Hi Katya,
Sorry for the late response. :-(
On 2018-10-29 21:24, Ekaterina Pavlova wrote:
Vladimir,
I added "-XX:+UseAOTStrictLoading" to "java -version", see
make/RunTests.gmk.
Also added "--compile-with-assertion" to jaotc flags in case "-ea"
flag was present in vm options
(otherwise AOTed library will warn that it does not have java
assertions in compiled code).
This code:
+ ifneq ($$(findstring -ea, $$($1_VM_OPTIONS)), )
Using findstring is somewhat brittle here, since it will find substrings
as well. Better to use filter which will match an exact word.
I will add -XX:+UseAOTStrictLoading and
"-Xlog:aot+class+fingerprint=trace -Xlog:aot+class+load=trace"
directly to AOT testing tasks in mach5 jobs.
Magnus,
I changed
"$$(addprefix -vmoption:, $$($1_AOT_OPTIONS))"
to
+ ifneq ($$($1_AOT_OPTIONS), )
+ $1_JTREG_BASIC_OPTIONS += -vmoptions:"$$($1_AOT_OPTIONS)"
+ endif
Please review updated webrev:
http://cr.openjdk.java.net/~epavlova//8152988/webrev.01/
Some fallout from my and Erik's discussion still needs implementing in
RunTests.gmk:
$$(addprefix --compile-commands , $$($1_AOT_CCLIST)) \
Please rewrite as:
$$(addprefix --compile-commands$(SPACE), $$($1_AOT_CCLIST)) \
I'm still not happy with the always printed "java -version". :( Vladimir
agreed that it would be reasonable to redirect this into a file. This
can be done by e.g.
+ $$(call ExecuteWithLog, $$@.check, \
+ $$(FIXPATH) $$(JDK_IMAGE_DIR)/bin/java \
+ $$($1_VM_OPTIONS) -XX:+PrintAOT -XX:+UseAOTStrictLoading
-XX:AOTLibrary=$$@ -version \
+ > $$@.verify-aot
+ )
This will redirect the output to $($1_AOT_LIB).verify-aot in
test-support/aot.
I still would like to see the if statement around SetupAot moved out,
like this:
+ ifneq ($$(GTEST_AOT_MODULES), ) + $$(eval $$(call SetupAot, $1, \
+ MODULES := $$(GTEST_AOT_MODULES), \
+ VM_OPTIONS := $$(GTEST_VM_OPTIONS) $$(GTEST_JAVA_OPTIONS), \
+ )) + endif
and similarly for JTREG. This is coupled with removing the "ifneq
($$($1_MODULES), )" in SetupAotBody (and do not forget to un-indent two
spaces).
/Magnus
thanks,
-katya
On 10/23/18 4:40 PM, Vladimir Kozlov wrote:
On 10/22/18 9:36 AM, Ekaterina Pavlova wrote:
Vladimir,
thanks a lot for the review.
"-XX:+PrintAOT" was only passed to "java -version" which is done
before the testing to verify that AOTed libraries work.
If it is only for -version then using -XX:+PrintAOT is okay.
It also perhaps could be useful debug info in case some tests starts
fail.
But I can change -XX:+PrintAOT to -XX:+UseAOTStrictLoading if you
think it is more than enough.
-XX:+UseAOTStrictLoading is more important for tests which have
different flags specified and may invalidate AOTLibrary
configuration. We would want to catch tests which will not use AOT
libraries. At least when we test these your changes.
Do you suggest to add "-Xlog:aot+class+fingerprint=trace
-Xlog:aot+class+load=trace" to "java -version" only
or to java flags used during tests execution as well?
These options are next step after -XX:+UseAOTStrictLoading. We were
asked before how we can guarantee that AOTed methods are used by
tests. One way is PrintAOT which should show published AOT methods.
An other way is Xlogs flags which have less output but provide at
least some level of information that AOTed classes are used and not
skipped.
Thanks,
Vladimir
-katya
On 10/19/18 10:40 AM, Vladimir Kozlov wrote:
I share concern about output of -XX:+PrintAOT - it prints all AOT
classes and methods. For AOTed java.base the output will be almost
the same for all tests (depends which are java.base classes are used).
I would suggest instead to use -XX:+UseAOTStrictLoading which
exit(1) VM if it can't load AOT library for some reason (mismatched
configuration - different GCs).
Also -Xlog:aot+class+fingerprint=trace to trace cases of
mismatching class fingerprint.
Also -Xlog:aot+class+load=trace -to trace AOT class loading or
skipping them if they are not matching.
And I agree to redirect this into file.
Thanks,
Vladimir
On 10/19/18 9:04 AM, Erik Joelsson wrote:
Hello,
I will answer the parts I'm responsible for.
