Re: [PATCH] qemu-xen-3.4-testing: Fix read-only image file handling

[Michal Novotny <minovotn@redhat.com>]

On 06/04/2010 03:16 PM, Keir Fraser wrote:
> On 04/06/2010 14:13, "Michal Novotny"<minovotn@redhat.com>  wrote:
>
>    
>> Keir,
>> I tried it and it's already up to date:
>>
>> $ git pull -a
>> Already up-to-date. Yeeah!
>> $
>>      
> Don't believe you, or it's not the one that's actually being built. :-) If
> you grep in the tree for xc_interface_open() you should see it is correctly
> passing three args now.

Sorry, my bad, something was wrong with my git tree so I deleted it and 
pulled again.

Thanks,
Michal

> And if it is, then that will build okay. If it's not
> passing three args, the tree is stale.
>
>   -- Keir
>
>    
>> -- Michal
>>
>> On 06/04/2010 03:11 PM, Keir Fraser wrote:
>>      
>>> You need to bring your qemu repo up to date. Throw it away, or go into it
>>> and 'git pull -a'.
>>>
>>>    -- Keir
>>>
>>> On 04/06/2010 14:00, "Michal Novotny"<minovotn@redhat.com>   wrote:
>>>
>>>
>>>        
>>>> Well, now the kernel is booting but when I'm trying to recompile tools
>>>> and stubdom images it's not OK returning those errors:
>>>>
>>>> ...
>>>> qemu successfuly configured for Xen qemu-dm build
>>>> make -C ioemu-dir install
>>>> === PCI passthrough capability has been enabled ===
>>>> make[3]: Entering directory
>>>> `/home2/shared/xen-unstable.hg/tools/ioemu-remote'
>>>> === PCI passthrough capability has been enabled ===
>>>> === PCI passthrough capability has been enabled ===
>>>> make[4]: Entering directory
>>>> `/home2/shared/xen-unstable.hg/tools/ioemu-remote/i386-dm'
>>>> CC i386-dm/vl.o
>>>> /home2/shared/xen-unstable.hg/tools/ioemu-dir/vl.c: In function Œmain¹:
>>>> /home2/shared/xen-unstable.hg/tools/ioemu-dir/vl.c:5729: warning:
>>>> implicit declaration of function Œdma_helper_init¹
>>>> /home2/shared/xen-unstable.hg/tools/ioemu-dir/vl.c:5731: error: too few
>>>> arguments to function Œxc_interface_open¹
>>>> make[4]: *** [vl.o] Error 1
>>>> make[4]: Leaving directory
>>>> `/home2/shared/xen-unstable.hg/tools/ioemu-remote/i386-dm'
>>>> make[3]: *** [subdir-i386-dm] Error 2
>>>> make[3]: Leaving directory
>>>> `/home2/shared/xen-unstable.hg/tools/ioemu-remote'
>>>> make[2]: *** [subdir-install-ioemu-dir] Error 2
>>>> make[2]: Leaving directory `/home2/shared/xen-unstable.hg/tools'
>>>> make[1]: *** [subdirs-install] Error 2
>>>> make[1]: Leaving directory `/home2/shared/xen-unstable.hg/tools'
>>>> make: *** [install-tools] Error 2
>>>> #
>>>>
>>>> The command I used for compilation is `make install-xen&&   make
>>>> install-tools&&   make install-stubdom&&   make install-docs` on the
>>>> latest updated xen-unstable staging mercurial tree so I need to
>>>> investigate this further. Nevertheless, the dom0 2.6.31.13 is booting on
>>>> RHEL-5 which is good. That links you gave me Pasi helped, thanks! Now I
>>>> need to resolve those tools issues.
>>>>
>>>> Michal
>>>>
>>>> On 06/04/2010 12:57 PM, Michal Novotny wrote:
>>>>
>>>>          
>>>>> On 06/04/2010 12:42 PM, Pasi Kärkkäinen wrote:
>>>>>
>>>>>            
