rpms/gdb
7
0
Fork 0
Code Issues Pull Requests Releases Activity
e19c0f8fb7
gdb/_gdb.spec.Patch.include
Andrew Burgess e19c0f8fb7 Rewrite (and rename) gdb-libexec-add-index.patch 
It has been observed that the changes added by
gdb-libexec-add-index.patch will result in GDB testing hanging when
the tests are being run using an in-tree GDB; that is when using 'make
check'.  One test that is known to fail is gdb.base/with-mf.exp,
though any test that calls the gdb-add-index.sh script will also hang.

The problem is that when the gdb-add-index.sh script is run, the GDB
testsuite passes the GDB command to use within the GDB environment
variable.  For in-tree testing this will be something like:

  GDB="/path/to/gdb -data-directory /path/to/data-directory"

Notice that the environment variable contains both an executable and
an argument.

Our changes to gdb-add-index.sh add this:

  GDB2=/usr/libexec/gdb
  if test -x $GDB2 && ! which $GDB &>/dev/null; then
      GDB=$GDB2
  fi

The problem then is that '-data-directory' is treated as a set of
options to 'which'.  Many of these options are not known to 'which',
but the '-i' option is known.  The documentation of '-i' says:

    --read-alias, -i
        Read aliases from stdin, reporting matching ones on
        stdout. This is useful in combination with using an alias for
        which itself. For example
        alias which=´alias | which -i´.

And here's the problem; this option causes 'which' to read from
stdin.  As the GDB testsuite doesn't send any additional input on
stdin then the which command will never complete, and the test will
hang.

The solution I think is to avoid calling 'which' like this on a user
supplied GDB environment variable.

The changes in the gdb-libexec-add-index.patch were really about what
the _default_ GDB executable should be.  The upstream version of this
script does this:

  GDB=${GDB:=gdb}

That is, the default is just 'gdb'.  However, for RH this is not good
enough.  We want to handle two additional cases, first, when only the
gdb-minimal package is installed, in which case the default should be
/usr/bin/gdb.minimal.  Then we also want to handle the case where the
user doesn't have 'gdb' itself in their $PATH, but does have the 'gdb'
executable installed in /usr/libexec/gdb.

The code as it currently stands also has a problem where, if
gdb.minimal is installed on the machine this will _always_ be used in
preference to the user supplied GDB value (assuming the code worked at
all) this means that when doing in-tree testing we wouldn't actually
be using the in-tree GDB to build the index, which isn't ideal.

So in this commit I propose that we rework our gdb-add-index.sh
changes.  Now, we only use the RH special values in the case that
there is no GDB environment variable set.  I believe this handles all
the required use cases:

  1. When doing in-tree testing GDB environment variable will be set,
  and this will always be used as is, with no special processing,

  2. When gdb-add-index.sh is used and GDB environment variable is not
  set then we will use the first of the following as the default:

    (a) /usr/bin/gdb.minimal if this file exists and is executable,
    (b) The first gdb executable that can be found in the $PATH,
    (c) /usr/libexec/gdb if this file exists and is executable.

While I was changing this patch anyway I've removed the libexec part
of the patch name -- this no longer seemed relevant, I suspect this
related to an older version of this patch.
2023-05-04 15:11:19 +01:00

