Age | Commit message (Collapse) | Author |
|
* commit '5b1efce4aeea83c7b18dfed5d3366d6b84874c23':
Add cold versions of benchmarks.
|
|
|
|
- Created macros for one and two buffer benchmarks to simplify
addition of new benchmarks.
- Created cold versions of all benchmarks that work on buffer benchmarks.
This means that the benchmark is given buffers that are not in the cache.
- Verified the cold versions are working as intended by
changing it so that the same buffer is written in each loop
instead of the strided buffer. Using this method, the cold and
normal benchmark run at the same speed.
Merge from internal master.
(cherry-picked from 991bcd464d67c3cff76477140331619ff138aafc)
Change-Id: I4b1b55510d90cc9a941fd68244971cb08cfba553
|
|
- Created macros for one and two buffer benchmarks to simplify
addition of new benchmarks.
- Created cold versions of all benchmarks that work on buffer benchmarks.
This means that the benchmark is given buffers that are not in the cache.
- Verified the cold versions are working as intended by
changing it so that the same buffer is written in each loop
instead of the strided buffer. Using this method, the cold and
normal benchmark run at the same speed.
Change-Id: Ie5d1b015dd70d3935cfa2830a82955429a2f5b3c
|
|
|
|
|
|
Change-Id: I90ca712463681c1c59533bd3d4d45509647ca3ff
Signed-off-by: Rom Lemarchand <romlem@google.com>
|
|
|
|
Change-Id: I3d2bfacbe2f0e338712babf81efcecb79e9758ad
|
|
Change-Id: I5d27f5f50275017134a48eed8d89e452353777f1
|
|
Change-Id: I6c29d6e90c5bc17b3070595c89fad9ca5c3f75f9
|
|
* commit 'dd9c23b6589351ed981e19c636761dc92ad2769c':
Add a static version of the test.
|
|
* commit '23062f994c2181d5de4f6f5b764c83d613465bba':
Add a static version of the test.
|
|
|
|
Merge from internal master.
(cherry-picked from 4052fe181a499f52729b7178f11ae91212823682)
Change-Id: I90d1320e8d8032d82909a2a7435fda9ee3ec1fdb
|
|
and strcat."
* commit '1f5c5c76aff9d7bb1723f1582b975b190dccc2f6':
Add tests for strcpy and strcat.
|
|
* commit 'f23b47e914b94faaa562306a37109125b7b90901':
Add tests for strcpy and strcat.
|
|
* commit 'd8f4bfc728c95bcd4df4f0ea35d43246d2d26af0':
Add tests for strcpy and strcat.
|
|
|
|
Change-Id: I222d441fb1f6946b5f2cad3c75354a8632603fb9
|
|
|
|
Also convert all of the different string length defines to a single
string length macro.
Move the string alignment loop to the top and make all alignment structures
const.
Merge from internal master.
(cherry-picked from 1c33ae7ce88efc23b627f5fdb74afc461fc8e4f5)
Change-Id: I973795cfff55975574dd1b17cc9b25d9f6932873
|
|
benchmarks."
* commit 'eb7e166fcc774250523d0a6444080c8ee60ead45':
Adding strlen, strcat, strcpy benchmarks.
|
|
* commit '1868728e5968e3106fe6ebb0e173b9f0fc9235a9':
Adding strlen, strcat, strcpy benchmarks.
|
|
* commit 'ecd2c8f2fdebcfcbbfda1727c09d221d76de1ce7':
Adding strlen, strcat, strcpy benchmarks.
|
|
|
|
"Add strlen tests. DO NOT MERGE"
* commit '29bbea5002fabf9631310bc61b425b3356cd9e82':
Add strlen tests. DO NOT MERGE
|
|
NOT MERGE"
* commit '38273ce3a20537c366dadac6047c77914c498364':
Add strlen tests. DO NOT MERGE
|
|
* commit 'ac2518d13b67d682a59bb070f9971549dcca7249':
Add strlen tests. DO NOT MERGE
|
|
Merge from internal master.
(cherry-picked from 7401bc1263d9b9d9f7605a3c5bdaaf3171c7e9c0)
Change-Id: I7139a96022540c3604dd38f7f7836c8823e540dc
|
|
|
|
Merge from internal master.
(cherry-picked from from commit a63c049f23c49b6345bf2c845a415109c5471ebe)
Change-Id: I443a866a4b146a8043dd84cf6b7c090ecd529519
|
|
|
|
|
|
Also convert all of the different string length defines to a single
string length macro.
Move the string alignment loop to the top and make all alignment structures
const.
Change-Id: Idc7121315406a8ba872522ef90024b66d67c6058
|
|
Change-Id: I31542ecdec95c2d1d3fb9b2795581ad4d5e13c5c
|
|
|
|
Change-Id: I05079dd20765276e9f8f492fc1a0531742d5f1f2
|
|
|
|
|
|
|
|
Because USER builds do not have root access, the new -u
option modifies the script to create a 512 Mbyte file
in /data/local/tmp/testfile, and use that for testing.
Read tests are disabled (because the inability to drop
the caches makes them pointless), and the write tests
now go through the filesystem layer, so it is not fair
to compare these results to a non-userbuild run. However,
this test can help with before/after tests to see if
a change has improved the I/O performance of the system.
Change-Id: Id724f12e4564af7ced858de1df7c2d45b0a440b6
|
|
Change-Id: I557983ded74a355df6fb9e14a01de25628d1eefb
|
|
Display addresses as ranges instead of individual addresses when applicable
Change-Id: I5cb2f162a86c45e60874f2cde1d649b5c7cc86b5
Signed-off-by: Rom Lemarchand <romlem@google.com>
|
|
Add an option to sort pages by usage count in ksminfo.
Change-Id: I115ed7d8f8e99b8eed636e88605421c1b20b8d77
Signed-off-by: Rom Lemarchand <romlem@google.com>
|
|
Add the ksminfo utility that prints information about the KSM pages
of a process
Change-Id: Ib6f5291cc28d57ef05adc861e71860aa83b56465
|
|
mmc_trace_reduce will convert a log of mmc trace events from
the kernel into a format usable by some perf tools.
Change-Id: Ia468c803b928bce54bf913846d4ed5422f818046
|
|
* commit '13285bd6a37469c4d335f8368762b4de0b47050b':
Add vldr vstr type benchmarking.
|
|
* commit 'a8f6f3218aabdd5a691679d06034dbb9a7812854':
Add vldr vstr type benchmarking.
|
|
|