2023-05-09 09:30:29 +00:00
# beacon_chain
2024-01-06 06:18:28 +00:00
# Copyright (c) 2020-2024 Status Research & Development GmbH
2023-05-09 09:30:29 +00:00
# Licensed and distributed under either of
# * MIT license (license terms in the root directory or at https://opensource.org/licenses/MIT).
# * Apache v2 license (license terms in the root directory or at https://www.apache.org/licenses/LICENSE-2.0).
# at your option. This file may not be copied, modified, or distributed except according to those terms.
2024-01-06 06:18:28 +00:00
import std/strutils
2021-05-19 06:38:13 +00:00
2022-10-04 23:39:12 +00:00
--noNimblePath
2022-10-21 19:48:43 +00:00
const currentDir = currentSourcePath()[0 .. ^(len("config.nims") + 1)]
2022-10-04 23:39:12 +00:00
if getEnv("NIMBUS_BUILD_SYSTEM") == "yes" and
2022-10-21 19:48:43 +00:00
# BEWARE
# In Nim 1.6, config files are evaluated with a working directory
# matching where the Nim command was invocated. This means that we
# must do all file existance checks with full absolute paths:
system.fileExists(currentDir & "nimbus-build-system.paths"):
2022-10-04 23:39:12 +00:00
include "nimbus-build-system.paths"
2021-02-14 20:04:54 +00:00
const nimCachePathOverride {.strdefine.} = ""
when nimCachePathOverride == "":
when defined(release):
let nimCachePath = "nimcache/release/" & projectName()
else:
let nimCachePath = "nimcache/debug/" & projectName()
2020-02-13 12:19:12 +00:00
else:
2021-02-14 20:04:54 +00:00
let nimCachePath = nimCachePathOverride
2020-10-27 12:09:03 +00:00
switch("nimcache", nimCachePath)
2019-07-03 07:35:05 +00:00
use LTO in release builds (#1661)
* use LTO in release builds
This significantly (40%) speeds up block replay and hashing - for example replaying first 1000
blocks, without/with LTO:
```
[arnetheduck@tempus ncli]$ ../env.sh nim c -d:release ncli_db
[arnetheduck@tempus ncli]$ ./ncli_db bench --db:db --network:medalla --slots:1000
Loaded 215006 blocks, head slot 307400
All time are ms
Average, StdDev, Min, Max, Samples, Test
Validation is turned off meaning that no BLS operations are performed
25468.481, 0.000, 25468.481, 25468.481, 1, Initialize DB
0.297, 0.516, 0.053, 13.645, 721, Load block from database
26.458, 0.000, 26.458, 26.458, 1, Load state from database
20.737, 8.288, 11.096, 199.325, 690, Apply block
333.069, 62.798, 45.225, 429.452, 31, Apply epoch block
0.000, 0.000, 0.000, 0.000, 0, Database block store
```
```
[arnetheduck@tempus ncli]$ ../env.sh nim c -d:release --passc:-flto --passl:-flto --stacktrace:off ncli_db
[arnetheduck@tempus ncli]$ ./ncli_db bench --db:db --network:medalla --slots:1000
Loaded 215006 blocks, head slot 307400
All time are ms
Average, StdDev, Min, Max, Samples, Test
Validation is turned off meaning that no BLS operations are performed
23903.006, 0.000, 23903.006, 23903.006, 1, Initialize DB
0.253, 0.122, 0.047, 0.731, 721, Load block from database
24.455, 0.000, 24.455, 24.455, 1, Load state from database
18.734, 7.062, 10.346, 167.397, 690, Apply block
194.869, 33.175, 29.311, 226.981, 31, Apply epoch block
```
Epoch processing is heavy on both arithmetics and hash caching, both of which get a
significant boost here.
This makes sense: nim creates lots of small functions spread out over many C files. A much
worse solution is to try to annotate code with `inline` - it copies functions to multiple
C files but still doesn't do intermodule optimizations significantly limiting the
compilers' ability to reason about the code, causing bloat and misrepresenting the usefulness
of a function to the call frequency analysis that drives actual (C-compiler) inlining and many
other optimizations.
In particular, many nim functions are part of `system` or the `C` backend - stack tracing,
memory allocation etc - nim's inlining system is pretty incomplete in that it does not deal
with these and many other cases.
* windows workaround
* skip LTO on windows for now
2020-09-24 16:40:28 +00:00
# `-flto` gives a significant improvement in processing speed, specially hash tree and state transition (basically any CPU-bound code implemented in nim)
# With LTO enabled, optimization flags should be passed to both compiler and linker!
2020-09-25 16:15:02 +00:00
if defined(release) and not defined(disableLTO):
2021-11-04 14:35:36 +00:00
# "-w" is not passed to the compiler during linking, so we need to disable
# some warnings by hand.
switch("passL", "-Wno-stringop-overflow -Wno-stringop-overread")
use LTO in release builds (#1661)
* use LTO in release builds
This significantly (40%) speeds up block replay and hashing - for example replaying first 1000
blocks, without/with LTO:
```
[arnetheduck@tempus ncli]$ ../env.sh nim c -d:release ncli_db
[arnetheduck@tempus ncli]$ ./ncli_db bench --db:db --network:medalla --slots:1000
Loaded 215006 blocks, head slot 307400
All time are ms
Average, StdDev, Min, Max, Samples, Test
Validation is turned off meaning that no BLS operations are performed
25468.481, 0.000, 25468.481, 25468.481, 1, Initialize DB
0.297, 0.516, 0.053, 13.645, 721, Load block from database
26.458, 0.000, 26.458, 26.458, 1, Load state from database
20.737, 8.288, 11.096, 199.325, 690, Apply block
333.069, 62.798, 45.225, 429.452, 31, Apply epoch block
0.000, 0.000, 0.000, 0.000, 0, Database block store
```
```
[arnetheduck@tempus ncli]$ ../env.sh nim c -d:release --passc:-flto --passl:-flto --stacktrace:off ncli_db
[arnetheduck@tempus ncli]$ ./ncli_db bench --db:db --network:medalla --slots:1000
Loaded 215006 blocks, head slot 307400
All time are ms
Average, StdDev, Min, Max, Samples, Test
Validation is turned off meaning that no BLS operations are performed
23903.006, 0.000, 23903.006, 23903.006, 1, Initialize DB
0.253, 0.122, 0.047, 0.731, 721, Load block from database
24.455, 0.000, 24.455, 24.455, 1, Load state from database
18.734, 7.062, 10.346, 167.397, 690, Apply block
194.869, 33.175, 29.311, 226.981, 31, Apply epoch block
```
Epoch processing is heavy on both arithmetics and hash caching, both of which get a
significant boost here.
This makes sense: nim creates lots of small functions spread out over many C files. A much
worse solution is to try to annotate code with `inline` - it copies functions to multiple
C files but still doesn't do intermodule optimizations significantly limiting the
compilers' ability to reason about the code, causing bloat and misrepresenting the usefulness
of a function to the call frequency analysis that drives actual (C-compiler) inlining and many
other optimizations.
In particular, many nim functions are part of `system` or the `C` backend - stack tracing,
memory allocation etc - nim's inlining system is pretty incomplete in that it does not deal
with these and many other cases.
* windows workaround
* skip LTO on windows for now
2020-09-24 16:40:28 +00:00
if defined(macosx): # Clang
switch("passC", "-flto=thin")
2020-10-27 12:09:03 +00:00
switch("passL", "-flto=thin -Wl,-object_path_lto," & nimCachePath & "/lto")
use LTO in release builds (#1661)
* use LTO in release builds
This significantly (40%) speeds up block replay and hashing - for example replaying first 1000
blocks, without/with LTO:
```
[arnetheduck@tempus ncli]$ ../env.sh nim c -d:release ncli_db
[arnetheduck@tempus ncli]$ ./ncli_db bench --db:db --network:medalla --slots:1000
Loaded 215006 blocks, head slot 307400
All time are ms
Average, StdDev, Min, Max, Samples, Test
Validation is turned off meaning that no BLS operations are performed
25468.481, 0.000, 25468.481, 25468.481, 1, Initialize DB
0.297, 0.516, 0.053, 13.645, 721, Load block from database
26.458, 0.000, 26.458, 26.458, 1, Load state from database
20.737, 8.288, 11.096, 199.325, 690, Apply block
333.069, 62.798, 45.225, 429.452, 31, Apply epoch block
0.000, 0.000, 0.000, 0.000, 0, Database block store
```
```
[arnetheduck@tempus ncli]$ ../env.sh nim c -d:release --passc:-flto --passl:-flto --stacktrace:off ncli_db
[arnetheduck@tempus ncli]$ ./ncli_db bench --db:db --network:medalla --slots:1000
Loaded 215006 blocks, head slot 307400
All time are ms
Average, StdDev, Min, Max, Samples, Test
Validation is turned off meaning that no BLS operations are performed
23903.006, 0.000, 23903.006, 23903.006, 1, Initialize DB
0.253, 0.122, 0.047, 0.731, 721, Load block from database
24.455, 0.000, 24.455, 24.455, 1, Load state from database
18.734, 7.062, 10.346, 167.397, 690, Apply block
194.869, 33.175, 29.311, 226.981, 31, Apply epoch block
```
Epoch processing is heavy on both arithmetics and hash caching, both of which get a
significant boost here.
This makes sense: nim creates lots of small functions spread out over many C files. A much
worse solution is to try to annotate code with `inline` - it copies functions to multiple
C files but still doesn't do intermodule optimizations significantly limiting the
compilers' ability to reason about the code, causing bloat and misrepresenting the usefulness
of a function to the call frequency analysis that drives actual (C-compiler) inlining and many
other optimizations.
In particular, many nim functions are part of `system` or the `C` backend - stack tracing,
memory allocation etc - nim's inlining system is pretty incomplete in that it does not deal
with these and many other cases.
* windows workaround
* skip LTO on windows for now
2020-09-24 16:40:28 +00:00
elif defined(linux):
2022-11-24 20:56:02 +00:00
switch("passC", "-flto=auto")
switch("passL", "-flto=auto")
2020-11-20 10:00:22 +00:00
switch("passC", "-finline-limit=100000")
switch("passL", "-finline-limit=100000")
use LTO in release builds (#1661)
* use LTO in release builds
This significantly (40%) speeds up block replay and hashing - for example replaying first 1000
blocks, without/with LTO:
```
[arnetheduck@tempus ncli]$ ../env.sh nim c -d:release ncli_db
[arnetheduck@tempus ncli]$ ./ncli_db bench --db:db --network:medalla --slots:1000
Loaded 215006 blocks, head slot 307400
All time are ms
Average, StdDev, Min, Max, Samples, Test
Validation is turned off meaning that no BLS operations are performed
25468.481, 0.000, 25468.481, 25468.481, 1, Initialize DB
0.297, 0.516, 0.053, 13.645, 721, Load block from database
26.458, 0.000, 26.458, 26.458, 1, Load state from database
20.737, 8.288, 11.096, 199.325, 690, Apply block
333.069, 62.798, 45.225, 429.452, 31, Apply epoch block
0.000, 0.000, 0.000, 0.000, 0, Database block store
```
```
[arnetheduck@tempus ncli]$ ../env.sh nim c -d:release --passc:-flto --passl:-flto --stacktrace:off ncli_db
[arnetheduck@tempus ncli]$ ./ncli_db bench --db:db --network:medalla --slots:1000
Loaded 215006 blocks, head slot 307400
All time are ms
Average, StdDev, Min, Max, Samples, Test
Validation is turned off meaning that no BLS operations are performed
23903.006, 0.000, 23903.006, 23903.006, 1, Initialize DB
0.253, 0.122, 0.047, 0.731, 721, Load block from database
24.455, 0.000, 24.455, 24.455, 1, Load state from database
18.734, 7.062, 10.346, 167.397, 690, Apply block
194.869, 33.175, 29.311, 226.981, 31, Apply epoch block
```
Epoch processing is heavy on both arithmetics and hash caching, both of which get a
significant boost here.
