On Fri, 3 Nov 2023 14:46:39 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to save
> A bug in GCC causes shared libraries linked with -ffast-math to disable
> denormal arithmetic. This breaks Java's floating-point semantics.
>
> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>
> One solution is to save and restore the floating-point control word around
> System
> A bug in GCC causes shared libraries linked with -ffast-math to disable
> denormal arithmetic. This breaks Java's floating-point semantics.
>
> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>
> One solution is to save and restore the floating-point control word around
> System
On Fri, 27 Oct 2023 11:59:59 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
On Fri, 27 Oct 2023 11:59:59 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
On Fri, 27 Oct 2023 11:59:59 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
On Thu, 26 Oct 2023 15:41:35 GMT, Thomas Stuefe wrote:
> This looks good to me.
>
> One suggestion: to reduce code duplication and to make the code a bit safer
> against accidental returns prior to fesetenv, I would have used a mark object
> like this:
Thanks. I take your point, but I think t
On Fri, 27 Oct 2023 11:49:16 GMT, Andrew Haley wrote:
> > One more thought, it would be good to add the FTZ_mode_enabled check to
> > `os::run_periodic_checks()`.
> > We already do signal handler checks there, and it is the right place to
> > check for "global things third party native code may
> A bug in GCC causes shared libraries linked with -ffast-math to disable
> denormal arithmetic. This breaks Java's floating-point semantics.
>
> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>
> One solution is to save and restore the floating-point control word around
> System
On Thu, 26 Oct 2023 17:42:39 GMT, Vladimir Ivanov wrote:
>> Andrew Haley has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> Remove accidental include
>
> make/test/JtregNativeHotspot.gmk line 854:
>
>> 852: BUILD_HOTSPOT_JTREG_EXECUTABLE
On Thu, 26 Oct 2023 15:59:08 GMT, Thomas Stuefe wrote:
> One more thought, it would be good to add the FTZ_mode_enabled check to
> `os::run_periodic_checks()`.
>
> We already do signal handler checks there, and it is the right place to check
> for "global things third party native code may mes
> A bug in GCC causes shared libraries linked with -ffast-math to disable
> denormal arithmetic. This breaks Java's floating-point semantics.
>
> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>
> One solution is to save and restore the floating-point control word around
> System
On Thu, 26 Oct 2023 17:41:52 GMT, Vladimir Ivanov wrote:
> > do you have any suggestion about where to move the remaining code that's in
> > stubRoutines.cpp?
>
> As it is now, `globalDefinitions.hpp/cpp` looks the most appropriate one to
> me.
That seems okay to me too and much better than `
On Thu, 26 Oct 2023 16:46:04 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
On Thu, 26 Oct 2023 10:39:04 GMT, Andrew Haley wrote:
> do you have any suggestion about where to move the remaining code that's in
> stubRoutines.cpp?
As it is now, `globalDefinitions.hpp/cpp` looks the most appropriate one to me.
You could create new header file for it (under `share/utiliti
> A bug in GCC causes shared libraries linked with -ffast-math to disable
> denormal arithmetic. This breaks Java's floating-point semantics.
>
> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>
> One solution is to save and restore the floating-point control word around
> System
On Thu, 26 Oct 2023 14:17:02 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
On Thu, 26 Oct 2023 14:17:02 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
> A bug in GCC causes shared libraries linked with -ffast-math to disable
> denormal arithmetic. This breaks Java's floating-point semantics.
>
> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>
> One solution is to save and restore the floating-point control word around
> System
> A bug in GCC causes shared libraries linked with -ffast-math to disable
> denormal arithmetic. This breaks Java's floating-point semantics.
>
> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>
> One solution is to save and restore the floating-point control word around
> System
On Wed, 18 Oct 2023 19:13:40 GMT, Vladimir Ivanov wrote:
> > Meta-question and apologies if this was covered before, but why is this
> > logic being added to stubRoutines.cpp?
>
> Because tha'ts where @iwanowww asked me to put it. I don't much care.
Hi @iwanowww , do you have any suggestion ab
> A bug in GCC causes shared libraries linked with -ffast-math to disable
> denormal arithmetic. This breaks Java's floating-point semantics.
