* [patch] pci: type may be unused in pci_access_init()
@ 2007-04-30 14:34 David Rientjes
2007-05-01 2:22 ` Andrew Morton
0 siblings, 1 reply; 4+ messages in thread
From: David Rientjes @ 2007-04-30 14:34 UTC (permalink / raw)
To: Greg Kroah-Hartman; +Cc: linux-kernel
The automatic 'type' variable is unused in !CONFIG_PCI_DIRECT and
!CONFIG_PCI_MMCONFIG.
Cc: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: David Rientjes <rientjes@google.com>
---
arch/i386/pci/init.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/arch/i386/pci/init.c b/arch/i386/pci/init.c
--- a/arch/i386/pci/init.c
+++ b/arch/i386/pci/init.c
@@ -6,7 +6,7 @@
in the right sequence from here. */
static __init int pci_access_init(void)
{
- int type = 0;
+ int type __attribute__((unused)) = 0;
#ifdef CONFIG_PCI_DIRECT
type = pci_direct_probe();
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [patch] pci: type may be unused in pci_access_init()
2007-04-30 14:34 [patch] pci: type may be unused in pci_access_init() David Rientjes
@ 2007-05-01 2:22 ` Andrew Morton
2007-05-01 2:46 ` David Rientjes
0 siblings, 1 reply; 4+ messages in thread
From: Andrew Morton @ 2007-05-01 2:22 UTC (permalink / raw)
To: David Rientjes; +Cc: Greg Kroah-Hartman, linux-kernel
On Mon, 30 Apr 2007 07:34:52 -0700 (PDT) David Rientjes <rientjes@google.com> wrote:
> The automatic 'type' variable is unused in !CONFIG_PCI_DIRECT and
> !CONFIG_PCI_MMCONFIG.
>
> Cc: Greg Kroah-Hartman <gregkh@suse.de>
> Signed-off-by: David Rientjes <rientjes@google.com>
> ---
> arch/i386/pci/init.c | 2 +-
> 1 files changed, 1 insertions(+), 1 deletions(-)
>
> diff --git a/arch/i386/pci/init.c b/arch/i386/pci/init.c
> --- a/arch/i386/pci/init.c
> +++ b/arch/i386/pci/init.c
> @@ -6,7 +6,7 @@
Please use `diff -p'.
> in the right sequence from here. */
> static __init int pci_access_init(void)
> {
> - int type = 0;
> + int type __attribute__((unused)) = 0;
We have __attribute_used__ for this, but the implementation looks whacky:
it's there for gcc-3 but not for gcc-4 and the intel compiler.
But it looks like the gcc-3 version is there to iron over a gcc
implementation glitch.
> #ifdef CONFIG_PCI_DIRECT
> type = pci_direct_probe();
Of course, one could do
int type = pci_direct_probe();
here instead, but that's a bit fragile.
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [patch] pci: type may be unused in pci_access_init()
2007-05-01 2:22 ` Andrew Morton
@ 2007-05-01 2:46 ` David Rientjes
2007-05-01 3:20 ` Andrew Morton
0 siblings, 1 reply; 4+ messages in thread
From: David Rientjes @ 2007-05-01 2:46 UTC (permalink / raw)
To: Andrew Morton; +Cc: Greg Kroah-Hartman, linux-kernel
On Mon, 30 Apr 2007, Andrew Morton wrote:
> On Mon, 30 Apr 2007 07:34:52 -0700 (PDT) David Rientjes <rientjes@google.com> wrote:
>
> > The automatic 'type' variable is unused in !CONFIG_PCI_DIRECT and
> > !CONFIG_PCI_MMCONFIG.
