seg fault on missing dwarf
Commit Message
Pedro,
this is Alex's patch of
https://sourceware.org/ml/gdb-patches/2017-09/msg00745.html cleaned up
with changelog formatting. Alex is a fellow FaceBooker.
The patch looks obvious to me, particularly as a few lines down we have:
fprintf_unfiltered (gdb_stdlog,
" %s CUs, %s TUs\n",
pulongest (dwp_file->cus ? dwp_file->cus->nr_units : 0),
pulongest (dwp_file->tus ? dwp_file->tus->nr_units : 0));
to protect dump output from exactly this problem.
ok?
nathan
Comments
Hi Nathan,
On 09/25/2017 09:33 PM, Nathan Sidwell wrote:
> Pedro,
> this is Alex's patch of
> https://sourceware.org/ml/gdb-patches/2017-09/msg00745.html cleaned up
> with changelog formatting. Alex is a fellow FaceBooker.
FYI, Doug reviewed Alex's patch:
https://sourceware.org/ml/gdb-patches/2017-09/msg00779.html
Thanks,
Pedro Alves
On 09/26/2017 07:27 AM, Pedro Alves wrote:
> Hi Nathan,
>
> On 09/25/2017 09:33 PM, Nathan Sidwell wrote:
>> Pedro,
>> this is Alex's patch of
>> https://sourceware.org/ml/gdb-patches/2017-09/msg00745.html cleaned up
>> with changelog formatting. Alex is a fellow FaceBooker.
>
> FYI, Doug reviewed Alex's patch:
> https://sourceware.org/ml/gdb-patches/2017-09/msg00779.html
Ah, I see the state changed between me checking and actually sending
email :) Alex, do you need me to commit this on your behalf?
nathan
Many thanks to everyone for looking at my patch,
i sent the final version (updated following Doug Evans suggestion) this
morning
https://sourceware.org/ml/gdb-patches/2017-09/msg00814.html
- so yeah, i would be very grateful if smb could commit this for me.
Regarding the separate issue with the dependence on the order of sections
(unrelated to this patch) - i'm currently preparing a bug report - will
file it to bugzilla soon.
Kind regards,
Alexander Shaposhnikov
On Tue, Sep 26, 2017 at 4:05 PM, Nathan Sidwell <nathan@acm.org> wrote:
> On 09/26/2017 07:27 AM, Pedro Alves wrote:
>
>> Hi Nathan,
>>
>> On 09/25/2017 09:33 PM, Nathan Sidwell wrote:
>>
>>> Pedro,
>>> this is Alex's patch of
>>> https://sourceware.org/ml/gdb-patches/2017-09/msg00745.html cleaned up
>>> with changelog formatting. Alex is a fellow FaceBooker.
>>>
>>
>> FYI, Doug reviewed Alex's patch:
>> https://sourceware.org/ml/gdb-patches/2017-09/msg00779.html
>>
>
> Ah, I see the state changed between me checking and actually sending email
> :) Alex, do you need me to commit this on your behalf?
>
> nathan
>
> --
> Nathan Sidwell
>
2017-09-25 Alexander Shaposhnikov <alexander.v.shaposhnikov@gmail.com>
* dwarf2read.c (open_and_init_dwp_file): Protect against dwp_file
having NULL cus or tus.
@@ -11185,7 +11185,8 @@ open_and_init_dwp_file (void)
dwp_file->tus = create_dwp_hash_table (dwp_file, 1);
/* The DWP file version is stored in the hash table. Oh well. */
- if (dwp_file->cus->version != dwp_file->tus->version)
+ if (dwp_file->cus && dwp_file->tus
+ && dwp_file->cus->version != dwp_file->tus->version)
{
/* Technically speaking, we should try to limp along, but this is
pretty bizarre. We use pulongest here because that's the established
@@ -11195,7 +11196,7 @@ open_and_init_dwp_file (void)
pulongest (dwp_file->cus->version),
pulongest (dwp_file->tus->version), dwp_name.c_str ());
}
- dwp_file->version = dwp_file->cus->version;
+ dwp_file->version = dwp_file->cus ? dwp_file->cus->version : 0;
if (dwp_file->version == 2)
bfd_map_over_sections (dwp_file->dbfd, dwarf2_locate_v2_dwp_sections,