>
> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>
> One solution is to save and restore the floating-point control word around
> System
On Thu, 19 Oct 2023 09:31:36 GMT, Andrew Haley wrote:
>> src/hotspot/os/linux/os_linux.cpp line 1853:
>>
>>> 1851:
>>> 1852: #ifndef IA32
>>> 1853: // Quickly test to make sure denormals are correctly handled.
>>
>> Nit: I recommend using "subnormal" rather than "denormal" for general
>>
> A bug in GCC causes shared libraries linked with -ffast-math to disable
> denormal arithmetic. This breaks Java's floating-point semantics.
>
> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>
> One solution is to save and restore the floating-point control word around
> System
On Thu, 19 Oct 2023 09:33:52 GMT, Andrew Haley wrote:
> I took them out because of a potential backwards-compatibility breakage.
Ok, I checked the removed code
(https://github.com/openjdk/jdk/pull/10661/commits/b817d4757c78594be5960ee0be27013e2588d30a)
and agree it is not needed here.
`Resto
On Thu, 19 Oct 2023 01:26:43 GMT, Joe Darcy wrote:
>> Andrew Haley has updated the pull request incrementally with three
>> additional commits since the last revision:
>>
>> - Review feedback
>> - Merge branch 'JDK-8295159' of https://github.com/theRealAph/jdk into
>> JDK-8295159
>> - Remov
On Wed, 18 Oct 2023 08:50:13 GMT, Andrew Haley wrote:
>> Meta-question and apologies if this was covered before, but why is this
>> logic being added to stubRoutines.cpp?
>
>> Meta-question and apologies if this was covered before, but why is this
>> logic being added to stubRoutines.cpp?
>
>
On Tue, 17 Oct 2023 11:43:59 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
On Wed, 18 Oct 2023 08:50:13 GMT, Andrew Haley wrote:
>> Meta-question and apologies if this was covered before, but why is this
>> logic being added to stubRoutines.cpp?
> Because tha'ts where @iwanowww asked me to put it. I don't much care.
The constants were duplicated in multiple places (i
On Wed, 18 Oct 2023 00:01:20 GMT, David Holmes wrote:
> Meta-question and apologies if this was covered before, but why is this logic
> being added to stubRoutines.cpp?
Because tha'ts where @iwanowww asked me to put it. I don't much care.
-
PR Comment: https://git.openjdk.org/jdk/
On Tue, 17 Oct 2023 11:43:59 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
On Tue, 17 Oct 2023 10:56:33 GMT, Andrew Haley wrote:
>> src/hotspot/cpu/x86/macroAssembler_x86.cpp line 5169:
>>
>>> 5167: // Perform a little arithmetic to make sure that denormal
>>> 5168: // numbers are handled correctly, i.e. that the "Denormals Are
>>> 5169: // Zeros" fla
> A bug in GCC causes shared libraries linked with -ffast-math to disable
> denormal arithmetic. This breaks Java's floating-point semantics.
>
> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>
> One solution is to save and restore the floating-point control word around
> System
On Tue, 17 Oct 2023 02:15:45 GMT, David Holmes wrote:
>> Andrew Haley has updated the pull request incrementally with two additional
>> commits since the last revision:
>>
>> - Comments only.
>> - Review feedback
>
> src/hotspot/cpu/x86/macroAssembler_x86.cpp line 5169:
>
>> 5167: // P
On Mon, 16 Oct 2023 16:22:27 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
> A bug in GCC causes shared libraries linked with -ffast-math to disable
> denormal arithmetic. This breaks Java's floating-point semantics.
>
> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>
> One solution is to save and restore the floating-point control word around
> System
On Wed, 11 Oct 2023 18:08:55 GMT, Vladimir Ivanov wrote:
>> Andrew Haley has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> Add TestDenormalDouble.java
>
> test/hotspot/jtreg/compiler/floatingpoint/libfast-math.c line 26:
>
>> 24: #include
On Wed, 11 Oct 2023 19:51:34 GMT, Vladimir Ivanov wrote:
> And I was confused at first why there's a volatile on `tresh`. A short
> comment describing the intentions would definitely help here.
