<html><head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body>
<p>It looks like <span class="pl-en">io_uring_wait_cqe [1] takes as
second argument a pointer to a pointer, but you pass it an
pointer to a struct.</span></p>
<div>I think it s<span class="pl-en">hould be:</span><br>
</div>
<p> // struct io_uring_cqe **<span class="pl-en">cqeRef =
malloc(sizeof *cqeRef);</span><br>
<span class="pl-en"><span class="pl-smi"> MemorySegment</span>
cqeRef = <span class="pl-s1">MemorySegment</span>.<span class="pl-en">allocateNative</span>(C_POINTER, <span class="pl-s1">session</span>);<br>
</span> io_uring_wait_cqe(ring, <span class="pl-en">cqeRef</span>);<br>
// struct io_uring_cqe *cqe = *cqeRef;<br>
MemoryAddress cqe = <span class="pl-en">cqeRef.get(C_POINTER,
0)</span><span class="pl-en">;</span><br>
</p>
<p> MemoryAddress user_data = io_uring_cqe_get_data(cqe);<br>
MemorySegment user_data_segment =
MemorySegment.ofAddress(user_data, myStructLayout.byteSize(),
session);</p>
<div>Note that MemorySegment::asSlice doesn't perform any
dereference, so it is not equivalent to `cqes[0]`.<br>
</div>
<div><br>
</div>
<div>Jorn<br>
</div>
<div><br>
</div>
<div>[1]:
<a class="moz-txt-link-freetext" href="https://man.archlinux.org/man/extra/liburing/io_uring_wait_cqe.3.en">https://man.archlinux.org/man/extra/liburing/io_uring_wait_cqe.3.en</a><br>
</div>
<div><br>
</div>
<div class="moz-cite-prefix">On 23/09/2022 22:15, Gavin Ray wrote:<br>
</div>
<blockquote type="cite" cite="mid:CAFtvWZPEw=jAvgv19Z=wjyrNFtGa6HwyOcDvPkZu0R9D0YoCtg@mail.gmail.com">
<div dir="ltr">
<div dir="ltr">
<div dir="ltr">
<div dir="ltr">
<div dir="ltr">Okay, it seems I've gotten the "hello
world" almost working
<div><br>
</div>
<div>The only problem left is that trying to read
the "user_data" pointer returned gives garbage.</div>
<div>When I submit the event, I attach a struct to it,
like this:</div>
<div><br>
</div>
<div>
<div> MemorySegment myStruct =
MemorySegment.allocateNative(myStructLayout,
session);</div>
<div> myStruct$foo.set(myStruct, 1);</div>
<div> myStruct$bar.set(myStruct, 2);</div>
<div> io_uring_sqe_set_data(sqePtr, myStruct);</div>
</div>
<div><br>
</div>
<div>Then when the results come back, I am doing this to
read the userdata pointer:</div>
<div><br>
</div>
<div>
<div> MemorySegment cqes =
io_uring_cqe.allocateArray(1, allocator);</div>
<div> io_uring_wait_cqe(ring, cqes);</div>
<div> // cqes[0]</div>
<div> MemorySegment cqe = cqes.asSlice(0,
io_uring_cqe.$LAYOUT().byteSize());</div>
<div> // cqes[0]->user_data</div>
<div> MemoryAddress user_data =
MemoryAddress.ofLong(io_uring_cqe.user_data$get(cqe));</div>
<div> MemorySegment user_data_segment =
MemorySegment.ofAddress(user_data,
myStructLayout.byteSize(), session);</div>
</div>
<div><br>
</div>
<div>I try to print it out, comparing the "foo" and
"bar" field against the original values of "1" and "2"</div>
<div><br>
</div>
<div>
<div> user_data->foo=1480561120</div>
<div> user_data->bar=32545</div>
</div>
<div><br>
</div>
<div>Does anyone know/see what I am doing wrong here? 🤔</div>
<div>
<div><a href="https://gist.github.com/GavinRay97/8ea0997693c00f8df61968a98ba30135#file-io_uring-java-L48-L55" moz-do-not-send="true" class="moz-txt-link-freetext">https://gist.github.com/GavinRay97/8ea0997693c00f8df61968a98ba30135#file-io_uring-java-L48-L55</a></div>
</div>
</div>
</div>
</div>
</div>
</div>
<br>
<div class="gmail_quote">
<div dir="ltr" class="gmail_attr">On Wed, Sep 21, 2022 at 8:18
PM Gavin Ray <<a href="mailto:ray.gavin97@gmail.com" moz-do-not-send="true" class="moz-txt-link-freetext">ray.gavin97@gmail.com</a>>
wrote:<br>
</div>
<blockquote class="gmail_quote" style="margin:0px 0px 0px
0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
<div dir="ltr">
<div dir="ltr">@Daniel, I didn't know the lwjgl folks were
looking into this area, I've had a brief look but it's
definitely something I want to check out more!
