linux-kernel - Re: Segfault running a binary in a compressed folder

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <b65e13cd-7479-4146-a6bf-cdbadab6795a@paragon-software.com>
Date: Wed, 19 Jun 2024 10:54:51 +0300
From: Konstantin Komarov <almaz.alexandrovich@...agon-software.com>
To: Linux regressions mailing list <regressions@...ts.linux.dev>, Giovanni
 Santini <giovannisantini93@...oo.it>
CC: <ntfs3@...ts.linux.dev>, LKML <linux-kernel@...r.kernel.org>,
	<stable@...r.kernel.org>
Subject: Re: Segfault running a binary in a compressed folder

On 11.06.2024 14:15, Linux regression tracking (Thorsten Leemhuis) wrote:
> On 11.06.24 12:55, Giovanni Santini wrote:
>> Hi Thorsten, nice to chat again!
> :-D
>
>> I am sorry for the lack of information,
> Happens.
>
>> this is my second bug report to
>> the kernel; the first one was via Bugzilla and I filled more information.
>>
>> Now, the missing information is:
>>
>> OS: ArchLinux
>>
>> Tested kernels: both latest Linux stable (6.9.3) and mainline (6.10rc3)
>>
>> Regression: no, I believe that this issue has been present forever.
> Thx. Okay, in that case anyone that replies in this thread consider
> dropping the stable and the regression lists to avoid confusion and
> spare the subscribes of those lists a few cycles.
>
>> I realized it may have been compression-related only recently.
>> I do remember testing ntfs3 long ago and having the same issues with a
>> Ruby vendoring folder.
>>
>> Please let me know if you need more information!
> That's up to Konstantin (or others on the ntfs3 list), who is known to
> sometimes reply quickly, while other times only replies after quite a
> while. We'll see what it will be here. :-D
>
> Ciao, Thorsten
>
>> On 2024-06-11 12:04, Linux regression tracking (Thorsten Leemhuis) wrote:
>>> On 11.06.24 11:19, Giovanni Santini wrote:
>>>> I am writing to report the issue mentioned in the subject.
>>>>
>>>> Essentially, when running an executable from a compressed folder in an
>>>> NTFS partition mounted via ntfs3 I get a segfault.
>>>>
>>>> The error line I get in dmesg is:
>>>>
>>>> ntfs3: nvme0n1p5: ino=c3754, "hello" mmap(write) compressed not
>>>> supported
>>>>
>>>> I've attached a terminal script where I show my source, Makefile and how
>>>> the error appears.
>>> You CCed the regression and the stable list, but that looks odd, as you
>>> don't even mention which kernel version you used (or which worked).
>>> Could you clarify? And ideally state if mainline (e.g. 6.10-rc3) is
>>> affected as well, as the answer to the question "who is obliged to look
>>> into this" depends on it.
>>>
>>> Ciao, Thorsten

Hi Giovanni,

ntfs3 currently does not support mmap write to compressed files,
as indicated by the driver with

   "mmap(write) compressed not supported."

We are aware of this limitation and will implement this functionality in
the future.

Regards, Konstantin