany   fax:   [phone]