linux-kernel - Re: [PATCH v2] scsi: ufs: mcq: Limit the amount of inflight requests

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]

Date:   Mon, 3 Apr 2023 09:52:33 +0100
From:   John Garry <john.g.garry@...cle.com>
To:     Avri Altman <avri.altman@....com>,
        "James E . J . Bottomley" <jejb@...ux.vnet.ibm.com>,
        "Martin K . Petersen" <martin.petersen@...cle.com>
Cc:     Asutosh Das <quic_asutoshd@...cinc.com>, quic_cang@...cinc.com,
        linux-scsi@...r.kernel.org, linux-kernel@...r.kernel.org,
        Johannes Thumshirn <Johannes.Thumshirn@....com>,
        Bart Van Assche <bvanassche@....org>
Subject: Re: [PATCH v2] scsi: ufs: mcq: Limit the amount of inflight requests

On 31/03/2023 08:46, Avri Altman wrote:
> in UFS, each request is designated via the triplet <iid, lun, task tag>.
> 
> In UFS4.0 the Initiator ID field is 8 bits wide, comprised of the
> EXT_IID and IID fields. Together with the task tag (single byte), they
> limit the driver's hw queues capacity.
> 
> ---
> v1 -> v2:
> Attend Johannes's and Bart's comments
> 
> Signed-off-by: Avri Altman <avri.altman@....com>
> ---
>   drivers/ufs/core/ufshcd.c | 5 +++++
>   1 file changed, 5 insertions(+)
> 
> diff --git a/drivers/ufs/core/ufshcd.c b/drivers/ufs/core/ufshcd.c
> index 35a3bd95c5e4..cac7c9918c5b 100644
> --- a/drivers/ufs/core/ufshcd.c
> +++ b/drivers/ufs/core/ufshcd.c
> @@ -8468,6 +8468,11 @@ static int ufshcd_alloc_mcq(struct ufs_hba *hba)
>   	if (ret)
>   		goto err;
>   
> +	if (hba->nutrs * hba->nr_hw_queues > SZ_64K - 1) {

If shost->host_tagset is set - which it seems to be for this driver - 
then the number of HW queues would not influence how many IOs the host 
may be sent. Rather this is just limited by just the HW queue depth.

Thanks,
John

> +		dev_info(hba->dev, "there can be at most 64K inflight requests\n");
> +		goto err;
> +	}
> +
>   	/*
>   	 * Previously allocated memory for nutrs may not be enough in MCQ mode.
>   	 * Number of supported tags in MCQ mode may be larger than SDB mode.