This makes sense: nim creates lots of small functions spread out over many C files. A much
worse solution is to try to annotate code with `inline` - it copies functions to multiple
C files but still doesn't do intermodule optimizations significantly limiting the
compilers' ability to reason about the code, causing bloat and misrepresenting the usefulness
of a function to the call frequency analysis that drives actual (C-compiler) inlining and many
other optimizations.
In particular, many nim functions are part of `system` or the `C` backend - stack tracing,
memory allocation etc - nim's inlining system is pretty incomplete in that it does not deal
with these and many other cases.
* windows workaround
* skip LTO on windows for now
2020-09-24 16:40:28 +00:00
else:
2020-10-27 12:09:03 +00:00
# On windows, LTO needs more love and attention so "gcc-ar" and "gcc-ranlib" are
# used for static libraries.
use LTO in release builds (#1661)
* use LTO in release builds
This significantly (40%) speeds up block replay and hashing - for example replaying first 1000
blocks, without/with LTO:
```
[arnetheduck@tempus ncli]$ ../env.sh nim c -d:release ncli_db
[arnetheduck@tempus ncli]$ ./ncli_db bench --db:db --network:medalla --slots:1000
Loaded 215006 blocks, head slot 307400
All time are ms
Average, StdDev, Min, Max, Samples, Test
Validation is turned off meaning that no BLS operations are performed
25468.481, 0.000, 25468.481, 25468.481, 1, Initialize DB
0.297, 0.516, 0.053, 13.645, 721, Load block from database
26.458, 0.000, 26.458, 26.458, 1, Load state from database
20.737, 8.288, 11.096, 199.325, 690, Apply block
333.069, 62.798, 45.225, 429.452, 31, Apply epoch block
0.000, 0.000, 0.000, 0.000, 0, Database block store
```
```
[arnetheduck@tempus ncli]$ ../env.sh nim c -d:release --passc:-flto --passl:-flto --stacktrace:off ncli_db
[arnetheduck@tempus ncli]$ ./ncli_db bench --db:db --network:medalla --slots:1000
Loaded 215006 blocks, head slot 307400
All time are ms
Average, StdDev, Min, Max, Samples, Test
Validation is turned off meaning that no BLS operations are performed
23903.006, 0.000, 23903.006, 23903.006, 1, Initialize DB
0.253, 0.122, 0.047, 0.731, 721, Load block from database
24.455, 0.000, 24.455, 24.455, 1, Load state from database
18.734, 7.062, 10.346, 167.397, 690, Apply block
194.869, 33.175, 29.311, 226.981, 31, Apply epoch block
```
Epoch processing is heavy on both arithmetics and hash caching, both of which get a
significant boost here.
This makes sense: nim creates lots of small functions spread out over many C files. A much
worse solution is to try to annotate code with `inline` - it copies functions to multiple
C files but still doesn't do intermodule optimizations significantly limiting the
compilers' ability to reason about the code, causing bloat and misrepresenting the usefulness
of a function to the call frequency analysis that drives actual (C-compiler) inlining and many
other optimizations.
In particular, many nim functions are part of `system` or the `C` backend - stack tracing,
memory allocation etc - nim's inlining system is pretty incomplete in that it does not deal
with these and many other cases.
* windows workaround
* skip LTO on windows for now
2020-09-24 16:40:28 +00:00
discard
2023-08-03 08:36:45 +00:00
# Hidden visibility allows for better position-independent codegen - it also
# resolves a build issue in BLST where otherwise private symbols would require
# an unsupported relocation on PIE-enabled distros such as ubuntu - BLST itself
# solves this via a linker script which is messy
switch("passC", "-fvisibility=hidden")
2021-11-04 14:35:36 +00:00
# show C compiler warnings
if defined(cwarnings):
let common_gcc_options = "-Wno-discarded-qualifiers -Wno-incompatible-pointer-types"
if defined(windows):
put("gcc.options.always", "-mno-ms-bitfields " & common_gcc_options)
put("clang.options.always", "-mno-ms-bitfields " & common_gcc_options)
else:
put("gcc.options.always", common_gcc_options)
put("clang.options.always", common_gcc_options)
if defined(limitStackUsage):
# This limits stack usage of each individual function to 1MB - the option is
# available on some GCC versions but not all - run with `-d:limitStackUsage`
2021-11-05 17:11:25 +00:00
# and look for .su files in "./build/", "./nimcache/" or $TMPDIR that list the
# stack size of each function.
2021-11-04 14:35:36 +00:00
switch("passC", "-fstack-usage -Werror=stack-usage=1048576")
switch("passL", "-fstack-usage -Werror=stack-usage=1048576")
2020-02-13 12:19:12 +00:00
if defined(windows):
2019-08-24 20:07:04 +00:00
# disable timestamps in Windows PE headers - https://wiki.debian.org/ReproducibleBuilds/TimestampsInPEBinaries
2020-02-13 12:19:12 +00:00
switch("passL", "-Wl,--no-insert-timestamp")
2019-04-11 21:30:26 +00:00
# increase stack size
2020-02-13 12:19:12 +00:00
switch("passL", "-Wl,--stack,8388608")
2019-04-24 23:49:41 +00:00
# https://github.com/nim-lang/Nim/issues/4057
--tlsEmulation:off
2020-02-13 12:19:12 +00:00
if defined(i386):
2019-11-12 14:34:12 +00:00
# set the IMAGE_FILE_LARGE_ADDRESS_AWARE flag so we can use PAE, if enabled, and access more than 2 GiB of RAM
2020-02-13 12:19:12 +00:00
switch("passL", "-Wl,--large-address-aware")
2019-10-25 17:03:55 +00:00
2019-10-07 08:29:39 +00:00
# The dynamic Chronicles output currently prevents us from using colors on Windows
# because these require direct manipulations of the stdout File object.
2020-02-13 12:19:12 +00:00
switch("define", "chronicles_colors=off")
2021-12-17 03:05:47 +00:00
# Avoid some rare stack corruption while using exceptions with a SEH-enabled
# toolchain: https://github.com/status-im/nimbus-eth2/issues/3121
switch("define", "nimRawSetjmp")
2019-02-05 17:48:34 +00:00
2021-12-05 18:02:58 +00:00
# https://github.com/status-im/nimbus-eth2/blob/stable/docs/cpu_features.md#ssse3-supplemental-sse3
# suggests that SHA256 hashing with SSSE3 is 20% faster than without SSSE3, so
# given its near-ubiquity in the x86 installed base, it renders a distribution
# build more viable on an overall broader range of hardware.
#
2021-12-17 03:10:30 +00:00
if defined(disableMarchNative):
2021-01-07 09:19:29 +00:00
if defined(i386) or defined(amd64):
2022-02-02 15:24:41 +00:00
if defined(macosx):
