Hi Jan,
On 12/04/2021 09:53, Jan Beulich wrote:
On 11.04.2021 04:12, osstest service owner wrote:
branch xen-unstable
xenbranch xen-unstable
job test-armhf-armhf-xl-multivcpu
testid guest-start/debian.repeat
Tree: linux git://xenbits.xen.org/linux-pvops.git
Tree: linuxfirmware git://xenbits.xen.org/osstest/linux-firmware.git
Tree: qemuu git://xenbits.xen.org/qemu-xen.git
Tree: xen git://xenbits.xen.org/xen.git
*** Found and reproduced problem changeset ***
Bug is in tree: xen git://xenbits.xen.org/xen.git
Bug introduced: 9617d5f9c19d1d157629e1e436791509526e0ce5
Bug not present: 5c3c410bd2ea8d2cc520e8e8f83ad143c9c5cff7
Last fail repro: http://logs.test-lab.xenproject.org/osstest/logs/160931/
commit 9617d5f9c19d1d157629e1e436791509526e0ce5
Author: Julien Grall <jgr...@amazon.com>
Date: Sat Feb 20 17:54:13 2021 +0000
xen/arm: mm: flush_page_to_ram() only need to clean to PoC
At the moment, flush_page_to_ram() is both cleaning and invalidate to
PoC the page.
The goal of flush_page_to_ram() is to prevent corruption when the guest
has disabled the cache (the cache line may be dirty) and the guest to
read previous content.
Per this definition, the invalidating the line is not necessary. So
invalidating the cache is unnecessary. In fact, it may be counter-
productive as the line may be (speculatively) accessed a bit after.
So this will incurr an expensive access to the memory.
More generally, we should avoid interferring too much with cache.
Therefore, flush_page_to_ram() is updated to only clean to PoC the page.
The performance impact of this change will depend on your
workload/processor.
Signed-off-by: Julien Grall <jgr...@amazon.com>
Reviewed-by: Bertrand Marquis <bertrand.marq...@arm.com>
Acked-by: Stefano Stabellini <sstabell...@kernel.org>
Is it possible that other code (guest one in particular considering the
failure pattern, but possibly not limited to that) has developed a
dependency on the prior behavior?
This is not a problem in the guest. I overlooked that
flush_page_to_ram() is also used when emulating the instruction to
invalidate by set/way the data cache.
We would need to tell flush_page_to_ram() which type of operation we
want to do. I will not have time to work on it right now, so I have
reverted the patch to unblock OssTest.
Sorry for the breakage.
Cheers,
--
Julien Grall