OK.
-
PR Review Comment: https://git.openjdk.org/jdk/pull/10661#discussion_r1356952291
On Wed, 11 Oct 2023 17:57:27 GMT, Vladimir Ivanov wrote:
>> Andrew Haley has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> Add TestDenormalDouble.java
>
> src/hotspot/os/bsd/os_bsd.cpp line 977:
>
>> 975:
>> 976: void *os::Bsd::dlopen_he
On Thu, 12 Oct 2023 09:23:08 GMT, Florian Weimer wrote:
>> Andrew Haley has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> Add TestDenormalDouble.java
>
> test/hotspot/jtreg/compiler/floatingpoint/libfast-math.c line 37:
>
>> 35:
>> 36: #
On Wed, 11 Oct 2023 18:41:08 GMT, Thomas Stuefe wrote:
>> What difference does it make if it does?
>
> I was trying to understand the BSD+gcc combination. We use clang on MacOS, so
> the only platform I thought would be affected would be one of the BSDs.
>
> But I thought those also moved to c
On Wed, 11 Oct 2023 17:29:11 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
On Wed, 11 Oct 2023 17:29:11 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
On Wed, 11 Oct 2023 17:29:11 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
On Wed, 11 Oct 2023 18:43:31 GMT, Thomas Stuefe wrote:
>> `thresh` is volatile. What more is needed?
>
> Argh, sorry, missed that volatile.
And I was confused at first why there's a volatile on `tresh`. A short comment
describing the intentions would definitely help here.
-
PR Rev
On Wed, 11 Oct 2023 17:20:18 GMT, Andrew Haley wrote:
>> src/hotspot/os/bsd/os_bsd.cpp line 976:
>>
>>> 974: // same architecture as Hotspot is running on
>>> 975:
>>> 976: void *os::Bsd::dlopen_helper(const char *filename, int mode) {
>>
>> I thought BSD is switching to clang.
>
> What differ
> A bug in GCC causes shared libraries linked with -ffast-math to disable
> denormal arithmetic. This breaks Java's floating-point semantics.
>
> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>
> One solution is to save and restore the floating-point control word around
> System
On Wed, 11 Oct 2023 13:54:05 GMT, Thomas Stuefe wrote:
> > > I'm seeing one automated test failure on Linux x86, which I don't
> > > understand because I've excluded that test for generic-i586. If anyone
> > > understands this, please shout up.
> >
> >
> > For avoidance of doubt, the test doe
On Wed, 11 Oct 2023 14:03:24 GMT, Thomas Stuefe wrote:
>> Andrew Haley has updated the pull request with a new target base due to a
>> merge or a rebase. The pull request now contains 28 commits:
>>
>> - Merge branch 'JDK-8295159' of https://github.com/theRealAph/jdk into
>> JDK-8295159
>> -
On Wed, 11 Oct 2023 13:36:01 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
On Wed, 11 Oct 2023 13:38:57 GMT, Andrew Haley wrote:
> > I'm seeing one automated test failure on Linux x86, which I don't
> > understand because I've excluded that test for generic-i586. If anyone
> > understands this, please shout up.
>
> For avoidance of doubt, the test doesn't run locally
On Wed, 27 Sep 2023 07:49:48 GMT, Magnus Ihse Bursie wrote:
>> Andrew Haley has updated the pull request with a new target base due to a
>> merge or a rebase. The pull request now contains 28 commits:
>>
>> - Merge branch 'JDK-8295159' of https://github.com/theRealAph/jdk into
>> JDK-8295159
On Wed, 11 Oct 2023 13:31:22 GMT, Andrew Haley wrote:
> I'm seeing one automated test failure on Linux x86, which I don't understand
> because I've excluded that test for generic-i586. If anyone understands this,
> please shout up.
For avoidance of doubt, the test doesn't run locally, only on
On Wed, 11 Oct 2023 13:31:22 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
On Wed, 12 Oct 2022 17:00:15 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
> A bug in GCC causes shared libraries linked with -ffast-math to disable
> denormal arithmetic. This breaks Java's floating-point semantics.
