qemu-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Qemu-devel] [PATCH] hostmem-file: add a property 'notrunc' to avoid


From: Haozhong Zhang
Subject: Re: [Qemu-devel] [PATCH] hostmem-file: add a property 'notrunc' to avoid data corruption
Date: Thu, 20 Oct 2016 21:47:28 +0800
User-agent: NeoMutt/20160827 (1.7.0)

On 10/20/16 11:34 -0200, Eduardo Habkost wrote:
On Thu, Oct 20, 2016 at 09:11:38PM +0800, Haozhong Zhang wrote:
On 10/20/16 14:34 +0200, Igor Mammedov wrote:
> On Thu, 20 Oct 2016 14:13:01 +0800
> Haozhong Zhang <address@hidden> wrote:
>
> > If a file is used as the backend of memory-backend-file and its size is
> > not identical to the property 'size', the file will be truncated. For a
> > file used as the backend of vNVDIMM, its data is expected to be
> > persistent and the truncation may corrupt the existing data.
> I wonder if it's possible just skip 'size' property in your case instead
> 'notrunc' property. That way if size is not present one'd get actual size
> using get_file_size() and set 'size' to it?
> And if 'size' is provided and 'size' != file_size then error out.
>

I don't know how this can be implemented in QEMU. Specially, how does
the memory-backend-file know it's used for vNVDIMM, so that it can
skip the 'size' property?

Besides, it cannot cover the case that only a part of file is used as
the backend of vNVDIMM, which is allowed by the current implementation.

I see 3 possible cases:

* If desired RAM size is the same as file size, 'size' can be
 optional.
* If desired RAM size is smaller than file size, we don't need to
 truncate the file at all. In this case, we could even use the
 size specified in the frontend config options (-numa and/or
 -m), instead of requiring size to be specified on the backend
 object.
* If desired RAM size is larger than file size, the only option
 is to specify size explicitly on the backend object and extend
 it using ftruncate().

None of those cases would require a new option.


One reason I added an additional notrunc is to make it to possible to
preserve the previous truncation behavior (when nontrunc is
off). However, as you listed here, it's not necessary to preserve the
previous truncation behavior, so notrunc is not necessary neither.

Haozhong




reply via email to

[Prev in Thread] Current Thread [Next in Thread]