On Sat, 6 Sep 2014 15:38:45 -0400 Sasha Levin <sasha.levin@xxxxxxxxxx> wrote:
> Very similar to VM_BUG_ON_PAGE and VM_BUG_ON_VMA, dump struct_mm
> when the bug is hit.
>
> ...
>
> +void dump_mm(const struct mm_struct *mm)
> +{
> + printk(KERN_ALERT
I'm not sure why we should use KERN_ALERT here - KERN_EMERG is for
"system is unusable", which is a fair descrition of a post-BUG kernel,
yes?
> + "mm %p mmap %p seqnum %d task_size %lu\n"
> +#ifdef CONFIG_MMU
> + "get_unmapped_area %p\n"
> +#endif
This printk is rather hilarious. I can't think of a better way apart
from a great string of individual printks.
And maybe we should use individual printks - dump_mm() presently uses
114 bytes of stack for that printk and that's somewhat of a concern
considering the situations when it will be called.
How's this look?
From: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
Subject: mm/debug.c: use pr_emerg()
- s/KERN_ALERT/pr_emerg/: we're going BUG so let's maximize the changes
of getting the message out.
- convert debug.c to pr_foo()
Cc: Sasha Levin <sasha.levin@xxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
mm/debug.c | 21 +++++++++------------
1 file changed, 9 insertions(+), 12 deletions(-)
diff -puN mm/debug.c~mm-debugc-use-pr_emerg mm/debug.c
--- a/mm/debug.c~mm-debugc-use-pr_emerg
+++ a/mm/debug.c
@@ -57,7 +57,7 @@ static void dump_flags(unsigned long fla
unsigned long mask;
int i;
- printk(KERN_ALERT "flags: %#lx(", flags);
+ pr_emerg("flags: %#lx(", flags);
/* remove zone id */
flags &= (1UL << NR_PAGEFLAGS) - 1;
@@ -69,24 +69,23 @@ static void dump_flags(unsigned long fla
continue;
flags &= ~mask;
- printk("%s%s", delim, names[i].name);
+ pr_cont("%s%s", delim, names[i].name);
delim = "|";
}
/* check for left over flags */
if (flags)
- printk("%s%#lx", delim, flags);
+ pr_cont("%s%#lx", delim, flags);
- printk(")\n");
+ pr_cont(")\n");
}
void dump_page_badflags(struct page *page, const char *reason,
unsigned long badflags)
{
- printk(KERN_ALERT
- "page:%p count:%d mapcount:%d mapping:%p index:%#lx\n",
- page, atomic_read(&page->_count), page_mapcount(page),
- page->mapping, page->index);
+ pr_emerg("page:%p count:%d mapcount:%d mapping:%p index:%#lx\n",
+ page, atomic_read(&page->_count), page_mapcount(page),
+ page->mapping, page->index);
BUILD_BUG_ON(ARRAY_SIZE(pageflag_names) != __NR_PAGEFLAGS);
dump_flags(page->flags, pageflag_names, ARRAY_SIZE(pageflag_names));
if (reason)
@@ -152,8 +151,7 @@ static const struct trace_print_flags vm
void dump_vma(const struct vm_area_struct *vma)
{
- printk(KERN_ALERT
- "vma %p start %p end %p\n"
+ pr_emerg("vma %p start %p end %p\n"
"next %p prev %p mm %p\n"
"prot %lx anon_vma %p vm_ops %p\n"
"pgoff %lx file %p private_data %p\n",
@@ -168,8 +166,7 @@ EXPORT_SYMBOL(dump_vma);
void dump_mm(const struct mm_struct *mm)
{
- printk(KERN_ALERT
- "mm %p mmap %p seqnum %d task_size %lu\n"
+ pr_emerg("mm %p mmap %p seqnum %d task_size %lu\n"
#ifdef CONFIG_MMU
"get_unmapped_area %p\n"
#endif
_
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>