b358bde020 randomenv: consolidate WIN32 #ifdefs (fanquake)
fff80cd248 random: remove windows-only compat.h include in randomenv (fanquake)
Pull request description:
Similar to #26814.
Having a windows-only include of compat.h is confusing, not-only because it's already included globally via util/time.h, but also because it's unclear why compat.h is included (neither of the required headers are included there).
This change is related to removing the use of compat.h as a miscellaneous catch-all for unclear/platform specific includes. Somewhat prompted by IWYU-related discussion here: https://github.com/bitcoin/bitcoin/pull/26763/files#r1058861693.
ACKs for top commit:
hebasto:
ACK b358bde020.
TheCharlatan:
ACK b358bde020
Tree-SHA512: d46dffe36a17ad0f9374a55e0ecaf2d60d0b473c8fc9ad6f3005142014c08a7c10bae4948856531abb443f5e0bd6062958fe574197e282dad22ae50134d71e5f
Note that this was probably only here to indirectly receive windows.h
via another include in compat.h (windows.h or winreg.h aren't included
there).
Also note that compat.h is already pulled in here for everyone via
util/time.h, so including inside a windows only ifdef is secondarily
redundant.
It was [pointed out in #23030](https://github.com/bitcoin/bitcoin/pull/23030#issuecomment-922893367) that we might be able to get rid of our weak linking of [`getauxval()`](https://man7.org/linux/man-pages/man3/getauxval.3.html) (`HAVE_WEAK_GETAUXVAL`) entirely, with only Android being a potential holdout:
> I wonder if it's time to get rid of HAVE_WEAK_GETAUXVAL. I think it's confusing. Either we build against a C library that has this functionality, or not. We don't do this weak linking thing for any other symbols and recently got rid of the other glibc backwards compatibility stuff.
> Unless there is still a current platform that really needs it (Android?), I'd prefer to remove it from the build system, it has caused enough issues.
After looking at Android further, it would seem that given we are moving to using `std::filesystem`, which [requires NDK version 22 and later](https://github.com/android/ndk/wiki/Changelog-r22), and `getauxval` has been available in the since [API version 18](https://developer.android.com/ndk/guides/cpu-features#features_using_libcs_getauxval3), that shouldn't really be an issue. Support for API levels < 19 will be dropped with the NDK 24 release, and according to [one website](https://apilevels.com/), supporting API level 18+ will cover ~99% of devices. Note that in the CI we currently build with NDK version 22 and API level 28.
The other change in this PR is removing the include of headers for ARM intrinsics, from the check for strong `getauxval()` support in configure, as they shouldn't be needed. Including these headers also meant that the check would basically only succeed when building for ARM. This would be an issue if we remove weak linking, as we wouldn't detect `getauxval()` as supported on other platforms. Note that we also use `getauxval()` in our RNG when it's available.
I've checked that with these changes we detect support for strong `getauxval()` on Alpine (muslibc). On Linux, previously we'd be detecting support for weak getauxval(), now we detect strong support. Note that we already require glibc 2.17, and `getauxval()` was introduced in `2.16`.
This is an alternative / supersedes #23030.
Check for HAVE_STRONG_GETAUXVAL or HAVE_WEAK_GETAUXVAL before using
getauxval to avoid a build failure on uclibc
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
clock_gettime(), CLOCK_MONOTONIC and CLOCK_REALTIME are all available for use on
macOS (now that we require macOS >=10.12). Use them rather than the deprecated
mach_timespec_t time API.
master:
2019-12-23T20:49:43Z Feeding 216 bytes of dynamic environment data into RNG
2019-12-23T20:50:43Z Feeding 216 bytes of dynamic environment data into RNG
this commit:
2019-12-23T20:32:41Z Feeding 232 bytes of dynamic environment data into RNG
2019-12-23T20:33:42Z Feeding 232 bytes of dynamic environment data into RNG
On the ::SLOW path we would use OpenSSL as an additional source of
random bytes. This commit removes that functionality. Note that this was
always only an additional source, and that we never checked the return
value
RAND_bytes(): https://www.openssl.org/docs/manmaster/man3/RAND_bytes.html
RAND_bytes() puts num cryptographically strong pseudo-random bytes into buf.