<div><br>
</div>
<div>@Maurizio, the idea about defining a second library
that depends on it is pretty clever.</div>
<div><br>
</div>
<div>It turns out that you can pull liburing in with CMake
without so much effort</div>
<div>To set that scenario up was just the below code, this
should work decently 👍</div>
<div><br>
</div>
<div>=============================</div>
<div>
<div>ExternalProject_Add(</div>
<div> liburing_git</div>
<div> GIT_REPOSITORY <a href="http://github.com/axboe/liburing.git" target="_blank" moz-do-not-send="true" class="moz-txt-link-freetext">http://github.com/axboe/liburing.git</a></div>
<div> GIT_TAG liburing-2.2</div>
<div> BUILD_IN_SOURCE 1</div>
<div> BUILD_BYPRODUCTS
"<SOURCE_DIR>/src/liburing.a"</div>
<div> BUILD_COMMAND make</div>
<div>)</div>
<div>ExternalProject_Get_Property(liburing_git
SOURCE_DIR)</div>
<div>add_library(liburing INTERFACE)</div>
<div>add_dependencies(liburing liburing_git)</div>
<div>target_include_directories(liburing INTERFACE
${SOURCE_DIR}/src/include)</div>
<div>target_link_libraries(liburing INTERFACE
${SOURCE_DIR}/src/liburing.a)</div>
<div><br>
</div>
<div># NOW LINK IT TO SECOND LIBRARY</div>
<div>add_library(liburing_extras SHARED
liburing_extras/main.c)</div>
<div>target_link_libraries(liburing_extras liburing)</div>
<div>add_dependencies(liburing_extras liburing)</div>
</div>
<div><br>
</div>
</div>
</div>
<br>
<div class="gmail_quote">
<div dir="ltr" class="gmail_attr">On Wed, Sep 21, 2022 at
7:24 PM Maurizio Cimadamore <<a href="mailto:maurizio.cimadamore@oracle.com" target="_blank" moz-do-not-send="true" class="moz-txt-link-freetext">maurizio.cimadamore@oracle.com</a>>
wrote:<br>
</div>
<blockquote class="gmail_quote" style="margin:0px 0px 0px
0.8ex;border-left:1px solid
rgb(204,204,204);padding-left:1ex"><br>
On 21/09/2022 23:39, Gavin Ray wrote:<br>
> That makes sense -- I will look at patching liburing
with a dirty <br>
> one-liner that strips "static inline" from the
definitions that have <br>
> them then, ty =)<br>
<br>
Sorry, hit send too fast, and forgot to reply to this.<br>
<br>
There are few approaches that can be tried.<br>
<br>
First, you could define another shared library that
depends on io_uring <br>
- e.g. a C file which includes the io_uring header. This
new file would <br>
contain _new_ exported symbols for all the static inline
functions, <br>
whose implementation just delegates to the underlying
static inline <br>
functions (in the included io_uring header). Then you load
up the <br>
library with Panama and, if you have set up things
correctly, you will <br>
see both the static inlines and the exported symbols in IO
uring. (in <br>
fact, we do a similar trick to expose system library
symbols like <br>
"printf" on Windows, which implements them as macros).<br>
<br>
Another (more fun?) option would be to just keep the
original library, <br>
and implement the static inline functions in Java
directly, using the <br>
FFM API. Most of the functions I see have only 2-3 lines
in them, so it <br>
shouldn't be hard.<br>
<br>
Maurizio<br>
<br>
</blockquote>
</div>
</blockquote>
</div>
</blockquote>
</body>
</html>