The test HWAddressSanitizer-aarch64 :: TestCases/stack-uar-dynamic.c fails already at commit a647100b4320923b0e9d156cc3872b3be470ad98 which is before mine.
As such can't be caused by mine, I'll re-commit the patch.

Buildbot has shown a test failure in hwasan with reverting this. Investigating.

Ah I forgot the context.. Thanks @eugenis !

Ping. Sorry for the early ping, but would be good to get bootstrap working again.

reverted to diff that removed

@MaskRay was that a you prefer it that way or?

In D102059#2749302, @MaskRay wrote:

The Linux kernel has removed the interface to cyclades from the latest kernel headers[1] due to them being orphaned for the past 13 years.

libsanitizer uses this header when compiling against glibc, but glibcs itself doesn't seem to have any references to cyclades.

Sounds like cyclades interceptors should just be deleted?

Update diff

In D102059#2744070, @jakubjelinek wrote:

Alternative to this would be

#ifdef __has_include
# if __has_include(<linux/cyclades.h>)
#  include <linux/cyclades.h>
#  define SANITIZER_LINUX_CYCLADES 1
# endif
#endif

For GCC this is guard-page - 1k. The reasoning is that with any outgoing argument larger than 1k we would probe such that we maintain the invariant, but probing that 1k means we have a whole guard-size -1k left that we can use without probing. These sizes were chose as they cover about 99% of all programs (for a subset of all :)).

I don't know enough about LLVM to comment on the actual code so I will only comment on the output I see generated from the testcases.

Aha thanks for the explanation

This commit adds the parameter and update call sites
accordingly, fixing one of Flake8's F821 warning (undefined name).

was it being indirectly imported before?

good job!

In D94791#2501557, @thopre wrote:

In D94791#2501415, @tnfchris wrote:

This seems rather random compared with D94789... so the spacing rule isn't consistent for assignments?

I'm not too familiar with the rules but I fail to see the inconsistency since this change does not modify any space around assignment, only spaces before parenthesis.

This seems rather random compared with D94789... so the spacing rule isn't consistent for assignments?

LGTM

Nice cleanup!

I agree, this makes more sense so we don't break cPerf again.

use package relative import syntax

Updated comment

was landed in 60343998f7907b77e24f01c94c48ea6c753eca7e

Add more context, to diff.