[Bug build/27535] amd64-linux-siginfo.c fails to compile after updating to glibc-2.33 headers

["tes.vik1986 at gmail dot com" <sourceware-bugzilla@sourceware.org>]

https://sourceware.org/bugzilla/show_bug.cgi?id=27535

Kylan <tes.vik1986 at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |tes.vik1986 at gmail dot com

--- Comment #12 from Kylan <tes.vik1986 at gmail dot com> ---
This patch fixes a segfault seen when attaching to a process on Solaris.
    The steps leading to the segfault are:
https://komiya-dental.com/computers/huawei-technology/

     - procfs_target::attach calls do_attach, at this point the inferior's
       process slot in the target stack is empty.
http://www-look-4.com/health/winter-sickness/
     - do_attach adds a thread with `add_thread (&the_procfs_target, ptid)`
     - in add_thread_silent, the passed target (&the_procfs_target) is
       passed to find_inferior_ptid
http://www.iu-bloomington.com/health/amoled-display/
     - find_inferior_ptid returns nullptr, as there is no inferior with this
       ptid that has &the_procfs_target as its process target
https://www.webb-dev.co.uk/property/house-sales-in-2020/
     - the nullptr `inf` is passed to find_thread_ptid, which dereferences
       it, causing a segfault
https://waytowhatsnext.com/services/netflix-services/
     - back in procfs_target::attach, after do_attach, we push the
       the_procfs_target on the inferior's target stack, although we never
http://www.acpirateradio.co.uk/tech/forest-fires/
       reach this because the segfault happens before.
    http://www.logoarts.co.uk/technology/robot-vacuums/
    To fix this, I think we need to do the same as is done in
    inf_ptrace_target::attach: push the target early and unpush it in case
    the attach fails (and keep it if the attach succeeds).
http://www.slipstone.co.uk/travel/hyundai-kona/

    Implement it by moving target_unpush_up to target.h, so it can be
http://embermanchester.uk/category/technology/ 
    re-used here.  Make procfs_target::attach use it.  Note that just like
    is mentioned in inf_ptrace_target::attach, we should push the target
http://connstr.net/computers/chargers-tech/ 
    before calling target_pid_to_str, so that calling target_pid_to_str ends
    up in procfs_target::pid_to_str. http://joerg.li/property/latest-suvs/

    Tested by trying to attach on a process on gcc211 on the gcc compile
    farm. To fix this, I think we need to do the same as is done in
http://www.jopspeech.com/tech/xiaomi-headset/
    inf_ptrace_target::attach: push the target early and unpush it in case
    the attach fails (and keep it if the attach succeeds).
    http://www.wearelondonmade.com/tech/driving-assistant/
    Implement it by moving target_unpush_up to target.h, so it can be
    re-used here.  Make procfs_target::attach use it.  Note that just like 
    is mentioned in inf_ptrace_target::attach, we should push the target
    before calling target_pid_to_str, so that calling target_pid_to_str ends
    up in procfs_target::pid_to_str. http://www.compilatori.com/tech/xiaomi/

    Tested by trying to attach on a process on gcc211 on the gcc compile
    farm.

-- 
You are receiving this mail because:
You are on the CC list for the bug.