[wip] more perf data collection for jemalloc + rustc #55217

alexcrichton · 2018-10-20T03:03:55Z

Filling in details on #36963 ...

rust-highfive · 2018-10-20T03:03:58Z

r? @cramertj

(rust_highfive has picked a reviewer for you, use r? to override)

alexcrichton · 2018-10-20T03:04:01Z

@bors: try

alexcrichton · 2018-10-20T03:04:09Z

r? @alexcrichton

bors · 2018-10-20T03:04:13Z

⌛ Trying commit 70437bd with merge 6073f1b...

[wip] more perf data collection for jemalloc + rustc Filling in details on #36963 ...

rust-highfive · 2018-10-20T04:46:48Z

The job mingw-check of your PR failed on Travis (raw log). Through arcane magic we have determined that the following fragments from the build log may contain information about the problem.

Click to expand the log.

[00:07:23] configure: build.locked-deps    := True
[00:07:23] configure: llvm.ccache          := sccache
[00:07:23] configure: build.openssl-static := True
[00:07:23] configure: dist.missing-tools   := True
[00:07:23] configure: build.configure-args := ['--enable-sccache', '--disable-manage-submodu ...
[00:07:23] configure: writing `config.toml` in current directory
[00:07:23] configure: 
[00:07:23] configure: run `python /checkout/x.py --help`
[00:07:23] configure: 
---
[00:09:26] --- stdout
[00:09:26] OPT_LEVEL = Some("2")
[00:09:26] TARGET = Some("i686-pc-windows-gnu")
[00:09:26] HOST = Some("x86_64-unknown-linux-gnu")
[00:09:26] CC_i686-pc-windows-gnu = Some("sccache i686-w64-mingw32-gcc")
[00:09:26] CFLAGS_i686-pc-windows-gnu = Some("-ffunction-sections -fdata-sections -m32 -fno-omit-frame-pointer")
[00:09:26] DEBUG = Some("false")
[00:09:26] running: "sh" "/cargo/registry/src/jackfan.us.kg-1ecc6299db9ec823/jemalloc-sys-0.1.4/jemalloc/configure" "--with-jemalloc-prefix=_rjem_" "--host=i686-pc-w64-mingw32" "--build=x86_64-unknown-linux-gnu" "--prefix=/checkout/obj/build/x86_64-unknown-linux-gnu/stage0-rustc/i686-pc-windows-gnu/release/build/jemalloc-sys-dbcfc730584ebc05/out"
[00:09:26] checking for xsltproc... false
[00:09:26] checking for i686-pc-w64-mingw32-gcc... i686-w64-mingw32-gcc
[00:09:26] checking for C compiler default output file name... a.exe
[00:09:26] checking for suffix of executables... .exe
[00:09:26] checking whether we are cross compiling... yes
[00:09:26] checking for suffix of object files... o
[00:09:26] checking for suffix of object files... o
[00:09:26] checking whether we are using the GNU C compiler... yes
[00:09:26] checking whether i686-w64-mingw32-gcc accepts -g... yes
[00:09:26] checking for i686-w64-mingw32-gcc option to accept ISO C89... none needed
[00:09:26] checking whether compiler is cray... no
[00:09:26] checking whether compiler supports -std=gnu11... yes
[00:09:26] checking whether compiler supports -Wall... yes
[00:09:26] checking whether compiler supports -Werror=declaration-after-statement... yes
[00:09:26] checking whether compiler supports -Wshorten-64-to-32... no
[00:09:26] checking whether compiler supports -Wsign-compare... yes
[00:09:26] checking whether compiler supports -g3... yes
[00:09:26] checking how to run the C preprocessor... i686-w64-mingw32-gcc -E
[00:09:26] checking for grep that handles long lines and -e... /bin/grep
[00:09:26] checking for egrep... /bin/grep -E
---
[00:09:26] checking for strings.h... yes
[00:09:26] checking for inttypes.h... yes
[00:09:26] checking for stdint.h... yes
[00:09:26] checking for unistd.h... yes
[00:09:26] checking whether byte ordering is bigendian... no
[00:09:26] checking size of void *... 4
[00:09:26] checking size of int... 4
[00:09:26] checking size of long... 4
[00:09:26] checking size of long long... 8
[00:09:26] checking size of intmax_t... 8
[00:09:26] checking host system type... 
[00:09:26] --- stderr
[00:09:26] --- stderr
[00:09:26] Invalid configuration `i686-pc-w64-mingw32': machine `i686-pc-w64' not recognized
[00:09:26] configure: error: /bin/bash /cargo/registry/src/jackfan.us.kg-1ecc6299db9ec823/jemalloc-sys-0.1.4/jemalloc/build-aux/config.sub i686-pc-w64-mingw32 failed
[00:09:26] thread 'main' panicked at 'command did not execute successfully: "sh" "/cargo/registry/src/jackfan.us.kg-1ecc6299db9ec823/jemalloc-sys-0.1.4/jemalloc/configure" "--with-jemalloc-prefix=_rjem_" "--host=i686-pc-w64-mingw32" "--build=x86_64-unknown-linux-gnu" "--prefix=/checkout/obj/build/x86_64-unknown-linux-gnu/stage0-rustc/i686-pc-windows-gnu/release/build/jemalloc-sys-dbcfc730584ebc05/out"
[00:09:26] expected success, got: exit code: 1', /cargo/registry/src/jackfan.us.kg-1ecc6299db9ec823/jemalloc-sys-0.1.4/build.rs:105:9
[00:09:26] 
[00:09:26] warning: build failed, waiting for other jobs to finish...
[00:09:27] error: build failed
[00:09:27] error: build failed
[00:09:27] command did not execute successfully: "/checkout/obj/build/x86_64-unknown-linux-gnu/stage0/bin/cargo" "check" "--target" "i686-pc-windows-gnu" "-j" "4" "--release" "--locked" "--color" "always" "--features" " jemalloc" "--manifest-path" "/checkout/src/rustc/Cargo.toml" "--message-format" "json"
[00:09:27] expected success, got: exit code: 101
[00:09:27] thread 'main' panicked at 'cargo must succeed', bootstrap/compile.rs:1115:9
[00:09:27] travis_fold:end:stage0-rustc

[00:09:27] travis_time:end:stage0-rustc:start=1540005358889866859,finish=1540005379684106728,duration=20794239869

---
travis_time:end:11800f50:start=1540005380369558572,finish=1540005380374163932,duration=4605360
travis_fold:end:after_failure.3
travis_fold:start:after_failure.4
travis_time:start:09df4640
$ ln -s . checkout && for CORE in obj/cores/core.*; do EXE=$(echo $CORE | sed 's|obj/cores/core\.[0-9]*\.!checkout!\(.*\)|\1|;y|!|/|'); if [ -f "$EXE" ]; then printf travis_fold":start:crashlog\n\033[31;1m%s\033[0m\n" "$CORE"; gdb --batch -q -c "$CORE" "$EXE" -iex 'set auto-load off' -iex 'dir src/' -iex 'set sysroot .' -ex bt -ex q; echo travis_fold":"end:crashlog; fi; done || true
travis_fold:end:after_failure.4
travis_fold:start:after_failure.5
travis_time:start:018050d2
travis_time:start:018050d2
$ cat ./obj/build/x86_64-unknown-linux-gnu/native/asan/build/lib/asan/clang_rt.asan-dynamic-i386.vers || true
cat: ./obj/build/x86_64-unknown-linux-gnu/native/asan/build/lib/asan/clang_rt.asan-dynamic-i386.vers: No such file or directory
travis_fold:end:after_failure.5
travis_fold:start:after_failure.6
travis_time:start:096bd2bc
$ dmesg | grep -i kill

I'm a bot! I can only do what humans tell me to, so if this was not helpful or you have suggestions for improvements, please ping or otherwise contact @TimNN. (Feature Requests)

bors · 2018-10-20T05:29:34Z

☀️ Test successful - status-travis
State: approved= try=True

bjorn3 · 2018-10-20T08:20:47Z

src/librustc_driver/lib.rs

@@ -30,6 +30,8 @@

 #![recursion_limit="256"]

+extern crate jemalloc_sys;


Should this be in libsyntax instead, so that rustc plugins don't use the system allocator?

In theory it should be ok here, but let's let the data find out!

alexcrichton · 2018-10-20T12:38:49Z

@rust-timer build 6073f1b

rust-timer · 2018-10-20T12:38:50Z

Success: Queued 6073f1b with parent 42dde96, comparison URL.

rust-timer · 2018-10-20T14:57:47Z

Finished benchmarking try commit 6073f1b

bjorn3 · 2018-10-20T15:54:23Z

Four benchmarks -10%
One +9%
Most +1-2%

alexcrichton · 2018-10-20T19:56:41Z

@bors: try

Oh I think I measured the wrong thing, that's the exact same as the previous profile

bors · 2018-10-20T19:56:53Z

⌛ Trying commit 76afdbc with merge 10c95ccfa7a7adc12f4e608621ca29f9b98eed29...

bors · 2018-10-20T22:18:25Z

☀️ Test successful - status-travis
State: approved= try=True

alexcrichton · 2018-10-21T00:05:56Z

@rust-timer build 10c95ccfa7a7adc12f4e608621ca29f9b98eed29

rust-timer · 2018-10-21T00:05:57Z

Success: Queued 10c95ccfa7a7adc12f4e608621ca29f9b98eed29 with parent 22cc2ae, comparison URL.

rust-timer · 2018-10-21T01:50:41Z

Finished benchmarking try commit 10c95ccfa7a7adc12f4e608621ca29f9b98eed29

alexcrichton · 2018-10-21T01:53:19Z

Now that's what I wanted to see. Gonna close this and will post a more official version soon.

Link to alloc_system by default

635cb31

rust-highfive assigned cramertj Oct 20, 2018

rust-highfive added the S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. label Oct 20, 2018

rust-highfive assigned alexcrichton and unassigned cramertj Oct 20, 2018

bors added a commit that referenced this pull request Oct 20, 2018

Auto merge of #55217 - alexcrichton:alloc-system-and-jemalloc, r=<try>

6073f1b

[wip] more perf data collection for jemalloc + rustc Filling in details on #36963 ...

brson mentioned this pull request Oct 20, 2018

Switch the default global allocator to System, remove alloc_jemalloc, use jemallocator in rustc #36963

Closed

bjorn3 reviewed Oct 20, 2018

View reviewed changes

Add jemalloc-sys to rustc_driver

76afdbc

alexcrichton force-pushed the alloc-system-and-jemalloc branch from 70437bd to 76afdbc Compare October 20, 2018 19:56

alexcrichton closed this Oct 21, 2018

alexcrichton deleted the alloc-system-and-jemalloc branch October 21, 2018 01:53

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[wip] more perf data collection for jemalloc + rustc #55217

[wip] more perf data collection for jemalloc + rustc #55217

alexcrichton commented Oct 20, 2018

rust-highfive commented Oct 20, 2018

alexcrichton commented Oct 20, 2018

alexcrichton commented Oct 20, 2018

bors commented Oct 20, 2018

rust-highfive commented Oct 20, 2018

bors commented Oct 20, 2018

bjorn3 Oct 20, 2018

alexcrichton Oct 20, 2018

alexcrichton commented Oct 20, 2018

rust-timer commented Oct 20, 2018

rust-timer commented Oct 20, 2018

bjorn3 commented Oct 20, 2018

alexcrichton commented Oct 20, 2018

bors commented Oct 20, 2018

bors commented Oct 20, 2018

alexcrichton commented Oct 21, 2018

rust-timer commented Oct 21, 2018

rust-timer commented Oct 21, 2018

alexcrichton commented Oct 21, 2018

		@@ -30,6 +30,8 @@

		#![recursion_limit="256"]

		extern crate jemalloc_sys;

[wip] more perf data collection for jemalloc + rustc #55217

[wip] more perf data collection for jemalloc + rustc #55217

Conversation

alexcrichton commented Oct 20, 2018

rust-highfive commented Oct 20, 2018

alexcrichton commented Oct 20, 2018

alexcrichton commented Oct 20, 2018

bors commented Oct 20, 2018

rust-highfive commented Oct 20, 2018

bors commented Oct 20, 2018

bjorn3 Oct 20, 2018

Choose a reason for hiding this comment

alexcrichton Oct 20, 2018

Choose a reason for hiding this comment

alexcrichton commented Oct 20, 2018

rust-timer commented Oct 20, 2018

rust-timer commented Oct 20, 2018

bjorn3 commented Oct 20, 2018

alexcrichton commented Oct 20, 2018

bors commented Oct 20, 2018

bors commented Oct 20, 2018

alexcrichton commented Oct 21, 2018

rust-timer commented Oct 21, 2018

rust-timer commented Oct 21, 2018

alexcrichton commented Oct 21, 2018