<feed xmlns='http://www.w3.org/2005/Atom'>
<title>delta/go-git.git/src/runtime/sys_linux_arm.s, branch dev.boringcrypto.go1.12</title>
<subtitle>github.com: golang/go
</subtitle>
<link rel='alternate' type='text/html' href='http://git.baserock.org/cgit/delta/go-git.git/'/>
<entry>
<title>runtime: use MADV_FREE on Linux if available</title>
<updated>2018-09-18T15:41:44+00:00</updated>
<author>
<name>Tobias Klauser</name>
<email>tklauser@distanz.ch</email>
</author>
<published>2018-09-14T07:57:06+00:00</published>
<link rel='alternate' type='text/html' href='http://git.baserock.org/cgit/delta/go-git.git/commit/?id=77f9b2728eb08456899e6500328e00ec4829dddf'/>
<id>77f9b2728eb08456899e6500328e00ec4829dddf</id>
<content type='text'>
On Linux, sysUnused currently uses madvise(MADV_DONTNEED) to signal the
kernel that a range of allocated memory contains unneeded data. After a
successful call, the range (but not the data it contained before the
call to madvise) is still available but the first access to that range
will unconditionally incur a page fault (needed to 0-fill the range).

A faster alternative is MADV_FREE, available since Linux 4.5. The
mechanism is very similar, but the page fault will only be incurred if
the kernel, between the call to madvise and the first access, decides to
reuse that memory for something else.

In sysUnused, test whether MADV_FREE is supported and fall back to
MADV_DONTNEED in case it isn't. This requires making the return value of
the madvise syscall available to the caller, so change runtime.madvise
to return it.

Fixes #23687

Change-Id: I962c3429000dd9f4a00846461ad128b71201bb04
Reviewed-on: https://go-review.googlesource.com/135395
Run-TryBot: Tobias Klauser &lt;tobias.klauser@gmail.com&gt;
TryBot-Result: Gobot Gobot &lt;gobot@golang.org&gt;
Reviewed-by: Ian Lance Taylor &lt;iant@golang.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
On Linux, sysUnused currently uses madvise(MADV_DONTNEED) to signal the
kernel that a range of allocated memory contains unneeded data. After a
successful call, the range (but not the data it contained before the
call to madvise) is still available but the first access to that range
will unconditionally incur a page fault (needed to 0-fill the range).

A faster alternative is MADV_FREE, available since Linux 4.5. The
mechanism is very similar, but the page fault will only be incurred if
the kernel, between the call to madvise and the first access, decides to
reuse that memory for something else.

In sysUnused, test whether MADV_FREE is supported and fall back to
MADV_DONTNEED in case it isn't. This requires making the return value of
the madvise syscall available to the caller, so change runtime.madvise
to return it.

Fixes #23687

Change-Id: I962c3429000dd9f4a00846461ad128b71201bb04
Reviewed-on: https://go-review.googlesource.com/135395
Run-TryBot: Tobias Klauser &lt;tobias.klauser@gmail.com&gt;
TryBot-Result: Gobot Gobot &lt;gobot@golang.org&gt;
Reviewed-by: Ian Lance Taylor &lt;iant@golang.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>runtime: use tgkill for raise</title>
<updated>2018-09-07T22:50:28+00:00</updated>
<author>
<name>Michael Pratt</name>
<email>mpratt@google.com</email>
</author>
<published>2018-09-07T00:21:59+00:00</published>
<link rel='alternate' type='text/html' href='http://git.baserock.org/cgit/delta/go-git.git/commit/?id=43f54c8d2e3bddfc6ad7887286eb6564986cb6ad'/>
<id>43f54c8d2e3bddfc6ad7887286eb6564986cb6ad</id>
<content type='text'>
raise uses tkill to send a signal to the current thread. For this use,
tgkill is functionally equivalent to tkill expect that it also takes the
pid as the first argument.

Using tgkill makes it simpler to run a Go program in a strict sandbox.
With kill and tgkill, the sandbox policy (e.g., seccomp) can prevent the
program from sending signals to other processes by checking that the
first argument == getpid().

With tkill, the policy must whitelist all tids in the process, which is
effectively impossible given Go's dynamic thread creation.

Fixes #27548

Change-Id: I8ed282ef1f7215b02ef46de144493e36454029ea
Reviewed-on: https://go-review.googlesource.com/133975
Run-TryBot: Michael Pratt &lt;mpratt@google.com&gt;
TryBot-Result: Gobot Gobot &lt;gobot@golang.org&gt;
Reviewed-by: Ian Lance Taylor &lt;iant@golang.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
raise uses tkill to send a signal to the current thread. For this use,
tgkill is functionally equivalent to tkill expect that it also takes the
pid as the first argument.

Using tgkill makes it simpler to run a Go program in a strict sandbox.
With kill and tgkill, the sandbox policy (e.g., seccomp) can prevent the
program from sending signals to other processes by checking that the
first argument == getpid().

With tkill, the policy must whitelist all tids in the process, which is
effectively impossible given Go's dynamic thread creation.

Fixes #27548

Change-Id: I8ed282ef1f7215b02ef46de144493e36454029ea
Reviewed-on: https://go-review.googlesource.com/133975
Run-TryBot: Michael Pratt &lt;mpratt@google.com&gt;
TryBot-Result: Gobot Gobot &lt;gobot@golang.org&gt;
Reviewed-by: Ian Lance Taylor &lt;iant@golang.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>runtime: use native CAS and memory barrier on ARMv7</title>
<updated>2018-05-03T21:35:39+00:00</updated>
<author>
<name>Cherry Zhang</name>
<email>cherryyz@google.com</email>
</author>
<published>2018-02-14T19:51:39+00:00</published>
<link rel='alternate' type='text/html' href='http://git.baserock.org/cgit/delta/go-git.git/commit/?id=150b728675c64addd24d79ad3bb68fec4c137940'/>
<id>150b728675c64addd24d79ad3bb68fec4c137940</id>
<content type='text'>
This gets us around the kernel helpers on ARMv7.

It is slightly faster than using the kernel helper.

name           old time/op  new time/op  delta
AtomicLoad-4   72.5ns ± 0%  69.5ns ± 0%  -4.08%  (p=0.000 n=9+9)
AtomicStore-4  57.6ns ± 1%  54.4ns ± 0%  -5.58%  (p=0.000 n=10+9)
[Geo mean]     64.6ns       61.5ns       -4.83%

If performance is really critical, we can even do compiler intrinsics
on GOARM=7.

Fixes #23792.

Change-Id: I36497d880890b26bdf01e048b542bd5fd7b17d23
Reviewed-on: https://go-review.googlesource.com/94076
Run-TryBot: Cherry Zhang &lt;cherryyz@google.com&gt;
TryBot-Result: Gobot Gobot &lt;gobot@golang.org&gt;
Reviewed-by: Austin Clements &lt;austin@google.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
This gets us around the kernel helpers on ARMv7.

It is slightly faster than using the kernel helper.

name           old time/op  new time/op  delta
AtomicLoad-4   72.5ns ± 0%  69.5ns ± 0%  -4.08%  (p=0.000 n=9+9)
AtomicStore-4  57.6ns ± 1%  54.4ns ± 0%  -5.58%  (p=0.000 n=10+9)
[Geo mean]     64.6ns       61.5ns       -4.83%

If performance is really critical, we can even do compiler intrinsics
on GOARM=7.

Fixes #23792.

Change-Id: I36497d880890b26bdf01e048b542bd5fd7b17d23
Reviewed-on: https://go-review.googlesource.com/94076
Run-TryBot: Cherry Zhang &lt;cherryyz@google.com&gt;
TryBot-Result: Gobot Gobot &lt;gobot@golang.org&gt;
Reviewed-by: Austin Clements &lt;austin@google.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>runtime: change GNU/Linux usleep to use nanosleep</title>
<updated>2018-04-24T05:01:13+00:00</updated>
<author>
<name>Ian Lance Taylor</name>
<email>iant@golang.org</email>
</author>
<published>2018-04-20T22:30:52+00:00</published>
<link rel='alternate' type='text/html' href='http://git.baserock.org/cgit/delta/go-git.git/commit/?id=665b9b3476ad0a6dc4e578e42e6c63012e23aaa0'/>
<id>665b9b3476ad0a6dc4e578e42e6c63012e23aaa0</id>
<content type='text'>
Ever since we added sleep to the runtime back in 2008, we've
implemented it on GNU/Linux with the select (or pselect or pselect6)
system call. But the Linux kernel has a nanosleep system call,
which should be a tiny bit more efficient since it doesn't have to
check to see whether there are any file descriptors. So use it.

Change-Id: Icc3430baca46b082a4d33f97c6c47e25fa91cb9a
Reviewed-on: https://go-review.googlesource.com/108538
Run-TryBot: Ian Lance Taylor &lt;iant@golang.org&gt;
TryBot-Result: Gobot Gobot &lt;gobot@golang.org&gt;
Reviewed-by: Brad Fitzpatrick &lt;bradfitz@golang.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Ever since we added sleep to the runtime back in 2008, we've
implemented it on GNU/Linux with the select (or pselect or pselect6)
system call. But the Linux kernel has a nanosleep system call,
which should be a tiny bit more efficient since it doesn't have to
check to see whether there are any file descriptors. So use it.

Change-Id: Icc3430baca46b082a4d33f97c6c47e25fa91cb9a
Reviewed-on: https://go-review.googlesource.com/108538
Run-TryBot: Ian Lance Taylor &lt;iant@golang.org&gt;
TryBot-Result: Gobot Gobot &lt;gobot@golang.org&gt;
Reviewed-by: Brad Fitzpatrick &lt;bradfitz@golang.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>runtime: get traceback from VDSO code</title>
<updated>2018-03-07T23:35:25+00:00</updated>
<author>
<name>Ian Lance Taylor</name>
<email>iant@golang.org</email>
</author>
<published>2018-02-26T22:03:47+00:00</published>
<link rel='alternate' type='text/html' href='http://git.baserock.org/cgit/delta/go-git.git/commit/?id=419c06455a91c54a0552e1eb1565c397dd6fa763'/>
<id>419c06455a91c54a0552e1eb1565c397dd6fa763</id>
<content type='text'>
Currently if a profiling signal arrives while executing within a VDSO
the profiler will report _ExternalCode, which is needlessly confusing
for a pure Go program. Change the VDSO calling code to record the
caller's PC/SP, so that we can do a traceback from that point. If that
fails for some reason, report _VDSO rather than _ExternalCode, which
should at least point in the right direction.

This adds some instructions to the code that calls the VDSO, but the
slowdown is reasonably negligible:

name                                  old time/op  new time/op  delta
ClockVDSOAndFallbackPaths/vDSO-8      40.5ns ± 2%  41.3ns ± 1%  +1.85%  (p=0.002 n=10+10)
ClockVDSOAndFallbackPaths/Fallback-8  41.9ns ± 1%  43.5ns ± 1%  +3.84%  (p=0.000 n=9+9)
TimeNow-8                             41.5ns ± 3%  41.5ns ± 2%    ~     (p=0.723 n=10+10)

Fixes #24142

Change-Id: Iacd935db3c4c782150b3809aaa675a71799b1c9c
Reviewed-on: https://go-review.googlesource.com/97315
Run-TryBot: Ian Lance Taylor &lt;iant@golang.org&gt;
TryBot-Result: Gobot Gobot &lt;gobot@golang.org&gt;
Reviewed-by: Austin Clements &lt;austin@google.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Currently if a profiling signal arrives while executing within a VDSO
the profiler will report _ExternalCode, which is needlessly confusing
for a pure Go program. Change the VDSO calling code to record the
caller's PC/SP, so that we can do a traceback from that point. If that
fails for some reason, report _VDSO rather than _ExternalCode, which
should at least point in the right direction.

This adds some instructions to the code that calls the VDSO, but the
slowdown is reasonably negligible:

name                                  old time/op  new time/op  delta
ClockVDSOAndFallbackPaths/vDSO-8      40.5ns ± 2%  41.3ns ± 1%  +1.85%  (p=0.002 n=10+10)
ClockVDSOAndFallbackPaths/Fallback-8  41.9ns ± 1%  43.5ns ± 1%  +3.84%  (p=0.000 n=9+9)
TimeNow-8                             41.5ns ± 3%  41.5ns ± 2%    ~     (p=0.723 n=10+10)

Fixes #24142

Change-Id: Iacd935db3c4c782150b3809aaa675a71799b1c9c
Reviewed-on: https://go-review.googlesource.com/97315
Run-TryBot: Ian Lance Taylor &lt;iant@golang.org&gt;
TryBot-Result: Gobot Gobot &lt;gobot@golang.org&gt;
Reviewed-by: Austin Clements &lt;austin@google.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>runtime: fix stack switch check in walltime/nanotime on linux/arm</title>
<updated>2018-03-06T14:24:19+00:00</updated>
<author>
<name>Tobias Klauser</name>
<email>tklauser@distanz.ch</email>
</author>
<published>2018-03-06T13:30:26+00:00</published>
<link rel='alternate' type='text/html' href='http://git.baserock.org/cgit/delta/go-git.git/commit/?id=9745397e1d5e84258887995fc3af6d9d3a5b0409'/>
<id>9745397e1d5e84258887995fc3af6d9d3a5b0409</id>
<content type='text'>
CL 98095 got the check wrong. We should be testing
'getg() == getg().m.curg', not 'getg().m == getg().m.curg'.

Change-Id: I32f6238b00409b67afa8efe732513d542aec5bc7
Reviewed-on: https://go-review.googlesource.com/98855
Run-TryBot: Tobias Klauser &lt;tobias.klauser@gmail.com&gt;
TryBot-Result: Gobot Gobot &lt;gobot@golang.org&gt;
Reviewed-by: Ian Lance Taylor &lt;iant@golang.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
CL 98095 got the check wrong. We should be testing
'getg() == getg().m.curg', not 'getg().m == getg().m.curg'.

Change-Id: I32f6238b00409b67afa8efe732513d542aec5bc7
Reviewed-on: https://go-review.googlesource.com/98855
Run-TryBot: Tobias Klauser &lt;tobias.klauser@gmail.com&gt;
TryBot-Result: Gobot Gobot &lt;gobot@golang.org&gt;
Reviewed-by: Ian Lance Taylor &lt;iant@golang.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>runtime: rename vdso symbols to use camel case</title>
<updated>2018-03-05T19:12:32+00:00</updated>
<author>
<name>Ian Lance Taylor</name>
<email>iant@golang.org</email>
</author>
<published>2018-03-05T06:10:04+00:00</published>
<link rel='alternate' type='text/html' href='http://git.baserock.org/cgit/delta/go-git.git/commit/?id=7178267b59b238a40fcf839a31264b2e65fa6f70'/>
<id>7178267b59b238a40fcf839a31264b2e65fa6f70</id>
<content type='text'>
This was originally C code using names with underscores, which were
retained when the code was rewritten into Go. Change the code to use
Go-like camel case names.

The names that come from the ELF ABI are left unchanged.

Change-Id: I181bc5dd81284c07bc67b7df4635f4734b41d646
Reviewed-on: https://go-review.googlesource.com/98520
Run-TryBot: Ian Lance Taylor &lt;iant@golang.org&gt;
TryBot-Result: Gobot Gobot &lt;gobot@golang.org&gt;
Reviewed-by: Tobias Klauser &lt;tobias.klauser@gmail.com&gt;
Reviewed-by: Brad Fitzpatrick &lt;bradfitz@golang.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
This was originally C code using names with underscores, which were
retained when the code was rewritten into Go. Change the code to use
Go-like camel case names.

The names that come from the ELF ABI are left unchanged.

Change-Id: I181bc5dd81284c07bc67b7df4635f4734b41d646
Reviewed-on: https://go-review.googlesource.com/98520
Run-TryBot: Ian Lance Taylor &lt;iant@golang.org&gt;
TryBot-Result: Gobot Gobot &lt;gobot@golang.org&gt;
Reviewed-by: Tobias Klauser &lt;tobias.klauser@gmail.com&gt;
Reviewed-by: Brad Fitzpatrick &lt;bradfitz@golang.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>runtime: remove unused SYS_* definitions on Linux</title>
<updated>2018-03-05T18:32:08+00:00</updated>
<author>
<name>Tobias Klauser</name>
<email>tklauser@distanz.ch</email>
</author>
<published>2018-03-05T09:56:11+00:00</published>
<link rel='alternate' type='text/html' href='http://git.baserock.org/cgit/delta/go-git.git/commit/?id=5f80e709125a1589ae52c9a33ff188e6c5224227'/>
<id>5f80e709125a1589ae52c9a33ff188e6c5224227</id>
<content type='text'>
Also fix the indentation of the SYS_* definitions in sys_linux_mipsx.s
and order them numerically.

Change-Id: I0c454301c329a163e7db09dcb25d4e825149858c
Reviewed-on: https://go-review.googlesource.com/98448
Run-TryBot: Tobias Klauser &lt;tobias.klauser@gmail.com&gt;
TryBot-Result: Gobot Gobot &lt;gobot@golang.org&gt;
Reviewed-by: Brad Fitzpatrick &lt;bradfitz@golang.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Also fix the indentation of the SYS_* definitions in sys_linux_mipsx.s
and order them numerically.

Change-Id: I0c454301c329a163e7db09dcb25d4e825149858c
Reviewed-on: https://go-review.googlesource.com/98448
Run-TryBot: Tobias Klauser &lt;tobias.klauser@gmail.com&gt;
TryBot-Result: Gobot Gobot &lt;gobot@golang.org&gt;
Reviewed-by: Brad Fitzpatrick &lt;bradfitz@golang.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>runtime: use vDSO for clock_gettime on linux/arm</title>
<updated>2018-03-03T12:12:58+00:00</updated>
<author>
<name>Tobias Klauser</name>
<email>tklauser@distanz.ch</email>
</author>
<published>2018-03-02T10:27:15+00:00</published>
<link rel='alternate' type='text/html' href='http://git.baserock.org/cgit/delta/go-git.git/commit/?id=51b027116c2c90e7cb938362b0134ff710fea54e'/>
<id>51b027116c2c90e7cb938362b0134ff710fea54e</id>
<content type='text'>
Use the __vdso_clock_gettime fast path via the vDSO on linux/arm to
speed up nanotime and walltime. This results in the following
performance improvement for time.Now on a RaspberryPi 3 (running
32bit Raspbian, i.e. GOOS=linux/GOARCH=arm):

name     old time/op  new time/op  delta
TimeNow  0.99µs ± 0%  0.39µs ± 1%  -60.74%  (p=0.000 n=12+20)

Change-Id: I3598278a6c88d7f6a6ce66c56b9d25f9dd2f4c9a
Reviewed-on: https://go-review.googlesource.com/98095
Reviewed-by: Ian Lance Taylor &lt;iant@golang.org&gt;
Run-TryBot: Ian Lance Taylor &lt;iant@golang.org&gt;
TryBot-Result: Gobot Gobot &lt;gobot@golang.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Use the __vdso_clock_gettime fast path via the vDSO on linux/arm to
speed up nanotime and walltime. This results in the following
performance improvement for time.Now on a RaspberryPi 3 (running
32bit Raspbian, i.e. GOOS=linux/GOARCH=arm):

name     old time/op  new time/op  delta
TimeNow  0.99µs ± 0%  0.39µs ± 1%  -60.74%  (p=0.000 n=12+20)

Change-Id: I3598278a6c88d7f6a6ce66c56b9d25f9dd2f4c9a
Reviewed-on: https://go-review.googlesource.com/98095
Reviewed-by: Ian Lance Taylor &lt;iant@golang.org&gt;
Run-TryBot: Ian Lance Taylor &lt;iant@golang.org&gt;
TryBot-Result: Gobot Gobot &lt;gobot@golang.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>runtime: remove unused getrlimit function</title>
<updated>2018-02-17T18:35:41+00:00</updated>
<author>
<name>Tobias Klauser</name>
<email>tklauser@distanz.ch</email>
</author>
<published>2018-02-16T10:26:14+00:00</published>
<link rel='alternate' type='text/html' href='http://git.baserock.org/cgit/delta/go-git.git/commit/?id=1b1c8b34d129eefcdbad234914df999581e62b2f'/>
<id>1b1c8b34d129eefcdbad234914df999581e62b2f</id>
<content type='text'>
Follow CL 93655 which removed the (commented-out) usage of this
function.

Also remove unused constant _RLIMIT_AS and type rlimit.

Change-Id: Ifb6e6b2104f4c2555269f8ced72bfcae24f5d5e9
Reviewed-on: https://go-review.googlesource.com/94775
Run-TryBot: Tobias Klauser &lt;tobias.klauser@gmail.com&gt;
TryBot-Result: Gobot Gobot &lt;gobot@golang.org&gt;
Reviewed-by: Austin Clements &lt;austin@google.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Follow CL 93655 which removed the (commented-out) usage of this
function.

Also remove unused constant _RLIMIT_AS and type rlimit.

Change-Id: Ifb6e6b2104f4c2555269f8ced72bfcae24f5d5e9
Reviewed-on: https://go-review.googlesource.com/94775
Run-TryBot: Tobias Klauser &lt;tobias.klauser@gmail.com&gt;
TryBot-Result: Gobot Gobot &lt;gobot@golang.org&gt;
Reviewed-by: Austin Clements &lt;austin@google.com&gt;
</pre>
</div>
</content>
</entry>
</feed>
