Commit Graph

6687 Commits

Author SHA1 Message Date
bors
8d53314233 Auto merge of #2318 - RalfJung:nodoc, r=RalfJung
there are no doctests in ui_test

It tries to run doctests which leads to an error when testing against a locally built rustc (that has no rustdoc).
2022-07-03 15:03:19 +00:00
Ralf Jung
2fc9278818 no doctests in ui_test 2022-07-03 10:45:39 -04:00
bors
cfad9d12f3 Auto merge of #1935 - saethlin:optimize-sb, r=RalfJung
Optimizing Stacked Borrows (part 1?): Cache locations of Tags in a Borrow Stack

Before this PR, a profile of Miri under almost any workload points quite squarely at these regions of code as being incredibly hot (each being ~40% of cycles):

dadcbebfbd/src/stacked_borrows.rs (L259-L269)

dadcbebfbd/src/stacked_borrows.rs (L362-L369)

This code is one of at least three reasons that stacked borrows analysis is super-linear: These are both linear in the number of borrows in the stack and they are positioned along the most commonly-taken paths.

I'm addressing the first loop (which is in `Stack::find_granting`) by adding a very very simple sort of LRU cache implemented on a `VecDeque`, which maps recently-looked-up tags to their position in the stack. For `Untagged` access we fall back to the same sort of linear search. But as far as I can tell there are never enough `Untagged` items to be significant.

I'm addressing the second loop by keeping track of the region of stack where there could be items granting `Permission::Unique`. This optimization is incredibly effective because `Read` access tends to dominate and many trips through this code path now skip the loop entirely.

These optimizations result in pretty enormous improvements:
Without raw pointer tagging, `mse` 34.5s -> 2.4s, `serde1` 5.6s -> 3.6s
With raw pointer tagging, `mse` 35.3s -> 2.4s, `serde1` 5.7s -> 3.6s

And there is hardly any impact on memory usage:
Memory usage on `mse` 844 MB -> 848 MB, `serde1` 184 MB -> 184 MB (jitter on these is a few MB).
2022-07-03 14:39:22 +00:00
bors
5382f463a6 Auto merge of #2316 - domenicquirl:readme-track-tag, r=RalfJung
Clarify the effect of the `-Zmiri-track-pointer-tag` flag in the README

Edit the README to explicitly say that the `-Zmiri-track-pointer-tag` flag also tracks the creation of tags, not just when they are popped/invalidated.

Related to https://github.com/rust-lang/miri/pull/2308 / https://github.com/Manishearth/triomphe/issues/38.
2022-07-03 14:07:27 +00:00
DQ
1b9e19a39b
clarify that a (tracked) tag may never be popped
Co-authored-by: Ralf Jung <post@ralfj.de>
2022-07-03 15:47:39 +02:00
bors
e86cd1a246 Auto merge of #2317 - RalfJung:this-is-not-a-std-test-suite, r=RalfJung
move arc_drop test to miri-test-libstd

That's where we have a bunch of slow and unlikely-to-regress tests already. Miri's test suite should primarily test Miri; we sometimes add libstd tests when they are cheap and/or exercise interesting code paths in Miri. https://github.com/rust-lang/miri/pull/2248 already ensures that we allow code like `Arc` with more direct tests.
2022-07-03 12:36:04 +00:00
Ralf Jung
e71f8b0630 fix ./miri run with MIRI_TEST_TARGET 2022-07-03 08:35:44 -04:00
Ralf Jung
c6f44c1639 another one for the trophy case 2022-07-03 08:18:03 -04:00
Ralf Jung
5f7bc9739b move arc_drop test to miri-test-libstd
and make weak memory consistency test a bit faster again
2022-07-03 08:13:58 -04:00
DQ
672d721544
Clarify -Zmiri-track-pointer-tag
to explicitly say it also tracks tag creation
2022-07-03 11:32:36 +02:00
Ben Kimock
b004a03bdb Typo 2022-07-02 20:45:27 -04:00
Ben Kimock
f3b479d556 Explain cache behavior a bit better, clean up diff 2022-07-02 19:44:55 -04:00
bors
c674f2604b Auto merge of #2312 - RalfJung:misc, r=RalfJung
put call to stacked borrows end_call in a more sensible place

This looks like a refactoring accident.
2022-07-02 22:38:36 +00:00
Ralf Jung
dcdf4fb6f6 readme tweaks 2022-07-02 18:38:07 -04:00
Ralf Jung
d9c441c5ab put call to stacked borrows end_call in a more sensible place 2022-07-02 18:38:07 -04:00
bors
c9925ff7f4 Auto merge of #2311 - RalfJung:many-seeds, r=RalfJung
./miri many-seeds: also print the seed before we try it

When using `cargo miri`, we otherwise have no way of even seeing which seed it is currently on.
2022-07-02 20:08:28 +00:00
Ralf Jung
733b141789 ./miri many-seeds: also print the seed before we try it 2022-07-02 16:06:23 -04:00
bors
f76ebd6feb Auto merge of #2309 - RalfJung:read_link, r=RalfJung
test fs::read_link surface function

The shim got implemented a while ago but we never ensured that that is actually what std uses.
2022-07-02 17:32:57 +00:00
Ralf Jung
b4520e4591 test fs::read_link surface function 2022-07-02 13:32:04 -04:00
Ben Kimock
7cdbf98f57 Explain the behavior of the cache upon clear 2022-07-02 13:22:22 -04:00
bors
c9d5c3f88c Auto merge of #2308 - RalfJung:ptr-tacking, r=RalfJung
pointer tag tracking: on creation, log the offsets it is created for

Hopefully this makes things like https://github.com/Manishearth/triomphe/issues/38 easier to diagnose.
2022-07-02 15:34:02 +00:00
Ralf Jung
98254f67af pointer tag tracking: on creation, log the offsets it is created for 2022-07-02 11:33:29 -04:00
bors
428245072e Auto merge of #2306 - RalfJung:unix, r=RalfJung
make some things available for all Unixes
2022-07-02 13:45:27 +00:00
Ralf Jung
5ba2c1e6be posix_fadvise is not Linux-specific 2022-07-02 09:45:00 -04:00
Ralf Jung
ae51998191 male libc.rs at least build on FreeBSD 2022-07-02 09:45:00 -04:00
Ralf Jung
98d20490f4 move some POSIX file shims from linux to unix module 2022-07-02 09:45:00 -04:00
bors
c17e68d9be Auto merge of #2300 - InfRandomness:freebsd-environ-extern-impl, r=RalfJung
Extend `environ` linux extern implementation to freebsd

This fixes the `env` test on freebsd, and enables the CI test

Signed-off-by: InfRandomness <infrandomness@gmail.com>
2022-07-02 12:57:14 +00:00
bors
24c5eaf8bc Auto merge of #2305 - RalfJung:rustup, r=RalfJung
rustup
2022-07-02 12:38:40 +00:00
Ralf Jung
c0cbb662d2 rustup 2022-07-02 08:38:07 -04:00
InfRandomness
a26be7ea5f
Enable env test on freebsd
Signed-off-by: InfRandomness <infrandomness@gmail.com>
2022-07-02 12:42:14 +02:00
InfRandomness
5ec25359ff
Add environ extern implementation to freebsd
Signed-off-by: InfRandomness <infrandomness@gmail.com>
2022-07-02 12:42:14 +02:00
bors
602a3dcefd Auto merge of #2304 - LegNeato:shim-fstat64-mac, r=oli-obk
Support (stat/fstat/lstat)64 on macos

"In order to accommodate advanced capabilities of newer file systems,
     the struct stat, struct statfs, and struct dirent data structures
     were updated in Mac OSX 10.5."

"TRANSITIONAL DESCRIPTION (NOW DEPRECATED)
     The fstat64, lstat64 and stat64 routines are equivalent to their
     corresponding non-64-suffixed routine, when 64-bit inodes are in
     effect.  They were added before there was support for the symbol
     variants, and so are now deprecated.  Instead of using these, set
     the _DARWIN_USE_64_BIT_INODE macro before including header files to
     force 64-bit inode support. The stat64 structure used by these deprecated routines is the same
     as the stat structure when 64-bit inodes are in effect (see above)."

"HISTORY
     An lstat() function call appeared in 4.2BSD.  The stat64(),
     fstat64(), and lstat64() system calls first appeared in Mac OS X
     10.5 (Leopard) and are now deprecated in favor of the corresponding
     symbol variants.  The fstatat() system call appeared in OS X 10.10"
2022-07-02 07:31:16 +00:00
Christian Legnitto
a1473ea8f5 Support (stat/fstat/lstat)64 on macos
"In order to accommodate advanced capabilities of newer file systems,
     the struct stat, struct statfs, and struct dirent data structures
     were updated in Mac OSX 10.5."

"TRANSITIONAL DESCRIPTION (NOW DEPRECATED)
     The fstat64, lstat64 and stat64 routines are equivalent to their
     corresponding non-64-suffixed routine, when 64-bit inodes are in
     effect.  They were added before there was support for the symbol
     variants, and so are now deprecated.  Instead of using these, set
     the _DARWIN_USE_64_BIT_INODE macro before including header files to
     force 64-bit inode support.

     The stat64 structure used by these deprecated routines is the same
     as the stat structure when 64-bit inodes are in effect (see above)."

"HISTORY
     An lstat() function call appeared in 4.2BSD.  The stat64(),
     fstat64(), and lstat64() system calls first appeared in Mac OS X
     10.5 (Leopard) and are now deprecated in favor of the corresponding
     symbol variants.  The fstatat() system call appeared in OS X 10.10"
2022-07-01 20:10:17 -06:00
Ben Kimock
17acc3f71c Document implementation a bit, add some fast paths 2022-07-01 20:15:34 -04:00
bors
d118a949f7 Auto merge of #2302 - RalfJung:datarace, r=RalfJung
avoid copying thread manager state in data race detector

When doing https://github.com/rust-lang/miri/pull/2047 I did not realize that there is some redundant state here that we can now remove from the data race detector.

Also this removes the vector clocks from the data race errors since those don't really help diagnose the problem.
2022-07-02 00:14:06 +00:00
Ralf Jung
d09db1660b fix and slightly improve data race reports 2022-07-01 20:13:44 -04:00
Ben Kimock
3bbcafe3b5 Cache lookups into the borrow stack
This adds a very simple LRU-like cache which stores the locations of
often-used tags. While the implementation is very simple, the cache hit
rate is incredible at ~99.9% on most programs, and often the element at
position 0 in the cache has a hit rate of 90%. So the sub-optimality of
this cache basicaly vanishes into the noise in a profile.

Additionally, we keep a range which denotes where there might be an item
granting Unique permission in the stack, so that when we invalidate
Uniques we do not need to scan much of the stack, and often scan nothing
at all.
2022-07-01 17:51:16 -04:00
Ralf Jung
dfdedae840 avoid copying thread manager state in data race detector 2022-07-01 17:07:29 -04:00
Ralf Jung
f238513efa rename some data_race types for more clarity 2022-07-01 16:21:48 -04:00
bors
70b97a5d7b Auto merge of #2299 - RalfJung:bsd, r=RalfJung
test more things on FreeBSD
2022-07-01 18:14:23 +00:00
Ralf Jung
ab0005bc38 run data_race tests on bsd 2022-07-01 14:13:36 -04:00
Ralf Jung
734e8903f2 more current_dir tests for freebsd 2022-07-01 14:12:25 -04:00
bors
c1a37b87aa Auto merge of #2298 - InfRandomness:shim-error, r=RalfJung
Add `__error` to freebsd shims

Signed-off-by: InfRandomness <infrandomness@gmail.com>
2022-07-01 17:41:10 +00:00
InfRandomness
dc47649d21
Add current_dir_with_isolation to freebsd tests list
Signed-off-by: InfRandomness <infrandomness@gmail.com>
2022-07-01 18:52:32 +02:00
InfRandomness
dca2bb68b3
Add __error to freebsd shims
Signed-off-by: InfRandomness <infrandomness@gmail.com>
2022-07-01 18:42:33 +02:00
bors
c768f3c03c Auto merge of #2297 - RalfJung:rustup, r=RalfJung
rustup
2022-07-01 16:26:25 +00:00
Ralf Jung
12d04ac4c4 make clippy happy 2022-07-01 12:25:35 -04:00
bors
5815d8d81c Auto merge of #2296 - RalfJung:miri-script, r=RalfJung
./miri improvements

I have needed to run something with many different seeds often enough that I would like an easier way to do it. ;) So now we have `./miri many-seeds`.

Also I made the script less dependent on the working directory, so calling it from a different directory should work properly now even if that other directory does not have the same rustup override as the one where Miri lives.
2022-07-01 14:46:30 +00:00
Ralf Jung
af39709a9c rustup 2022-07-01 10:14:31 -04:00
Ralf Jung
9bc7938bc2 more tweaks 2022-07-01 10:01:00 -04:00