>>>>>> On Fri, Jun 04, 2010 at 11:37:51AM +0200, Michal Novotny wrote:
>>>>>>
>>>>>>              
>>>>>>> Oh, one more thing I've discovered now. Since the codebase for
>>>>>>> qemu-xen-unstable is looking the same like for qemu-xen-3.4-testing and
>>>>>>> the patch is applicable without any modifications it could be working
>>>>>>> for xen-4.1 unstable as well, unfortunately I was not able to boot
>>>>>>> PVOPS
>>>>>>> kernel since it always ends up in kernel panic, maybe the missing
>>>>>>> drivers or something like that.
>>>>>>>
>>>>>>>
>>>>>>>                
>>>>>> What kernel version? Did you use xen/stable-2.6.32.x branch from
>>>>>> xen.git?
>>>>>>
>>>>>>              
>>>>> It's the one that's being automatically downloaded from git as
>>>>> linux-2.6-pvops.git. The URL is
>>>>> git://git.kernel.org/pub/scm/linux/kernel/git/jeremy/xen.git with the
>>>>> "fetch" being set to "+refs/heads/*:refs/remotes/xen/*". This is the
>>>>> .git/config automatically created by `make world` in xen-unstable tree:
>>>>>
>>>>> [core]
>>>>> repositoryformatversion = 0
>>>>> filemode = true
>>>>> bare = false
>>>>> logallrefupdates = true
>>>>> [remote "xen"]
>>>>> url = git://git.kernel.org/pub/scm/linux/kernel/git/jeremy/xen.git
>>>>> fetch = +refs/heads/*:refs/remotes/xen/*
>>>>> [branch "xen/stable-2.6.31.x"]
>>>>> remote = xen
>>>>> merge = refs/heads/xen/stable-2.6.31.x
>>>>>
>>>>> By default it's using uhci-hcd/ehci-hcd and ohci-hcd as builtins but I
>>>>> needed to change it to be modules since otherwise it's being stuck.
>>>>> I'm trying this on RHEL-5 dom0 (I'm having second partition RHEL-5
>>>>> with upstream xen).
>>>>>
>>>>>            
>>>>>> See my example .config files from:
>>>>>> http://wiki.xensource.com/xenwiki/XenParavirtOps in the
>>>>>> "troubleshooting" section.
>>>>>>
>>>>>>              
>>>>> I'll have a look, thanks for the link.
>>>>>
>>>>>
>>>>>            
>>>>>> Works on F12/F13 dom0's.
>>>>>>
>>>>>> For el5 dom0 you might need these:
>>>>>> http://wiki.xensource.com/xenwiki/2.6.18-to-2.6.31-and-higher
>>>>>>
>>>>>>              
>>>>> Oh, ok, this may be very useful since I'm trying this on RHEL-5. Thanks!
>>>>>
>>>>>
>>>>>            
>>>>>> aka
>>>>>> CONFIG_SYSFS_DEPRECATED=y
>>>>>> CONFIG_SYSFS_DEPRECATED_V2=y
>>>>>>
>>>>>> Also atm I think you need CONFIG_XEN_XENBUS_FRONTEND=y in the .config
>>>>>> because of the recently added pv-on-hvm drivers (you'll get build
>>>>>> failure without that option),
>>>>>> so you need XEN_PCIDEV_FRONTEND XEN_BLKDEV_FRONTEND or
>>>>>> XEN_NETDEV_FRONTEND or XEN_FBDEV_FRONTEND needs to be =y.
>>>>>>
>>>>>>
>>>>>>              
>>>>> Great! Thanks for your help Pasi! If anything I'll let you know.
>>>>>
>>>>> Michal
>>>>>
>>>>>
>>>>>            
>>>>>> -- Pasi
>>>>>>
>>>>>>
>>>>>>
>>>>>>              
>>>>>>> On 06/03/2010 04:12 PM, Michal Novotny wrote:
>>>>>>>
>>>>>>>                
>>>>>>>> Oh, just one more thing that should be mentioned:
>>>>>>>>
>>>>>>>> When you want to mount an image that is set as read-only in the domain
>>>>>>>> configuration file but you omit to set mode to read-only it results
