linux-kernel - Re: [PATCH] [SCSI] sg: late O

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: Windows password security audit tool. GUI, reports in PDF.

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

Message-ID: <5265037B.4050108@oracle.com>
Date:	Mon, 21 Oct 2013 18:35:39 +0800
From:	Vaughan Cao <vaughan.cao@...cle.com>
To:	dgilbert@...erlog.com, Bart Van Assche <bvanassche@....org>,
	SCSI development list <linux-scsi@...r.kernel.org>,
	Madper Xie <cxie@...hat.com>,
	James Bottomley <james.bottomley@...senpartnership.com>
CC:	linux-kernel <linux-kernel@...r.kernel.org>, vaughan.cao@...cle.com
Subject: Re: [PATCH] [SCSI] sg: late O_EXCL fix for lk 3.12-rc


On 2013年10月21日 07:00, Douglas Gilbert wrote:
> On 13-10-20 01:31 PM, Bart Van Assche wrote:
>> On 10/20/13 18:09, Douglas Gilbert wrote:
>>> Given that lk 3.12.0 release is not far away, the safest path
>>> may still be to revert Vaughan Cao's patch. I'll leave that
>>> decision to the maintainers.
>>
>> Hello Doug,
>>
>> Thanks for looking into this. But I would appreciate it if you could 
>> address the
>> whitespace errors reported by checkpatch:
>>
>> ERROR: space prohibited after that '!' (ctx:BxW)
>> #24: FILE: drivers/scsi/sg.c:241:
>> + (excl_case ? (! sdp->exclude) : sfds_list_empty(sdp))));
>> ^
>>
>> ERROR: space prohibited after that '!' (ctx:BxW)
>> #55: FILE: drivers/scsi/sg.c:289:
>> + if (! alone) {
>> ^
>>
>> ERROR: code indent should use tabs where possible
>> #59: FILE: drivers/scsi/sg.c:292:
>> + }$
>>
>> WARNING: please, no spaces at the start of a line
>> #59: FILE: drivers/scsi/sg.c:292:
>> + }$
>>
>> ERROR: space prohibited after that '!' (ctx:BxW)
>> #73: FILE: drivers/scsi/sg.c:301:
>> + while (! alone) {
>> ^
>>
>> WARNING: suspect code indent for conditional statements (8, 12)
>> #144: FILE: drivers/scsi/sg.c:375:
>> + if (excl || sfds_list_empty(sdp))
>> + wake_up_interruptible(&sdp->open_wait);
>>
>
> I'd prefer people to test the patch or find logical flaws.
>
> Doug Gilbert
>
Hi Doug,

Will the lines below conflict with the meaning of NONBLOCK?
 >+ down(&sdp->or_sem);
 >+ alone = sfds_list_empty(sdp);
 > if (!((flags & O_NONBLOCK) ||
 > scsi_block_when_processing_errors(sdp->device))) {

Assume one thread holds the or_sem and waiting in 
scsi_block_when_processing_errors for the underlying scsi device to 
complete error recovery,
another thread with O_NONBLOCK call sg_open().

I'm also curious why we can skip checking _processing_errors() when 
called with O_NONBLOCK?...Though it has been there from the very beginning.
In other words, since scsi device may go into a error recovery state at 
random time, why we only check here?

Thanks,
Vaughan

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/