>
> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>
> One solution is to save and restore the floating-point control word around
> System
On Wed, 12 Oct 2022 15:59:48 GMT, Aleksey Shipilev wrote:
>> Andrew Haley has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> 8295159: DSO created with -ffast-math breaks Java floating-point arithmetic
>
> test/hotspot/jtreg/compiler/floatin
On Wed, 12 Oct 2022 17:00:15 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
On Wed, 12 Oct 2022 17:00:15 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
On Wed, 12 Oct 2022 17:00:15 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
On Wed, 12 Oct 2022 17:00:15 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
On Mon, 31 Oct 2022 18:25:57 GMT, Jorn Vernee wrote:
> > Anyway, I plan to
> > a. Restore the FPU CR after calls to dlopen(3).
> > b. Detect FPU CR corruption at safepoints, and print a warning. At least
> > the user might find out that something is wrong.
>
> Doing (a) seems good. I can't say f
On Mon, 31 Oct 2022 17:44:07 GMT, Andrew Haley wrote:
> Anyway, I plan to
>
> a. Restore the FPU CR after calls to dlopen(3).
> b. Detect FPU CR corruption at safepoints, and print a warning. At least
> the user might find out that something is wrong.
Doing (a) seems good. I can't say for sure
On Wed, 12 Oct 2022 17:00:15 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
On 10/31/22 15:49, Joe Darcy wrote:
In terms of the overhead of using floating-point expression evaluation
as a guard, are there still platforms where operating on subnormal
values is pathologically slower? Some generations of SPARC chips had
that behavior where a subnormal multiply would take, s
On 10/25/2022 7:27 AM, Andrew Haley wrote:
On Thu, 20 Oct 2022 20:26:47 GMT, Vladimir Ivanov wrote:
The GCC bugs with `-ffast-math` only corrupts `FTZ` and `DAZ`.
But `RC` and exception masks may be corrupted as well the same way and I
believe the consequences are be similar (silent divergen
On Wed, 12 Oct 2022 17:00:15 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
On Wed, 12 Oct 2022 17:00:15 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
On Wed, 12 Oct 2022 17:00:15 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
On Wed, 26 Oct 2022 17:10:00 GMT, David Holmes wrote:
> At most an expanded -Xcheck:jni check for FP-state manipulation, with an
enhanced/fixed RestoreMXCSROnJNICalls is in order IMVHO.
FTR both `-Xcheck:jni` and `-XX:+RestoreMXCSROnJNICalls` already catch the
problematic case being discussed.
On Wed, 12 Oct 2022 17:00:15 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
On Tue, 25 Oct 2022 23:13:08 GMT, Vladimir Ivanov wrote:
> …very modest impact while still being able to catch important types of MXCSR
> corruption. I fully support having it turned on by default for JNI calls.
I guess I agree. With the clever test for the bad mode Java cares about, the
over
On 26/10/2022 1:02 am, Andrew Haley wrote:
On Tue, 25 Oct 2022 14:46:57 GMT, Florian Weimer wrote:
Sorry, I feel like this has gone a bit off track. It started as some hardening
for `loadLibrary`, but now it's about making all JNI calls a bit slower? Is
there any data to suggest that this is
On Wed, 12 Oct 2022 17:00:15 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
On Tue, 18 Oct 2022 07:46:35 GMT, Florian Weimer wrote:
> I wonder if something that focuses on diagnostic tools might be better here,
> particularly if there hasn't been any reported breakage. The `dlopen`
> protection is of course very incomplete because any JNI call can change the
> state i
On Wed, 12 Oct 2022 17:00:15 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
On Tue, 25 Oct 2022 14:46:57 GMT, Florian Weimer wrote:
> Sorry, I feel like this has gone a bit off track. It started as some
> hardening for `loadLibrary`, but now it's about making all JNI calls a bit
> slower? Is there any data to suggest that this is necessary?
>
> Would it be possible to
On Wed, 12 Oct 2022 17:00:15 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
On Tue, 25 Oct 2022 13:29:52 GMT, Andrew Haley wrote:
> Enabling `-XX:+RestoreMXCSROnJNICalls` makes the overhead much worse:
>
> ```
> Benchmark Mode Cnt Score Error Units
> CallOverheadConstant.jni_blank avgt 40 14.741 ± 0.031 ns/op
> CallOverheadConstant
On Thu, 20 Oct 2022 20:26:47 GMT, Vladimir Ivanov wrote:
> The GCC bugs with `-ffast-math` only corrupts `FTZ` and `DAZ`.
