From: Alan Hayward <Alan.Hayward@arm.com>
To: GDB Patches <gdb-patches@sourceware.org>
Cc: nd <nd@arm.com>
Subject: [PING][PATCH v2 0/2] Aarch64: Fix segfault when casting dummy calls
Date: Tue, 09 Oct 2018 08:26:00 -0000 [thread overview]
Message-ID: <3DF26781-3D5E-4A95-9427-7369E2DEFA01@arm.com> (raw)
In-Reply-To: <20181001155255.14859-1-alan.hayward@arm.com>
Ping.
> On 1 Oct 2018, at 16:52, Alan Hayward <Alan.Hayward@arm.com> wrote:
>
> This is a reworking of a patch I posted in March.
> V1 had a long discussion which was then paused to wait for
> Pedro's IFUNC rewrite.
>
>
> Prevent the int cast in the following causing a segfault on aarch64:
> (gdb) b foo if (int)strcmp(name,"abc") == 0
> (gdb) run
>
>
> This is because to aarch64_push_dummy_call determines the return type
> of the function and then does not check for null pointer.
>
> A null pointer for the return type means either 1) the call has a
> cast or 2) an error has occured.
> You can see this in infcall.c:call_function_by_hand_dummy():
>
> CORE_ADDR funaddr = find_function_addr (function, &values_type, &ftype);
>
> if (values_type == NULL)
> values_type = default_return_type;
> if (values_type == NULL)
> {
> const char *name = get_function_name (funaddr,
> name_buf, sizeof (name_buf));
> error (_("'%s' has unknown return type; "
> "cast the call to its declared return type"),
> name);
> }
>
> In aarch64_push_dummy_call we do not have default_return_type, so cannot
> determine between the two cases.
>
> (In addition, aarch64_push_dummy_call incorrectly resolves the return
> type for IFUNC).
>
>
> However, aarch64_push_dummy_call only requires the return value in order
> to calculate lang_struct_return ... which has previously been calculated
> in the caller:
>
> This is slightly awkward, ideally the flag "lang_struct_return"
> would be passed to the targets implementation of push_dummy_call.
> Rather that change the target interface we call the language code
> directly ourselves.
>
> The fix is simple:
> Patch 1: Update gdbarch interface to pass lang_struct_return.
> Patch 2: Remove incorrect code and use the passed in lang_struct_return.
>
> Built on x86 target all build and tested on aarch64.
>
>
>
> Alan Hayward (2):
> Add lang_struct_return to _push_dummy_call
> Aarch64: Fix segfault when casting dummy calls
>
> gdb/aarch64-tdep.c | 29 +-----
> gdb/alpha-tdep.c | 3 +-
> gdb/amd64-tdep.c | 3 +-
> gdb/amd64-windows-tdep.c | 3 +-
> gdb/arc-tdep.c | 2 +-
> gdb/arm-tdep.c | 2 +-
> gdb/avr-tdep.c | 3 +-
> gdb/bfin-tdep.c | 1 +
> gdb/cris-tdep.c | 3 +-
> gdb/csky-tdep.c | 3 +-
> gdb/frv-tdep.c | 3 +-
> gdb/gdbarch.c | 4 +-
> gdb/gdbarch.h | 4 +-
> gdb/gdbarch.sh | 2 +-
> gdb/h8300-tdep.c | 3 +-
> gdb/hppa-tdep.c | 6 +-
> gdb/i386-darwin-tdep.c | 3 +-
> gdb/i386-tdep.c | 2 +-
> gdb/ia64-tdep.c | 3 +-
> gdb/infcall.c | 3 +-
> gdb/iq2000-tdep.c | 3 +-
> gdb/lm32-tdep.c | 3 +-
> gdb/m32c-tdep.c | 2 +-
> gdb/m32r-tdep.c | 2 +-
> gdb/m68hc11-tdep.c | 3 +-
> gdb/m68k-tdep.c | 2 +-
> gdb/mep-tdep.c | 2 +-
> gdb/mips-tdep.c | 15 +--
> gdb/mn10300-tdep.c | 1 +
> gdb/msp430-tdep.c | 3 +-
> gdb/nds32-tdep.c | 3 +-
> gdb/nios2-tdep.c | 3 +-
> gdb/or1k-tdep.c | 3 +-
> gdb/ppc-sysv-tdep.c | 6 +-
> gdb/ppc-tdep.h | 2 +
> gdb/riscv-tdep.c | 1 +
> gdb/rl78-tdep.c | 3 +-
> gdb/rs6000-aix-tdep.c | 3 +-
> gdb/rs6000-lynx178-tdep.c | 3 +-
> gdb/rx-tdep.c | 2 +-
> gdb/s390-tdep.c | 3 +-
> gdb/score-tdep.c | 3 +-
> gdb/sh-tdep.c | 2 +
> gdb/sparc-tdep.c | 3 +-
> gdb/sparc64-tdep.c | 3 +-
> gdb/spu-tdep.c | 3 +-
> gdb/testsuite/gdb.base/condbreak-solib-lib.cc | 21 +++++
> .../gdb.base/condbreak-solib-main.cc | 33 +++++++
> gdb/testsuite/gdb.base/condbreak-solib.exp | 93 +++++++++++++++++++
> gdb/tic6x-tdep.c | 3 +-
> gdb/tilegx-tdep.c | 1 +
> gdb/v850-tdep.c | 1 +
> gdb/vax-tdep.c | 2 +-
> gdb/xstormy16-tdep.c | 1 +
> gdb/xtensa-tdep.c | 1 +
> 55 files changed, 246 insertions(+), 77 deletions(-)
> create mode 100644 gdb/testsuite/gdb.base/condbreak-solib-lib.cc
> create mode 100644 gdb/testsuite/gdb.base/condbreak-solib-main.cc
> create mode 100644 gdb/testsuite/gdb.base/condbreak-solib.exp
>
> --
> 2.17.1 (Apple Git-112)
>
next prev parent reply other threads:[~2018-10-09 8:26 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-10-01 15:53 [PATCH " Alan Hayward
2018-10-01 15:53 ` [PATCH v2 2/2] " Alan Hayward
2018-10-09 16:15 ` Pedro Alves
2018-10-01 15:53 ` [PATCH v2 1/2] Add lang_struct_return to _push_dummy_call Alan Hayward
2018-10-09 16:14 ` Pedro Alves
2018-10-10 11:54 ` Alan Hayward
2018-10-09 8:26 ` Alan Hayward [this message]
2018-10-09 16:10 ` [PATCH v2 0/2] Aarch64: Fix segfault when casting dummy calls Pedro Alves
2018-10-09 17:50 ` Alan Hayward
2018-10-10 8:23 ` Pedro Alves
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=3DF26781-3D5E-4A95-9427-7369E2DEFA01@arm.com \
--to=alan.hayward@arm.com \
--cc=gdb-patches@sourceware.org \
--cc=nd@arm.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox