diff mbox

[v1] IB: Move PCI dependency from root KConfig to HW's KConfigs

Message ID 1504585010.3360.10.camel@wdc.com (mailing list archive)
State Superseded
Headers show

Commit Message

Bart Van Assche Sept. 5, 2017, 4:16 a.m. UTC 
On Tue, 2017-09-05 at 07:08 +0300, Leon Romanovsky wrote:
> Why do we need to be dependent on CONFIG_BROKEN?

> Is it still valid for IB stack?


Hello Leon,

As far as I know "&& BROKEN" makes sense but "|| BROKEN" not. The "|| BROKEN" 
part was introduced through the following patch:


From a2b2f45be7e9138bde7fcba3b8e9257fea04d087 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@www.linux.org.uk>

Date: Tue, 23 Aug 2005 22:45:41 +0100
Subject: [PATCH] [PATCH] Kconfig fix (infiniband and PCI)

infiniband uses PCI helpers all over the place (including the core parts) and
won't build without PCI.

Signed-off-by: Al Viro <viro@parcelfarce.linux.theplanet.co.uk>

Signed-off-by: Linus Torvalds <torvalds@osdl.org>

---
 drivers/infiniband/Kconfig | 1 +
 1 file changed, 1 insertion(+)

Comments

Leon Romanovsky Sept. 5, 2017, 4:34 a.m. UTC | #1 
On Tue, Sep 05, 2017 at 04:16:51AM +0000, Bart Van Assche wrote:
> On Tue, 2017-09-05 at 07:08 +0300, Leon Romanovsky wrote:
> > Why do we need to be dependent on CONFIG_BROKEN?
> > Is it still valid for IB stack?
>
> Hello Leon,

Thanks Bart,

It was 12 years ago and it is worth to try to build without BROKEN at all.

>
> As far as I know "&& BROKEN" makes sense but "|| BROKEN" not. The "|| BROKEN"
> part was introduced through the following patch:
>
>
> From a2b2f45be7e9138bde7fcba3b8e9257fea04d087 Mon Sep 17 00:00:00 2001
> From: Al Viro <viro@www.linux.org.uk>
> Date: Tue, 23 Aug 2005 22:45:41 +0100
> Subject: [PATCH] [PATCH] Kconfig fix (infiniband and PCI)
>
> infiniband uses PCI helpers all over the place (including the core parts) and
> won't build without PCI.
>
> Signed-off-by: Al Viro <viro@parcelfarce.linux.theplanet.co.uk>
> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
> ---
>  drivers/infiniband/Kconfig | 1 +
>  1 file changed, 1 insertion(+)
>
> diff --git a/drivers/infiniband/Kconfig b/drivers/infiniband/Kconfig
> index 79c8e2dd9c33..32cdfb30e9b4 100644
> --- a/drivers/infiniband/Kconfig
> +++ b/drivers/infiniband/Kconfig
> @@ -1,6 +1,7 @@
>  menu "InfiniBand support"
>
>  config INFINIBAND
> +	depends on PCI || BROKEN
>  	tristate "InfiniBand support"
>  	---help---
>  	  Core support for InfiniBand (IB).  Make sure to also select
Yuval Shaia Sept. 5, 2017, 8:29 a.m. UTC | #2 
On Tue, Sep 05, 2017 at 04:16:51AM +0000, Bart Van Assche wrote:
> On Tue, 2017-09-05 at 07:08 +0300, Leon Romanovsky wrote:
> > Why do we need to be dependent on CONFIG_BROKEN?
> > Is it still valid for IB stack?
> 
> Hello Leon,
> 
> As far as I know "&& BROKEN" makes sense but "|| BROKEN" not. The "|| BROKEN" 
> part was introduced through the following patch:

Can you elaborate more on (1) why BROKEN is needed and (2) why "&&" make
sense and "||" not?

> 
> 
> From a2b2f45be7e9138bde7fcba3b8e9257fea04d087 Mon Sep 17 00:00:00 2001
> From: Al Viro <viro@www.linux.org.uk>
> Date: Tue, 23 Aug 2005 22:45:41 +0100
> Subject: [PATCH] [PATCH] Kconfig fix (infiniband and PCI)
> 
> infiniband uses PCI helpers all over the place (including the core parts) and
> won't build without PCI.
> 
> Signed-off-by: Al Viro <viro@parcelfarce.linux.theplanet.co.uk>
> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
> ---
>  drivers/infiniband/Kconfig | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/drivers/infiniband/Kconfig b/drivers/infiniband/Kconfig
> index 79c8e2dd9c33..32cdfb30e9b4 100644
> --- a/drivers/infiniband/Kconfig
> +++ b/drivers/infiniband/Kconfig
> @@ -1,6 +1,7 @@
>  menu "InfiniBand support"
>  
>  config INFINIBAND
> +	depends on PCI || BROKEN
>  	tristate "InfiniBand support"
>  	---help---
>  	  Core support for InfiniBand (IB).  Make sure to also select
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Bart Van Assche Sept. 5, 2017, 2:06 p.m. UTC | #3 
On Tue, 2017-09-05 at 11:29 +0300, Yuval Shaia wrote:
> On Tue, Sep 05, 2017 at 04:16:51AM +0000, Bart Van Assche wrote:

> > On Tue, 2017-09-05 at 07:08 +0300, Leon Romanovsky wrote:

> > > Why do we need to be dependent on CONFIG_BROKEN?

> > > Is it still valid for IB stack?

> > 

> > Hello Leon,

> > 

> > As far as I know "&& BROKEN" makes sense but "|| BROKEN" not. The "|| BROKEN" 

> > part was introduced through the following patch:

> 

> Can you elaborate more on (1) why BROKEN is needed and (2) why "&&" make

> sense and "||" not?


Hello Yuval,

Sorry if I was not clear enough. What I meant is that for broken drivers "&& BROKEN"
should be used instead of "|| BROKEN".  My opinion is that "|| BROKEN" is neither
needed nor that it makes sense for InfiniBand.

Bart.
diff mbox

Patch

diff --git a/drivers/infiniband/Kconfig b/drivers/infiniband/Kconfig
index 79c8e2dd9c33..32cdfb30e9b4 100644
--- a/drivers/infiniband/Kconfig
+++ b/drivers/infiniband/Kconfig
@@ -1,6 +1,7 @@ 
 menu "InfiniBand support"
 
 config INFINIBAND
+	depends on PCI || BROKEN
 	tristate "InfiniBand support"
 	---help---
 	  Core support for InfiniBand (IB).  Make sure to also select