Message ID | 1476790457-7776-1-git-send-email-mark.rutland@arm.com |
---|---|
State | New |
Headers | show |
On 10/18/2016 01:34 PM, Mark Rutland wrote: > Recently, printk was reworked in commit: > > 4bcc595ccd80decb ("printk: reinstate KERN_CONT for printing continuation > lines") > > As of this commit, printk calls missing KERN_CONT will have a linebreak > inserted implicitly. > > In show_free_areas, we miss KERN_CONT in a few cases, and as a result > prints are unexpectedly split over a number of lines, making them > difficult to read (in v4.9-rc1). > > This patch uses pr_cont (with uits implicit KERN_CONT) to mark all > continued prints that occur withing a show_free_areas() call. Note that > show_migration_types() is only called by show_free_areas(). > Depending on CONFIG_NUMA a printk after show_node() may or may not be a > continuation, but follows an explicit newline if not (and thus marking > it as a continuation should not be harmful). I think this was already fixed: http://marc.info/?l=linux-mm&m=147623910031630&w=2 > Signed-off-by: Mark Rutland <mark.rutland@arm.com> > Cc: Andrew Morton <akpm@linux-foundation.org> > Cc: Vlastimil Babka <vbabka@suse.cz> > Cc: Mel Gorman <mgorman@techsingularity.net> > Cc: linux-mm@kvack.org > Cc: linux-kernel@vger.kernel.org > --- > mm/page_alloc.c | 14 +++++++------- > 1 file changed, 7 insertions(+), 7 deletions(-) > > diff --git a/mm/page_alloc.c b/mm/page_alloc.c > index 2b3bf67..833f271 100644 > --- a/mm/page_alloc.c > +++ b/mm/page_alloc.c > @@ -4224,7 +4224,7 @@ static void show_migration_types(unsigned char type) > } > > *p = '\0'; > - printk("(%s) ", tmp); > + pr_cont("(%s) ", tmp); > } > > /* > @@ -4335,7 +4335,7 @@ void show_free_areas(unsigned int filter) > free_pcp += per_cpu_ptr(zone->pageset, cpu)->pcp.count; > > show_node(zone); > - printk("%s" > + pr_cont("%s" > " free:%lukB" > " min:%lukB" > " low:%lukB" > @@ -4382,8 +4382,8 @@ void show_free_areas(unsigned int filter) > K(zone_page_state(zone, NR_FREE_CMA_PAGES))); > printk("lowmem_reserve[]:"); > for (i = 0; i < MAX_NR_ZONES; i++) > - printk(" %ld", zone->lowmem_reserve[i]); > - printk("\n"); > + pr_cont(" %ld", zone->lowmem_reserve[i]); > + pr_cont("\n"); > } > > for_each_populated_zone(zone) { > @@ -4394,7 +4394,7 @@ void show_free_areas(unsigned int filter) > if (skip_free_areas_node(filter, zone_to_nid(zone))) > continue; > show_node(zone); > - printk("%s: ", zone->name); > + pr_cont("%s: ", zone->name); > > spin_lock_irqsave(&zone->lock, flags); > for (order = 0; order < MAX_ORDER; order++) { > @@ -4412,11 +4412,11 @@ void show_free_areas(unsigned int filter) > } > spin_unlock_irqrestore(&zone->lock, flags); > for (order = 0; order < MAX_ORDER; order++) { > - printk("%lu*%lukB ", nr[order], K(1UL) << order); > + pr_cont("%lu*%lukB ", nr[order], K(1UL) << order); > if (nr[order]) > show_migration_types(types[order]); > } > - printk("= %lukB\n", K(total)); > + pr_cont("= %lukB\n", K(total)); > } > > hugetlb_show_meminfo(); >
On Tue, Oct 18, 2016 at 03:00:40PM +0200, Vlastimil Babka wrote: > On 10/18/2016 01:34 PM, Mark Rutland wrote: > >In show_free_areas, we miss KERN_CONT in a few cases, and as a result > >prints are unexpectedly split over a number of lines, making them > >difficult to read (in v4.9-rc1). > > > >This patch uses pr_cont (with uits implicit KERN_CONT) to mark all > >continued prints that occur withing a show_free_areas() call. Note that > >show_migration_types() is only called by show_free_areas(). > >Depending on CONFIG_NUMA a printk after show_node() may or may not be a > >continuation, but follows an explicit newline if not (and thus marking > >it as a continuation should not be harmful). > > I think this was already fixed: > > http://marc.info/?l=linux-mm&m=147623910031630&w=2 So it was; thanks for the pointer! Thanks, Mark.
