3rd-party/linux-kernel-module-cheat
1
0
Fork 0
mirror of https://github.com/cirosantilli/linux-kernel-module-cheat.git synced 2026-01-23 02:05:57 +01:00
Code Issues Releases Wiki Activity

userland/arch/aarch64/freestanding/linux/gem5_exit.S to benchmark immediate gem5 exit

Browse Source 
ldxr_stxr does an ldxr on thread 0 to fall in the ISA case where event is certain.

Also add ldxr_str which tests yet another case where event is certain: non conditional
write.
This commit is contained in:
Ciro Santilli 六四事件 法轮功
2020-01-10 00:00:00 +00:00
parent ae758f589a
commit 3b3f073e2c
5 changed files with 102 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files

63
README.adoc
View File

@@ -9951,7 +9951,9 @@ The `--verbose` is optional, but shows clearly each GCC build command so that yo
The build outputs are automatically stored in a different directories for optimized and debug builds, which prevents `debug` files from overwriting `opt` ones. Therefore, `--gem5-build-id` is not required.
The price to pay for debuggability is high however: a Linux kernel boot was about 3x slower in QEMU and 14 times slower in gem5 debug compared to opt, see benchmarks at: xref:benchmark-linux-kernel-boot[xrefstyle=full]
The price to pay for debuggability is high however: a Linux kernel boot was about 3x slower in QEMU and 14 times slower in gem5 debug compared to opt, see benchmarks at: xref:benchmark-linux-kernel-boot[xrefstyle=full].
Similar slowdowns can be observed at: xref:benchmark-emulators-on-userland-executables[xrefstyle=full].
When in <<qemu-text-mode>>, using `--debug-vm` makes Ctrl-C not get passed to the QEMU guest anymore: it is instead captured by GDB itself, so allow breaking. So e.g. you won't be able to easily quit from a guest program like:
@@ -14069,7 +14071,9 @@ Programs under link:userland/cpp/[] are examples of https://en.wikipedia.org/wik
* random
** link:userland/cpp/random.cpp[]
* containers
** link:userland/cpp/set.cpp[]: `std::set` contains unique keys
** associative
*** <<algorithms>> contains a benchmark comparison of different c++ containers
*** link:userland/cpp/set.cpp[]: `std::set` contains unique keys
[[cpp-multithreading]]
==== C++ multithreading
@@ -18715,7 +18719,10 @@ For how userland spinlocks and mutexes are implemented see <<userland-mutex-impl
====== ARM WFE global monitor events
link:userland/arch/aarch64/inline_asm/wfe_ldxr_stxr.cpp[]
Examples:
* link:userland/arch/aarch64/inline_asm/wfe_ldxr_stxr.cpp[]
* link:userland/arch/aarch64/inline_asm/wfe_ldxr_str.cpp[]
SEV is not the only thing that can wake up a WFE, it is only an explicit software way to do it.
@@ -19544,9 +19551,9 @@ cd -
./bench-all -A
....
=== Continuous integraion
=== Continuous integration
We have exploreed a few Continuous integration solutions.
We have explored a few Continuous integration solutions.
We haven't setup any of them yet.
@@ -19556,7 +19563,7 @@ We tried to automate it on Travis with link:.travis.yml[] but it hits the curren
==== CircleCI
This setup sucessfully built gem5 on every commit: link:.circleci/config.yml[]
This setup successfully built gem5 on every commit: link:.circleci/config.yml[]
Enabling it is however blocked on: https://github.com/cirosantilli/linux-kernel-module-cheat/issues/79 so we disabled the builds on the web UI.
@@ -19714,6 +19721,14 @@ Summary of manually collected results on <<p51>> at LKMC a18f28e263c91362519ef55
|1.10018162 * 10^8
|1
|a18f28e263c91362519ef550150b5c9d75fa3679 + 1
|link:userland/gcc/busy_loop.c[] `-O0`
|`gem5 --arch aarch64 --gem5-build-id debug`
|10^5
|32
|2.528728 * 10^6
|0.08
|a18f28e263c91362519ef550150b5c9d75fa3679 + 1
|link:userland/gcc/busy_loop.c[] `-O0`
|`+gem5 --arch aarch64 -- --cpu-type MinorCPU --caches+`
@@ -19755,7 +19770,7 @@ Summary of manually collected results on <<p51>> at LKMC a18f28e263c91362519ef55
|1.6
|ab6f7331406b22f8ab6e2df5f8b8e464fb35b611
|link:userland/c/m5ops.c[] `-O0`
|glibc C pre-main link:userland/c/m5ops.c[] `-O0`
|`gem5 --arch aarch64 --userland-args e`
|1
|2
@@ -19763,13 +19778,45 @@ Summary of manually collected results on <<p51>> at LKMC a18f28e263c91362519ef55
|0.05
|ab6f7331406b22f8ab6e2df5f8b8e464fb35b611
|link:userland/cpp/m5ops.cpp[] `-O0`
|glibc C pre-main link:userland/c/m5ops.c[] `-O0`
|`gem5 --arch aarch64 --userland-args e --gem5-build-type debug`
|1
|2
|1.26479 * 10^5
|0.05
|ab6f7331406b22f8ab6e2df5f8b8e464fb35b611
|glibc C++ pre-main link:userland/cpp/m5ops.cpp[] `-O0`
|`gem5 --arch aarch64 --userland-args e`
|1
|2
|2.385012 * 10^6
|1
|ab6f7331406b22f8ab6e2df5f8b8e464fb35b611
|glibc C++ pre-main link:userland/cpp/m5ops.cpp[] `-O0`
|`gem5 --arch aarch64 --userland-args e --gem5-build-type debug`
|1
|25
|2.385012 * 10^6
|0.1
|ab6f7331406b22f8ab6e2df5f8b8e464fb35b611
|immediate exit link:userland/arch/aarch64/freestanding/linux/gem5_exit.S[] `-O0`
|`gem5 --arch aarch64`
|1
|1
|
|
|ab6f7331406b22f8ab6e2df5f8b8e464fb35b611
|immediate exit link:userland/arch/aarch64/freestanding/linux/gem5_exit.S[] `-O0`
|`gem5 --arch aarch64 --gem5-build-type debug`
|1
|1
|
|
|===
The first step is to determine a number of loops that will run long enough to have meaningful results, but not too long that we will get bored, so about 1 minute.