>>>>>>>> into I/O errors when processing the requests. Remounting as read-only
>>>>>>>> or unmounting and remounting using the `mount /dev/* /path/to/mount -o
>>>>>>>> ro` shall do the mounting the correct way, i.e. with no I/O errors, so
>>>>>>>> make sure you mount those disks as read-only otherwise you can be
>>>>>>>> getting errors like:
>>>>>>>>
>>>>>>>> end_request: I/O error, dev hdb, sector 52
>>>>>>>>
>>>>>>>> Buffer I/O error on device hdb1, logical block 1
>>>>>>>>
>>>>>>>> lost page write due to I/O error on hdb1
>>>>>>>>
>>>>>>>>
>>>>>>>> and for IDE devices you'll be getting several additional DeviceFault
>>>>>>>> errors since mounting the device read-write (default setting) writes
>>>>>>>> some data onto a disk at the mount-time.
>>>>>>>>
>>>>>>>> Michal
>>>>>>>>
>>>>>>>> On 06/03/2010 04:04 PM, Michal Novotny wrote:
>>>>>>>>
>>>>>>>>                  
>>>>>>>>> [Well, I did send an e-mail to the list using git but it's not here
>>>>>>>>> so I'm forwarding the e-mail to the list for sure:]
>>>>>>>>>
>>>>>>>>> Hi,
>>>>>>>>> this is the patch for qemu-xen-3.4-testing to fix the read-only
>>>>>>>>> image file handling since the image file was always treated as
>>>>>>>>> read-write which means that all the HVM guests were able to
>>>>>>>>> write to all the disk images available in domain configuration
>>>>>>>>> file no matter what the mode of the image was defined. This
>>>>>>>>> patch fixes this functionality to honor the O_RDONLY in the
>>>>>>>>> BDRV_O_ACCESS flag in block.c and also fixes the IDE and SCSI
>>>>>>>>> interfaces that uses it.
>>>>>>>>>
>>>>>>>>> It's been tested on RHEL-5 with xen-3.4-testing version of
>>>>>>>>> upstream xen with xen-3.4-testing qemu implementation.
>>>>>>>>>
>>>>>>>>> For SCSI devices the DATA PROTECT request sense has been added
>>>>>>>>> as found at: http://en.wikipedia.org/wiki/SCSI_Request_Sense_Command
>>>>>>>>>
>>>>>>>>> file no matter what the mode of the image was defined. This
>>>>>>>>> patch fixes this functionality to honor the O_RDONLY in the
>>>>>>>>> BDRV_O_ACCESS flag in block.c and also fixes the IDE and SCSI
>>>>>>>>> interfaces that uses it.
>>>>>>>>>
>>>>>>>>> It's been tested on RHEL-5 with xen-3.4-testing version of
>>>>>>>>> Michal
>>>>>>>>>
>>>>>>>>> Signed-off-by: Michal Novotny<minovotn@redhat.com>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> _______________________________________________
>>>>>>>>> Xen-devel mailing list
>>>>>>>>> Xen-devel@lists.xensource.com
>>>>>>>>> http://lists.xensource.com/xen-devel
>>>>>>>>>
>>>>>>>>>                    
>>>>>>>>
>>>>>>>>                  
>>>>>>> -- 
>>>>>>> Michal Novotny<minovotn@redhat.com>, RHCE
>>>>>>> Virtualization Team (xen userspace), Red Hat
>>>>>>>
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> Xen-devel mailing list
>>>>>>> Xen-devel@lists.xensource.com
>>>>>>> http://lists.xensource.com/xen-devel
>>>>>>>
>>>>>>>                
>>>>>
>>>>>            
>>>>
>>>>          
>>>
>>>        
>>      
>
>    


-- 
Michal Novotny<minovotn@redhat.com>, RHCE
Virtualization Team (xen userspace), Red Hat