<feed xmlns='http://www.w3.org/2005/Atom'>
<title>linux-toradex.git/arch/mips, branch v5.11-rc5</title>
<subtitle>Linux kernel for Apalis and Colibri modules</subtitle>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/'/>
<entry>
<title>mips/mm/highmem: use set_pte() for kmap_local()</title>
<updated>2021-01-24T18:34:52+00:00</updated>
<author>
<name>Thomas Gleixner</name>
<email>tglx@linutronix.de</email>
</author>
<published>2021-01-24T05:02:07+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=8c0d5d78f3596e203e9cd27563a8380649c03ad0'/>
<id>8c0d5d78f3596e203e9cd27563a8380649c03ad0</id>
<content type='text'>
set_pte_at() on MIPS invokes update_cache() which might recurse into
kmap_local().

Use set_pte() like the original MIPS highmem implementation did.

Link: https://lkml.kernel.org/r/20210112170411.187513575@linutronix.de
Fixes: a4c33e83bca1 ("mips/mm/highmem: Switch to generic kmap atomic")
Signed-off-by: Thomas Gleixner &lt;tglx@linutronix.de&gt;
Reported-by: Paul Cercueil &lt;paul@crapouillou.net&gt;
Reported-by: Thomas Bogendoerfer &lt;tsbogend@alpha.franken.de&gt;
Acked-by: Thomas Bogendoerfer &lt;tsbogend@alpha.franken.de&gt;
Cc: Andreas Larsson &lt;andreas@gaisler.com&gt;
Cc: "David S. Miller" &lt;davem@davemloft.net&gt;
Cc: Michael Ellerman &lt;mpe@ellerman.id.au&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Signed-off-by: Andrew Morton &lt;akpm@linux-foundation.org&gt;
Signed-off-by: Linus Torvalds &lt;torvalds@linux-foundation.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
set_pte_at() on MIPS invokes update_cache() which might recurse into
kmap_local().

Use set_pte() like the original MIPS highmem implementation did.

Link: https://lkml.kernel.org/r/20210112170411.187513575@linutronix.de
Fixes: a4c33e83bca1 ("mips/mm/highmem: Switch to generic kmap atomic")
Signed-off-by: Thomas Gleixner &lt;tglx@linutronix.de&gt;
Reported-by: Paul Cercueil &lt;paul@crapouillou.net&gt;
Reported-by: Thomas Bogendoerfer &lt;tsbogend@alpha.franken.de&gt;
Acked-by: Thomas Bogendoerfer &lt;tsbogend@alpha.franken.de&gt;
Cc: Andreas Larsson &lt;andreas@gaisler.com&gt;
Cc: "David S. Miller" &lt;davem@davemloft.net&gt;
Cc: Michael Ellerman &lt;mpe@ellerman.id.au&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Signed-off-by: Andrew Morton &lt;akpm@linux-foundation.org&gt;
Signed-off-by: Linus Torvalds &lt;torvalds@linux-foundation.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>Merge tag 'mips_fixes_5.11.1' of git://git.kernel.org/pub/scm/linux/kernel/git/mips/linux</title>
<updated>2021-01-15T21:07:33+00:00</updated>
<author>
<name>Linus Torvalds</name>
<email>torvalds@linux-foundation.org</email>
</author>
<published>2021-01-15T21:07:33+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=f288c895622f881324cf71ef9eec3af60ac8a197'/>
<id>f288c895622f881324cf71ef9eec3af60ac8a197</id>
<content type='text'>
Pull MIPS fixes from Thomas Bogendoerfer:

 - fix coredumps on 64bit kernels

 - fix for alignment bugs preventing booting

 - fix checking for failed irq_alloc_desc calls

* tag 'mips_fixes_5.11.1' of git://git.kernel.org/pub/scm/linux/kernel/git/mips/linux:
  MIPS: OCTEON: fix unreachable code in octeon_irq_init_ciu
  MIPS: relocatable: fix possible boot hangup with KASLR enabled
  MIPS: Fix malformed NT_FILE and NT_SIGINFO in 32bit coredumps
  MIPS: boot: Fix unaligned access with CONFIG_MIPS_RAW_APPENDED_DTB
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Pull MIPS fixes from Thomas Bogendoerfer:

 - fix coredumps on 64bit kernels

 - fix for alignment bugs preventing booting

 - fix checking for failed irq_alloc_desc calls

* tag 'mips_fixes_5.11.1' of git://git.kernel.org/pub/scm/linux/kernel/git/mips/linux:
  MIPS: OCTEON: fix unreachable code in octeon_irq_init_ciu
  MIPS: relocatable: fix possible boot hangup with KASLR enabled
  MIPS: Fix malformed NT_FILE and NT_SIGINFO in 32bit coredumps
  MIPS: boot: Fix unaligned access with CONFIG_MIPS_RAW_APPENDED_DTB
</pre>
</div>
</content>
</entry>
<entry>
<title>MIPS: OCTEON: fix unreachable code in octeon_irq_init_ciu</title>
<updated>2021-01-13T09:59:32+00:00</updated>
<author>
<name>Menglong Dong</name>
<email>dong.menglong@zte.com.cn</email>
</author>
<published>2021-01-11T13:27:25+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=7b490a8ab0f2d3ab8d838a4ff22ae86edafd34a1'/>
<id>7b490a8ab0f2d3ab8d838a4ff22ae86edafd34a1</id>
<content type='text'>
The type of 'r' in octeon_irq_init_ciu is 'unsigned int', so 'r &lt; 0'
can't be true.

Fix this by change the type of 'r' and 'i' from 'unsigned int'
to 'int'. As 'i' won't be negative, this change works.

Fixes: 99fbc70f8547 ("MIPS: Octeon: irq: Alloc desc before configuring IRQ")
Signed-off-by: Menglong Dong &lt;dong.menglong@zte.com.cn&gt;
Reviewed-by: Alexander Sverdlin &lt;alexander.sverdlin@nokia.com&gt;
Signed-off-by: Thomas Bogendoerfer &lt;tsbogend@alpha.franken.de&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
The type of 'r' in octeon_irq_init_ciu is 'unsigned int', so 'r &lt; 0'
can't be true.

Fix this by change the type of 'r' and 'i' from 'unsigned int'
to 'int'. As 'i' won't be negative, this change works.

Fixes: 99fbc70f8547 ("MIPS: Octeon: irq: Alloc desc before configuring IRQ")
Signed-off-by: Menglong Dong &lt;dong.menglong@zte.com.cn&gt;
Reviewed-by: Alexander Sverdlin &lt;alexander.sverdlin@nokia.com&gt;
Signed-off-by: Thomas Bogendoerfer &lt;tsbogend@alpha.franken.de&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>MIPS: relocatable: fix possible boot hangup with KASLR enabled</title>
<updated>2021-01-13T09:59:04+00:00</updated>
<author>
<name>Alexander Lobakin</name>
<email>alobakin@pm.me</email>
</author>
<published>2021-01-10T14:21:05+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=69e976831cd53f9ba304fd20305b2025ecc78eab'/>
<id>69e976831cd53f9ba304fd20305b2025ecc78eab</id>
<content type='text'>
LLVM-built Linux triggered a boot hangup with KASLR enabled.

arch/mips/kernel/relocate.c:get_random_boot() uses linux_banner,
which is a string constant, as a random seed, but accesses it
as an array of unsigned long (in rotate_xor()).
When the address of linux_banner is not aligned to sizeof(long),
such access emits unaligned access exception and hangs the kernel.

Use PTR_ALIGN() to align input address to sizeof(long) and also
align down the input length to prevent possible access-beyond-end.

Fixes: 405bc8fd12f5 ("MIPS: Kernel: Implement KASLR using CONFIG_RELOCATABLE")
Cc: stable@vger.kernel.org # 4.7+
Signed-off-by: Alexander Lobakin &lt;alobakin@pm.me&gt;
Tested-by: Nathan Chancellor &lt;natechancellor@gmail.com&gt;
Reviewed-by: Kees Cook &lt;keescook@chromium.org&gt;
Signed-off-by: Thomas Bogendoerfer &lt;tsbogend@alpha.franken.de&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
LLVM-built Linux triggered a boot hangup with KASLR enabled.

arch/mips/kernel/relocate.c:get_random_boot() uses linux_banner,
which is a string constant, as a random seed, but accesses it
as an array of unsigned long (in rotate_xor()).
When the address of linux_banner is not aligned to sizeof(long),
such access emits unaligned access exception and hangs the kernel.

Use PTR_ALIGN() to align input address to sizeof(long) and also
align down the input length to prevent possible access-beyond-end.

Fixes: 405bc8fd12f5 ("MIPS: Kernel: Implement KASLR using CONFIG_RELOCATABLE")
Cc: stable@vger.kernel.org # 4.7+
Signed-off-by: Alexander Lobakin &lt;alobakin@pm.me&gt;
Tested-by: Nathan Chancellor &lt;natechancellor@gmail.com&gt;
Reviewed-by: Kees Cook &lt;keescook@chromium.org&gt;
Signed-off-by: Thomas Bogendoerfer &lt;tsbogend@alpha.franken.de&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>local64.h: make &lt;asm/local64.h&gt; mandatory</title>
<updated>2020-12-29T23:36:49+00:00</updated>
<author>
<name>Randy Dunlap</name>
<email>rdunlap@infradead.org</email>
</author>
<published>2020-12-29T23:14:49+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=87dbc209ea04645fd2351981f09eff5d23f8e2e9'/>
<id>87dbc209ea04645fd2351981f09eff5d23f8e2e9</id>
<content type='text'>
Make &lt;asm-generic/local64.h&gt; mandatory in include/asm-generic/Kbuild and
remove all arch/*/include/asm/local64.h arch-specific files since they
only #include &lt;asm-generic/local64.h&gt;.

This fixes build errors on arch/c6x/ and arch/nios2/ for
block/blk-iocost.c.

Build-tested on 21 of 25 arch-es.  (tools problems on the others)

Yes, we could even rename &lt;asm-generic/local64.h&gt; to
&lt;linux/local64.h&gt; and change all #includes to use
&lt;linux/local64.h&gt; instead.

Link: https://lkml.kernel.org/r/20201227024446.17018-1-rdunlap@infradead.org
Signed-off-by: Randy Dunlap &lt;rdunlap@infradead.org&gt;
Suggested-by: Christoph Hellwig &lt;hch@infradead.org&gt;
Reviewed-by: Masahiro Yamada &lt;masahiroy@kernel.org&gt;
Cc: Jens Axboe &lt;axboe@kernel.dk&gt;
Cc: Ley Foon Tan &lt;ley.foon.tan@intel.com&gt;
Cc: Mark Salter &lt;msalter@redhat.com&gt;
Cc: Aurelien Jacquiot &lt;jacquiot.aurelien@gmail.com&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Arnd Bergmann &lt;arnd@arndb.de&gt;
Signed-off-by: Andrew Morton &lt;akpm@linux-foundation.org&gt;
Signed-off-by: Linus Torvalds &lt;torvalds@linux-foundation.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Make &lt;asm-generic/local64.h&gt; mandatory in include/asm-generic/Kbuild and
remove all arch/*/include/asm/local64.h arch-specific files since they
only #include &lt;asm-generic/local64.h&gt;.

This fixes build errors on arch/c6x/ and arch/nios2/ for
block/blk-iocost.c.

Build-tested on 21 of 25 arch-es.  (tools problems on the others)

Yes, we could even rename &lt;asm-generic/local64.h&gt; to
&lt;linux/local64.h&gt; and change all #includes to use
&lt;linux/local64.h&gt; instead.

Link: https://lkml.kernel.org/r/20201227024446.17018-1-rdunlap@infradead.org
Signed-off-by: Randy Dunlap &lt;rdunlap@infradead.org&gt;
Suggested-by: Christoph Hellwig &lt;hch@infradead.org&gt;
Reviewed-by: Masahiro Yamada &lt;masahiroy@kernel.org&gt;
Cc: Jens Axboe &lt;axboe@kernel.dk&gt;
Cc: Ley Foon Tan &lt;ley.foon.tan@intel.com&gt;
Cc: Mark Salter &lt;msalter@redhat.com&gt;
Cc: Aurelien Jacquiot &lt;jacquiot.aurelien@gmail.com&gt;
Cc: Peter Zijlstra &lt;peterz@infradead.org&gt;
Cc: Arnd Bergmann &lt;arnd@arndb.de&gt;
Signed-off-by: Andrew Morton &lt;akpm@linux-foundation.org&gt;
Signed-off-by: Linus Torvalds &lt;torvalds@linux-foundation.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>MIPS: Fix malformed NT_FILE and NT_SIGINFO in 32bit coredumps</title>
<updated>2020-12-28T22:26:17+00:00</updated>
<author>
<name>Al Viro</name>
<email>viro@zeniv.linux.org.uk</email>
</author>
<published>2020-12-24T19:44:38+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=698222457465ce343443be81c5512edda86e5914'/>
<id>698222457465ce343443be81c5512edda86e5914</id>
<content type='text'>
Patches that introduced NT_FILE and NT_SIGINFO notes back in 2012
had taken care of native (fs/binfmt_elf.c) and compat (fs/compat_binfmt_elf.c)
coredumps; unfortunately, compat on mips (which does not go through the
usual compat_binfmt_elf.c) had not been noticed.

As the result, both N32 and O32 coredumps on 64bit mips kernels
have those sections malformed enough to confuse the living hell out of
all gdb and readelf versions (up to and including the tip of binutils-gdb.git).

Longer term solution is to make both O32 and N32 compat use the
regular compat_binfmt_elf.c, but that's too much for backports.  The minimal
solution is to do in arch/mips/kernel/binfmt_elf[on]32.c the same thing
those patches have done in fs/compat_binfmt_elf.c

Cc: stable@kernel.org # v3.7+
Signed-off-by: Al Viro &lt;viro@zeniv.linux.org.uk&gt;
Signed-off-by: Thomas Bogendoerfer &lt;tsbogend@alpha.franken.de&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Patches that introduced NT_FILE and NT_SIGINFO notes back in 2012
had taken care of native (fs/binfmt_elf.c) and compat (fs/compat_binfmt_elf.c)
coredumps; unfortunately, compat on mips (which does not go through the
usual compat_binfmt_elf.c) had not been noticed.

As the result, both N32 and O32 coredumps on 64bit mips kernels
have those sections malformed enough to confuse the living hell out of
all gdb and readelf versions (up to and including the tip of binutils-gdb.git).

Longer term solution is to make both O32 and N32 compat use the
regular compat_binfmt_elf.c, but that's too much for backports.  The minimal
solution is to do in arch/mips/kernel/binfmt_elf[on]32.c the same thing
those patches have done in fs/compat_binfmt_elf.c

Cc: stable@kernel.org # v3.7+
Signed-off-by: Al Viro &lt;viro@zeniv.linux.org.uk&gt;
Signed-off-by: Thomas Bogendoerfer &lt;tsbogend@alpha.franken.de&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>MIPS: boot: Fix unaligned access with CONFIG_MIPS_RAW_APPENDED_DTB</title>
<updated>2020-12-28T22:26:16+00:00</updated>
<author>
<name>Paul Cercueil</name>
<email>paul@crapouillou.net</email>
</author>
<published>2020-12-16T23:39:56+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=4d4f9c1a17a3480f8fe523673f7232b254d724b7'/>
<id>4d4f9c1a17a3480f8fe523673f7232b254d724b7</id>
<content type='text'>
The compressed payload is not necesarily 4-byte aligned, at least when
compiling with Clang. In that case, the 4-byte value appended to the
compressed payload that corresponds to the uncompressed kernel image
size must be read using get_unaligned_le32().

This fixes Clang-built kernels not booting on MIPS (tested on a Ingenic
JZ4770 board).

Fixes: b8f54f2cde78 ("MIPS: ZBOOT: copy appended dtb to the end of the kernel")
Cc: &lt;stable@vger.kernel.org&gt; # v4.7
Signed-off-by: Paul Cercueil &lt;paul@crapouillou.net&gt;
Reviewed-by: Nick Desaulniers &lt;ndesaulniers@google.com&gt;
Reviewed-by: Philippe Mathieu-Daudé &lt;f4bug@amsat.org&gt;
Signed-off-by: Thomas Bogendoerfer &lt;tsbogend@alpha.franken.de&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
The compressed payload is not necesarily 4-byte aligned, at least when
compiling with Clang. In that case, the 4-byte value appended to the
compressed payload that corresponds to the uncompressed kernel image
size must be read using get_unaligned_le32().

This fixes Clang-built kernels not booting on MIPS (tested on a Ingenic
JZ4770 board).

Fixes: b8f54f2cde78 ("MIPS: ZBOOT: copy appended dtb to the end of the kernel")
Cc: &lt;stable@vger.kernel.org&gt; # v4.7
Signed-off-by: Paul Cercueil &lt;paul@crapouillou.net&gt;
Reviewed-by: Nick Desaulniers &lt;ndesaulniers@google.com&gt;
Reviewed-by: Philippe Mathieu-Daudé &lt;f4bug@amsat.org&gt;
Signed-off-by: Thomas Bogendoerfer &lt;tsbogend@alpha.franken.de&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>epoll: fix compat syscall wire up of epoll_pwait2</title>
<updated>2020-12-20T18:01:38+00:00</updated>
<author>
<name>Heiko Carstens</name>
<email>hca@linux.ibm.com</email>
</author>
<published>2020-12-20T09:58:30+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=450f68e2425e63228d91f346547714076fc5f3e8'/>
<id>450f68e2425e63228d91f346547714076fc5f3e8</id>
<content type='text'>
Commit b0a0c2615f6f ("epoll: wire up syscall epoll_pwait2") wired up
the 64 bit syscall instead of the compat variant in a couple of places.

Fixes: b0a0c2615f6f ("epoll: wire up syscall epoll_pwait2")
Signed-off-by: Heiko Carstens &lt;hca@linux.ibm.com&gt;
Acked-by: Arnd Bergmann &lt;arnd@arndb.de&gt;
Cc: Willem de Bruijn &lt;willemb@google.com&gt;
Cc: Al Viro &lt;viro@zeniv.linux.org.uk&gt;
Cc: Arnd Bergmann &lt;arnd@arndb.de&gt;
Cc: Matthew Wilcox (Oracle) &lt;willy@infradead.org&gt;
Cc: Catalin Marinas &lt;catalin.marinas@arm.com&gt;
Cc: Will Deacon &lt;will@kernel.org&gt;
Cc: Thomas Bogendoerfer &lt;tsbogend@alpha.franken.de&gt;
Cc: Vasily Gorbik &lt;gor@linux.ibm.com&gt;
Cc: Christian Borntraeger &lt;borntraeger@de.ibm.com&gt;
Cc: "David S. Miller" &lt;davem@davemloft.net&gt;
Signed-off-by: Linus Torvalds &lt;torvalds@linux-foundation.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Commit b0a0c2615f6f ("epoll: wire up syscall epoll_pwait2") wired up
the 64 bit syscall instead of the compat variant in a couple of places.

Fixes: b0a0c2615f6f ("epoll: wire up syscall epoll_pwait2")
Signed-off-by: Heiko Carstens &lt;hca@linux.ibm.com&gt;
Acked-by: Arnd Bergmann &lt;arnd@arndb.de&gt;
Cc: Willem de Bruijn &lt;willemb@google.com&gt;
Cc: Al Viro &lt;viro@zeniv.linux.org.uk&gt;
Cc: Arnd Bergmann &lt;arnd@arndb.de&gt;
Cc: Matthew Wilcox (Oracle) &lt;willy@infradead.org&gt;
Cc: Catalin Marinas &lt;catalin.marinas@arm.com&gt;
Cc: Will Deacon &lt;will@kernel.org&gt;
Cc: Thomas Bogendoerfer &lt;tsbogend@alpha.franken.de&gt;
Cc: Vasily Gorbik &lt;gor@linux.ibm.com&gt;
Cc: Christian Borntraeger &lt;borntraeger@de.ibm.com&gt;
Cc: "David S. Miller" &lt;davem@davemloft.net&gt;
Signed-off-by: Linus Torvalds &lt;torvalds@linux-foundation.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>epoll: wire up syscall epoll_pwait2</title>
<updated>2020-12-19T19:18:38+00:00</updated>
<author>
<name>Willem de Bruijn</name>
<email>willemb@google.com</email>
</author>
<published>2020-12-18T22:05:41+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=b0a0c2615f6f199a656ed8549d7dce625d77aa77'/>
<id>b0a0c2615f6f199a656ed8549d7dce625d77aa77</id>
<content type='text'>
Split off from prev patch in the series that implements the syscall.

Link: https://lkml.kernel.org/r/20201121144401.3727659-4-willemdebruijn.kernel@gmail.com
Signed-off-by: Willem de Bruijn &lt;willemb@google.com&gt;
Cc: Al Viro &lt;viro@zeniv.linux.org.uk&gt;
Cc: Arnd Bergmann &lt;arnd@arndb.de&gt;
Cc: Matthew Wilcox (Oracle) &lt;willy@infradead.org&gt;
Signed-off-by: Andrew Morton &lt;akpm@linux-foundation.org&gt;
Signed-off-by: Linus Torvalds &lt;torvalds@linux-foundation.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Split off from prev patch in the series that implements the syscall.

Link: https://lkml.kernel.org/r/20201121144401.3727659-4-willemdebruijn.kernel@gmail.com
Signed-off-by: Willem de Bruijn &lt;willemb@google.com&gt;
Cc: Al Viro &lt;viro@zeniv.linux.org.uk&gt;
Cc: Arnd Bergmann &lt;arnd@arndb.de&gt;
Cc: Matthew Wilcox (Oracle) &lt;willy@infradead.org&gt;
Signed-off-by: Andrew Morton &lt;akpm@linux-foundation.org&gt;
Signed-off-by: Linus Torvalds &lt;torvalds@linux-foundation.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>Merge tag 'tif-task_work.arch-2020-12-14' of git://git.kernel.dk/linux-block</title>
<updated>2020-12-16T20:33:35+00:00</updated>
<author>
<name>Linus Torvalds</name>
<email>torvalds@linux-foundation.org</email>
</author>
<published>2020-12-16T20:33:35+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=005b2a9dc819a1265a8c765595f8f6d88d6173d9'/>
<id>005b2a9dc819a1265a8c765595f8f6d88d6173d9</id>
<content type='text'>
Pull TIF_NOTIFY_SIGNAL updates from Jens Axboe:
 "This sits on top of of the core entry/exit and x86 entry branch from
  the tip tree, which contains the generic and x86 parts of this work.

  Here we convert the rest of the archs to support TIF_NOTIFY_SIGNAL.

  With that done, we can get rid of JOBCTL_TASK_WORK from task_work and
  signal.c, and also remove a deadlock work-around in io_uring around
  knowing that signal based task_work waking is invoked with the sighand
  wait queue head lock.

  The motivation for this work is to decouple signal notify based
  task_work, of which io_uring is a heavy user of, from sighand. The
  sighand lock becomes a huge contention point, particularly for
  threaded workloads where it's shared between threads. Even outside of
  threaded applications it's slower than it needs to be.

  Roman Gershman &lt;romger@amazon.com&gt; reported that his networked
  workload dropped from 1.6M QPS at 80% CPU to 1.0M QPS at 100% CPU
  after io_uring was changed to use TIF_NOTIFY_SIGNAL. The time was all
  spent hammering on the sighand lock, showing 57% of the CPU time there
  [1].

  There are further cleanups possible on top of this. One example is
  TIF_PATCH_PENDING, where a patch already exists to use
  TIF_NOTIFY_SIGNAL instead. Hopefully this will also lead to more
  consolidation, but the work stands on its own as well"

[1] https://github.com/axboe/liburing/issues/215

* tag 'tif-task_work.arch-2020-12-14' of git://git.kernel.dk/linux-block: (28 commits)
  io_uring: remove 'twa_signal_ok' deadlock work-around
  kernel: remove checking for TIF_NOTIFY_SIGNAL
  signal: kill JOBCTL_TASK_WORK
  io_uring: JOBCTL_TASK_WORK is no longer used by task_work
  task_work: remove legacy TWA_SIGNAL path
  sparc: add support for TIF_NOTIFY_SIGNAL
  riscv: add support for TIF_NOTIFY_SIGNAL
  nds32: add support for TIF_NOTIFY_SIGNAL
  ia64: add support for TIF_NOTIFY_SIGNAL
  h8300: add support for TIF_NOTIFY_SIGNAL
  c6x: add support for TIF_NOTIFY_SIGNAL
  alpha: add support for TIF_NOTIFY_SIGNAL
  xtensa: add support for TIF_NOTIFY_SIGNAL
  arm: add support for TIF_NOTIFY_SIGNAL
  microblaze: add support for TIF_NOTIFY_SIGNAL
  hexagon: add support for TIF_NOTIFY_SIGNAL
  csky: add support for TIF_NOTIFY_SIGNAL
  openrisc: add support for TIF_NOTIFY_SIGNAL
  sh: add support for TIF_NOTIFY_SIGNAL
  um: add support for TIF_NOTIFY_SIGNAL
  ...
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Pull TIF_NOTIFY_SIGNAL updates from Jens Axboe:
 "This sits on top of of the core entry/exit and x86 entry branch from
  the tip tree, which contains the generic and x86 parts of this work.

  Here we convert the rest of the archs to support TIF_NOTIFY_SIGNAL.

  With that done, we can get rid of JOBCTL_TASK_WORK from task_work and
  signal.c, and also remove a deadlock work-around in io_uring around
  knowing that signal based task_work waking is invoked with the sighand
  wait queue head lock.

  The motivation for this work is to decouple signal notify based
  task_work, of which io_uring is a heavy user of, from sighand. The
  sighand lock becomes a huge contention point, particularly for
  threaded workloads where it's shared between threads. Even outside of
  threaded applications it's slower than it needs to be.

  Roman Gershman &lt;romger@amazon.com&gt; reported that his networked
  workload dropped from 1.6M QPS at 80% CPU to 1.0M QPS at 100% CPU
  after io_uring was changed to use TIF_NOTIFY_SIGNAL. The time was all
  spent hammering on the sighand lock, showing 57% of the CPU time there
  [1].

  There are further cleanups possible on top of this. One example is
  TIF_PATCH_PENDING, where a patch already exists to use
  TIF_NOTIFY_SIGNAL instead. Hopefully this will also lead to more
  consolidation, but the work stands on its own as well"

[1] https://github.com/axboe/liburing/issues/215

* tag 'tif-task_work.arch-2020-12-14' of git://git.kernel.dk/linux-block: (28 commits)
  io_uring: remove 'twa_signal_ok' deadlock work-around
  kernel: remove checking for TIF_NOTIFY_SIGNAL
  signal: kill JOBCTL_TASK_WORK
  io_uring: JOBCTL_TASK_WORK is no longer used by task_work
  task_work: remove legacy TWA_SIGNAL path
  sparc: add support for TIF_NOTIFY_SIGNAL
  riscv: add support for TIF_NOTIFY_SIGNAL
  nds32: add support for TIF_NOTIFY_SIGNAL
  ia64: add support for TIF_NOTIFY_SIGNAL
  h8300: add support for TIF_NOTIFY_SIGNAL
  c6x: add support for TIF_NOTIFY_SIGNAL
  alpha: add support for TIF_NOTIFY_SIGNAL
  xtensa: add support for TIF_NOTIFY_SIGNAL
  arm: add support for TIF_NOTIFY_SIGNAL
  microblaze: add support for TIF_NOTIFY_SIGNAL
  hexagon: add support for TIF_NOTIFY_SIGNAL
  csky: add support for TIF_NOTIFY_SIGNAL
  openrisc: add support for TIF_NOTIFY_SIGNAL
  sh: add support for TIF_NOTIFY_SIGNAL
  um: add support for TIF_NOTIFY_SIGNAL
  ...
</pre>
</div>
</content>
</entry>
</feed>
