From: Luis Gonzalez Fernandez <hidden> Date: 2012-09-04 09:22:38
frag_start(), frag_next(), frag_stop(), walk_zones_in_node() throws
compilation warnings (-Wunused-function) even when are currently used.
This patchs fix the compilation warnings in vmstat.c
Signed-off-by: Luis Gonzalez Fernandez <redacted>
---
mm/vmstat.c | 11 +++++++----
1 file changed, 7 insertions(+), 4 deletions(-)
@@ -639,12 +641,13 @@ static void *frag_next(struct seq_file *m, void *arg, loff_t *pos)returnnext_online_pgdat(pgdat);}-staticvoidfrag_stop(structseq_file*m,void*arg)+staticvoid__attribute__((unused))frag_stop(structseq_file*m,void*arg){}/* Walk all the zones in a node and print using a callback */-staticvoidwalk_zones_in_node(structseq_file*m,pg_data_t*pgdat,+staticvoid__attribute__((unused))walk_zones_in_node(structseq_file*m,+pg_data_t*pgdat,void(*print)(structseq_file*m,pg_data_t*,structzone*)){structzone*zone;
--
1.7.9.5
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
Hi Luis,
On Tue, Sep 4, 2012 at 6:22 AM, Luis Gonzalez Fernandez
[off-list ref] wrote:
frag_start(), frag_next(), frag_stop(), walk_zones_in_node() throws
compilation warnings (-Wunused-function) even when are currently used.
This is very odd. I don't get that warning, and (as you said) there's
no reason to get it,
since those functions are used.
What compiler are you using?
Thanks,
Ezequiel.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
Hi Ezequiel:
I'm using GCC 4.6.3
2012/9/4 Ezequiel Garcia [off-list ref]:
Hi Luis,
On Tue, Sep 4, 2012 at 6:22 AM, Luis Gonzalez Fernandez
[off-list ref] wrote:
quoted
frag_start(), frag_next(), frag_stop(), walk_zones_in_node() throws
compilation warnings (-Wunused-function) even when are currently used.
This is very odd. I don't get that warning, and (as you said) there's
no reason to get it,
since those functions are used.
What compiler are you using?
Thanks,
Ezequiel.
--
--
Luis Gonzalez Fernandez
Telf: 661772374
E-Mail: luisgf@gmail.com
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
Hi Luis,
On Tue, Sep 4, 2012 at 6:51 AM, Luis G.F [off-list ref] wrote:
Hi Ezequiel:
I'm using GCC 4.6.3
Please, avoid top posting as it makes very difficult to follow the discussion
(and people around here hate it).
Also, in the future when fixing warnings you may want to add the warning message
to the commit message.
Anyway, I don't really know why are you getting that (wrong) warning,
but I don't think the solution is to add the 'unused' attribute.
Hope this helps,
Ezequiel.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
Also, in the future when fixing warnings you may want to add the warning message
to the commit message.
Yes, please always quote the messages in the changelog.
Anyway, I don't really know why are you getting that (wrong) warning,
but I don't think the solution is to add the 'unused' attribute.
And yes, let's not work around compiler problems too eagerly. We _do_
occasionally work around bogus warnings, but only long-established ones
which we see no other way of fixing.
In this case, it might be that these functions are indeed unused with
certain Kconfig combinations. For example and from inspection,
CONFIG_PROCFS=n, CONFIG_DEBUG_FS=n, CONFIG_COMPACTION=y might cause
such a warning?
Also, please don't directly use __attribute__((unused)) - we have
various helper macros in include/linux/compiler*.h for this.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
Also, in the future when fixing warnings you may want to add the warning message
to the commit message.
Yes, please always quote the messages in the changelog.
quoted
Anyway, I don't really know why are you getting that (wrong) warning,
but I don't think the solution is to add the 'unused' attribute.
And yes, let's not work around compiler problems too eagerly. We _do_
occasionally work around bogus warnings, but only long-established ones
which we see no other way of fixing.
In this case, it might be that these functions are indeed unused with
certain Kconfig combinations. For example and from inspection,
CONFIG_PROCFS=n, CONFIG_DEBUG_FS=n, CONFIG_COMPACTION=y might cause
such a warning?
I generate a complete random conf (with make randconfig) and the
problem with warnings is that
CONFIG_PROC_FS is undefined but CONFIG_COMPACTION=y (as you say).That's create
certain scenario where the variables are defined but never used.
Also, please don't directly use __attribute__((unused)) - we have
various helper macros in include/linux/compiler*.h for this.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>