RE: 2.6: unused code under drivers/message/fusion/

RE: 2.6: unused code under drivers/message/fusion/

[Moore, Eric Dean]

We need to hold off on this change. Yes, there are 
customers of LSI Logic using mptstm.c, as
part of the target-mode drivers.  

The proposed generic target mode drivers proposal is yet part
of the kernel.  
http://scst.sourceforge.net/
We are looking into supporting this once its available.

Eric Moore
LSI Logic


On Sunday, November 07, 2004 1:18 PM, Adrian Bunk wrote:
> 
> I know this patch is wrong, but I'd like to note that it actually 
> changes nothing since all code removed or made static 
> currently has no 
> in-kernel users (there seems to be a mptstm.c missing?).
> 
> Please comment on how to correctly handle this.
> 
> 
> diffstat output:
>  drivers/message/fusion/mptbase.c  |    7 ++-----
>  drivers/message/fusion/mptbase.h  |    2 --
>  drivers/message/fusion/mptscsih.c |    2 +-
>  3 files changed, 3 insertions(+), 8 deletions(-)
> 
> 
> Signed-off-by: Adrian Bunk <bunk@stusta.de>
> 
> --- 
> linux-2.6.10-rc1-mm3-full/drivers/message/fusion/mptscsi
> h.c.old	2004-11-07 20:56:13.000000000 +0100
> +++ 
> linux-2.6.10-rc1-mm3-full/drivers/message/fusion/mptscsih.c	
> 2004-11-07 20:56:48.000000000 +0100
> @@ -98,7 +98,7 @@
>  
>  /* Set string for command line args from insmod */
>  #ifdef MODULE
> -char *mptscsih = NULL;
> +static char *mptscsih = NULL;
>  module_param(mptscsih, charp, 0);
>  #endif
>  
> --- 
> linux-2.6.10-rc1-mm3-full/drivers/message/fusion/mptbas
> e.h.old	2004-11-07 20:57:21.000000000 +0100
> +++ 
> linux-2.6.10-rc1-mm3-full/drivers/message/fusion/mptbase.h	
> 2004-11-07 20:57:37.000000000 +0100
> @@ -984,10 +984,8 @@
>   *  Public data decl's...
>   */
>  extern struct list_head	  ioc_list;
> -extern struct proc_dir_entry	*mpt_proc_root_dir;
>  
>  extern int		  mpt_lan_index;	/* needed by mptlan.c */
> -extern int		  mpt_stm_index;	/* needed by mptstm.c */
>  
>  
> /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
> =-=-=-=-=-=-=-=*/
>  #endif		/* } __KERNEL__ */
> --- 
> linux-2.6.10-rc1-mm3-full/drivers/message/fusion/mptbas
> e.c.old	2004-11-07 20:57:46.000000000 +0100
> +++ 
> linux-2.6.10-rc1-mm3-full/drivers/message/fusion/mptbase.c	
> 2004-11-07 20:58:54.000000000 +0100
> @@ -133,9 +133,8 @@
>   *  Public data...
>   */
>  int mpt_lan_index = -1;
> -int mpt_stm_index = -1;
>  
> -struct proc_dir_entry *mpt_proc_root_dir;
> +static struct proc_dir_entry *mpt_proc_root_dir;
>  
>  #define WHOINIT_UNKNOWN		0xAA
>  
> @@ -355,7 +354,7 @@
>  			dmfprintk((MYIOC_s_INFO_FMT "Got TURBO 
> reply req_idx=%08x\n", ioc->name, pa));
>  			type = (pa >> MPI_CONTEXT_REPLY_TYPE_SHIFT);
>  			if (type == 
> MPI_CONTEXT_REPLY_TYPE_SCSI_TARGET) {
> -				cb_idx = mpt_stm_index;
> +				cb_idx = -1;
>  				mf = NULL;
>  				mr = (MPT_FRAME_HDR *) 
> CAST_U32_TO_PTR(pa);
>  			} else if (type == MPI_CONTEXT_REPLY_TYPE_LAN) {
> @@ -5868,7 +5867,6 @@
>  
>  
> /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
> =-=-=-=-=-=-=-=*/
>  EXPORT_SYMBOL(ioc_list);
> -EXPORT_SYMBOL(mpt_proc_root_dir);
>  EXPORT_SYMBOL(mpt_register);
>  EXPORT_SYMBOL(mpt_deregister);
>  EXPORT_SYMBOL(mpt_event_register);
> @@ -5886,7 +5884,6 @@
>  EXPORT_SYMBOL(mpt_GetIocState);
>  EXPORT_SYMBOL(mpt_print_ioc_summary);
>  EXPORT_SYMBOL(mpt_lan_index);
> -EXPORT_SYMBOL(mpt_stm_index);
>  EXPORT_SYMBOL(mpt_HardResetHandler);
>  EXPORT_SYMBOL(mpt_config);
>  EXPORT_SYMBOL(mpt_toolbox);
> 
> 
> 

Re: 2.6: unused code under drivers/message/fusion/

[Vladislav Bolkhovitin]

Moore, Eric Dean wrote:
> We need to hold off on this change. Yes, there are 
> customers of LSI Logic using mptstm.c, as
> part of the target-mode drivers.  
> 
> The proposed generic target mode drivers proposal is yet part
> of the kernel.  
> http://scst.sourceforge.net/
> We are looking into supporting this once its available.