232 lines
8.2 KiB
Plaintext
Raw Blame History

# Check distro name is included in the version output.
Patch001: gdb-6.3-rh-testversion-20041202.patch
# Add a wrapper script to GDB that implements pstack using the
# --readnever option.
#=push
Patch002: gdb-6.3-gstack-20050411.patch
# Support TLS symbols (+`errno' suggestion if no pthread is found) (BZ 185337).
#=push+jan: It should be replaced by Infinity project.
Patch003: gdb-6.5-bz185337-resolve-tls-without-debuginfo-v2.patch
# Fix TLS symbols resolving for shared libraries with a relative pathname.
# The testsuite needs `gdb-6.5-tls-of-separate-debuginfo.patch'.
#=fedoratest: One should recheck if it is really fixed upstream.
Patch004: gdb-6.5-sharedlibrary-path.patch
# Improved testsuite results by the testsuite provided by the courtesy of BEA.
#=fedoratest: For upstream it should be rewritten as a dejagnu test, the test of no "??" was useful.
Patch005: gdb-6.5-BEA-testsuite.patch
# Test sideeffects of skipping ppc .so libs trampolines (BZ 218379).
#=fedoratest
Patch006: gdb-6.5-bz218379-ppc-solib-trampoline-test.patch
# Allow running `/usr/bin/gcore' with provided but inaccessible tty (BZ 229517).
#=fedoratest
Patch007: gdb-6.6-bz229517-gcore-without-terminal.patch
# Avoid too long timeouts on failing cases of "annota1.exp annota3.exp".
#=fedoratest
Patch008: gdb-6.6-testsuite-timeouts.patch
# Support for stepping over PPC atomic instruction sequences (BZ 237572).
#=fedoratest
Patch009: gdb-6.6-bz237572-ppc-atomic-sequence-test.patch
# Test kernel VDSO decoding while attaching to an i386 process.
#=fedoratest
Patch010: gdb-6.3-attach-see-vdso-test.patch
# Test leftover zombie process (BZ 243845).
#=fedoratest
Patch011: gdb-6.5-bz243845-stale-testing-zombie-test.patch
# New locating of the matching binaries from the pure core file (build-id).
#=push+jan
Patch012: gdb-6.6-buildid-locate.patch
# Fix loading of core files without build-ids but with build-ids in executables.
# Load strictly build-id-checked core files only if no executable is specified
# (Jan Kratochvil, RH BZ 1339862).
#=push+jan
Patch013: gdb-6.6-buildid-locate-solib-missing-ids.patch
#=push+jan
Patch014: gdb-6.6-buildid-locate-rpm.patch
# Test PPC hiding of call-volatile parameter register.
#=fedoratest
Patch015: gdb-6.7-ppc-clobbered-registers-O2-test.patch
# Testsuite fixes for more stable/comparable results.
#=fedoratest
Patch016: gdb-6.7-testsuite-stable-results.patch
# Test ia64 memory leaks of the code using libunwind.
#=fedoratest
Patch017: gdb-6.5-ia64-libunwind-leak-test.patch
# Test hiding unexpected breakpoints on intentional step commands.
#=fedoratest
Patch018: gdb-6.5-missed-trap-on-step-test.patch
# Test gcore memory and time requirements for large inferiors.
#=fedoratest
Patch019: gdb-6.5-gcore-buffer-limit-test.patch
# Test GCORE for shmid 0 shared memory mappings.
#=fedoratest: But it is broken anyway, sometimes the case being tested is not reproducible.
Patch020: gdb-6.3-mapping-zero-inode-test.patch
# Test a crash on libraries missing the .text section.
#=fedoratest
Patch021: gdb-6.5-section-num-fixup-test.patch
# Fix resolving of variables at locations lists in prelinked libs (BZ 466901).
#=fedoratest
Patch022: gdb-6.8-bz466901-backtrace-full-prelinked.patch
# New test for step-resume breakpoint placed in multiple threads at once.
#=fedoratest
Patch023: gdb-simultaneous-step-resume-breakpoint-test.patch
# Fix GNU/Linux core open: Can't read pathname for load map: Input/output error.
# Fix regression of undisplayed missing shared libraries caused by a fix for.
#=fedoratest: It should be in glibc: libc-alpha: <20091004161706.GA27450@.*>
Patch024: gdb-core-open-vdso-warning.patch
# Workaround ccache making lineno non-zero for command-line definitions.
#=fedoratest: ccache is rarely used and it is even fixed now.
Patch025: gdb-ccache-workaround.patch
# Testcase for "Do not make up line information" fix by Daniel Jacobowitz.
#=fedoratest
Patch026: gdb-lineno-makeup-test.patch
# Test power7 ppc disassembly.
#=fedoratest
Patch027: gdb-ppc-power7-test.patch
# Fix follow-exec for C++ programs (bugreported by Martin Stransky).
#=fedoratest
Patch028: gdb-archer-next-over-throw-cxx-exec.patch
# Workaround librpm BZ 643031 due to its unexpected exit() calls (BZ 642879).
#=push+jan
Patch029: gdb-6.6-buildid-locate-rpm-librpm-workaround.patch
# [delayed-symfile] Test a backtrace regression on CFIs without DIE (BZ 614604).
#=fedoratest
Patch030: gdb-test-bt-cfi-without-die.patch
# Verify GDB Python built-in function gdb.solib_address exists (BZ # 634108).
#=fedoratest
Patch031: gdb-bz634108-solib_address.patch
# New test gdb.arch/x86_64-pid0-core.exp for kernel PID 0 cores (BZ 611435).
#=fedoratest
Patch032: gdb-test-pid0-core.patch
# [archer-tromey-delayed-symfile] New test gdb.dwarf2/dw2-aranges.exp.
#=fedoratest
Patch033: gdb-test-dw2-aranges.patch
# Workaround PR libc/14166 for inferior calls of strstr.
#=fedoratest: Compatibility with RHELs (unchecked which ones).
Patch034: gdb-glibc-strstr-workaround.patch
# Include testcase for `Unable to see a variable inside a module (XLF)' (BZ 823789).
#=fedoratest
Patch035: gdb-rhel5.9-testcase-xlf-var-inside-mod.patch
# Testcase for `Setting solib-absolute-prefix breaks vDSO' (BZ 818343).
#=fedoratest
Patch036: gdb-rhbz-818343-set-solib-absolute-prefix-testcase.patch
# Import regression test for `gdb/findvar.c:417: internal-error:
# read_var_value: Assertion `frame' failed.' (RH BZ 947564) from RHEL 6.5.
#=fedoratest
Patch037: gdb-rhbz947564-findvar-assertion-frame-failed-testcase.patch
# Fix 'memory leak in infpy_read_memory()' (RH BZ 1007614)
#=fedoratest
Patch038: gdb-rhbz1007614-memleak-infpy_read_memory-test.patch
# Fix 'gdb gives highly misleading error when debuginfo pkg is present,
# but not corresponding binary pkg' (RH BZ 981154).
#=push+jan
Patch039: gdb-6.6-buildid-locate-misleading-warning-missing-debuginfo-rhbz981154.patch
# Testcase for '[SAP] Recursive dlopen causes SAP HANA installer to
# crash.' (RH BZ 1156192).
#=fedoratest
Patch040: gdb-rhbz1156192-recursive-dlopen-test.patch
# Fix '`catch syscall' doesn't work for parent after `fork' is called'
# (Philippe Waroquiers, RH BZ 1149205).
#=fedoratest
Patch041: gdb-rhbz1149205-catch-syscall-after-fork-test.patch
# Fix 'backport GDB 7.4 fix to RHEL 6.6 GDB' [Original Sourceware bug
# description: 'C++ (and objc): Internal error on unqualified name
# re-set', PR 11657] (RH BZ 1186476).
#=fedoratest
Patch042: gdb-rhbz1186476-internal-error-unqualified-name-re-set-test.patch
# Fix '[ppc64] and [s390x] wrong prologue skip on -O2 -g code' (Jan
# Kratochvil, RH BZ 1084404).
#=fedoratest
Patch043: gdb-rhbz1084404-ppc64-s390x-wrong-prologue-skip-O2-g-3of3.patch
# Force libncursesw over libncurses to match the includes (RH BZ 1270534).
#=push+jan
Patch044: gdb-fedora-libncursesw.patch
# [aarch64] Fix hardware watchpoints (RH BZ 1261564).
#=fedoratest
Patch045: gdb-rhbz1261564-aarch64-hw-watchpoint-test.patch
# Add messages suggesting more recent RHEL gdbserver (RH BZ 1321114).
#=fedora
Patch046: gdb-container-rh-pkg.patch
# [dts+el7] [x86*] Bundle linux_perf.h for libipt (RH BZ 1256513).
#=fedora
Patch047: gdb-linux_perf-bundle.patch
# Update gdb-add-index.sh such that, when the GDB environment
# variable is not set, the script is smarter than just looking for
# 'gdb' in the $PATH.
#
# The actual search order is now: /usr/bin/gdb.minimal, gdb (in the
# $PATH), then /usr/libexec/gdb.
#
# For the rationale of looking for gdb.minimal see:
#
# https://fedoraproject.org/wiki/Changes/Minimal_GDB_in_buildroot
#
#=fedora
Patch048: gdb-add-index.patch
# [s390x] Backport arch12 instructions decoding (RH BZ 1553104).
# =fedoratest
Patch049: gdb-rhbz1553104-s390x-arch12-test.patch
# [aarch64] Backport fix from Luis Machado for RH BZ 2177655.
Patch050: gdb-rhbz2177655-aarch64-pauth-valid-regcache.patch
# Backport "Fix crash in inside_main_func"
# (Tom Tromey, RHBZ 2183595)
Patch051: gdb-rhbz2183595-rustc-inside_main.patch
# Backport "Fix a potential illegal memory access in the BFD library..."
# (Nick Clifton, binutils/29988)
Patch052: gdb-binutils29988-read_indexed_address.patch
# Backport upstream patch fixing a "dangling pointer" build problem
# first seen when building with GCC 13.1.1 20230426 (Red Hat ;; 13.1.1-1).
Patch053: gdb-rhbz2192105-ftbs-dangling-pointer
Reference in New Issue View Git Blame Copy Permalink