netdev - Re: [PATCH net-next 04/12] ibmvnic: Introduce xmit_more support using batched subCRQ hcalls

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]

Date:   Mon, 16 Nov 2020 12:40:57 -0600
From:   Thomas Falcon <tlfalcon@...ux.ibm.com>
To:     Jakub Kicinski <kuba@...nel.org>
Cc:     netdev@...r.kernel.org, linuxppc-dev@...ts.ozlabs.org,
        dnbanerg@...ibm.com, brking@...ux.vnet.ibm.com, pradeep@...ibm.com,
        drt@...ux.vnet.ibm.com, sukadev@...ux.vnet.ibm.com,
        ljp@...ux.vnet.ibm.com, cforno12@...ux.ibm.com,
        ricklind@...ux.ibm.com
Subject: Re: [PATCH net-next 04/12] ibmvnic: Introduce xmit_more support using
 batched subCRQ hcalls


On 11/14/20 5:46 PM, Jakub Kicinski wrote:
> On Thu, 12 Nov 2020 13:09:59 -0600 Thomas Falcon wrote:
>> Include support for the xmit_more feature utilizing the
>> H_SEND_SUB_CRQ_INDIRECT hypervisor call which allows the sending
>> of multiple subordinate Command Response Queue descriptors in one
>> hypervisor call via a DMA-mapped buffer. This update reduces hypervisor
>> calls and thus hypervisor call overhead per TX descriptor.
>>
>> Signed-off-by: Thomas Falcon <tlfalcon@...ux.ibm.com>
> The common bug with xmit_more is not flushing the already queued
> notifications when there is a drop. Any time you drop a skb you need
> to check it's not an skb that was the end of an xmit_more train and
> if so flush notifications (or just always flush on error).
>
> Looking at the driver e.g. this starting goto:
>
>          if (ibmvnic_xmit_workarounds(skb, netdev)) {
>                  tx_dropped++;
>                  tx_send_failed++;
>                  ret = NETDEV_TX_OK;
>                  goto out;
>          }
>
> Does not seem to hit any flush on its way out AFAICS.

Hi, I included those updates in a later patch to ease review but see now 
that that was a mistake. I will merge those bits back into this patch 
and resubmit.

Thanks!