linux-2.5.3-pre4/drivers/acenic.c: pci_unmap_addr_set not defined for x86

linux-2.5.3-pre4/drivers/acenic.c: pci_unmap_addr_set not defined for x86

[Adam J. Richter]

	linux-2.5.3-pre4/drivers/acenic.c uses pci_unmap_addr_set,
which is defined for most architectures in include/asm-*/pci.h, but
not for i386.  For i386 this results in undefined references.  I
imagine that this is the result of a missed file (include/asm-i386/pci.h?)
from an Acenic update patch.

Adam J. Richter     __     ______________   4880 Stevens Creek Blvd, Suite 104
adam@yggdrasil.com     \ /                  San Jose, California 95129-1034
+1 408 261-6630         | g g d r a s i l   United States of America
fax +1 408 261-6631      "Free Software For The Rest Of Us."

Re: linux-2.5.3-pre4/drivers/acenic.c: pci_unmap_addr_set not defined for x86

[David S. Miller]

   From: "Adam J. Richter" <adam@yggdrasil.com>
   Date: Thu, 24 Jan 2002 02:01:55 -0800

   	linux-2.5.3-pre4/drivers/acenic.c uses pci_unmap_addr_set,
   which is defined for most architectures in include/asm-*/pci.h, but
   not for i386.  For i386 this results in undefined references.  I
   imagine that this is the result of a missed file (include/asm-i386/pci.h?)
   from an Acenic update patch.
   
No, just a dumb typo:

--- include/asm-i386/pci.h.~1~	Tue Jan 15 10:59:36 2002
+++ include/asm-i386/pci.h	Thu Jan 24 05:32:28 2002
@@ -118,7 +118,7 @@
 #define DECLARE_PCI_UNMAP_ADDR(ADDR_NAME)
 #define DECLARE_PCI_UNMAP_LEN(LEN_NAME)
 #define pci_unmap_addr(PTR, ADDR_NAME)		(0)
-#define pci_unmap_addr_SET(PTR, ADDR_NAME, VAL)	do { } while (0)
+#define pci_unmap_addr_set(PTR, ADDR_NAME, VAL)	do { } while (0)
 #define pci_unmap_len(PTR, LEN_NAME)		(0)
 #define pci_unmap_len_set(PTR, LEN_NAME, VAL)	do { } while (0)
 

Re: linux-2.5.3-pre4/drivers/acenic.c: pci_unmap_addr_set not defined for x86

[Narayan Desai]

I think that this change is necessary for 2.4.18pre6 as well.
 -nld

Re: linux-2.5.3-pre4/drivers/acenic.c: pci_unmap_addr_set not defined for x86

[David S. Miller]

   From: "Narayan Desai" <desai@mcs.anl.gov>
   Date: Thu, 24 Jan 2002 09:41:26 -0600

   I think that this change is necessary for 2.4.18pre6 as well.

Already sent to Marcelo.

Re: linux-2.5.3-pre4/drivers/acenic.c: pci_unmap_addr_set not defined for x86

[Jes Sorensen]

>>>>> "Adam" == Adam J Richter <adam@yggdrasil.com> writes:

Adam> 	linux-2.5.3-pre4/drivers/acenic.c uses pci_unmap_addr_set, which
Adam> is defined for most architectures in include/asm-*/pci.h, but not
Adam> for i386.  For i386 this results in undefined references.  I
Adam> imagine that this is the result of a missed file
Adam> (include/asm-i386/pci.h?)  from an Acenic update patch.

Hi Adam

I haven't had a chance to look at it yet. The patch wasn't done by me
and whoever submitted it didn't seem to think it was worth the effort of
Cc'ing me a copy of it ;-(

I'll take a look.

Thanks,
Jes

Re: linux-2.5.3-pre4/drivers/acenic.c: pci_unmap_addr_set not defined for x86

[Jes Sorensen]

>>>>> "David" == David S Miller <davem@redhat.com> writes:

>    From: Jes Sorensen <jes@wildopensource.com> Date: Thu, 24 Jan
> 2002 08:44:44 -0500
   
>    I haven't had a chance to look at it yet. The patch wasn't
> done by me and whoever submitted it didn't seem to think it was
> worth the effort of Cc'ing me a copy of it ;-(

David> I was changing APIs, do I have to CC: every driver author on the
David> planet when I do this?

Considering a) it's just a few keystrokes to add a CC: line, b) it's the
driver authors who are the first to get the bug reports, then yes it
seems like a very reasonable request.

Jes

Re: linux-2.5.3-pre4/drivers/acenic.c: pci_unmap_addr_set not defined for x86

[Jes Sorensen]

>>>>> "David" == David S Miller <davem@redhat.com> writes:

David>    From: Jes Sorensen <jes@wildopensource.com> Date: Thu, 24 Jan
David> 2002 09:15:42 -0500
   
>    Considering a) it's just a few keystrokes to add a CC: line,
> b) it's the driver authors who are the first to get the bug
> reports, then yes it seems like a very reasonable request.

David> This list actually is the first place the reports typically go
David> to.

Actually thats not the case in my experience, very often thing like that
go to the driver specific mailing lists or the driver maintainer in
private email. In fact this more common than a report going
linux-kernel.

David> I'm a responsible maintainer, when I add a problem, I always go
David> back and fix it or revert my changes.  I never disappear after
David> making these kinds of changes.

I never claimed you don't, but some reports you never receive.

All I am suggesting is a CC, no fairytales or other stories to go with
it that actually takes time to write.

Jes

Re: linux-2.5.3-pre4/drivers/acenic.c: pci_unmap_addr_set not defined for x86

[David S. Miller]

   From: Jes Sorensen <jes@wildopensource.com>
   Date: Thu, 24 Jan 2002 08:44:44 -0500
   
   I haven't had a chance to look at it yet. The patch wasn't done by me
   and whoever submitted it didn't seem to think it was worth the effort of
   Cc'ing me a copy of it ;-(

I was changing APIs, do I have to CC: every driver author on the
planet when I do this?

Re: linux-2.5.3-pre4/drivers/acenic.c: pci_unmap_addr_set not defined for x86

[David S. Miller]

   From: Jes Sorensen <jes@wildopensource.com>
   Date: Thu, 24 Jan 2002 09:15:42 -0500
   
   Considering a) it's just a few keystrokes to add a CC: line, b) it's the
   driver authors who are the first to get the bug reports, then yes it
   seems like a very reasonable request.

This list actually is the first place the reports typically
go to.

I'm a responsible maintainer, when I add a problem, I always go back
and fix it or revert my changes.  I never disappear after making
these kinds of changes.