Re: [PATCH] s390/zcrypt: optimize memory allocation in online_show()

[Yunseong Kim <yskelg@gmail.com>]

Hi Dan,

On 7/9/24 12:55 오전, Dan Carpenter wrote:
> Hi,
> 
> kernel test robot noticed the following build warnings:
> 
> https://git-scm.com/docs/git-format-patch#_base_tree_information]
> 
> url:    https://github.com/intel-lab-lkp/linux/commits/yskelg-gmail-com/s390-zcrypt-optimize-memory-allocation-in-online_show/20240626-071004
> base:   https://git.kernel.org/pub/scm/linux/kernel/git/s390/linux.git features
> patch link:    https://lore.kernel.org/r/20240623120147.35554-3-yskelg%40gmail.com
> patch subject: [PATCH] s390/zcrypt: optimize memory allocation in online_show()
> config: s390-randconfig-r081-20240707 (https://download.01.org/0day-ci/archive/20240707/202407071714.4AUEoeUD-lkp@intel.com/config)
> compiler: s390-linux-gcc (GCC) 13.2.0
> 
> If you fix the issue in a separate patch/commit (i.e. not just a new version of
> the same patch/commit), kindly add following tags
> | Reported-by: kernel test robot <lkp@intel.com>
> | Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
> | Closes: https://lore.kernel.org/r/202407071714.4AUEoeUD-lkp@intel.com/
> 
> smatch warnings:
> drivers/s390/crypto/zcrypt_card.c:103 online_store() warn: iterator used outside loop: 'zq'
> 
> vim +/zq +103 drivers/s390/crypto/zcrypt_card.c

Thank you for the review. I decided not to work on my patch anymore
after the review because my patch was problematic and not useful.

> df6f508c68dbc6 Harald Freudenberger 2021-04-13   90  	list_for_each_entry(zq, &zc->zqueues, list)
> 2ff6be56a27c2d Yunseong Kim         2024-06-23   91  		if (!!zq->online != !!online)
> df6f508c68dbc6 Harald Freudenberger 2021-04-13   92  			maxzqs++;
> df6f508c68dbc6 Harald Freudenberger 2021-04-13   93  	if (maxzqs > 0)
> 2ff6be56a27c2d Yunseong Kim         2024-06-23   94  		zq_uelist = kcalloc(maxzqs, sizeof(*zq_uelist), GFP_ATOMIC);
> e28d2af43614eb Ingo Tuchscherer     2016-08-25   95  	list_for_each_entry(zq, &zc->zqueues, list)
> df6f508c68dbc6 Harald Freudenberger 2021-04-13   96  		if (zcrypt_queue_force_online(zq, online))
> df6f508c68dbc6 Harald Freudenberger 2021-04-13   97  			if (zq_uelist) {
> df6f508c68dbc6 Harald Freudenberger 2021-04-13   98  				zcrypt_queue_get(zq);
> df6f508c68dbc6 Harald Freudenberger 2021-04-13   99  				zq_uelist[i++] = zq;
> df6f508c68dbc6 Harald Freudenberger 2021-04-13  100  			}
> e28d2af43614eb Ingo Tuchscherer     2016-08-25  101  	spin_unlock(&zcrypt_list_lock);
> 2ff6be56a27c2d Yunseong Kim         2024-06-23  102  	while (i--) {
> df6f508c68dbc6 Harald Freudenberger 2021-04-13 @103  		ap_send_online_uevent(&zq->queue->ap_dev, online);
>                                                                                        ^^
> zq is an invalid pointer here.  It's an offset off the list head.
> There used to be a "zq = zq_uelist[i];" before this function call but
> the patch deleted it.

My patch doesn't mean much to me at this point, but if it needed to be
fixed, I think it should have been.

> 2ff6be56a27c2d Yunseong Kim         2024-06-23  104  		zcrypt_queue_put(zq_uelist[i]);
> df6f508c68dbc6 Harald Freudenberger 2021-04-13  105  	}
> df6f508c68dbc6 Harald Freudenberger 2021-04-13  106  	kfree(zq_uelist);
> df6f508c68dbc6 Harald Freudenberger 2021-04-13  107  
> e28d2af43614eb Ingo Tuchscherer     2016-08-25  108  	return count;
> e28d2af43614eb Ingo Tuchscherer     2016-08-25  109  }


I've been doing some Linux kernel research on the IBM Cloud s390x VPC
and running syzkaller and Linux kernel tests tools to send out useful
patches. :)


Warm regards,

Yunseong kim