pmem: fail io-requests to known bad blocks

Check the sectors specified in a read bio to see if they hit a known bad
block, and return an error code pmem_do_bvec().

Note that the ->rw_page() is not in a position to return errors.  For
now, copy the same layering violation present in zram_rw_page() to avoid
crashes of the form:

 kernel BUG at mm/filemap.c:822!
 [..]
 Call Trace:
  [<ffffffff811c540e>] page_endio+0x1e/0x60
  [<ffffffff81290d29>] mpage_end_io+0x39/0x60
  [<ffffffff8141c4ef>] bio_endio+0x3f/0x60
  [<ffffffffa005c491>] pmem_make_request+0x111/0x230 [nd_pmem]

...i.e. unlock a page that was already unlocked via pmem_rw_page() =>
page_endio().

Reported-by: Vishal Verma <vishal.l.verma@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>

diff --git a/drivers/nvdimm/pmem.c b/drivers/nvdimm/pmem.c
index 2b1f3009f827029c5f2ea65cf412cd6bd71b9b01..d00c659d13049612121543809d6ffc205ca898d2 100644 (file)
--- a/drivers/nvdimm/pmem.c
+++ b/drivers/nvdimm/pmem.c
@@ -47,7 +47,20 @@ struct pmem_device {
 
 static int pmem_major;
 
-static void pmem_do_bvec(struct pmem_device *pmem, struct page *page,
+static bool is_bad_pmem(struct badblocks *bb, sector_t sector, unsigned int len)
+{
+       if (bb->count) {
+               sector_t first_bad;
+               int num_bad;
+
+               return !!badblocks_check(bb, sector, len / 512, &first_bad,
+                               &num_bad);
+       }
+
+       return false;
+}
+
+static int pmem_do_bvec(struct pmem_device *pmem, struct page *page,
                        unsigned int len, unsigned int off, int rw,
                        sector_t sector)
 {
@@ -56,6 +69,8 @@ static void pmem_do_bvec(struct pmem_device *pmem, struct page *page,
        void __pmem *pmem_addr = pmem->virt_addr + pmem_off;
 
        if (rw == READ) {
+               if (unlikely(is_bad_pmem(&pmem->bb, sector, len)))
+                       return -EIO;
                memcpy_from_pmem(mem + off, pmem_addr, len);
                flush_dcache_page(page);
        } else {
@@ -64,10 +79,12 @@ static void pmem_do_bvec(struct pmem_device *pmem, struct page *page,
        }
 
        kunmap_atomic(mem);
+       return 0;
 }
 
 static blk_qc_t pmem_make_request(struct request_queue *q, struct bio *bio)
 {
+       int rc = 0;
        bool do_acct;
        unsigned long start;
        struct bio_vec bvec;
@@ -76,9 +93,15 @@ static blk_qc_t pmem_make_request(struct request_queue *q, struct bio *bio)
        struct pmem_device *pmem = bdev->bd_disk->private_data;
 
        do_acct = nd_iostat_start(bio, &start);
-       bio_for_each_segment(bvec, bio, iter)
-               pmem_do_bvec(pmem, bvec.bv_page, bvec.bv_len, bvec.bv_offset,
-                               bio_data_dir(bio), iter.bi_sector);
+       bio_for_each_segment(bvec, bio, iter) {
+               rc = pmem_do_bvec(pmem, bvec.bv_page, bvec.bv_len,
+                               bvec.bv_offset, bio_data_dir(bio),
+                               iter.bi_sector);
+               if (rc) {
+                       bio->bi_error = rc;
+                       break;
+               }
+       }
        if (do_acct)
                nd_iostat_end(bio, start);
 
@@ -93,13 +116,22 @@ static int pmem_rw_page(struct block_device *bdev, sector_t sector,
                       struct page *page, int rw)
 {
        struct pmem_device *pmem = bdev->bd_disk->private_data;
+       int rc;
 
-       pmem_do_bvec(pmem, page, PAGE_CACHE_SIZE, 0, rw, sector);
+       rc = pmem_do_bvec(pmem, page, PAGE_CACHE_SIZE, 0, rw, sector);
        if (rw & WRITE)
                wmb_pmem();
-       page_endio(page, rw & WRITE, 0);
 
-       return 0;
+       /*
+        * The ->rw_page interface is subtle and tricky.  The core
+        * retries on any error, so we can only invoke page_endio() in
+        * the successful completion case.  Otherwise, we'll see crashes
+        * caused by double completion.
+        */
+       if (rc == 0)
+               page_endio(page, rw & WRITE, 0);
+
+       return rc;
 }
 
 static long pmem_direct_access(struct block_device *bdev, sector_t sector,