Re: scsi-next fail to be compiled

Re: scsi-next fail to be compiled

[James Bottomley]

On Tue, 2010-07-27 at 11:56 -0700, Yinghai Lu wrote:
> LD      drivers/scsi/built-in.o
> drivers/scsi/qla4xxx/built-in.o: In function `qla82xx_idc_unlock':
> (.text+0x8814): multiple definition of `qla82xx_idc_unlock'
> drivers/scsi/qla2xxx/built-in.o:(.text+0x2ea14): first defined here
> drivers/scsi/qla4xxx/built-in.o: In function `qla82xx_pci_config':
>      (.text+0x344e): multiple definition of `qla82xx_pci_config'
> drivers/scsi/qla2xxx/built-in.o:(.text+0x2cdfd): first defined here
> drivers/scsi/qla4xxx/built-in.o: In function `qla82xx_pci_mem_write_2M':
> (.text+0x9b15): multiple definition of `qla82xx_pci_mem_write_2M'
> drivers/scsi/qla2xxx/built-in.o:(.text+0x2f0fb): first defined here
> drivers/scsi/qla4xxx/built-in.o: In function `qla82xx_pci_get_crb_addr_2M':
> (.text+0x80c0): multiple definition of `qla82xx_pci_get_crb_addr_2M'
> drivers/scsi/qla2xxx/built-in.o:(.text+0x2c2ec): first defined here
> drivers/scsi/qla4xxx/built-in.o: In function `qla82xx_wr_32':
> (.text+0x88b3): multiple definition of `qla82xx_wr_32'
> drivers/scsi/qla2xxx/built-in.o:(.text+0x2d23a): first defined here
> drivers/scsi/qla4xxx/built-in.o: In function `qla82xx_pci_mem_read_2M':
> (.text+0x9620): multiple definition of `qla82xx_pci_mem_read_2M'
> drivers/scsi/qla2xxx/built-in.o:(.text+0x2eded): first defined here
> drivers/scsi/qla4xxx/built-in.o: In function `qla82xx_msix_rsp_q':
> (.text+0x7d24): multiple definition of `qla82xx_msix_rsp_q'
> drivers/scsi/qla2xxx/built-in.o:(.text+0x2caca): first defined here
> drivers/scsi/qla4xxx/built-in.o: In function `qla82xx_rd_32':
> (.text+0x8786): multiple definition of `qla82xx_rd_32'
> drivers/scsi/qla2xxx/built-in.o:(.text+0x2d14f): first defined here
> drivers/scsi/qla4xxx/built-in.o: In function `qla82xx_watchdog':
> (.text+0x14ea): multiple definition of `qla82xx_watchdog'
> drivers/scsi/qla2xxx/built-in.o:(.text+0x2de2f): first defined here
> drivers/scsi/qla4xxx/built-in.o: In function `qla82xx_load_risc':
> (.text+0xafd4): multiple definition of `qla82xx_load_risc'
> drivers/scsi/qla2xxx/built-in.o:(.text+0x313d8): first defined here
> drivers/scsi/qla4xxx/built-in.o: In function `qla82xx_crb_win_lock':
> (.text+0x8b29): multiple definition of `qla82xx_crb_win_lock'
> drivers/scsi/qla2xxx/built-in.o:(.text+0x2d1e4): first defined here
> drivers/scsi/qla4xxx/built-in.o: In function `qla82xx_intr_handler':
> (.text+0x7e9b): multiple definition of `qla82xx_intr_handler'
> drivers/scsi/qla2xxx/built-in.o:(.text+0x2e068): first defined here
> drivers/scsi/qla4xxx/built-in.o: In function `qla82xx_clear_drv_active':
> (.text+0x9333): multiple definition of `qla82xx_clear_drv_active'
> drivers/scsi/qla2xxx/built-in.o:(.text+0x2e027): first defined here
> drivers/scsi/qla4xxx/built-in.o: In function `qla82xx_disable_intrs':
> (.text+0x8947): multiple definition of `qla82xx_disable_intrs'
> drivers/scsi/qla2xxx/built-in.o:(.text+0x2dd7d): first defined here
> drivers/scsi/qla4xxx/built-in.o: In function `qla82xx_iospace_config':
> (.text+0x1266): multiple definition of `qla82xx_iospace_config'
> drivers/scsi/qla2xxx/built-in.o:(.text+0x2ce36): first defined here
> drivers/scsi/qla4xxx/built-in.o: In function `qla82xx_enable_intrs':
> (.text+0x8a39): multiple definition of `qla82xx_enable_intrs'
> drivers/scsi/qla2xxx/built-in.o:(.text+0x2ddd6): first defined here
> drivers/scsi/qla4xxx/built-in.o: In function `qla82xx_device_state_handler':
> (.text+0x9fff): multiple definition of `qla82xx_device_state_handler'
> drivers/scsi/qla2xxx/built-in.o:(.text+0x30ccb): first defined here
> drivers/scsi/qla4xxx/built-in.o: In function `qla82xx_idc_lock':
> (.text+0x8829): multiple definition of `qla82xx_idc_lock'
> drivers/scsi/qla2xxx/built-in.o:(.text+0x30c56): first defined here
> make[1]: *** [drivers/scsi/built-in.o] Error 1
> make: *** [drivers/scsi/] Error 2

This is because exactly the same functions appear in the qla2xxx/ and
qla4xxx/ directories

Qlogic people, please get this sorted out now, otherwise I'll drop the
qla4xxx update before sending to Linus

The best option would be to separate out the 82xx functions so that both
2xxx and 4xxx can use them.  In the interim, you could add a
discriminator prefix.

James

RE: scsi-next fail to be compiled

[Ravi Anand]

>On Tue, 2010-07-27 James Bottomley wrote:
>On Tue, 2010-07-27 at 11:56 -0700, Yinghai Lu wrote:
> LD      drivers/scsi/built-in.o
> drivers/scsi/qla4xxx/built-in.o: In function `qla82xx_idc_unlock':
> (.text+0x8814): multiple definition of `qla82xx_idc_unlock'
> drivers/scsi/qla2xxx/built-in.o:(.text+0x2ea14): first defined here
> drivers/scsi/qla2xxx/built-in.o:(.text+0x30c56): first defined here
> make[1]: *** [drivers/scsi/built-in.o] Error 1
> make: *** [drivers/scsi/] Error 2

>This is because exactly the same functions appear in the qla2xxx/ and
>qla4xxx/ directories

>Qlogic people, please get this sorted out now, otherwise I'll drop the
>qla4xxx update before sending to Linus

We will try to resolve this ASAP.  When are you planning to send it to Linus ?

>The best option would be to separate out the 82xx functions so that both
>2xxx and 4xxx can use them.  In the interim, you could add a
>discriminator prefix.

We will put in discriminator prefix and  repost the patches.

Thanks
Ravi

RE: scsi-next fail to be compiled

[James Bottomley]

On Tue, 2010-07-27 at 13:45 -0700, Ravi Anand wrote:
> 
> >On Tue, 2010-07-27 James Bottomley wrote:
> >On Tue, 2010-07-27 at 11:56 -0700, Yinghai Lu wrote:
> > LD      drivers/scsi/built-in.o
> > drivers/scsi/qla4xxx/built-in.o: In function `qla82xx_idc_unlock':
> > (.text+0x8814): multiple definition of `qla82xx_idc_unlock'
> > drivers/scsi/qla2xxx/built-in.o:(.text+0x2ea14): first defined here
> > drivers/scsi/qla2xxx/built-in.o:(.text+0x30c56): first defined here
> > make[1]: *** [drivers/scsi/built-in.o] Error 1
> > make: *** [drivers/scsi/] Error 2
> 
> >This is because exactly the same functions appear in the qla2xxx/ and
> >qla4xxx/ directories
> 
> >Qlogic people, please get this sorted out now, otherwise I'll drop the
> >qla4xxx update before sending to Linus
> 
> We will try to resolve this ASAP.

Thanks

>   When are you planning to send it to Linus ?

Whenever he opens the merge window ... he's been making noises like it
will be this week.

> >The best option would be to separate out the 82xx functions so that both
> >2xxx and 4xxx can use them.  In the interim, you could add a
> >discriminator prefix.
> 
> We will put in discriminator prefix and  repost the patches.

