Thanks for the report. I compared builds 10158 and 10159. The additional breakpoint is for _dl_relocate_static_pie_ifunc in static-reloc.c:28. Before b6477e6c4ef, the function and its symbol were already present, but its IPA-ICF wrapper had no address-bearing subprogram DIE. With the change, GDB can now see the function when minsyms=False and set a breakpoint on it.
This is the behavior PR debug/126355 was intended to restore, so the result also provides useful confirmation that the GCC fix is effective. The failure comes from py-rbreak.exp using the unanchored regexp "func.*". It now also matches the "ifunc" suffix of _dl_relocate_static_pie_ifunc, changing the result from nine breakpoints to ten. GDB, glibc, binutils and Linux are unchanged between the two builds; the GCC revisions are this commit and its immediate parent. Changing the expected count to ten would make the test depend on the target startup objects. A more robust fix would be to narrow the two unscoped regexp uses in the exact-count and throttle checks, for example to "^e?func.*", so that they only select functions defined by the test. I therefore do not think this is a GCC regression. It looks like the GDB test needs to be made more specific. <[email protected]> 于2026年8月24日周一 11:47写道： > Dear contributor, > > Our automatic CI has detected problems related to your patch(es). Please > find some details below. > > In aarch64 native, after: > | commit gcc-17-3278-gb6477e6c4ef > | Author: Longjun Luo <[email protected]> > | Date: Thu Jul 23 00:01:50 2026 +0800 > | > | debug/126355 - preserve debug info for IPA ICF wrappers > | > | When IPA ICF keeps a distinct symbol for an address-taken function > by > | replacing its body with a wrapper, cgraph_node::create_wrapper > reuses the > | original function declaration and its early debug DIE. > | ... 28 lines of the commit log omitted. > > Produces 1 regression: > | > | regressions.sum: > | Running gdb:gdb.python/py-rbreak.exp ... > | FAIL: gdb.python/py-rbreak.exp: check number of returned breakpoints > is 9 > > Used configuration : > *CI config* tcwg_gnu_native_check_gdb aarch64-linux-gnu > *configure and test flags:* none, autodetected on > aarch64-unknown-linux-gnu--disable-multilib --enable-fix-cortex-a53-835769 > --enable-fix-cortex-a53-843419 --with-arch=armv8-a > > We track this bug report under > https://linaro.atlassian.net/browse/GNU-2005. (Require Linaro login) > Please let us know if you have a fix. > > If you have any questions regarding this report, please ask on > [email protected] mailing list. > > > -----------------8<--------------------------8<--------------------------8<-------------------------- > > The information below contains the details of the failures, and the ways > to reproduce a debug environment: > > You can find the failure logs in *.log.1.xz files in > * > http://54.172.246.49:9090/jobs/tcwg_gnu_native_check_gdb--master-aarch64-build/builds/10159/archive/artifacts/00-sumfiles/ > The full lists of regressions and improvements as well as configure and > make commands are in > * > http://54.172.246.49:9090/jobs/tcwg_gnu_native_check_gdb--master-aarch64-build/builds/10159/archive/artifacts/notify/ > The list of [ignored] baseline and flaky failures are in > * > http://54.172.246.49:9090/jobs/tcwg_gnu_native_check_gdb--master-aarch64-build/builds/10159/archive/artifacts/sumfiles/xfails.xfail > > Current build : > http://54.172.246.49:9090/jobs/tcwg_gnu_native_check_gdb--master-aarch64-build/builds/10159/archive/artifacts/ > Reference build : > http://54.172.246.49:9090/jobs/tcwg_gnu_native_check_gdb--master-aarch64-build/builds/10158/archive/artifacts/ > > Instruction to reproduce the build : > https://gitlab.com/LinaroLtd/tcwg/ci/interesting-commits/-/raw/master/gcc/sha1/b6477e6c4efd6df2719190cb048cf300ff31af46/tcwg_gnu_native_check_gdb/master-aarch64/reproduction_instructions.txt > > Full commit : See in git+ssh://[email protected]/git/gcc.git
