From: Martin Str|mberg Message-Id: <200101071903.UAA24570@father.ludd.luth.se> Subject: Re: Fw: Patch for statfs.c In-Reply-To: <5.0.2.1.0.20010107121357.00aa6580@pop5.banet.net> from "Peter J. Farley III" at "Jan 7, 2001 01:34:54 pm" To: djgpp-workers AT delorie DOT com Date: Sun, 7 Jan 2001 20:03:05 +0100 (MET) X-Mailer: ELM [version 2.4ME+ PL54 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Reply-To: djgpp-workers AT delorie DOT com Errors-To: nobody AT delorie DOT com X-Mailing-List: djgpp-workers AT delorie DOT com X-Unsubscribes-To: listserv AT delorie DOT com Precedence: bulk According to Peter J. Farley III: > Sorry, Martin. This version does not report CD size correctly on my > system. Here are the results I got, after rebuilding libc with your > new statfs and rebuilding fileutils 3.16 after that. Again, Y: is a > DVD-ROM, Z: is a CDRW, same discs loaded as in my prior test. > > Note that the rebuilt df reports too many 1024 blocks, 326552*1024 = > 334389248 while 325888*1024 = 333709312. The difference is 664 blocks, > which I cannot explain easily. > > O:\gnu\filutil3.16\src>.\df z: > Filesystem 1024-blocks Used Available Capacity Mounted on > Corel Linux 1.2 Sources > 326552 326552 0 100% z:/ > > O:\gnu\filutil3.16\src>\bin\df z: > Filesystem 1024-blocks Used Available Capacity Mounted on > Corel Linux 1.2 Sources > 325888 325888 0 100% z:/ This is really strange. Both the recently compiled df and df from fil316b.zip ought to pick the data from INT21 AX=1510, which ought to be the same. > I added some "#if TEST" print statements to produce more detailed > info. It sure looks to me like AX1510 is producing wrong results for > the CDRW drive and AX7303 is producing correct results. By checking The problem is the block size. If you have a CDROM it has block size 2048 and nothing else. If you see a different block size that is a lie. > whether AX1510 is greater than AX7303, Martin's code produces (I think) > the wrong results for CD's, though correct for DVD's. At least, it > produces results that are different from the Win98 "Properties" > value. Here are my results (patched statfs.c follows at the end). I think I and Eli agree on that WINDOZE properties values are not exactly correct all the time. Thank you for running the tests! Right, MartinS