On Mon, 4 Sep 2023 21:17:06 GMT, Martin Doerr wrote:
>> I've found a way to solve the remaining FFI problem on linux PPC64 Big
>> Endian. Large structs (>8 Bytes) which are passed in registers or on stack
>> require shifting the Bytes in the last slot if the size is not a multiple of
>> 8. Thi
On Mon, 4 Sep 2023 21:17:06 GMT, Martin Doerr wrote:
>> I've found a way to solve the remaining FFI problem on linux PPC64 Big
>> Endian. Large structs (>8 Bytes) which are passed in registers or on stack
>> require shifting the Bytes in the last slot if the size is not a multiple of
>> 8. Thi
On Mon, 4 Sep 2023 14:30:06 GMT, Maurizio Cimadamore
wrote:
>> @TheRealMDoerr We've been discussing the shifts in order to wrap our heads
>> around it, and we've ended up with this diagram in order to try and
>> visualize what happens:
>>
>> Let's say we have a struct with 3 ints:
>>
>>
>>
On Mon, 4 Sep 2023 14:30:06 GMT, Maurizio Cimadamore
wrote:
>> @TheRealMDoerr We've been discussing the shifts in order to wrap our heads
>> around it, and we've ended up with this diagram in order to try and
>> visualize what happens:
>>
>> Let's say we have a struct with 3 ints:
>>
>>
>>
On Mon, 4 Sep 2023 21:17:06 GMT, Martin Doerr wrote:
>> I've found a way to solve the remaining FFI problem on linux PPC64 Big
>> Endian. Large structs (>8 Bytes) which are passed in registers or on stack
>> require shifting the Bytes in the last slot if the size is not a multiple of
>> 8. Thi
> I've found a way to solve the remaining FFI problem on linux PPC64 Big
> Endian. Large structs (>8 Bytes) which are passed in registers or on stack
> require shifting the Bytes in the last slot if the size is not a multiple of
> 8. This PR adds the required functionality to the Java code.
>
>
On Mon, 4 Sep 2023 15:38:19 GMT, Martin Doerr wrote:
>> I've found a way to solve the remaining FFI problem on linux PPC64 Big
>> Endian. Large structs (>8 Bytes) which are passed in registers or on stack
>> require shifting the Bytes in the last slot if the size is not a multiple of
>> 8. Thi
On Mon, 4 Sep 2023 14:53:33 GMT, Jorn Vernee wrote:
>> Martin Doerr has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> Separate type cast from Shift bindings.
>
> src/java.base/share/classes/jdk/internal/foreign/abi/Binding.java line 425:
>
> I've found a way to solve the remaining FFI problem on linux PPC64 Big
> Endian. Large structs (>8 Bytes) which are passed in registers or on stack
> require shifting the Bytes in the last slot if the size is not a multiple of
> 8. This PR adds the required functionality to the Java code.
>
>
On Mon, 4 Sep 2023 14:56:18 GMT, Martin Doerr wrote:
>> I've found a way to solve the remaining FFI problem on linux PPC64 Big
>> Endian. Large structs (>8 Bytes) which are passed in registers or on stack
>> require shifting the Bytes in the last slot if the size is not a multiple of
>> 8. Thi
On Mon, 4 Sep 2023 14:56:18 GMT, Martin Doerr wrote:
>> I've found a way to solve the remaining FFI problem on linux PPC64 Big
>> Endian. Large structs (>8 Bytes) which are passed in registers or on stack
>> require shifting the Bytes in the last slot if the size is not a multiple of
>> 8. Thi
On Mon, 4 Sep 2023 14:54:05 GMT, Martin Doerr wrote:
> Do you see a good place for such a comment?
PPC CallArranger seems like a good place to me. We have a similar explanation
comment in the AArch64 CallArranger.
> Maybe it would be better to use a different size for the last chunk. Maybe
>
On Mon, 4 Sep 2023 14:56:18 GMT, Martin Doerr wrote:
>> I've found a way to solve the remaining FFI problem on linux PPC64 Big
>> Endian. Large structs (>8 Bytes) which are passed in registers or on stack
>> require shifting the Bytes in the last slot if the size is not a multiple of
>> 8. Thi
On Mon, 4 Sep 2023 14:50:37 GMT, Martin Doerr wrote:
>>> as it would no longer need an input type?
>>
>> Yes. Then both shift ops would always operate on `long`.
