| Message ID | 161052210936.1804207.17896246772670985157.stgit@dwillia2-desk3.amr.corp.intel.com (mailing list archive) |
|---|---|
| State | Accepted |
| Commit | 6694afe31dd67d186199a58d2252be5ea3472692 |
| Headers | show |
| Series | ndctl/test: softoffline, mremap, and misc fixups | expand |
diff --git a/test/dax-pmd.c b/test/dax-pmd.c index b1251db63041..7648e348b0a6 100644 --- a/test/dax-pmd.c +++ b/test/dax-pmd.c @@ -69,6 +69,11 @@ int test_dax_remap(struct ndctl_test *test, int dax_fd, unsigned long align, voi remap = mremap(addr, REMAP_SIZE, REMAP_SIZE, MREMAP_MAYMOVE|MREMAP_FIXED, anon); + if (remap == MAP_FAILED) { + fprintf(stderr, "%s: mremap failed, that's ok too\n", __func__); + return 0; + } + if (remap != anon) { rc = -ENXIO; perror("mremap");
The test_dax_remap() test is a regression check for mishandling of mremap() in the presence of pmd_devmap(). My understanding is that it was a fuzzing condition not something an application would want to do in practice. On recent kernels with commit 73d5e0629919 ("mremap: check if it's possible to split original vma"), the test fails for device-dax. That seems an equally acceptable result of attempting this remap, so update the test rather than ask the kernel to preserve the old behaviour. Signed-off-by: Dan Williams <dan.j.williams@intel.com> --- test/dax-pmd.c | 5 +++++ 1 file changed, 5 insertions(+)