2023-08-28 09:50:00 +00:00
# macOS Big Sur is EOL as of 2023-11
# https://support.apple.com/en-us/HT212551
# "macOS Monterey is compatible with these computers" of which the
# oldest is "Mac Pro (Late 2013)". All have Haswell or newer CPUs.
2022-10-19 21:55:01 +00:00
#
# This ensures AVX2, AES-NI, PCLMUL, BMI1, and BMI2 instruction set support.
switch("passC", "-march=haswell -mtune=generic")
switch("passL", "-march=haswell -mtune=generic")
2022-02-02 15:24:41 +00:00
else:
2023-05-11 10:05:52 +00:00
if defined(marchOptimized):
# https://github.com/status-im/nimbus-eth2/blob/stable/docs/cpu_features.md#bmi2--adx
switch("passC", "-march=broadwell -mtune=generic")
switch("passL", "-march=broadwell -mtune=generic")
else:
switch("passC", "-mssse3")
switch("passL", "-mssse3")
2021-12-17 03:10:30 +00:00
elif defined(macosx) and defined(arm64):
# Apple's Clang can't handle "-march=native" on M1: https://github.com/status-im/nimbus-eth2/issues/2758
2023-03-11 07:33:53 +00:00
switch("passC", "-mcpu=apple-m1")
switch("passL", "-mcpu=apple-m1")
2024-07-23 02:03:53 +00:00
elif defined(riscv64):
# riscv64 needs specification of ISA with extensions. 'gc' is widely supported
# and seems to be the minimum extensions needed to build.
switch("passC", "-march=rv64gc")
switch("passL", "-march=rv64gc")
2020-02-13 12:19:12 +00:00
else:
switch("passC", "-march=native")
use LTO in release builds (#1661)
* use LTO in release builds
This significantly (40%) speeds up block replay and hashing - for example replaying first 1000
blocks, without/with LTO:
```
[arnetheduck@tempus ncli]$ ../env.sh nim c -d:release ncli_db
[arnetheduck@tempus ncli]$ ./ncli_db bench --db:db --network:medalla --slots:1000
Loaded 215006 blocks, head slot 307400
All time are ms
Average, StdDev, Min, Max, Samples, Test
Validation is turned off meaning that no BLS operations are performed
25468.481, 0.000, 25468.481, 25468.481, 1, Initialize DB
0.297, 0.516, 0.053, 13.645, 721, Load block from database
26.458, 0.000, 26.458, 26.458, 1, Load state from database
20.737, 8.288, 11.096, 199.325, 690, Apply block
333.069, 62.798, 45.225, 429.452, 31, Apply epoch block
0.000, 0.000, 0.000, 0.000, 0, Database block store
```
```
[arnetheduck@tempus ncli]$ ../env.sh nim c -d:release --passc:-flto --passl:-flto --stacktrace:off ncli_db
[arnetheduck@tempus ncli]$ ./ncli_db bench --db:db --network:medalla --slots:1000
Loaded 215006 blocks, head slot 307400
All time are ms
Average, StdDev, Min, Max, Samples, Test
Validation is turned off meaning that no BLS operations are performed
23903.006, 0.000, 23903.006, 23903.006, 1, Initialize DB
0.253, 0.122, 0.047, 0.731, 721, Load block from database
24.455, 0.000, 24.455, 24.455, 1, Load state from database
18.734, 7.062, 10.346, 167.397, 690, Apply block
194.869, 33.175, 29.311, 226.981, 31, Apply epoch block
```
Epoch processing is heavy on both arithmetics and hash caching, both of which get a
significant boost here.
This makes sense: nim creates lots of small functions spread out over many C files. A much
worse solution is to try to annotate code with `inline` - it copies functions to multiple
C files but still doesn't do intermodule optimizations significantly limiting the
compilers' ability to reason about the code, causing bloat and misrepresenting the usefulness
of a function to the call frequency analysis that drives actual (C-compiler) inlining and many
other optimizations.
In particular, many nim functions are part of `system` or the `C` backend - stack tracing,
memory allocation etc - nim's inlining system is pretty incomplete in that it does not deal
with these and many other cases.
* windows workaround
* skip LTO on windows for now
2020-09-24 16:40:28 +00:00
switch("passL", "-march=native")
2020-02-13 12:19:12 +00:00
if defined(windows):
2020-02-11 23:36:54 +00:00
# https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65782
2020-02-12 13:23:49 +00:00
# ("-fno-asynchronous-unwind-tables" breaks Nim's exception raising, sometimes)
2020-09-24 11:45:34 +00:00
switch("passC", "-mno-avx512f")
use LTO in release builds (#1661)
* use LTO in release builds
This significantly (40%) speeds up block replay and hashing - for example replaying first 1000
blocks, without/with LTO:
```
[arnetheduck@tempus ncli]$ ../env.sh nim c -d:release ncli_db
[arnetheduck@tempus ncli]$ ./ncli_db bench --db:db --network:medalla --slots:1000
Loaded 215006 blocks, head slot 307400
All time are ms
Average, StdDev, Min, Max, Samples, Test
Validation is turned off meaning that no BLS operations are performed
25468.481, 0.000, 25468.481, 25468.481, 1, Initialize DB
0.297, 0.516, 0.053, 13.645, 721, Load block from database
26.458, 0.000, 26.458, 26.458, 1, Load state from database
20.737, 8.288, 11.096, 199.325, 690, Apply block
333.069, 62.798, 45.225, 429.452, 31, Apply epoch block
0.000, 0.000, 0.000, 0.000, 0, Database block store
```
```
[arnetheduck@tempus ncli]$ ../env.sh nim c -d:release --passc:-flto --passl:-flto --stacktrace:off ncli_db
[arnetheduck@tempus ncli]$ ./ncli_db bench --db:db --network:medalla --slots:1000
Loaded 215006 blocks, head slot 307400
All time are ms
Average, StdDev, Min, Max, Samples, Test
Validation is turned off meaning that no BLS operations are performed
23903.006, 0.000, 23903.006, 23903.006, 1, Initialize DB
0.253, 0.122, 0.047, 0.731, 721, Load block from database
24.455, 0.000, 24.455, 24.455, 1, Load state from database
18.734, 7.062, 10.346, 167.397, 690, Apply block
194.869, 33.175, 29.311, 226.981, 31, Apply epoch block
```
Epoch processing is heavy on both arithmetics and hash caching, both of which get a
significant boost here.
This makes sense: nim creates lots of small functions spread out over many C files. A much
worse solution is to try to annotate code with `inline` - it copies functions to multiple
C files but still doesn't do intermodule optimizations significantly limiting the
compilers' ability to reason about the code, causing bloat and misrepresenting the usefulness
of a function to the call frequency analysis that drives actual (C-compiler) inlining and many
other optimizations.
In particular, many nim functions are part of `system` or the `C` backend - stack tracing,
memory allocation etc - nim's inlining system is pretty incomplete in that it does not deal
with these and many other cases.
* windows workaround
* skip LTO on windows for now
2020-09-24 16:40:28 +00:00
switch("passL", "-mno-avx512f")
2020-02-05 17:20:05 +00:00
2020-12-16 13:07:48 +00:00
# omitting frame pointers in nim breaks the GC
# https://github.com/nim-lang/Nim/issues/10625
switch("passC", "-fno-omit-frame-pointer")
switch("passL", "-fno-omit-frame-pointer")
2019-08-24 20:07:04 +00:00
--threads:on
--opt:speed
2023-08-08 00:04:41 +00:00
--mm:refc
2019-08-24 20:07:04 +00:00
--excessiveStackTrace:on
2019-10-17 13:21:45 +00:00
# enable metric collection
2020-02-13 12:19:12 +00:00
--define:metrics
2021-11-02 17:06:36 +00:00
--define:chronicles_line_numbers # These are disabled for release binaries
2020-03-28 22:04:43 +00:00
# for heap-usage-by-instance-type metrics and object base-type strings
--define:nimTypeNames
2019-11-12 23:22:21 +00:00
2020-06-11 16:41:43 +00:00
switch("define", "nim_compiler_path=" & currentDir & "env.sh nim")
2020-11-13 15:00:45 +00:00
switch("define", "withoutPCRE")
2020-06-11 16:41:43 +00:00
2020-03-24 11:13:07 +00:00
switch("import", "testutils/moduletests")
2022-03-05 14:40:08 +00:00
when not defined(disable_libbacktrace):
2020-06-04 13:36:44 +00:00
--define:nimStackTraceOverride
switch("import", "libbacktrace")
else:
--stacktrace:on
--linetrace:on
2021-05-19 06:38:13 +00:00
var canEnableDebuggingSymbols = true
if defined(macosx):