On 2018-10-19 00:21, Magnus Ihse Bursie wrote:
In RunTests.gmk, the following line is repeated:
# Note, this could not be done during JDK build time.
In the options:
$1_JAOTC_OPTS := \
-J-Xmx4g --info \
-Xmx4g is huge, much larger than what we typically use. Is this
intentional? This will risk crashing at machines with too little
free memory. Will AOT fail to work with less memory?
There's an extra space before the comma here:
$$(addprefix --compile-commands , $$($1_AOT_CCLIST)) \
Question (to Erik, I presume): the idea of addprefix here is that
AOT_CCLIST can be empty, and this was a convenient way to just
add "--compile-commands <the list file>" if it exists? I was a
bit confounded at first since normally we only use it for
distributing a prefix over multiple items, and I could see no way
for AOT_CCLIST to ever be more than one item.
Yes, that was the intention. I often use addprefix that way and
find it convenient that it seamlessly handles 0-N number of
elements in the argument without the need for cumbersome
conditionals. The extra space is needed in this case but should
perhaps be explicit using $(SPACE).
About this:
# Note, ExecuteWithLog doesn't play well with with two calls in
the same recipe.
$$(info $$(JDK_IMAGE_DIR)/bin/java $$($1_JVM_OPTIONS)
-XX:+PrintAOT -XX:AOTLibrary=$$@ -version)
$$(FIXPATH) $$(JDK_IMAGE_DIR)/bin/java \
$$($1_JVM_OPTIONS) \
-XX:+PrintAOT -XX:AOTLibrary=$$@ -version
First, what is the purpose of this? Is it to verify that the
generated AOT library works? This will result in a lot of "java
-version" being printed at the time when running tests. Perhaps
the output should be redirected? (Assuming that java exists with
a non-zero exit value if things fail, but if it does not, I'm not
sure how this would help otherwise.)
My guess is that they want diagnostics saved in case tests fail.
Second, there's no problem to use multiple ExecuteWithLog in the
same recipe, however, the base variable needs to be different. E.g:
$$(call ExecuteWithLog, $$@_verify, \
$$(FIXPATH) $$(JDK_IMAGE_DIR)/bin/java \
...
The issue I'm referring to is that the $(info) line in
ExecuteWithLog will print the commands first, then execute the
first line of the recipe. This makes the output confusing since
both of those commands print (what I'm assuming is) relevant
diagnostics information. This is simply a consequence of using
$(info) for printing the command line. Make will evaluate all the
recipe lines first, then execute them one by one. To be clear, it
ends up like this:
Executing [jaotc ...]
Executing [java ...]
<output from jaotc>
<output from java>
Further,
SetupAot = $(NamedParamsMacroTemplate)
define SetupAotBody
ifneq ($$($1_MODULES), )
...
If think it would be clearer if the if-test is on the call site
for SetupAot instead. Now it's unconditionally called, but does
nothing if AOT is not used. I think that looks odd.
My goal was to minimize the repeated code at each call site. While
developing this, I had a lot more duplication at first, and it was
a hassle to update each location every time I needed to change
something. Other than that I don't feel strongly for either
construct.
And here,
- run-test-$1: clean-workdir-$1 $(TEST_PREREQS)
+ run-test-$1: clean-workdir-$1 $$($1_AOT_TARGETS)
I don't think you can just remove TEST_PREREQS like that..? (The
same goes for the other run-test-$1 instance.)
I added the TEST_PREREQS variable to support the CDS archive
generation, which was recently removed. With the way SetupAot
turned out to need separate calls from each SetupTestRun, the
global TEST_PREREQS wasn't suitable. I don't know of any current
need for a general TEST_PREREQS.
Finally:
+ $$(addprefix -vmoption:, $$($1_AOT_OPTIONS)) \
I'm not really comfortable with this use of addprefix. I think a
construct like:
ifneq ($$($1_AOT_OPTIONS), )
$1_JTREG_BASIC_OPTIONS += -vmoption:$$($1_AOT_OPTIONS)
endif
would be much clearar. Ideally, I'd like to see the same
construct in the SetupAotModule function as well, but that's not
as important.
I agree.
/Erik
The rest of the build changes look good. (I can't say anything
about the test/**/*-list.txt files.)
/Magnus
testing: Tested by running subset of hotspot, jdk and jck tests
with TEST_OPTS_AOT_MODULES=java.base jdk.internal.vm.compiler
with "make run-test" and in Mach5.
thanks,
-katya
p.s.
Thanks a lot Erik for huge help with porting original patch
done in old testing framework (test/TestCommon.gmk)
to new one (make/RunTests*).
