Mail Archives: djgpp-workers/2000/02/27/09:14:08
At 19:26 21/02/00 +0200, Eli wrote:
>
>> I include here a diff to version 2.03 of crt0.S file that also fixes the
>> problem for the C library!
>
>Pierre, I don't understand this patch: it seems that you store
>something in sbrk16_interrupt_state, but never use that value:
>
>> *** 552,557 ****
>> --- 567,576 ----
>> movw ___djgpp_base_address+2, %cx
>> int $0x31
>>
>> + /* Now we can reenable interrupts PM */
>> + movl %eax,sbrk16_interrupt_state
>> + int $0x31
>> +
>
>Shouldn't this be the other way around: "movl sbrk16_interrupt_state,%eax"?
Of course it should !! I am sometimes still interverting Intel and ATT
syntax :(
In the original code its done just by pushing it
on the stack and popping it back into %eax before calling int $0x31
again, as it is now postponed to later, I needed a variable to store the
return value of function 0x902.
To Charles Sandmann, about the problem of locking :
shouldn't the DPMI server copy the current lock flag for each page
that is move by the UNIXY SBRK function ?
In that case, the only think would be to ensure that no sbrk call is made
inside the lock_data and lock_code functions, no ?
- Raw text -