# The default open files limit is too low on macOS (512), breaking the
# "--debugger:native" build. It can be increased with `ulimit -n 1024`.
let openFilesLimitTarget = 1024
var openFilesLimit = 0
try:
openFilesLimit = staticExec("ulimit -n").strip(chars = Whitespace + Newlines).parseInt()
if openFilesLimit < openFilesLimitTarget:
echo "Open files limit too low to enable debugging symbols and lightweight stack traces."
echo "Increase it with \"ulimit -n " & $openFilesLimitTarget & "\""
canEnableDebuggingSymbols = false
except:
echo "ulimit error"
# We ignore this resource limit on Windows, where a default `ulimit -n` of 256
# in Git Bash is apparently ignored by the OS, and on Linux where the default of
# 1024 is good enough for us.
if canEnableDebuggingSymbols:
2019-11-12 17:05:05 +00:00
# add debugging symbols and original files and line numbers
--debugger:native
2020-02-13 12:19:12 +00:00
--define:nimOldCaseObjects # https://github.com/status-im/nim-confutils/issues/9
# `switch("warning[CaseTransition]", "off")` fails with "Error: invalid command line option: '--warning[CaseTransition]'"
switch("warning", "CaseTransition:off")
2019-08-24 20:07:04 +00:00
2020-06-03 12:49:32 +00:00
# The compiler doth protest too much, methinks, about all these cases where it can't
# do its (N)RVO pass: https://github.com/nim-lang/RFCs/issues/230
switch("warning", "ObservableStores:off")
2023-02-24 19:50:46 +00:00
# Too many right now to read compiler output. Warnings are legitimate, but
# should be fixed out-of-band of `unstable` branch.
2023-05-09 09:30:29 +00:00
switch("warning", "BareExcept:off")
2023-02-24 19:50:46 +00:00
2022-11-24 20:56:02 +00:00
# Too many of these because of Defect compat in 1.2
2023-02-03 15:12:11 +00:00
switch("hint", "XCannotRaiseY:off")
2022-11-24 20:56:02 +00:00
2020-08-08 19:24:14 +00:00
# Useful for Chronos metrics.
2020-09-11 15:07:16 +00:00
#--define:chronosFutureTracking
2020-10-07 10:02:00 +00:00
2023-03-23 09:47:04 +00:00
--define:kzgExternalBlst
2020-10-07 10:02:00 +00:00
# ############################################################
#
# No LTO for crypto
#
# ############################################################
# This applies per-file compiler flags to C files
2022-07-06 10:33:02 +00:00
# which do not support {.localPassC: "-fno-lto".}
2020-10-07 10:02:00 +00:00
# Unfortunately this is filename based instead of path-based
# Assumes GCC
# BLST
put("server.always", "-fno-lto")
put("assembly.always", "-fno-lto")
# Secp256k1
2024-07-03 22:06:10 +00:00
# -fomit-frame-pointer for https://github.com/status-im/nimbus-eth2/issues/6324
put("secp256k1.always", "-fno-lto -fomit-frame-pointer")
2020-10-07 10:02:00 +00:00
# BearSSL - only RNGs
put("aesctr_drbg.always", "-fno-lto")
put("hmac_drbg.always", "-fno-lto")
put("sysrng.always", "-fno-lto")
# ############################################################
#
# Spurious warnings
#
# ############################################################
# sqlite3.c: In function ‘ sqlite3SelectNew’ :
# vendor/nim-sqlite3-abi/sqlite3.c:124500: warning: function may return address of local variable [-Wreturn-local-addr]
put("sqlite3.always", "-fno-lto") # -Wno-return-local-addr