[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Xen-devel] [qemu-upstream-4.11-testing test] 136184: regressions - FAIL
On 6/4/19 6:39 PM, Stefano Stabellini wrote:
On Tue, 4 Jun 2019, Julien Grall wrote:
Hi Stefano,
On 6/4/19 6:09 PM, Stefano Stabellini wrote:
On Tue, 4 Jun 2019, Julien Grall wrote:
Hi Jan,
On 6/4/19 8:06 AM, Jan Beulich wrote:
On 03.06.19 at 19:15, <anthony.perard@xxxxxxxxxx> wrote:
On Tue, May 21, 2019 at 05:52:12PM +0100, Julien Grall wrote:
The same error cannot be reproduced on laxton*. Looking at the test
history,
it looks like qemu-upstream-4.12-testing flight has run successfully
a
few
times on rochester*. So we may have fixed the error in Xen 4.12.
Potential candidates would be:
- 00c96d7742 "xen/arm: mm: Set-up page permission for Xen
mappings
earlier on"
- f60658c6ae "xen/arm: Stop relocating Xen"
Ian, is it something the bisector could automatically look at?
If not, I will need to find some time and borrow the board to bisect
the
issues.
I attempted to do that bisection myself, and the first commit that git
wanted to try, a common commit to both branches, boots just fine.
Thanks for doing this!
One thing that, for now, completely escapes me: How come the
main 4.11 branch has progressed fine, but the qemuu one has
got stalled like this?
Because Xen on Arm today does not fully respect the Arm Arm when modifying
the
page-tables. This may result to TLB conflict and break of coherency.
Yes, I follow your reasoning, but it is still quite strange that it only
happens with the qemu testing branch. Maybe it is because laxton was
picked instead of rochester to run the tests for this branch? Otherwise,
there must be a difference in the Xen configuration between the normal
branch and the qemu testing branch, aside from QEMU of course, that
shouldn't make any differences.
Per the discussion before, the .config is different between the 2 flights.
QEMU testing is not selecting CONFIG_LIVEPATCH while staging-4.11 is.
Has anybody tried to start selecting CONFIG_LIVEPATCH in the QEMU testing
branch? Is it possible to give it a try?
I don't know and I am not sure how this would help here it is pretty
clear that backporting 00c96d7742 "xen/arm: mm: Set-up page permission
for Xen mappings earlier on" is actually going to help booting.
So it is very unlikely that CONFIG_LIVEPATCH is the problem.
It turns out that the first commit that fails to boot on rochester is
e202feb713 xen/cmdline: Fix buggy strncmp(s, LITERAL, ss - s)
construct
(even with the "eb8acba82a xen: Fix backport of .." applied)
Now that's particularly odd a regression candidate. It doesn't
touch any Arm code at all (nor does the fixup commit). And the
common code changes don't look "risky" either; the one thing that
jumps out as the most likely of all the unlikely candidates would
seem to be the xen/common/efi/boot.c change, but if there was
a problem there then the EFI boot on Arm would be latently
broken in other ways as well. Plus, of course, you say that the
same change is no problem on 4.12.
Of course the commit itself could be further "bisected" - all
changes other than the introduction of cmdline_strcmp() are
completely independent of one another.
I think this is just a red-herring. The commit is probably modifying
enough
the layout of Xen that TLB conflict will appear.
Anthony said backporting 00c96d7742 "xen/arm: mm: Set-up page permission
for
Xen mappings earlier on" makes staging-4.11 boots. This patch removes some
of
the potential cause of TLB conflict.
I haven't suggested a backport of this patch so far, because there are
still
TLB conflict possible within the function modified. It might also be
possible
that it exposes more of TLB conflict as more work in Xen is needed (see my
MM-PARTn series).
I don't know whether backporting this patch is worth it compare to the
risk it
introduces.
I think we should backport 00c96d7742. We don't need to fix all issues,
we only need to make improvements without introducing more bugs.
From that standpoints, I think 00c96d7742 is doable. I'll backport it now to
4.11.
You don't seem to assess/acknowledge any risk I mention in this thread.
Note that I am not suggesting to not backport it. I am trying to understand
how you came to your conclusion here.
Based on the fact that by code inspection the patch should be risk
decremental in terms of Arm Arm violations, which is consistent with the
fact that Anthony found it "fixing" the regression. Do you foresee cases
where the patch increments the risk of failure?
Well yes and no. I guess you haven't read what I wrote on the separate
thread.
Yes, two potential source of TLB conflict is removed by avoiding
replacing 4KB entries with 2MB block entry (and vice versa) without
respecting the Break-Before-Make.
No, this patch introducing another source of TLB conflict if the
processor is caching intermediate translation (this is implementation
defined).
The bug reported by osstest actually taught me that even if Xen may boot
today on a given platform, this may not be the case tomorrow because of
the slight change in the code ordering (and therefore memory access).
/!\ Below is my interpretation and does not imply I am correct ;)
However, such Arm Arm violations are mostly gathered around boot and
shouldn't affect runtime. IOW, Xen would stop booting on those platforms
rather than making unrealiable. So it would not be too bad.
/!\ End
We just have to be aware of the risk we are taking with backporting the
patch.
What about the other older stanging branches?
The only one we could consider is 4.10, but AFAICT Jan already did cut the
last release for it.
So I wouldn't consider any backport unless we begin to see the branch failing.
If Jan already made the last release for 4.10, then little point in
backporting it to it. However, it is not ideal to have something like
00c96d7742 in some still-maintained staging branches but not all.
Cheers,
--
Julien Grall
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/xen-devel
|