Just the one patch that adds the 82xx support, if possible ... that way
I can just swap the bad one out.

James

RE: scsi-next fail to be compiled

[Ravi Anand]

> >On Tue, 2010-07-27 James Bottomley wrote:
>On Tue, 2010-07-27 at 13:45 -0700, Ravi Anand wrote:
>
> >On Tue, 2010-07-27 James Bottomley wrote:
> >On Tue, 2010-07-27 at 11:56 -0700, Yinghai Lu wrote:
> > LD      drivers/scsi/built-in.o
> > drivers/scsi/qla4xxx/built-in.o: In function `qla82xx_idc_unlock':
> > (.text+0x8814): multiple definition of `qla82xx_idc_unlock'
> > drivers/scsi/qla2xxx/built-in.o:(.text+0x2ea14): first defined here
> > drivers/scsi/qla2xxx/built-in.o:(.text+0x30c56): first defined here
> > make[1]: *** [drivers/scsi/built-in.o] Error 1
> > make: *** [drivers/scsi/] Error 2
>
> >This is because exactly the same functions appear in the qla2xxx/ and
> >qla4xxx/ directories
>
> >Qlogic people, please get this sorted out now, otherwise I'll drop the
> >qla4xxx update before sending to Linus
>
> We will try to resolve this ASAP.

>Thanks

>   When are you planning to send it to Linus ?

>Whenever he opens the merge window ... he's been making noises like it
>will be this week.

> >The best option would be to separate out the 82xx functions so that both
> >2xxx and 4xxx can use them.  In the interim, you could add a
> >discriminator prefix.
>
> We will put in discriminator prefix and  repost the patches.

>Just the one patch that adds the 82xx support, if possible ... that way
>I can just swap the bad one out.

Most likely yes - it will be a single 82xx patch which you can swap out.

Thanks
Ravi

Re: scsi-next fail to be compiled

[Yinghai Lu]

On 07/27/2010 01:56 PM, Ravi Anand wrote:
> 
>>> On Tue, 2010-07-27 James Bottomley wrote:
>> On Tue, 2010-07-27 at 13:45 -0700, Ravi Anand wrote:
>>
>>> On Tue, 2010-07-27 James Bottomley wrote:
>>> On Tue, 2010-07-27 at 11:56 -0700, Yinghai Lu wrote:
>>> LD      drivers/scsi/built-in.o
>>> drivers/scsi/qla4xxx/built-in.o: In function `qla82xx_idc_unlock':
>>> (.text+0x8814): multiple definition of `qla82xx_idc_unlock'
>>> drivers/scsi/qla2xxx/built-in.o:(.text+0x2ea14): first defined here
>>> drivers/scsi/qla2xxx/built-in.o:(.text+0x30c56): first defined here
>>> make[1]: *** [drivers/scsi/built-in.o] Error 1
>>> make: *** [drivers/scsi/] Error 2
>>
>>> This is because exactly the same functions appear in the qla2xxx/ and
>>> qla4xxx/ directories
>>
>>> Qlogic people, please get this sorted out now, otherwise I'll drop the
>>> qla4xxx update before sending to Linus
>>
>> We will try to resolve this ASAP.
> 
>> Thanks
> 
>>   When are you planning to send it to Linus ?
> 
>> Whenever he opens the merge window ... he's been making noises like it
>> will be this week.
> 
>>> The best option would be to separate out the 82xx functions so that both
>>> 2xxx and 4xxx can use them.  In the interim, you could add a
>>> discriminator prefix.
>>
>> We will put in discriminator prefix and  repost the patches.
> 
>> Just the one patch that adds the 82xx support, if possible ... that way
>> I can just swap the bad one out.
> 
> Most likely yes - it will be a single 82xx patch which you can swap out.
> 

so qa2xxx and qa4xxx could be used with 82xx at same time?