>
> I've changed it. Note that I need many more conversions because buffer
> load/store also use subtypes of `int`. Please take a loo
On Mon, 4 Sep 2023 12:22:00 GMT, Martin Doerr wrote:
>> I've found a way to solve the remaining FFI problem on linux PPC64 Big
>> Endian. Large structs (>8 Bytes) which are passed in registers or on stack
>> require shifting the Bytes in the last slot if the size is not a multiple of
>> 8. Thi
> I've found a way to solve the remaining FFI problem on linux PPC64 Big
> Endian. Large structs (>8 Bytes) which are passed in registers or on stack
> require shifting the Bytes in the last slot if the size is not a multiple of
> 8. This PR adds the required functionality to the Java code.
>
>
On Mon, 4 Sep 2023 12:44:51 GMT, Jorn Vernee wrote:
>> Martin Doerr has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> Split Shift into ShiftLeft and ShiftRight + minor improvements.
>
> src/java.base/share/classes/jdk/internal/foreign/abi/
On Mon, 4 Sep 2023 14:33:08 GMT, Jorn Vernee wrote:
>>> I think it's worth it in order to have a cleaner contract for the shift
>>> ops, should we want to use them for anything else in the future, but also
>>> just to make them easier to understand for future readers.
>>
>> I agree that having
On Mon, 4 Sep 2023 14:27:27 GMT, Maurizio Cimadamore
wrote:
> as it would no longer need an input type?
Yes. Then both shift ops would always operate on `long`.
-
PR Review Comment: https://git.openjdk.org/jdk/pull/15417#discussion_r1315016599
On Mon, 4 Sep 2023 14:18:44 GMT, Jorn Vernee wrote:
> If this struct is passed as an argument, then the load of the second 'half'
> of the struct would look like this:
It would perhaps be cleaner if in the MSB/LSB comments we said:
LSBs are zzz...z
LSBs are 000...0
(e.g. avoid to refer to MSB
On Mon, 4 Sep 2023 12:20:52 GMT, Martin Doerr wrote:
>> Sorry for the delay, I've been on vacation.
>
>> Sorry for the delay, I've been on vacation.
>
> No problem. Hope you had a good time! Thanks for your feedback.
@TheRealMDoerr We've been discussing the shifts in order to wrap our heads
ar
On Mon, 4 Sep 2023 12:22:00 GMT, Martin Doerr wrote:
>> I've found a way to solve the remaining FFI problem on linux PPC64 Big
>> Endian. Large structs (>8 Bytes) which are passed in registers or on stack
>> require shifting the Bytes in the last slot if the size is not a multiple of
>> 8. Thi
On Mon, 4 Sep 2023 12:19:42 GMT, Martin Doerr wrote:
>> src/java.base/share/classes/jdk/internal/foreign/abi/Binding.java line 398:
>>
>>> 396: bindings.add(Binding.cast(type, int.class));
>>> 397: type = int.class;
>>> 398: }
>>
>> Part of the casts
On Mon, 4 Sep 2023 07:26:36 GMT, Jorn Vernee wrote:
> Sorry for the delay, I've been on vacation.
No problem. Hope you had a good time! Thanks for your feedback.
> src/java.base/share/classes/jdk/internal/foreign/abi/Binding.java line 398:
>
>> 396: bindings.add(Binding.cast(ty
On Mon, 4 Sep 2023 07:06:39 GMT, Jorn Vernee wrote:
>> Martin Doerr has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> Remove unnecessary imports.
>
> src/java.base/share/classes/jdk/internal/foreign/abi/Binding.java line 695:
>
>> 693:
> I've found a way to solve the remaining FFI problem on linux PPC64 Big
> Endian. Large structs (>8 Bytes) which are passed in registers or on stack
> require shifting the Bytes in the last slot if the size is not a multiple of
> 8. This PR adds the required functionality to the Java code.
>
>
On Fri, 25 Aug 2023 09:24:15 GMT, Maurizio Cimadamore
wrote:
>> The ABI says: "An aggregate or union smaller than one doubleword in size is
>> padded so that it appears in the least significant bits of the doubleword.
>> All others are padded, if necessary, at their tail."
>> [https://refspec
On Fri, 25 Aug 2023 10:59:45 GMT, Martin Doerr wrote:
>> I've found a way to solve the remaining FFI problem on linux PPC64 Big
>> Endian. Large structs (>8 Bytes) which are passed in registers or on stack
>> require shifting the Bytes in the last slot if the size is not a multiple of
>> 8. Th
> I've found a way to solve the remaining FFI problem on linux PPC64 Big
> Endian. Large structs (>8 Bytes) which are passed in registers or on stack
> require shifting the Bytes in the last slot if the size is not a multiple of
> 8. This PR adds the required functionality to the Java code.
>
>
On Fri, 25 Aug 2023 09:01:43 GMT, Maurizio Cimadamore
wrote:
>> I had something like that, but another reviewer didn't like it, either.
>> Originally, I had thought that the v1 and v2 CallArrangers would get more
>> content, but they're still empty. Would it be better to remove these special
> I've found a way to solve the remaining FFI problem on linux PPC64 Big
> Endian. Large structs (>8 Bytes) which are passed in registers or on stack
> require shifting the Bytes in the last slot if the size is not a multiple of
> 8. This PR adds the required functionality to the Java code.
>
>
On Fri, 25 Aug 2023 07:54:51 GMT, Martin Doerr wrote:
>> src/java.base/share/classes/jdk/internal/foreign/abi/Binding.java line 717:
>>
>>> 715: public void interpret(Deque stack, StoreFunc storeFunc,
>>> 716: LoadFunc loadFunc, SegmentAllocator
>>> allocat
On Fri, 25 Aug 2023 07:48:19 GMT, Martin Doerr wrote:
>> If my assumption above is correct, then maybe another way to solve the
>> problem, would be to, instead of adding a new shift binding, to generalize
>> the VM store binding we have to allow writing a smaller value into a bigger
>> storag
On Fri, 25 Aug 2023 07:36:47 GMT, Martin Doerr wrote:
>> src/java.base/share/classes/jdk/internal/foreign/abi/ppc64/ABIv1CallArranger.java
>> line 33:
>>
>>> 31: * PPC64 CallArranger specialized for ABI v1.
>>> 32: */
>>> 33: public class ABIv1CallArranger extends CallArranger {
>>
>> Wouldn
On Thu, 24 Aug 2023 23:38:42 GMT, Maurizio Cimadamore
wrote:
>> I've found a way to solve the remaining FFI problem on linux PPC64 Big
>> Endian. Large structs (>8 Bytes) which are passed in registers or on stack
>> require shifting the Bytes in the last slot if the size is not a multiple of
On Fri, 25 Aug 2023 00:10:11 GMT, Maurizio Cimadamore
wrote:
>> I've found a way to solve the remaining FFI problem on linux PPC64 Big
>> Endian. Large structs (>8 Bytes) which are passed in registers or on stack
>> require shifting the Bytes in the last slot if the size is not a multiple of
On Thu, 24 Aug 2023 23:58:35 GMT, Maurizio Cimadamore
wrote:
>> Maybe I'm starting to see it - it's not a special rule, as much as it is a
>> consequence of the endianness. E.g. if you have a struct that is 64 + 32
>> bytes, you can store the first 64 bytes as a long. Then, there's an issue as
On Thu, 24 Aug 2023 23:33:23 GMT, Maurizio Cimadamore
wrote:
>> I've found a way to solve the remaining FFI problem on linux PPC64 Big
>> Endian. Large structs (>8 Bytes) which are passed in registers or on stack
>> require shifting the Bytes in the last slot if the size is not a multiple of
On Thu, 24 Aug 2023 13:56:12 GMT, Martin Doerr wrote:
> I've found a way to solve the remaining FFI problem on linux PPC64 Big
> Endian. Large structs (>8 Bytes) which are passed in registers or on stack
> require shifting the Bytes in the last slot if the size is not a multiple of
> 8. This P
On Thu, 24 Aug 2023 23:55:28 GMT, Maurizio Cimadamore
wrote:
>> src/java.base/share/classes/jdk/internal/foreign/abi/Binding.java line 695:
>>
>>> 693: * Negative [shiftAmount] shifts right and converts to int if
>>> needed.
>>> 694: */
>>> 695: record ShiftLeft(int shiftAmount
On Thu, 24 Aug 2023 23:36:22 GMT, Maurizio Cimadamore
wrote:
>> I've found a way to solve the remaining FFI problem on linux PPC64 Big
>> Endian. Large structs (>8 Bytes) which are passed in registers or on stack
>> require shifting the Bytes in the last slot if the size is not a multiple of
On Thu, 24 Aug 2023 13:56:12 GMT, Martin Doerr wrote:
> I've found a way to solve the remaining FFI problem on linux PPC64 Big
> Endian. Large structs (>8 Bytes) which are passed in registers or on stack
> require shifting the Bytes in the last slot if the size is not a multiple of
> 8. This P
On Thu, 24 Aug 2023 13:56:12 GMT, Martin Doerr wrote:
> I've found a way to solve the remaining FFI problem on linux PPC64 Big
> Endian. Large structs (>8 Bytes) which are passed in registers or on stack
> require shifting the Bytes in the last slot if the size is not a multiple of
> 8. This P
I've found a way to solve the remaining FFI problem on linux PPC64 Big Endian.
Large structs (>8 Bytes) which are passed in registers or on stack require
shifting the Bytes in the last slot if the size is not a multiple of 8. This PR
adds the required functionality to the Java code.
Please revi
44 matches
Mail list logo
