[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [PATCH] Journal support for ext2
From: |
Robert Millan |
Subject: |
Re: [PATCH] Journal support for ext2 |
Date: |
Thu, 20 Mar 2008 14:13:31 +0100 |
User-agent: |
Mutt/1.5.13 (2006-08-11) |
On Thu, Mar 20, 2008 at 02:03:45PM +0800, Bean wrote:
> On Thu, Mar 20, 2008 at 8:12 AM, Robert Millan <address@hidden> wrote:
> >
> > Hi!
> >
> > Is this patch supposed to fix > 32-bit disk addressing in filesystems? I
> > just setup a ~4 TB hard disk for testing (in qemu), and added a (gpt)
> > partition
> > at the end of it, surely above the 2 TiB barrier.
> >
> > What I found puzzling is that grub-emu is capable of listing files in that
> > partition _without_ this patch. Any idea why? This wasn't really supposed
> > to work was it?
> >
> > Note: grub itself can't read it yet, but this must be due to disk io
> > problems
> > (asides from the lba32 limit in bochsbios).
>
> Hi,
>
> the sector parameter in grub_disk_read is 64-bit, so it can read from
> disk larger than 4g.The fs interface is using 64-bit address as well,
> so if the implementation is all right, it can handle large partition.
Ah, that explains it. Note that the disk io part can't handle it yet,
though. I'll have a look into that (bochsbios itself limits it, but ata.mod
didn't work either).
> The problem is that fshelp function use int for offset and block
> number, which limit the maximum file size to 4g. You can verify this
> with loopback on a large file.
Ah ok.
--
Robert Millan
<GPLv2> I know my rights; I want my phone call!
<DRM> What use is a phone call… if you are unable to speak?
(as seen on /.)