> >
> > Cc: Greg Kroah-Hartman <gregkh@suse.de>
> > Signed-off-by: David Rientjes <rientjes@google.com>
> > ---
> > arch/i386/pci/init.c | 2 +-
> > 1 files changed, 1 insertions(+), 1 deletions(-)
> >
> > diff --git a/arch/i386/pci/init.c b/arch/i386/pci/init.c
> > --- a/arch/i386/pci/init.c
> > +++ b/arch/i386/pci/init.c
> > @@ -6,7 +6,7 @@
>
> Please use `diff -p'.
>
The patch would have appeared the same, pci_access_init() is the first
function (and there are no prototypes) in arch/i386/pci/init.c and at line
6 it had not yet begun.
diff -rupN a/init.c b/init.c
--- a/init.c 2007-04-30 19:28:59.887833000 -0700
+++ b/init.c 2007-04-30 19:29:26.615993000 -0700
@@ -6,7 +6,7 @@
in the right sequence from here. */
static __init int pci_access_init(void)
{
- int type = 0;
+ int type __attribute__((unused)) = 0;
#ifdef CONFIG_PCI_DIRECT
type = pci_direct_probe();
> > #ifdef CONFIG_PCI_DIRECT
> > type = pci_direct_probe();
>
> Of course, one could do
>
> int type = pci_direct_probe();
>
> here instead, but that's a bit fragile.
>
No, because 'type' can be referenced in the !CONFIG_PCI_DIRECT and
CONFIG_PCI_MMCONFIG case:
static __init int pci_access_init(void)
{
int type = 0;
#ifdef CONFIG_PCI_DIRECT
type = pci_direct_probe();
#endif
#ifdef CONFIG_PCI_MMCONFIG
pci_mmcfg_init(type);
#endif
...
So, due to the whacky implementation of __attribute_used__, we can just do
this:
---
arch/i386/pci/init.c | 2 ++
1 files changed, 2 insertions(+), 0 deletions(-)
diff --git a/arch/i386/pci/init.c b/arch/i386/pci/init.c
index b21b6da..508d6fd 100644
--- a/arch/i386/pci/init.c
+++ b/arch/i386/pci/init.c
@@ -6,7 +6,9 @@
in the right sequence from here. */
static __init int pci_access_init(void)
{
+#if defined(CONFIG_PCI_DIRECT) || defined(CONFIG_PCI_MMCONFIG)
int type = 0;
+#endif
#ifdef CONFIG_PCI_DIRECT
type = pci_direct_probe();
^ permalink raw reply related [flat|nested] 4+ messages in thread
* Re: [patch] pci: type may be unused in pci_access_init()
2007-05-01 2:46 ` David Rientjes
@ 2007-05-01 3:20 ` Andrew Morton
0 siblings, 0 replies; 4+ messages in thread
From: Andrew Morton @ 2007-05-01 3:20 UTC (permalink / raw)
To: David Rientjes; +Cc: Greg Kroah-Hartman, linux-kernel
On Mon, 30 Apr 2007 19:46:12 -0700 (PDT) David Rientjes <rientjes@google.com> wrote:
> So, due to the whacky implementation of __attribute_used__, we can just do
> this:
> ---
> arch/i386/pci/init.c | 2 ++
> 1 files changed, 2 insertions(+), 0 deletions(-)
>
> diff --git a/arch/i386/pci/init.c b/arch/i386/pci/init.c
> index b21b6da..508d6fd 100644
> --- a/arch/i386/pci/init.c
> +++ b/arch/i386/pci/init.c
> @@ -6,7 +6,9 @@
> in the right sequence from here. */
> static __init int pci_access_init(void)
> {
> +#if defined(CONFIG_PCI_DIRECT) || defined(CONFIG_PCI_MMCONFIG)
> int type = 0;
> +#endif
Drat, I was hoping someone would unwhacky __attribute_used__.
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2007-05-01 3:21 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2007-04-30 14:34 [patch] pci: type may be unused in pci_access_init() David Rientjes
2007-05-01 2:22 ` Andrew Morton
2007-05-01 2:46 ` David Rientjes
2007-05-01 3:20 ` Andrew Morton
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox