ACK/Cmnt: [PATCH 0/4][SRU F] Update Hisilicon crypto driver to use one wq

Stefan Bader stefan.bader at
Mon Mar 14 13:52:01 UTC 2022

On 02.03.22 09:53, Ike Panhc wrote:
> BugLink:
> BugLink:
> After checkbox CPU stress test, it takes 2min 30sec to login.
> Checks system resources and find out there are more then 900000
> processes in output of `ps aux` and most of them are [hisi_qm].
> In order to fix this issue, we need to backport patch for adding
> workqueue and using 1 wq per qm. Patch "Unify hardware error
> init/uninit into QM" is pulled in because of patch dependency.
> Another issue is that resetting crypto module can not work. To fix
> this issue we need to backport patch "Update SEC init and reset".
> Longfang Liu (1):
>    crypto: hisilicon/sec2 - update SEC initialization and reset
> Shukun Tan (2):
>    crypto: hisilicon - Unify hardware error init/uninit into QM
>    crypto: hisilicon - Use one workqueue per qm instead of per qp
> Ye Kai (1):
>    crypto: hisilicon/sec2 - Add workqueue for SEC driver.
>   drivers/crypto/hisilicon/hpre/hpre_main.c |  33 +++---
>   drivers/crypto/hisilicon/qm.c             | 125 +++++++++++++-------
>   drivers/crypto/hisilicon/qm.h             |  25 +++-
>   drivers/crypto/hisilicon/sec2/sec_main.c  | 138 ++++++++++++----------
>   drivers/crypto/hisilicon/zip/zip_main.c   |  58 ++++-----
>   6 files changed, 226 insertions(+), 155 deletions(-)

Like Tim I would like to note that it is not a good idea to have two bug reports 
mingled into a single submission. If those are independent, then better submit 
as two individual sets. If those are rather the same, then duplicate one against 
the other. Chances are just high that something goes wrong with this where not 
all of the patches have all the bugs linked. And if it is only an odd layout of 
the changelog...

Acked-by: Stefan Bader <stefan.bader at>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: OpenPGP_signature
Type: application/pgp-signature
Size: 833 bytes
Desc: OpenPGP digital signature
URL: <>

More information about the kernel-team mailing list