UbiquitousOS
/
OpenCloudOS-Kernel
mirror of https://gitee.com/OpenCloudOS/OpenCloudOS-Kernel.git
11
0
Fork 0
Code Issues Projects Releases Wiki Activity

dma-debug: fix debug_dma_assert_idle(), use rcu_read_lock() 

Since commit 2a9127fcf2 ("mm: rewrite wait_on_page_bit_common()
logic") improved unlock_page(), it has become more noticeable how
cow_user_page() in a kernel with CONFIG_DMA_API_DEBUG=y can create and
suffer from heavy contention on DMA debug's radix_lock in
debug_dma_assert_idle().

It is only doing a lookup: use rcu_read_lock() and rcu_read_unlock()
instead; though that does require the static ents[] to be moved
onstack...

...but, hold on, isn't that radix_tree_gang_lookup() and loop doing
quite the wrong thing: searching CACHELINES_PER_PAGE entries for an
exact match with the first cacheline of the page in question?
radix_tree_gang_lookup() is the right tool for the job, but we need
nothing more than to check the first entry it can find, reporting if
that falls anywhere within the page.

(Is RCU safe here? As safe as using the spinlock was. The entries are
never freed, so don't need to be freed by RCU. They may be reused, and
there is a faint chance of a race, with an offending entry reused while
printing its error info; but the spinlock did not prevent that either,
and I agree that it's not worth worrying about. ]

[ Side noe: this patch is a clear improvement to the status quo, but the
  next patch will be removing this debug function entirely.

  But just in case we decide we want to resurrect the debugging code
  some day, I'm first applying this improvement patch so that it doesn't
  get lost    - Linus ]

Fixes: 3b7a6418c7 ("dma debug: account for cachelines and read-only mappings in overlap tracking")
Signed-off-by: Hugh Dickins <hughd@google.com>
Acked-by: Dan Williams <dan.j.williams@intel.com>
Acked-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This commit is contained in:
Hugh Dickins 2020-08-12 20:17:00 -07:00 committed by Linus Torvalds
parent b923f1247b
commit a85ffd59bd
1 changed files with 9 additions and 18 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

27
kernel/dma/debug.c
View File

@ -565,11 +565,8 @@ static void active_cacheline_remove(struct dma_debug_entry *entry)
*/ */
void debug_dma_assert_idle(struct page *page) void debug_dma_assert_idle(struct page *page)
{ {
static struct dma_debug_entry *ents[CACHELINES_PER_PAGE]; struct dma_debug_entry *entry;
struct dma_debug_entry *entry = NULL; unsigned long pfn;
void **results = (void **) &ents;
unsigned int nents, i;
unsigned long flags;
phys_addr_t cln; phys_addr_t cln;
if (dma_debug_disabled()) if (dma_debug_disabled())
@ -578,20 +575,14 @@ void debug_dma_assert_idle(struct page *page)
if (!page) if (!page)
return; return;
cln = (phys_addr_t) page_to_pfn(page) << CACHELINE_PER_PAGE_SHIFT; pfn = page_to_pfn(page);
spin_lock_irqsave(&radix_lock, flags); cln = (phys_addr_t) pfn << CACHELINE_PER_PAGE_SHIFT;
nents = radix_tree_gang_lookup(&dma_active_cacheline, results, cln,
CACHELINES_PER_PAGE);
for (i = 0; i < nents; i++) {
phys_addr_t ent_cln = to_cacheline_number(ents[i]);
if (ent_cln == cln) { rcu_read_lock();
entry = ents[i]; if (!radix_tree_gang_lookup(&dma_active_cacheline, (void **) &entry,
break; cln, 1) || entry->pfn != pfn)
} else if (ent_cln >= cln + CACHELINES_PER_PAGE) entry = NULL;
break; rcu_read_unlock();
}
spin_unlock_irqrestore(&radix_lock, flags);
if (!entry) if (!entry)
return; return;