netdev - Re: [PATCH] vDPA/ifcvf: allow userspace to suspend a queue

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: <f224fc38-327c-6dac-d2cb-93f8b4ac5b82@intel.com>
Date:   Thu, 21 Apr 2022 19:05:50 +0800
From:   "Zhu, Lingshan" <lingshan.zhu@...el.com>
To:     Zhou Furong <furong.zhou@...ux.intel.com>, jasowang@...hat.com,
        mst@...hat.com
Cc:     virtualization@...ts.linux-foundation.org, netdev@...r.kernel.org
Subject: Re: [PATCH] vDPA/ifcvf: allow userspace to suspend a queue



On 4/12/2022 2:55 PM, Zhou Furong wrote:
> Hi,
>
>> +bool ifcvf_get_vq_ready(struct ifcvf_hw *hw, u16 qid)
>> +{
>> +    struct virtio_pci_common_cfg __iomem *cfg = hw->common_cfg;
>> +    bool queue_enable;
>> +
>> +    vp_iowrite16(qid, &cfg->queue_select);
>> +    queue_enable = vp_ioread16(&cfg->queue_enable);
>> +
>> +    return (bool)queue_enable;
> queue_enable is bool, why cast? looks like remove the variable is better.
> return vp_ioread16(&cfg->queue_enable);
Hi, thanks for your comments. This cast tries to make some static 
checkers happy.
Please correct me if I misunderstood it:
vp_ioread16() returns an u16, and bool is not an one bit variable, it is 
C99 _Bool.
C99 defines _Bool to be true(expends to int 1) or false(expends to int 0).
I think this implies a bool is actually capable to store an u16.
so I am afraid some checkers may complain about "queue_enable = 
vp_ioread16(&cfg->queue_enable);",
a cast here can help us silence the checkers.

see http://www.open-std.org/jtc1/sc22/wg14/www/docs/n1256.pdf section 7.16
and https://www.kernel.org/doc/Documentation/process/coding-style.rst 
section 17

Thanks,
Zhu Lingshan
>
>>   static bool ifcvf_vdpa_get_vq_ready(struct vdpa_device *vdpa_dev, 
>> u16 qid)
>>   {
>>       struct ifcvf_hw *vf = vdpa_to_vf(vdpa_dev);
>> +    bool ready;
>>   -    return vf->vring[qid].ready;
>> +    ready = ifcvf_get_vq_ready(vf, qid);
>> +
>> +    return ready;
> remove ready looks better
> return ifcvf_get_vq_ready(vf, qid);
>
>
> Best regards,
> Furong