Fix Darwin builds for non-standard environments (e.g. Nix)

[martinjlowm]

Issues:

Not reported.

Description of changes:

aws-ls-sys fails to build via Nix (crate2nix) due to missing typedefs, presumably because the Clang used from the Nix sandbox doesn't set -D_DARWIN_C_SOURCE.

The content of interest is https://github.com/alexey-lysiuk/macos-sdk/blob/main/MacOSX26.1.sdk/usr/include/sys/types.h#L83-L96 which doesn't evaluate unless _DARWIN_C_SOURCE is set. The first POSIX branch is set because __STDC_ALLOC_LIB__ isn't: https://github.com/aws/aws-lc/blob/main/crypto/err/err.c#L113C8-L117

I suppose _POSIX_C_SOURCE was set for a reason and I can't figure out why a normal Cargo build would work - I cannot find any trace of __STDC_ALLOC_LIB__ - apart from a comment in #610.

The absence of _DARWIN_C_SOURCE causes the CC builder to throw:

In file included from /nix/var/nix/builds/nix-37298-3130532386/aws-lc-sys-0.32.3/aws-lc/crypto/fipsmodule/bcm.c:78:
In file included from /nix/var/nix/builds/nix-37298-3130532386/aws-lc-sys-0.32.3/aws-lc/crypto/fipsmodule/cpucap/cpu_aarch64_apple.c:20:
In file included from /nix/store/6py65dicilvc6qxfra4mzw0xpxw2vc06-apple-sdk-11.3/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/usr/include/sys/sysctl.h:83:
/nix/store/6py65dicilvc6qxfra4mzw0xpxw2vc06-apple-sdk-11.3/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/usr/include/sys/ucred.h:101:2:
error: unknown type name 'u_int'
  101 |         u_int   cr_version;             /* structure layout version */
      |         ^
In file included from /nix/var/nix/builds/nix-37298-3130532386/aws-lc-sys-0.32.3/aws-lc/crypto/fipsmodule/bcm.c:78:
In file included from /nix/var/nix/builds/nix-37298-3130532386/aws-lc-sys-0.32.3/aws-lc/crypto/fipsmodule/cpucap/cpu_aarch64_apple.c:20:
In file included from /nix/store/6py65dicilvc6qxfra4mzw0xpxw2vc06-apple-sdk-11.3/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/usr/include/sys/sysctl.h:85:
/nix/store/6py65dicilvc6qxfra4mzw0xpxw2vc06-apple-sdk-11.3/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/usr/include/sys/proc.h:116:2: error: unknown type name 'u_int'
  116 |         u_int   p_estcpu;        /* Time averaged value of p_cpticks. */
      |         ^
/nix/store/6py65dicilvc6qxfra4mzw0xpxw2vc06-apple-sdk-11.3/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/usr/include/sys/proc.h:121:2: error: unknown type name 'u_int'
  121 |         u_int   p_swtime;        /* Time swapped in or out. */
      |         ^
/nix/store/6py65dicilvc6qxfra4mzw0xpxw2vc06-apple-sdk-11.3/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/usr/include/sys/proc.h:122:2: error: unknown type name 'u_int'
  122 |         u_int   p_slptime;       /* Time since last blocked. */
      |         ^
/nix/store/6py65dicilvc6qxfra4mzw0xpxw2vc06-apple-sdk-11.3/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/usr/include/sys/proc.h:136:2: error: unknown type name 'u_char'; did you mean 'char'?
  136 |         u_char  p_priority;     /* Process priority. */
      |         ^
/nix/store/6py65dicilvc6qxfra4mzw0xpxw2vc06-apple-sdk-11.3/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/usr/include/sys/proc.h:137:2: error: unknown type name 'u_char'; did you mean 'char'?
  137 |         u_char  p_usrpri;       /* User-priority based on p_cpu and p_nice. */
      |         ^
/nix/store/6py65dicilvc6qxfra4mzw0xpxw2vc06-apple-sdk-11.3/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/usr/include/sys/proc.h:142:2: error: unknown type name 'u_short'; did you mean 'short'?
  142 |         u_short p_xstat;        /* Exit status for wait; also stop signal. */
      |         ^
/nix/store/6py65dicilvc6qxfra4mzw0xpxw2vc06-apple-sdk-11.3/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/usr/include/sys/proc.h:143:2: error: unknown type name 'u_short'; did you mean 'short'?
  143 |         u_short p_acflag;       /* Accounting flags. */
      |         ^

Call-outs:

This is unfamiliar territory for me. I don't know if configuring _POSIX_C_SOURCE appropriately is the better option.

Testing:

N/A

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license and the ISC license.

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 92.56%. Comparing base (c358484) to head (ea4d131).
⚠️ Report is 295 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #941      +/-   ##
==========================================
- Coverage   95.80%   92.56%   -3.24%     
==========================================
  Files          61       74      +13     
  Lines        8143     9893    +1750     
  Branches        0     9893    +9893     
==========================================
+ Hits         7801     9157    +1356     
- Misses        342      455     +113     
- Partials        0      281     +281     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[justsmth]

@martinjlowm -- Sorry, I've not had time to follow up with adding a test. I did push an update to this PR to conditionally set the define in a couple of places.

Can you confirm whether this latest change fixes the problem for you?

[martinjlowm]

@martinjlowm -- Sorry, I've not had time to follow up with adding a test. I did push an update to this PR to conditionally set the define in a couple of places.

Can you confirm whether this latest change fixes the problem for you?

Hey! crate2nix doesn't support Cargo patches through prefetching, so I'd have to assume the minimal reproducible environment is sufficient for considering this change enough for fixing the problem. That being said, it builds by referencing a85d004 ✌️