<feed xmlns='http://www.w3.org/2005/Atom'>
<title>yuzu/src/core/arm/arm_interface.cpp, branch main</title>
<subtitle>Nintendo Switch emulator</subtitle>
<link rel='alternate' type='text/html' href='https://git.dog6.net/yuzu/'/>
<entry>
<title>arm: Skip duplicate consecutive addresses in backtrace output</title>
<updated>2025-05-11T11:17:03+00:00</updated>
<author>
<name>Zephyron</name>
<email>zephyron@citron-emu.org</email>
</author>
<published>2025-01-02T07:25:36+00:00</published>
<link rel='alternate' type='text/html' href='https://git.dog6.net/yuzu/commit/?id=001e337260ec4596f21f133109bd3402e93e22a9'/>
<id>001e337260ec4596f21f133109bd3402e93e22a9</id>
<content type='text'>
Adds logic to track and skip duplicate consecutive addresses when logging
backtraces. This improves log readability by removing redundant entries
that point to the same address in sequence.
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Adds logic to track and skip duplicate consecutive addresses when logging
backtraces. This improves log readability by removing redundant entries
that point to the same address in sequence.
</pre>
</div>
</content>
</entry>
<entry>
<title>arm: Skip duplicate consecutive addresses in backtrace output</title>
<updated>2024-12-20T15:24:58+00:00</updated>
<author>
<name>EmulationEnjoyer</name>
<email>emulationenjoyer@noreply.localhost</email>
</author>
<published>2024-11-12T21:38:09+00:00</published>
<link rel='alternate' type='text/html' href='https://git.dog6.net/yuzu/commit/?id=65202006a27cc35a6633f8b5092d51fe62dfd373'/>
<id>65202006a27cc35a6633f8b5092d51fe62dfd373</id>
<content type='text'>
ref: https://git.citron-emu.org/Citron/Citron/commit/aa56430f2a3179068f77b865d4fd0179aa5cc4c4
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
ref: https://git.citron-emu.org/Citron/Citron/commit/aa56430f2a3179068f77b865d4fd0179aa5cc4c4
</pre>
</div>
</content>
</entry>
<entry>
<title>kernel: instantiate memory separately for each guest process</title>
<updated>2023-12-23T02:52:49+00:00</updated>
<author>
<name>Liam</name>
<email>byteslice@airmail.cc</email>
</author>
<published>2023-12-12T01:21:23+00:00</published>
<link rel='alternate' type='text/html' href='https://git.dog6.net/yuzu/commit/?id=419055e484f0f0073d5832f7ded5fd3a3e5ad7de'/>
<id>419055e484f0f0073d5832f7ded5fd3a3e5ad7de</id>
<content type='text'>
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
</pre>
</div>
</content>
</entry>
<entry>
<title>core: refactor emulated cpu core activation</title>
<updated>2023-12-04T15:37:16+00:00</updated>
<author>
<name>Liam</name>
<email>byteslice@airmail.cc</email>
</author>
<published>2023-11-28T19:30:39+00:00</published>
<link rel='alternate' type='text/html' href='https://git.dog6.net/yuzu/commit/?id=45c87c7e6e841c11def43e5ab25160006dab6d77'/>
<id>45c87c7e6e841c11def43e5ab25160006dab6d77</id>
<content type='text'>
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
</pre>
</div>
</content>
</entry>
<entry>
<title>arm: Print backtrace on data abort</title>
<updated>2023-11-25T05:47:36+00:00</updated>
<author>
<name>GPUCode</name>
<email>geoster3d@gmail.com</email>
</author>
<published>2023-11-20T16:56:54+00:00</published>
<link rel='alternate' type='text/html' href='https://git.dog6.net/yuzu/commit/?id=1cde01c8c8f4c7272f7f2db3846ff7403caa49e1'/>
<id>1cde01c8c8f4c7272f7f2db3846ff7403caa49e1</id>
<content type='text'>
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
</pre>
</div>
</content>
</entry>
<entry>
<title>core: check for thread dpc before eret</title>
<updated>2023-11-13T16:28:23+00:00</updated>
<author>
<name>Liam</name>
<email>byteslice@airmail.cc</email>
</author>
<published>2023-11-13T16:28:23+00:00</published>
<link rel='alternate' type='text/html' href='https://git.dog6.net/yuzu/commit/?id=e3b510a4b4aa79e4fefadcbb3f33136016d35988'/>
<id>e3b510a4b4aa79e4fefadcbb3f33136016d35988</id>
<content type='text'>
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
</pre>
</div>
</content>
</entry>
<entry>
<title>kernel: update KProcess</title>
<updated>2023-10-22T00:03:41+00:00</updated>
<author>
<name>Liam</name>
<email>byteslice@airmail.cc</email>
</author>
<published>2023-10-21T20:47:43+00:00</published>
<link rel='alternate' type='text/html' href='https://git.dog6.net/yuzu/commit/?id=8c59543ee32c8bff575bab7ec1e70f76f8eda437'/>
<id>8c59543ee32c8bff575bab7ec1e70f76f8eda437</id>
<content type='text'>
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
</pre>
</div>
</content>
</entry>
<entry>
<title>Fixes and workarounds to make UBSan happier on macOS</title>
<updated>2023-07-15T19:00:28+00:00</updated>
<author>
<name>comex</name>
<email>comexk@gmail.com</email>
</author>
<published>2023-07-01T22:00:39+00:00</published>
<link rel='alternate' type='text/html' href='https://git.dog6.net/yuzu/commit/?id=d7c532d8894ce806c9af13b8dd3eec975642b348'/>
<id>d7c532d8894ce806c9af13b8dd3eec975642b348</id>
<content type='text'>
There are still some other issues not addressed here, but it's a start.

Workarounds for false-positive reports:

- `RasterizerAccelerated`: Put a gigantic array behind a `unique_ptr`,
  because UBSan has a [hardcoded limit](https://stackoverflow.com/questions/64531383/c-runtime-error-using-fsanitize-undefined-object-has-a-possibly-invalid-vp)
  of how big it thinks objects can be, specifically when dealing with
  offset-to-top values used with multiple inheritance.  Hopefully this
  doesn't have a performance impact.

- `QueryCacheBase::QueryCacheBase`: Avoid an operation that UBSan thinks
  is UB even though it at least arguably isn't.  See the link in the
  comment for more information.

Fixes for correct reports:

- `PageTable`, `Memory`: Use `uintptr_t` values instead of pointers to
  avoid UB from pointer overflow (when pointer arithmetic wraps around
  the address space).

- `KScheduler::Reload`: `thread-&gt;GetOwnerProcess()` can be `nullptr`;
  avoid calling methods on it in this case.  (The existing code returns
  a garbage reference to a field, which is then passed into
  `LoadWatchpointArray`, and apparently it's never used, so it's
  harmless in practice but still triggers UBSan.)

- `KAutoObject::Close`: This function calls `this-&gt;Destroy()`, which
  overwrites the beginning of the object with junk (specifically a free
  list pointer).  Then it calls `this-&gt;UnregisterWithKernel()`.  UBSan
  complains about a type mismatch because the vtable has been
  overwritten, and I believe this is indeed UB.  `UnregisterWithKernel`
  also loads `m_kernel` from the 'freed' object, which seems to be
  technically safe (the overwriting doesn't extend as far as that
  field), but seems dubious.  Switch to a `static` method and load
  `m_kernel` in advance.
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
There are still some other issues not addressed here, but it's a start.

Workarounds for false-positive reports:

- `RasterizerAccelerated`: Put a gigantic array behind a `unique_ptr`,
  because UBSan has a [hardcoded limit](https://stackoverflow.com/questions/64531383/c-runtime-error-using-fsanitize-undefined-object-has-a-possibly-invalid-vp)
  of how big it thinks objects can be, specifically when dealing with
  offset-to-top values used with multiple inheritance.  Hopefully this
  doesn't have a performance impact.

- `QueryCacheBase::QueryCacheBase`: Avoid an operation that UBSan thinks
  is UB even though it at least arguably isn't.  See the link in the
  comment for more information.

Fixes for correct reports:

- `PageTable`, `Memory`: Use `uintptr_t` values instead of pointers to
  avoid UB from pointer overflow (when pointer arithmetic wraps around
  the address space).

- `KScheduler::Reload`: `thread-&gt;GetOwnerProcess()` can be `nullptr`;
  avoid calling methods on it in this case.  (The existing code returns
  a garbage reference to a field, which is then passed into
  `LoadWatchpointArray`, and apparently it's never used, so it's
  harmless in practice but still triggers UBSan.)

- `KAutoObject::Close`: This function calls `this-&gt;Destroy()`, which
  overwrites the beginning of the object with junk (specifically a free
  list pointer).  Then it calls `this-&gt;UnregisterWithKernel()`.  UBSan
  complains about a type mismatch because the vtable has been
  overwritten, and I believe this is indeed UB.  `UnregisterWithKernel`
  also loads `m_kernel` from the 'freed' object, which seems to be
  technically safe (the overwriting doesn't extend as far as that
  field), but seems dubious.  Switch to a `static` method and load
  `m_kernel` in advance.
</pre>
</div>
</content>
</entry>
<entry>
<title>arm_interface: correct breakpoint rewind condition</title>
<updated>2023-07-09T16:03:25+00:00</updated>
<author>
<name>Liam</name>
<email>byteslice@airmail.cc</email>
</author>
<published>2023-07-09T16:03:25+00:00</published>
<link rel='alternate' type='text/html' href='https://git.dog6.net/yuzu/commit/?id=82568412f6916068d1b4fb10988f51cbd019b0dd'/>
<id>82568412f6916068d1b4fb10988f51cbd019b0dd</id>
<content type='text'>
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
</pre>
</div>
</content>
</entry>
<entry>
<title>core: decouple ARM interface from Dynarmic</title>
<updated>2023-06-13T02:11:51+00:00</updated>
<author>
<name>Liam</name>
<email>byteslice@airmail.cc</email>
</author>
<published>2023-06-13T01:34:25+00:00</published>
<link rel='alternate' type='text/html' href='https://git.dog6.net/yuzu/commit/?id=8506915208dc2d518c821f00e4286824d778642c'/>
<id>8506915208dc2d518c821f00e4286824d778642c</id>
<content type='text'>
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
</pre>
</div>
</content>
</entry>
</feed>