>
> But `RC` and exception masks may be corrupted as well the same way and I
> believe the consequences are be similar (silent divergence in results during
> FP computation
On Wed, 12 Oct 2022 17:00:15 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
On Thu, 20 Oct 2022 20:26:47 GMT, Vladimir Ivanov wrote:
> That sounds like a very interesting idea.
>
> It would be very helpful to get an understanding how much overhead `STMXCSR`
> plus a branch adds in JNI stub to decide whether it's worth optimizing for.
It's not just Intel's implementati
On Wed, 12 Oct 2022 17:00:15 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
PS And additional expressions could be crashed to rule out
flush-to-zero, treating all subnormals and zero, and other non-compliant
modes of operation.
-Joe
On 10/20/2022 1:05 PM, Joseph D. Darcy wrote:
On 10/20/2022 10:43 AM, Andrew Haley wrote:
On Wed, 12 Oct 2022 17:00:15 GMT, Andrew Hal
On 10/20/2022 10:43 AM, Andrew Haley wrote:
On Wed, 12 Oct 2022 17:00:15 GMT, Andrew Haley wrote:
A bug in GCC causes shared libraries linked with -ffast-math to disable
denormal arithmetic. This breaks Java's floating-point semantics.
The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?i
On Wed, 12 Oct 2022 17:00:15 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
On Wed, 12 Oct 2022 17:00:15 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
On Wed, 12 Oct 2022 17:00:15 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
On Wed, 12 Oct 2022 17:00:15 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
On Tue, 18 Oct 2022 07:43:13 GMT, Florian Weimer wrote:
>> Andrew Haley has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> 8295159: DSO created with -ffast-math breaks Java floating-point arithmetic
>
> src/hotspot/os/linux/os_linux.cpp lin
On Wed, 12 Oct 2022 17:00:15 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
On Wed, 12 Oct 2022 17:00:15 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
On Wed, 12 Oct 2022 17:00:15 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
On Wed, 12 Oct 2022 17:00:15 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
On 10/13/22 19:13, Vladimir Ivanov wrote:
On Thu, 13 Oct 2022 07:47:20 GMT, Andrew Haley wrote:
The problem is that this bug is very insidious: the user probably doesn't know
that there's anything wrong, and almost certainly has no idea that it's
anything to do with JNI.
I'm still trying t
On Thu, 13 Oct 2022 07:47:20 GMT, Andrew Haley wrote:
> The problem is that this bug is very insidious: the user probably doesn't
> know that there's anything wrong, and almost certainly has no idea that it's
> anything to do with JNI.
I'm still trying to grasp why the current problem is somet
On Wed, 12 Oct 2022 17:00:15 GMT, Andrew Haley wrote:
>> A bug in GCC causes shared libraries linked with -ffast-math to disable
>> denormal arithmetic. This breaks Java's floating-point semantics.
>>
>> The bug is https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55522
>>
>> One solution is to sav
On Wed, 12 Oct 2022 20:26:34 GMT, Jorn Vernee wrote:
> Following that logic: from our perspective `dlopen` violates its ABI in
> certain cases. Preserving the control bits across calls to `dlopen` seems
> like a pragmatic solution. I'm not sure how important it is to have an opt-in
> for the c
On Wed, 12 Oct 2022 20:35:09 GMT, Maurizio Cimadamore
wrote:
> Isn't it an illustration of a more general problem we have with native code
> where it can mess with FP environment at any time?
Yes.
> We already have similar problems with MXCSR register and provide verification
> logic (part o
On Wed, 12 Oct 2022 20:35:09 GMT, Maurizio Cimadamore
wrote:
> Note also that this specific problem is triggered by dlopen itself, because
> certain libraries might have some "bad" (from the perspective of JVM)
> initialization code.
... and it leads to another question whether the JVM itself
1 - 100 of 125 matches
Mail list logo
