[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [PATCH v4] block/raw-posix.c: Fixes raw_getlength() on
From: |
Programmingkid |
Subject: |
Re: [Qemu-devel] [PATCH v4] block/raw-posix.c: Fixes raw_getlength() on Mac OS X so that it reports the correct length of a real CD |
Date: |
Tue, 13 Jan 2015 12:52:05 -0500 |
On Jan 13, 2015, at 9:05 AM, Stefan Hajnoczi wrote:
> On Tue, Jan 6, 2015 at 3:04 PM, Programmingkid
> <address@hidden> wrote:
>>
>> On Jan 6, 2015, at 9:02 AM, Stefan Hajnoczi wrote:
>>
>>> On Fri, Jan 02, 2015 at 04:44:38PM -0500, Programmingkid wrote:
>>>> Removes redundant ret variable and renames sectorSize variable to meet
>>>> QEMU coding standards.
>>>
>>> This is a changelog item for v4 of this patch. Changelogs should go
>>> below the '---' line so they are not merged into git history.
>>>
>>> The rationale is that when a patch is merged into git, the changelog
>>> describing patch revisions that were posted on the mailing list is not
>>> relevant (we only see the final patch in git, not the revisions from the
>>> mailing list).
>>>
>>> Patches usually look like this:
>>>
>>> Subject: block/raw-posix: brief summary
>>>
>>> A longer description of the problem, maybe a command-line to reproduce a
>>> bug, and some rationale for this code change.
>>>
>>> Signed-off-by: Me <address@hidden>
>>> ---
>>> v2:
>>> * Fix int -> size_t for memory lengths [Requested by Bob]
>>>
>>> The changelog at the bottom is useful to code reviewers but won't get
>>> merged in the git history.
>>>
>>> Anyway, thanks for this patch. I have dropped this changelog line and
>>> merged it!
>>>
>>>>
>>>> Signed-off-by: John Arbuckle <address@hidden>
>>>>
>>>> ---
>>>> block/raw-posix.c | 18 +++++++++++++++++-
>>>> configure | 2 +-
>>>> 2 files changed, 18 insertions(+), 2 deletions(-)
>>>
>>> Thanks, applied to my block tree:
>>> https://github.com/stefanha/qemu/commits/block
>>>
>>> Stefan
>>
>> Thank you very much for accepting my patch.
>
> Hi,
> Unfortunately I had to drop this patch because it breaks -drive
> file=/dev/null,...
>
> /dev/null is a character device and we should not return -ENOTSUP when
> the CD-ROM ioctls fail.
>
> Please let it fail gracefully when the device is not a CD-ROM.
>
> Stefan
What is the exact command you use with QEMU involving the /dev/null device?
What value is suppose to be returned when using a device like /dev/null?