Re: [Qemu-devel] [PATCH] Pass the drive's readonly attribute to the guest OS

[Kevin Wolf]

Am 15.10.2009 12:01, schrieb Gleb Natapov:
> On Thu, Oct 15, 2009 at 11:55:20AM +0200, Kevin Wolf wrote:
>> Am 15.10.2009 11:54, schrieb Gleb Natapov:
>>> On Thu, Oct 15, 2009 at 11:50:39AM +0200, Kevin Wolf wrote:
>>>> Am 15.10.2009 11:43, schrieb Gleb Natapov:
>>>>> On Thu, Oct 15, 2009 at 11:36:59AM +0200, Kevin Wolf wrote:
>>>>>> Am 14.10.2009 18:40, schrieb Naphtali Sprei:
>>>>>>> Naphtali Sprei wrote:
>>>>>>> <snip>
>>>>>>>> I'm planning to investigate where qemu should check the read only 
>>>>>>>> attribute before exeuting any write command
>>>>>>>> to drives, would be sent in a different patch.
>>>>>>>
>>>>>>> revisiting it, if guest OS knows it's a read only device and tries to 
>>>>>>> modify it, anyhow, we don't really care about error reporting,
>>>>>>> as long as qemu doesn't crash (or modify the drive).
>>>>>>
>>>>>> If the right response to a write on a read-only device is defined in the
>>>>>> specification (and it most probably is), we should still give the right
>>>>>> response, even though the OS is doing something wrong.
>>>>>>
>>>>> And since our response to write error may be pausing a VM we shouldn't
>>>>> allow this to be triggered by a guest OS.
>>>>
>>>> I thought we only pause the VM if we get an host IO error? But if you do
>>>> want to stop it for all errors, you shouldn't start suppressing errors
>>>> so that it doesn't stop.
>>>>
>>> We pause only on host IO errors, but if we open underlying file as
>>> read only (do we?) and try to write into it we will get an IO error
>>> in the host.
>>
>> No, we'll return an error before a write request to the host is even issued.
>>
> Who is "we"? If "we" == "bdrv_write()/dma_bdrv_write()" then it's all the 
> same.
> Upper layers don't actually care why block driver failed.

Right, "we" is the qemu block layer. If the devices don't use the error
code returned, they better should be fixed, I think?

Kevin