Charles Manning wrote:
> On Monday 01 August 2005 11:04, Luc Van Oostenryck wrote:
>>- There is a problem related to truncation and rewrite of files but I
>>   have not yet investigated this one.
>>   May be related to not using USE_HEADER_SIZE ???
> 
> 
> If you can make a test case I will investigate further.
> 
> 

Yesterday, it's was 100% reproductible at the very beginning of the test,
but now it happen much later and not at the same place (no more just after a truncate).
I didn't had the time to really investigate it.

The test went good on nandemul2k (but the test must run much longer that I did)
but fail on nandsim.

The output of the program is:
	# fsx-linux -p -W -N 102400000 /mnt/nandsim

	...
	139315(51 mod 256): READ        0x6e73 thru 0x9ce4      (0x2e72 bytes)
	139316(52 mod 256): READ        0x8e3b thru 0x9ce4      (0xeaa bytes)
	139317(53 mod 256): MAPREAD     0x9326 thru 0x9ce4      (0x9bf bytes)
	139318(54 mod 256): MAPWRITE 0x3552c thru 0x3ffff       (0xaad4 bytes)
	139319(55 mod 256): MAPWRITE 0x33797 thru 0x3ec5f       (0xb4c9 bytes)
	139320(56 mod 256): READ        0x1209 thru 0x34bd      (0x22b5 bytes)
	139321(57 mod 256): WRITE       0x27adf thru 0x2f666    (0x7b88 bytes)
	139322(58 mod 256): MAPREAD     0x9b04 thru 0xaa67      (0xf64 bytes)
	139323(59 mod 256): WRITE       0x14804 thru 0x1bc3b    (0x7438 bytes)
	139324(60 mod 256): WRITE       0xc842 thru 0x1bf0f     (0xf6ce bytes)
	139325(61 mod 256): MAPREAD     0x12f69 thru 0x1409f    (0x1137 bytes)
	139326(62 mod 256): MAPREAD     0x14c53 thru 0x1657a    (0x1928 bytes)
	139327(63 mod 256): MAPREAD     0x17f1f thru 0x22ebc    (0xaf9e bytes)  ***RRRR***
	Correct content saved for comparison
	(maybe hexdump "/mnt/nandsim/fsx-test" vs "/mnt/nandsim/fsx-test.fsxgood")


I put in attachement the two test files, the log and the test program.


Luc