RFR: 8263236: runtime/os/TestTracePageSizes.java fails on old kernels [v5]

[Aleksey Shipilev]

On Thu, 6 May 2021 14:24:53 GMT, Aleksey Shipilev <shade at openjdk.org> wrote:

> Okay, I did a rough bisect over pre-built Debian kernels, and that points to `4.15` as the first kernel that works. I updated the PR changeset and PR description accordingly. I am still struggling to find a particular commit that solved it, so I can explain to myself why this works...

All right. Here is a kicker. Vanilla `4.14.17` kernel is passing the tests, while Debian's `4.14.17` fails it! Looking at Debian kernel patches, I see a few suspicious ones. Applying [these patches](https://salsa.debian.org/kernel-team/linux/-/commit/3f937de4507cf678ebfa53654bee5ffe20d76ad6) to vanilla `4.14.17` also makes the test fail. This hunk from that Debian's kernel patch is apparently the problem:


diff --git a/include/linux/mman.h b/include/linux/mman.h
index 7c87b6652244..f22c15d5e24c 100644
--- a/include/linux/mman.h
+++ b/include/linux/mman.h
@@ -87,7 +87,8 @@ calc_vm_flag_bits(unsigned long flags)
 {
 	return _calc_vm_trans(flags, MAP_GROWSDOWN,  VM_GROWSDOWN ) |
 	       _calc_vm_trans(flags, MAP_DENYWRITE,  VM_DENYWRITE ) |
-	       _calc_vm_trans(flags, MAP_LOCKED,     VM_LOCKED    );
+	       _calc_vm_trans(flags, MAP_LOCKED,     VM_LOCKED    ) |
+	       _calc_vm_trans(flags, MAP_FIXED,      VM_FIXED     );
 }


It copies the mmap tags to the VMA flag. Now, that thing is getting printed in [show_smap_vma_flags](https://github.com/torvalds/linux/blob/master/fs/proc/task_mmu.c#L603-L681). Notice how that method prepares itself for missing flag cases by [printing `??`](https://github.com/torvalds/linux/blob/master/fs/proc/task_mmu.c#L609-L612). Indeed, Debian patch misses that addition, so we see `??` for some entries in `/proc/smaps`.

How's that a problem for this test? Stare at this regexp:


        String smapsPatternString = "(\\w+)-(\\w+).*?" +
                                    "KernelPageSize:\\s*(\\d*) kB.*?" +
                                    "VmFlags: ([\\w ]*)";


`?` is _not_ `\w`, so the entire thing mismatches the moment the VMA flags have `??` in them. 

And this only happens on some Debian kernels, because you have to have that new unhandled flag. This also explains why `4.15` works, the [later commit](https://salsa.debian.org/kernel-team/linux/-/commit/7dd9b58675f9c285944cd6b12199481765c932a0#b23bf90ba1c0322d4a1bb8ace6fac27f98426b10_21_21) redefined that new flag to existing, handled value for `VM_ARCH`.

So this is deserves a simple fix in the regexp itself:


-                                    "VmFlags: ([\\w ]*)";
+                                    "VmFlags: ([\\w\? ]*)";


Testing that now...

-------------

PR: https://git.openjdk.java.net/jdk/pull/3415