Already taken care of by http://lkml.kernel.org/r/c7df37c8665134654a17aaeb8b9f6ace1d6db58b.1476239034.git.joe@perches.com maybe it just fell through the cracks... On Tue 18-10-16 12:34:17, Mark Rutland wrote: > Recently, printk was reworked in commit: > > 4bcc595ccd80decb ("printk: reinstate KERN_CONT for printing continuation > lines") > > As of this commit, printk calls missing KERN_CONT will have a linebreak > inserted implicitly. > > In show_free_areas, we miss KERN_CONT in a few cases, and as a result > prints are unexpectedly split over a number of lines, making them > difficult to read (in v4.9-rc1). > > This patch uses pr_cont (with uits implicit KERN_CONT) to mark all > continued prints that occur withing a show_free_areas() call. Note that > show_migration_types() is only called by show_free_areas(). > Depending on CONFIG_NUMA a printk after show_node() may or may not be a > continuation, but follows an explicit newline if not (and thus marking > it as a continuation should not be harmful). > > Signed-off-by: Mark Rutland <mark.rutland@arm.com> > Cc: Andrew Morton <akpm@linux-foundation.org> > Cc: Vlastimil Babka <vbabka@suse.cz> > Cc: Mel Gorman <mgorman@techsingularity.net> > Cc: linux-mm@kvack.org > Cc: linux-kernel@vger.kernel.org > --- > mm/page_alloc.c | 14 +++++++------- > 1 file changed, 7 insertions(+), 7 deletions(-) > > diff --git a/mm/page_alloc.c b/mm/page_alloc.c > index 2b3bf67..833f271 100644 > --- a/mm/page_alloc.c > +++ b/mm/page_alloc.c > @@ -4224,7 +4224,7 @@ static void show_migration_types(unsigned char type) > } > > *p = '\0'; > - printk("(%s) ", tmp); > + pr_cont("(%s) ", tmp); > } > > /* > @@ -4335,7 +4335,7 @@ void show_free_areas(unsigned int filter) > free_pcp += per_cpu_ptr(zone->pageset, cpu)->pcp.count; > > show_node(zone); > - printk("%s" > + pr_cont("%s" > " free:%lukB" > " min:%lukB" > " low:%lukB" > @@ -4382,8 +4382,8 @@ void show_free_areas(unsigned int filter) > K(zone_page_state(zone, NR_FREE_CMA_PAGES))); > printk("lowmem_reserve[]:"); > for (i = 0; i < MAX_NR_ZONES; i++) > - printk(" %ld", zone->lowmem_reserve[i]); > - printk("\n"); > + pr_cont(" %ld", zone->lowmem_reserve[i]); > + pr_cont("\n"); > } > > for_each_populated_zone(zone) { > @@ -4394,7 +4394,7 @@ void show_free_areas(unsigned int filter) > if (skip_free_areas_node(filter, zone_to_nid(zone))) > continue; > show_node(zone); > - printk("%s: ", zone->name); > + pr_cont("%s: ", zone->name); > > spin_lock_irqsave(&zone->lock, flags); > for (order = 0; order < MAX_ORDER; order++) { > @@ -4412,11 +4412,11 @@ void show_free_areas(unsigned int filter) > } > spin_unlock_irqrestore(&zone->lock, flags); > for (order = 0; order < MAX_ORDER; order++) { > - printk("%lu*%lukB ", nr[order], K(1UL) << order); > + pr_cont("%lu*%lukB ", nr[order], K(1UL) << order); > if (nr[order]) > show_migration_types(types[order]); > } > - printk("= %lukB\n", K(total)); > + pr_cont("= %lukB\n", K(total)); > } > > hugetlb_show_meminfo(); > -- > 1.9.1 -- Michal Hocko SUSE Labs
diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 2b3bf67..833f271 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -4224,7 +4224,7 @@ static void show_migration_types(unsigned char type) } *p = '\0'; - printk("(%s) ", tmp); + pr_cont("(%s) ", tmp); } /* @@ -4335,7 +4335,7 @@ void show_free_areas(unsigned int filter) free_pcp += per_cpu_ptr(zone->pageset, cpu)->pcp.count; show_node(zone); - printk("%s" + pr_cont("%s" " free:%lukB" " min:%lukB" " low:%lukB" @@ -4382,8 +4382,8 @@ void show_free_areas(unsigned int filter) K(zone_page_state(zone, NR_FREE_CMA_PAGES))); printk("lowmem_reserve[]:"); for (i = 0; i < MAX_NR_ZONES; i++) - printk(" %ld", zone->lowmem_reserve[i]); - printk("\n"); + pr_cont(" %ld", zone->lowmem_reserve[i]); + pr_cont("\n"); } for_each_populated_zone(zone) { @@ -4394,7 +4394,7 @@ void show_free_areas(unsigned int filter) if (skip_free_areas_node(filter, zone_to_nid(zone))) continue; show_node(zone); - printk("%s: ", zone->name); + pr_cont("%s: ", zone->name); spin_lock_irqsave(&zone->lock, flags); for (order = 0; order < MAX_ORDER; order++) { @@ -4412,11 +4412,11 @@ void show_free_areas(unsigned int filter) } spin_unlock_irqrestore(&zone->lock, flags); for (order = 0; order < MAX_ORDER; order++) { - printk("%lu*%lukB ", nr[order], K(1UL) << order); + pr_cont("%lu*%lukB ", nr[order], K(1UL) << order); if (nr[order]) show_migration_types(types[order]); } - printk("= %lukB\n", K(total)); + pr_cont("= %lukB\n", K(total)); } hugetlb_show_meminfo();
Recently, printk was reworked in commit: 4bcc595ccd80decb ("printk: reinstate KERN_CONT for printing continuation lines") As of this commit, printk calls missing KERN_CONT will have a linebreak inserted implicitly. In show_free_areas, we miss KERN_CONT in a few cases, and as a result prints are unexpectedly split over a number of lines, making them difficult to read (in v4.9-rc1). This patch uses pr_cont (with uits implicit KERN_CONT) to mark all continued prints that occur withing a show_free_areas() call. Note that show_migration_types() is only called by show_free_areas(). Depending on CONFIG_NUMA a printk after show_node() may or may not be a continuation, but follows an explicit newline if not (and thus marking it as a continuation should not be harmful). Signed-off-by: Mark Rutland <mark.rutland@arm.com> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: Mel Gorman <mgorman@techsingularity.net> Cc: linux-mm@kvack.org Cc: linux-kernel@vger.kernel.org --- mm/page_alloc.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) -- 1.9.1