Well, SCST is already available, stable and useful. People use it 
without considerable problems, except with inconvenient LUNs management, 
which we are going to fix in the next version. I don't expect it will be 
considering for the kernel inclusion at least until 2.7. So, you can 
start supporting it right now :-).

Vlad

> Eric Moore
> LSI Logic
> 
> 
> On Sunday, November 07, 2004 1:18 PM, Adrian Bunk wrote:
> 
>>I know this patch is wrong, but I'd like to note that it actually 
>>changes nothing since all code removed or made static 
>>currently has no 
>>in-kernel users (there seems to be a mptstm.c missing?).
>>
>>Please comment on how to correctly handle this.
>>
>>
>>diffstat output:
>> drivers/message/fusion/mptbase.c  |    7 ++-----
>> drivers/message/fusion/mptbase.h  |    2 --
>> drivers/message/fusion/mptscsih.c |    2 +-
>> 3 files changed, 3 insertions(+), 8 deletions(-)
>>
>>
>>Signed-off-by: Adrian Bunk <bunk@stusta.de>
>>
>>--- 
>>linux-2.6.10-rc1-mm3-full/drivers/message/fusion/mptscsi
>>h.c.old	2004-11-07 20:56:13.000000000 +0100
>>+++ 
>>linux-2.6.10-rc1-mm3-full/drivers/message/fusion/mptscsih.c	
>>2004-11-07 20:56:48.000000000 +0100
>>@@ -98,7 +98,7 @@
>> 
>> /* Set string for command line args from insmod */
>> #ifdef MODULE
>>-char *mptscsih = NULL;
>>+static char *mptscsih = NULL;
>> module_param(mptscsih, charp, 0);
>> #endif
>> 
>>--- 
>>linux-2.6.10-rc1-mm3-full/drivers/message/fusion/mptbas
>>e.h.old	2004-11-07 20:57:21.000000000 +0100
>>+++ 
>>linux-2.6.10-rc1-mm3-full/drivers/message/fusion/mptbase.h	
>>2004-11-07 20:57:37.000000000 +0100
>>@@ -984,10 +984,8 @@
>>  *  Public data decl's...
>>  */
>> extern struct list_head	  ioc_list;
>>-extern struct proc_dir_entry	*mpt_proc_root_dir;
>> 
>> extern int		  mpt_lan_index;	/* needed by mptlan.c */
>>-extern int		  mpt_stm_index;	/* needed by mptstm.c */
>> 
>> 
>>/*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
>>=-=-=-=-=-=-=-=*/
>> #endif		/* } __KERNEL__ */
>>--- 
>>linux-2.6.10-rc1-mm3-full/drivers/message/fusion/mptbas
>>e.c.old	2004-11-07 20:57:46.000000000 +0100
>>+++ 
>>linux-2.6.10-rc1-mm3-full/drivers/message/fusion/mptbase.c	
>>2004-11-07 20:58:54.000000000 +0100
>>@@ -133,9 +133,8 @@
>>  *  Public data...
>>  */
>> int mpt_lan_index = -1;
>>-int mpt_stm_index = -1;
>> 
>>-struct proc_dir_entry *mpt_proc_root_dir;
>>+static struct proc_dir_entry *mpt_proc_root_dir;
>> 
>> #define WHOINIT_UNKNOWN		0xAA
>> 
>>@@ -355,7 +354,7 @@
>> 			dmfprintk((MYIOC_s_INFO_FMT "Got TURBO 
>>reply req_idx=%08x\n", ioc->name, pa));
>> 			type = (pa >> MPI_CONTEXT_REPLY_TYPE_SHIFT);
>> 			if (type == 
>>MPI_CONTEXT_REPLY_TYPE_SCSI_TARGET) {
>>-				cb_idx = mpt_stm_index;
>>+				cb_idx = -1;
>> 				mf = NULL;
>> 				mr = (MPT_FRAME_HDR *) 
>>CAST_U32_TO_PTR(pa);
>> 			} else if (type == MPI_CONTEXT_REPLY_TYPE_LAN) {
>>@@ -5868,7 +5867,6 @@
>> 
>> 
>>/*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
>>=-=-=-=-=-=-=-=*/
>> EXPORT_SYMBOL(ioc_list);
>>-EXPORT_SYMBOL(mpt_proc_root_dir);
>> EXPORT_SYMBOL(mpt_register);
>> EXPORT_SYMBOL(mpt_deregister);
>> EXPORT_SYMBOL(mpt_event_register);
>>@@ -5886,7 +5884,6 @@
>> EXPORT_SYMBOL(mpt_GetIocState);
>> EXPORT_SYMBOL(mpt_print_ioc_summary);
>> EXPORT_SYMBOL(mpt_lan_index);
>>-EXPORT_SYMBOL(mpt_stm_index);
>> EXPORT_SYMBOL(mpt_HardResetHandler);
>> EXPORT_SYMBOL(mpt_config);
>> EXPORT_SYMBOL(mpt_toolbox);
>>
>>
>>
> 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
> 
> 

Re: 2.6: unused code under drivers/message/fusion/

[Adrian Bunk]

On Wed, Nov 10, 2004 at 11:11:51AM +0300, Vladislav Bolkhovitin wrote:
> 
> Moore, Eric Dean wrote:
> >We need to hold off on this change. Yes, there are 
> >customers of LSI Logic using mptstm.c, as
> >part of the target-mode drivers.  
> >
> >The proposed generic target mode drivers proposal is yet part
> >of the kernel.  
> >http://scst.sourceforge.net/
> >We are looking into supporting this once its available.
> 
> Well, SCST is already available, stable and useful. People use it 
> without considerable problems, except with inconvenient LUNs management, 
> which we are going to fix in the next version. I don't expect it will be 
> considering for the kernel inclusion at least until 2.7. So, you can 
> start supporting it right now :-).

With the current kernel development model, there is no 2.7 planned for 
the next years.

Linus and Andrew believe 6 was an odd number, so you could submit your 
code now. [1]

> Vlad

cu
Adrian

[1] this is a slightly abbreviated version of the development model
    Linus announced

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed

Re: 2.6: unused code under drivers/message/fusion/

[Vladislav Bolkhovitin]

Adrian Bunk wrote:
> On Wed, Nov 10, 2004 at 11:11:51AM +0300, Vladislav Bolkhovitin wrote:
> 
>>Moore, Eric Dean wrote:
>>
>>>We need to hold off on this change. Yes, there are 
>>>customers of LSI Logic using mptstm.c, as
>>>part of the target-mode drivers.  
>>>
>>>The proposed generic target mode drivers proposal is yet part
>>>of the kernel.  
>>>http://scst.sourceforge.net/
>>>We are looking into supporting this once its available.
>>
>>Well, SCST is already available, stable and useful. People use it 
>>without considerable problems, except with inconvenient LUNs management, 
>>which we are going to fix in the next version. I don't expect it will be 
>>considering for the kernel inclusion at least until 2.7. So, you can 
>>start supporting it right now :-).
> 
> 
> With the current kernel development model, there is no 2.7 planned for 
> the next years.
> 
> Linus and Andrew believe 6 was an odd number, so you could submit your 
> code now. [1]

OK, I'll prepare the next version as the kernel patch.

Thanks,
Vlad

>>Vlad
> 
> 
> cu
> Adrian
> 
> [1] this is a slightly abbreviated version of the development model
>     Linus announced
> 

Re: 2.6: unused code under drivers/message/fusion/

[Adrian Bunk]

On Wed, Nov 10, 2004 at 12:58:47PM +0300, Vladislav Bolkhovitin wrote:
> Adrian Bunk wrote:
> >On Wed, Nov 10, 2004 at 11:11:51AM +0300, Vladislav Bolkhovitin wrote:
> >
> >>Moore, Eric Dean wrote:
> >>
> >>>We need to hold off on this change. Yes, there are 
> >>>customers of LSI Logic using mptstm.c, as
> >>>part of the target-mode drivers.  
> >>>
> >>>The proposed generic target mode drivers proposal is yet part
> >>>of the kernel.  
> >>>http://scst.sourceforge.net/
> >>>We are looking into supporting this once its available.
> >>
> >>Well, SCST is already available, stable and useful. People use it 
> >>without considerable problems, except with inconvenient LUNs management, 
> >>which we are going to fix in the next version. I don't expect it will be 
> >>considering for the kernel inclusion at least until 2.7. So, you can 
> >>start supporting it right now :-).
> >
> >
> >With the current kernel development model, there is no 2.7 planned for 
> >the next years.
> >
> >Linus and Andrew believe 6 was an odd number, so you could submit your 
> >code now. [1]
> 
> OK, I'll prepare the next version as the kernel patch.

Any news regarding this?

> Thanks,
> Vlad

Thanks
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed

Re: 2.6: unused code under drivers/message/fusion/

[Vladislav Bolkhovitin]

Adrian Bunk wrote:
> On Wed, Nov 10, 2004 at 12:58:47PM +0300, Vladislav Bolkhovitin wrote:
> 
>>Adrian Bunk wrote:
>>
>>>On Wed, Nov 10, 2004 at 11:11:51AM +0300, Vladislav Bolkhovitin wrote:
>>>
>>>
>>>>Moore, Eric Dean wrote:
>>>>
>>>>
>>>>>We need to hold off on this change. Yes, there are 
>>>>>customers of LSI Logic using mptstm.c, as
>>>>>part of the target-mode drivers.  
>>>>>
>>>>>The proposed generic target mode drivers proposal is yet part
>>>>>of the kernel.  
>>>>>http://scst.sourceforge.net/
>>>>>We are looking into supporting this once its available.
>>>>
>>>>Well, SCST is already available, stable and useful. People use it 
>>>>without considerable problems, except with inconvenient LUNs management, 
>>>>which we are going to fix in the next version. I don't expect it will be 
>>>>considering for the kernel inclusion at least until 2.7. So, you can 
>>>>start supporting it right now :-).
>>>
>>>
>>>With the current kernel development model, there is no 2.7 planned for 
>>>the next years.
>>>
>>>Linus and Andrew believe 6 was an odd number, so you could submit your 
>>>code now. [1]
>>
>>OK, I'll prepare the next version as the kernel patch.
> 
> 
> Any news regarding this?

Unfortunately, I have not much time for it now, so I can't expect
release of 0.9.3 version, which is going to have the patch, earlier than
in May. But I'm ready to publish 0.9.3-pre1 as well as some bugfixes in
the Qlogic driver in the nearest future, as soon as I settle down some
untechnical stuff. If somebody is interested, I can send him a copy
privately. As before, it still has ZERO modifications of the kernel, so
it could live perfectly outside the tree.

Also, everybody who think that he/she can help in testing,
developing, bugfixing or just commenting/suggesting are welcome.

Vlad

Re: 2.6: unused code under drivers/message/fusion/

[Vladislav Bolkhovitin]

Adrian Bunk wrote:
> On Wed, Nov 10, 2004 at 12:58:47PM +0300, Vladislav Bolkhovitin wrote:
> 
>>Adrian Bunk wrote:
>>
>>>On Wed, Nov 10, 2004 at 11:11:51AM +0300, Vladislav Bolkhovitin wrote:
>>>
>>>
>>>>Moore, Eric Dean wrote:
>>>>
>>>>
>>>>>We need to hold off on this change. Yes, there are 
>>>>>customers of LSI Logic using mptstm.c, as
>>>>>part of the target-mode drivers.  
>>>>>
>>>>>The proposed generic target mode drivers proposal is yet part
>>>>>of the kernel.  
>>>>>http://scst.sourceforge.net/
>>>>>We are looking into supporting this once its available.
>>>>
>>>>Well, SCST is already available, stable and useful. People use it 
>>>>without considerable problems, except with inconvenient LUNs management, 
>>>>which we are going to fix in the next version. I don't expect it will be 
>>>>considering for the kernel inclusion at least until 2.7. So, you can 
>>>>start supporting it right now :-).
>>>
>>>
>>>With the current kernel development model, there is no 2.7 planned for 
>>>the next years.
>>>
>>>Linus and Andrew believe 6 was an odd number, so you could submit your 
>>>code now. [1]
>>
>>OK, I'll prepare the next version as the kernel patch.
> 
> 
> Any news regarding this?

Unfortunately, I have not much time for it now, so I can't expect
release of 0.9.3 version, which is going to have the patch, earlier than
in May. But I'm ready to publish 0.9.3-pre1 as well as some bugfixes in
the Qlogic driver in the nearest future, as soon as I settle down some
untechnical stuff. If somebody is interested, I can send him a copy
privately. As before, it still has ZERO modifications of the kernel, so
it could live perfectly outside the tree.

Also, everybody who think that he/she can help in testing,
developing, bugfixing or just commenting/suggesting are welcome.

Vlad

Re: 2.6: unused code under drivers/message/fusion/

[Adrian Bunk]

On Mon, Nov 08, 2004 at 04:12:45PM -0700, Moore, Eric Dean wrote:

> We need to hold off on this change. Yes, there are 
> customers of LSI Logic using mptstm.c, as
> part of the target-mode drivers.  
>...

Wy can't you submit them for inclusion into the kernel?

> Eric Moore
> LSI Logic

cu
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed