Commit graph

29110 commits

Author SHA1 Message Date
Elliott Hughes
a1fb15bb67 strptime: support everything that strftime supports.
Our strptime was missing `%F`, `%G`, `%g`, `%P`, `%u`, `%V`, and
`%v`. Most of these are already supported upstream (and I've just pulled
their current implementation), but some aren't. We're horribly out of
sync anyway, so I'll upstream the missing pieces and then try to get us
back in sync later.

Test: new tests, but originally found by toybox trying to use %F
Change-Id: Ib1a10801a7a3b9c9189440c3b300109bde535fd9
2019-03-26 19:07:40 -07:00
Elliott Hughes
e08288465e Merge "bionic tests: use GTEST_SKIP." 2019-03-27 02:05:57 +00:00
Christopher Ferris
8bfffe9261 Merge "Increase the allocation limit again." am: e990078f05 am: a423fed2ba
am: d4e97ace4f

Change-Id: Iebb9ce00424cdaea77022ad88452099873521c67
2019-03-26 18:33:53 -07:00
Christopher Ferris
d4e97ace4f Merge "Increase the allocation limit again." am: e990078f05
am: a423fed2ba

Change-Id: I3a3706dcfc7708213ff6a3c934a4a9bc09e236ef
2019-03-26 18:21:44 -07:00
Christopher Ferris
a423fed2ba Merge "Increase the allocation limit again."
am: e990078f05

Change-Id: I1c902d4781568a1ca8750eb0cdf7e4cc8a80de1c
2019-03-26 18:17:37 -07:00
Christopher Ferris
e990078f05 Merge "Increase the allocation limit again." 2019-03-27 01:04:36 +00:00
Elliott Hughes
36fa2647af Merge "Fix running the tests on the host again." am: b0c33ecb99 am: 1f63174243
am: b462788ee1

Change-Id: I87615ed7b66bc3efa0da2ac4b697b31754a646f4
2019-03-26 17:50:24 -07:00
Dan Albert
0650d04668 Merge "Typo fix in comment. O_CLOEXEC is e, not x." am: fe88d26aca am: 82ed244032
am: 169df88962

Change-Id: I7bcac6d3c2e6d64819f181e507f2bc45dc6274e7
2019-03-26 17:36:44 -07:00
Elliott Hughes
b462788ee1 Merge "Fix running the tests on the host again." am: b0c33ecb99
am: 1f63174243

Change-Id: I0d169dc59c2db2413d20de719aafcc412f91c30e
2019-03-26 17:26:09 -07:00
Dan Albert
169df88962 Merge "Typo fix in comment. O_CLOEXEC is e, not x." am: fe88d26aca
am: 82ed244032

Change-Id: I80c22c3e1967facc3f3ac74b6f92dc8dea463152
2019-03-26 17:05:18 -07:00
Elliott Hughes
1f63174243 Merge "Fix running the tests on the host again."
am: b0c33ecb99

Change-Id: Ie7aa9c329512b174e7f8801788fcf18166f3aeb1
2019-03-26 16:54:41 -07:00
Treehugger Robot
b0c33ecb99 Merge "Fix running the tests on the host again." 2019-03-26 23:44:19 +00:00
Dan Albert
82ed244032 Merge "Typo fix in comment. O_CLOEXEC is e, not x."
am: fe88d26aca

Change-Id: I2d6798a05e3ca8d0e34898ed959b11d0507b441f
2019-03-26 16:09:07 -07:00
Christopher Ferris
ce491abe24 Increase the allocation limit again.
Still getting a few flakes, so double the allowed tries.

Bug: 128872105

Test: Test passes.
Change-Id: I0fb3d74655eaece6660dec26a2a1d01430ef8fbc
2019-03-26 15:47:07 -07:00
Treehugger Robot
fe88d26aca Merge "Typo fix in comment. O_CLOEXEC is e, not x." 2019-03-26 22:00:31 +00:00
Elliott Hughes
8ceb95c506 Fix running the tests on the host again.
Bug: N/A
Test: ran the tests on the host under glibc
Change-Id: Id05a896bdd11b231304ed1a08ef24e1a661ee20d
2019-03-26 14:47:08 -07:00
Dan Albert
ba1151c761 Typo fix in comment. O_CLOEXEC is e, not x.
Test: None
Bug: None
Change-Id: I061fe1d3cac6307d878155f1fabdba70da8e9fd5
2019-03-26 13:01:22 -07:00
Elliott Hughes
6dcbf59245 Merge "docs: mention the fact that bionic now uses ifuncs." am: 2059021322 am: d1bba14064
am: 38dcc53f95

Change-Id: I7c51fa89e148055f937ae811aad143a959086a77
2019-03-26 11:32:33 -07:00
Elliott Hughes
38dcc53f95 Merge "docs: mention the fact that bionic now uses ifuncs." am: 2059021322
am: d1bba14064

Change-Id: Id701b6df9caee25e477962b69c52adab387d50b9
2019-03-26 11:12:16 -07:00
Elliott Hughes
d1bba14064 Merge "docs: mention the fact that bionic now uses ifuncs."
am: 2059021322

Change-Id: I4e64df2e56576a5445b08338ae69e02643ac0db9
2019-03-26 10:49:32 -07:00
Treehugger Robot
2059021322 Merge "docs: mention the fact that bionic now uses ifuncs." 2019-03-26 17:32:12 +00:00
Elliott Hughes
3770d93263 docs: mention the fact that bionic now uses ifuncs.
Bug: http://b/123296740
Test: N/A
Change-Id: I2cdc575270ebf32fd8ac1de21182150f5222cf9f
2019-03-26 08:52:07 -07:00
Elliott Hughes
a307b497b9 Merge "Fix internal uses of _PATH_BSHELL." am: b1ddbfdcdc am: b4c1555dcd
am: 3914996a7c

Change-Id: Ie44d7be1a47574775e93f552fba5a77366504ccc
2019-03-26 03:01:44 -07:00
Elliott Hughes
3914996a7c Merge "Fix internal uses of _PATH_BSHELL." am: b1ddbfdcdc
am: b4c1555dcd

Change-Id: Ice856f4e0b56eae26b22c26bc1acd77a17a553a7
2019-03-26 02:53:27 -07:00
Elliott Hughes
b4c1555dcd Merge "Fix internal uses of _PATH_BSHELL."
am: b1ddbfdcdc

Change-Id: I11ecf04c647871d154baac824e6db83db9d84f19
2019-03-26 02:45:21 -07:00
Treehugger Robot
b1ddbfdcdc Merge "Fix internal uses of _PATH_BSHELL." 2019-03-26 09:32:16 +00:00
Pirama Arumuga Nainar
ebe1ae7747 Reserve realtime signal for libprofile-extras
Bug: http://b/128524141

This signal is used by libprofile-extras to trigger flush of gcov
coverage data.

Test: libprofile-extras in system/extras/toolchain-extras
Change-Id: I77dd5b1d67371a9cfaad54ad60254bfb2dd18ba9
2019-03-25 22:09:19 -07:00
Nick Kralevich
d4b46b3896 Merge "Make powerof2 macro ubsan safe" am: 1b6517b601 am: cdfe9d7d96
am: 91479bb201

Change-Id: I7f3513a4c009879a909e0eaeecc06b3990e634d7
2019-03-25 21:05:45 -07:00
Nick Kralevich
91479bb201 Merge "Make powerof2 macro ubsan safe" am: 1b6517b601
am: cdfe9d7d96

Change-Id: I8f9099d09f850faf629c512f86e258471f6edf66
2019-03-25 20:57:07 -07:00
Nick Kralevich
cdfe9d7d96 Merge "Make powerof2 macro ubsan safe"
am: 1b6517b601

Change-Id: Iec86f326ab1dd2b79cb41dd750975da7ee10474d
2019-03-25 20:52:59 -07:00
Treehugger Robot
1b6517b601 Merge "Make powerof2 macro ubsan safe" 2019-03-26 03:43:23 +00:00
Elliott Hughes
886370c240 Fix internal uses of _PATH_BSHELL.
We regressed on this recently: code under the upstream-* directories has
_PATH_BSHELL defined as a call to __bionic_get_shell_path(). In our own
code, we may as well just call it directly.

Bug: https://issuetracker.google.com/129030706
Test: ran tests
Change-Id: Ic2423f521272be95e67f94771772fe8072636ef0
2019-03-25 17:28:22 -07:00
Elliott Hughes
364af39c83 Merge "popen: stop using _fwalk." am: 49e7b68ac2 am: db64bbbb6e
am: 6f767b5f85

Change-Id: Ibb142d471967876af633566750a5c6592a002bcd
2019-03-25 17:17:15 -07:00
Elliott Hughes
6f767b5f85 Merge "popen: stop using _fwalk." am: 49e7b68ac2
am: db64bbbb6e

Change-Id: I25cba6c4ccd079bc858d190f6de31baf1aa462b7
2019-03-25 17:08:51 -07:00
Elliott Hughes
db64bbbb6e Merge "popen: stop using _fwalk."
am: 49e7b68ac2

Change-Id: I5c1cf191d1156e03b46e64dadbd464844946f9af
2019-03-25 16:58:28 -07:00
Elliott Hughes
49e7b68ac2 Merge "popen: stop using _fwalk." 2019-03-25 23:48:47 +00:00
Nick Kralevich
c50b6a2b89 Make powerof2 macro ubsan safe
Subtracting one from the smallest value expressable by the provided
variable could cause an underflow operation. In particular, this is
problematic when code similar to:

  uint64_t foo = 0;
  if (powerof2(foo)) {
    ...;
  }

is run with integer sanitization enabled. The macro would subtract one
from zero, underflowing and triggering the sanitizer.

Make the powerof2() macro ubsan safe, by explicitly handling underflows.

Note: This change DOES NOT make powerof2() accurate. We continue to
falsely return "true" for 0 and negative numbers (see attached tests).

Found while investigating Bug: 122975762
Test: see added testcase
Test: atest ziparchive-tests

Change-Id: I5408ce5c18868d797bcae8f115ddb7c4c1ced81e
2019-03-25 13:51:33 -07:00
Elliott Hughes
137b6f36ef popen: stop using _fwalk.
We don't need this now that popen always uses O_CLOEXEC, and it's unsafe
because _fwalk takes a lock. (In <= P, the equivalent code walked the
list without a lock in the child.)

Bug: http://b/129156634
Test: ran tests
Change-Id: Ic9cee7eb59cfc9397f370d1dc47ea3d3326179ca
2019-03-25 11:18:36 -07:00
Evgenii Stepanov
a27b0180a9 Merge "Disable malloc_info test under HWASan." am: e187d92062 am: c1437e8bf0
am: 18c641bbb9

Change-Id: I0902f47bb0e22abf52a341b7d138909bc86c8b7c
2019-03-22 16:03:48 -07:00
Evgenii Stepanov
18c641bbb9 Merge "Disable malloc_info test under HWASan." am: e187d92062
am: c1437e8bf0

Change-Id: I83608f3b8d9522de4821312a471c025a581a32c3
2019-03-22 15:57:44 -07:00
Evgenii Stepanov
c1437e8bf0 Merge "Disable malloc_info test under HWASan."
am: e187d92062

Change-Id: I89c0dfe27d9aa027d19854f618f8c162bcd41f0c
2019-03-22 15:48:35 -07:00
Treehugger Robot
e187d92062 Merge "Disable malloc_info test under HWASan." 2019-03-22 22:24:11 +00:00
Evgenii Stepanov
8de6b46ad8 Disable malloc_info test under HWASan.
HWASan does not implement malloc_info.

Test: bionic-unit-tests with SANITIZE_TARGET=hwaddress
Change-Id: I61d50d9110fbca1784ed39950ed6c5540ca0c9a7
2019-03-22 13:23:18 -07:00
Steven Moreland
1e577dd10d Merge "Add /product/bin to path" am: 118ab10d75 am: c626794192
am: 513359fc0e

Change-Id: I52f5308c1fb91a703908e4805f888f6068aab7e9
2019-03-22 13:19:57 -07:00
Steven Moreland
513359fc0e Merge "Add /product/bin to path" am: 118ab10d75
am: c626794192

Change-Id: Ide7b616b56fc31c9ce681ab35cb98c6faaff3f8b
2019-03-22 13:14:48 -07:00
Steven Moreland
c626794192 Merge "Add /product/bin to path"
am: 118ab10d75

Change-Id: I7530b336feb2ddadbbdf0b83271387556a6b15c9
2019-03-22 13:10:44 -07:00
Steven Moreland
118ab10d75 Merge "Add /product/bin to path" 2019-03-22 19:59:55 +00:00
Vic Yang
6aebd166d4 Merge "Use std::move() for rvalue-reference setters and ctors" am: 4848b6d7c3 am: 24c8c6e397
am: 13833635ca

Change-Id: I3f553fb80529791c0884d5f4f72c4ec614c854d2
2019-03-21 12:34:49 -07:00
Vic Yang
13833635ca Merge "Use std::move() for rvalue-reference setters and ctors" am: 4848b6d7c3
am: 24c8c6e397

Change-Id: I40ba1748f8dbde0c5b0adb9881100e88239843af
2019-03-21 12:28:49 -07:00
Christopher Ferris
8c7b5ffdc2 [automerger skipped] Merge "Update kernel headers to v5.0.3." am: d992e1e9fa am: 79569bb9eb
am: 7ef82af86d -s ours
am skip reason: change_id I6ed881b5dc6dc7d0a6cc56edcd09b42ddd185b62 with SHA1 020d04d28c is in history

Change-Id: I0c1418ebb87b6599736683bb15ded1d5145da815
2019-03-21 12:25:34 -07:00