Hi Simon,

>Thanks for digging into the pipe issue, it is indeed reproducible on
>Linux.  Please see this small 2-patch series that hopefully fixes it for
>you too and makes testing less noisy

>https://urldefense.proofpoint.com/v2/url?u=https-3A__inbox.sourceware.org_gdb-2Dpatches_20261013171340.1748775-2D1-2Dsimon.marchi-40efficios.com_T_-23m50cf93bf2bbe1a9fd9c7ac7237519570df8a5ceb&d=DwICaQ&c=BSDicqBQBDjDI9RkVyTcHQ&r=f-oUQ8ByG1nZ71OI9p76qywCPh7mxzU69hBYnkP9Nis&m=a00eP242E_D4VAASU_j8zMu2t_Vf29AIAghQEZll_XzeU_APW-TVoCat0riBcaj3&s=82cvPgaYrtuANiE_ZnSQXU71WChpKrAdgNMXzRKKCwg&e=

Thanks for the patch.

I re-ran the full gdb.base testsuite in 32-bit mode with your

patches applied.  Here are the 32-bit numbers from before my series,

after my series, and after my series plus your patches:

                          32-bit

                                    before   after    after + your patches

  expected passes             41687   41541    41640

  unexpected failures         2302    2294     2308

  unexpected successes           1       1        1

  expected failures                     15      15       15

  known failures                   30      30       30

  unresolved testcases            11      86       11

  untested testcases           107     107      108

  unsupported tests             179     179      178

Will give more information in that patch thread.

Have a nice day ahead.

Regards,

Aditya.