mm: madvise: skip unmapped vma holes passed to process_madvise
authorCharan Teja Kalla <quic_charante@quicinc.com>
Tue, 22 Mar 2022 21:46:48 +0000 (14:46 -0700)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Fri, 8 Apr 2022 12:22:56 +0000 (14:22 +0200)
commit 08095d6310a7ce43256b4251577bc66a25c6e1a6 upstream.

The process_madvise() system call is expected to skip holes in vma passed
through 'struct iovec' vector list.  But do_madvise, which
process_madvise() calls for each vma, returns ENOMEM in case of unmapped
holes, despite the VMA is processed.

Thus process_madvise() should treat ENOMEM as expected and consider the
VMA passed to as processed and continue processing other vma's in the
vector list.  Returning -ENOMEM to user, despite the VMA is processed,
will be unable to figure out where to start the next madvise.

Link: https://lkml.kernel.org/r/4f091776142f2ebf7b94018146de72318474e686.1647008754.git.quic_charante@quicinc.com
Fixes: ecb8ac8b1f14("mm/madvise: introduce process_madvise() syscall: an external memory hinting API")
Signed-off-by: Charan Teja Kalla <quic_charante@quicinc.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Nadav Amit <nadav.amit@gmail.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
mm/madvise.c

index 0734db8d53a7a9e8ebb65112c6ae76e9c9b07f2d..b215f0e2ab409902792a3a83b29c98333ba58ad5 100644 (file)
@@ -1287,9 +1287,16 @@ SYSCALL_DEFINE5(process_madvise, int, pidfd, const struct iovec __user *, vec,
 
        while (iov_iter_count(&iter)) {
                iovec = iov_iter_iovec(&iter);
+               /*
+                * do_madvise returns ENOMEM if unmapped holes are present
+                * in the passed VMA. process_madvise() is expected to skip
+                * unmapped holes passed to it in the 'struct iovec' list
+                * and not fail because of them. Thus treat -ENOMEM return
+                * from do_madvise as valid and continue processing.
+                */
                ret = do_madvise(mm, (unsigned long)iovec.iov_base,
                                        iovec.iov_len, behavior);
-               if (ret < 0)
+               if (ret < 0 && ret != -ENOMEM)
                        break;
                iov_iter_advance(&iter, iovec.iov_len);
        }