Yinghai

Re: scsi-next fail to be compiled

[Vikas Chaudhary]

On Jul 28, 2010, at 2:22 AM, James Bottomley wrote:

> On Tue, 2010-07-27 at 13:45 -0700, Ravi Anand wrote:
>>
>>> On Tue, 2010-07-27 James Bottomley wrote:
>>> On Tue, 2010-07-27 at 11:56 -0700, Yinghai Lu wrote:
>>> LD      drivers/scsi/built-in.o
>>> drivers/scsi/qla4xxx/built-in.o: In function `qla82xx_idc_unlock':
>>> (.text+0x8814): multiple definition of `qla82xx_idc_unlock'
>>> drivers/scsi/qla2xxx/built-in.o:(.text+0x2ea14): first defined here
>>> drivers/scsi/qla2xxx/built-in.o:(.text+0x30c56): first defined here
>>> make[1]: *** [drivers/scsi/built-in.o] Error 1
>>> make: *** [drivers/scsi/] Error 2
>>
>>> This is because exactly the same functions appear in the qla2xxx/  
>>> and
>>> qla4xxx/ directories
>>
>>> Qlogic people, please get this sorted out now, otherwise I'll drop  
>>> the
>>> qla4xxx update before sending to Linus
>>
>> We will try to resolve this ASAP.
>
> Thanks
>
>>  When are you planning to send it to Linus ?
>
> Whenever he opens the merge window ... he's been making noises like it
> will be this week.
>
>>> The best option would be to separate out the 82xx functions so  
>>> that both
>>> 2xxx and 4xxx can use them.  In the interim, you could add a
>>> discriminator prefix.
>>
>> We will put in discriminator prefix and  repost the patches.
>
> Just the one patch that adds the 82xx support, if possible ... that  
> way
> I can just swap the bad one out.

I am sending new 82xx support patch in next mail which address this  
issue.
We added new discriminator prefix "qla4_8xxx_*" for qla4xxx drive  
module.

Please swap new 82xx patch with earlier patch.

Thanks,
Vikas.

>
> James
>
>
>

Re: scsi-next fail to be compiled

[James Bottomley]

On Wed, 2010-07-28 at 15:54 +0530, Vikas Chaudhary wrote:
> On Jul 28, 2010, at 2:22 AM, James Bottomley wrote:
> 
> > On Tue, 2010-07-27 at 13:45 -0700, Ravi Anand wrote:
> >>
> >>> On Tue, 2010-07-27 James Bottomley wrote:
> >>> On Tue, 2010-07-27 at 11:56 -0700, Yinghai Lu wrote:
> >>> LD      drivers/scsi/built-in.o
> >>> drivers/scsi/qla4xxx/built-in.o: In function `qla82xx_idc_unlock':
> >>> (.text+0x8814): multiple definition of `qla82xx_idc_unlock'
> >>> drivers/scsi/qla2xxx/built-in.o:(.text+0x2ea14): first defined here
> >>> drivers/scsi/qla2xxx/built-in.o:(.text+0x30c56): first defined here
> >>> make[1]: *** [drivers/scsi/built-in.o] Error 1
> >>> make: *** [drivers/scsi/] Error 2
> >>
> >>> This is because exactly the same functions appear in the qla2xxx/  
> >>> and
> >>> qla4xxx/ directories
> >>
> >>> Qlogic people, please get this sorted out now, otherwise I'll drop  
> >>> the
> >>> qla4xxx update before sending to Linus
> >>
> >> We will try to resolve this ASAP.
> >
> > Thanks
> >
> >>  When are you planning to send it to Linus ?
> >
> > Whenever he opens the merge window ... he's been making noises like it
> > will be this week.
> >
> >>> The best option would be to separate out the 82xx functions so  
> >>> that both
> >>> 2xxx and 4xxx can use them.  In the interim, you could add a
> >>> discriminator prefix.
> >>
> >> We will put in discriminator prefix and  repost the patches.
> >
> > Just the one patch that adds the 82xx support, if possible ... that  
> > way
> > I can just swap the bad one out.
> 
> I am sending new 82xx support patch in next mail which address this  
> issue.
> We added new discriminator prefix "qla4_8xxx_*" for qla4xxx drive  
> module.
> 
> Please swap new 82xx patch with earlier patch.

OK, I've done the patch swap and it all seems to be working now.

Thanks,

James