RHEL Engineering is moving the tracking of its product development work on RHEL 6 through RHEL 9 to Red Hat Jira (issues.redhat.com). If you're a Red Hat customer, please continue to file support cases via the Red Hat customer portal. If you're not, please head to the "RHEL project" in Red Hat Jira and file new tickets here. Individual Bugzilla bugs in the statuses "NEW", "ASSIGNED", and "POST" are being migrated throughout September 2023. Bugs of Red Hat partners with an assigned Engineering Partner Manager (EPM) are migrated in late September as per pre-agreed dates. Bugs against components "kernel", "kernel-rt", and "kpatch" are only migrated if still in "NEW" or "ASSIGNED". If you cannot log in to RH Jira, please consult article #7032570. That failing, please send an e-mail to the RH Jira admins at rh-issues@redhat.com to troubleshoot your issue as a user management inquiry. The email creates a ServiceNow ticket with Red Hat. Individual Bugzilla bugs that are migrated will be moved to status "CLOSED", resolution "MIGRATED", and set with "MigratedToJIRA" in "Keywords". The link to the successor Jira issue will be found under "Links", have a little "two-footprint" icon next to it, and direct you to the "RHEL project" in Red Hat Jira (issue links are of type "https://issues.redhat.com/browse/RHEL-XXXX", where "X" is a digit). This same link will be available in a blue banner at the top of the page informing you that that bug has been migrated.
Bug 2045594 - NIC driver (st_gmac) requires 'swiotlb=65536' on kernel command line
Summary: NIC driver (st_gmac) requires 'swiotlb=65536' on kernel command line
Keywords:
Status: CLOSED ERRATA
Alias: None
Product: Red Hat Enterprise Linux 9
Classification: Red Hat
Component: kernel
Version: 9.0
Hardware: x86_64
OS: Linux
unspecified
high
Target Milestone: rc
: 9.0
Assignee: Mark Salter
QA Contact: Dipali
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2022-01-25 18:03 UTC by Mark Salter
Modified: 2022-05-17 15:59 UTC (History)
13 users (show)

Fixed In Version: kernel-5.14.0-70.el9
Doc Type: If docs needed, set a value
Doc Text:
Clone Of: 1989540
Environment:
Last Closed: 2022-05-17 15:44:56 UTC
Type: Bug
Target Upstream Version:
Embargoed:
pm-rhel: mirror+

Attachments (Terms of Use)


Links
System ID Private Priority Status Summary Last Updated
Gitlab redhat/centos-stream/src/kernel centos-stream-9 merge_requests 484 0 None None None 2022-02-05 16:55:34 UTC
Red Hat Issue Tracker RHELPLAN-109653 0 None None None 2022-01-25 18:51:11 UTC
Red Hat Product Errata RHBA-2022:3907 0 None None None 2022-05-17 15:45:37 UTC

Description Mark Salter 2022-01-25 18:03:22 UTC 
+++ This bug was initially created as a clone of Bug #1989540 +++

Description of problem:

Without adding 'swiotlb=65536' on certain Intel platforms neither RHEL or an
upstream kernel is able to bring both stmmac interfaces online.  The following
errors are noted:

[   14.858941] intel-eth-pci 0000:00:1d.1: swiotlb buffer is full (sz: 346 bytes), total 32768 (slots), used 32768
 (slots)
[   14.869921] intel-eth-pci 0000:00:1d.1 enp0s29f1: Tx DMA map failed
[   14.879022] intel-eth-pci 0000:00:1d.1: swiotlb buffer is full (sz: 90 bytes), total 32768 (slots), used 32768 
(slots)
[   14.889909] intel-eth-pci 0000:00:1d.1 enp0s29f1: Tx DMA map failed
[   14.891756] intel-eth-pci 0000:00:1d.2 enp0s29f2: Link is Up - 100Mbps/Full - flow control rx/tx
[   14.933469] intel-eth-pci 0000:00:1d.2: swiotlb buffer is full (sz: 346 bytes), total 32768 (slots), used 32768
 (slots)
[   14.944454] intel-eth-pci 0000:00:1d.2 enp0s29f2: Tx DMA map failed
[   14.953957] intel-eth-pci 0000:00:1d.2: swiotlb buffer is full (sz: 90 bytes), total 32768 (slots), used 32768 

<snip>

Version-Release number of selected component (if applicable):

RHEL8.5, upstream 5.13


How reproducible:

100%

Steps to Reproduce:
1. Boot system


There is an upstream fix for this issue, so we should bring it into RHEL9:

884d2b845477 net: stmmac: Add GFP_DMA32 for rx buffers if no 64 capability
Comment 5 Dipali 2022-02-24 19:47:09 UTC 
verified fix on kernel 5.14.0-55.mr484_220205_1705.el9.x86_64. 

#dmesg 
no errors are seen for swiotlb.
Comment 11 Dipali 2022-02-28 19:30:46 UTC 
@David Acari - I wanted to use intel-elkhartlake-02.ml3.eng.bos.redhat.com for verifying this bug. Can you set it up ?

Thanks.
-Dipali
Comment 13 Dipali 2022-03-01 16:14:51 UTC 
Verified kernel 5.14.0-70.el9.x86_64 for no swiotlb errors in dmesg

cat /proc/cmdline
BOOT_IMAGE=(hd0,gpt2)/vmlinuz-5.14.0-70.el9.x86_64 root=/dev/mapper/cs_intel--elkhartlake--02-root ro rd.lvm.lv=cs_intel-elkhartlake-02/root rd.lvm.lv=cs_intel-elkhartlake-02/swap console=ttyS0,115200n81


# dmesg | grep swiotlb
#


[root@intel-elkhartlake-02 ~]# ip addr
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host 
       valid_lft forever preferred_lft forever
2: enp0s29f1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP group default qlen 1000
    link/ether 98:4f:ee:16:11:99 brd ff:ff:ff:ff:ff:ff
    inet 10.19.182.147/21 brd 10.19.183.255 scope global dynamic noprefixroute enp0s29f1
       valid_lft 60214sec preferred_lft 60214sec
    inet6 2620:52:0:13b0:9a4f:eeff:fe16:1199/64 scope global dynamic noprefixroute 
       valid_lft 2591880sec preferred_lft 604680sec
    inet6 fe80::9a4f:eeff:fe16:1199/64 scope link noprefixroute 
       valid_lft forever preferred_lft forever
3: enp0s29f2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP group default qlen 1000
    link/ether 98:4f:ee:16:11:9a brd ff:ff:ff:ff:ff:ff
    inet 10.19.182.106/21 brd 10.19.183.255 scope global dynamic noprefixroute enp0s29f2
       valid_lft 72075sec preferred_lft 72075sec
    inet6 2620:52:0:13b0:9a4f:eeff:fe16:119a/64 scope global dynamic noprefixroute 
       valid_lft 2591880sec preferred_lft 604680sec
    inet6 fe80::9a4f:eeff:fe16:119a/64 scope link noprefixroute 
       valid_lft forever preferred_lft forever
4: enp0s30f4: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP group default qlen 1000
    link/ether 98:4f:ee:16:11:9b brd ff:ff:ff:ff:ff:ff
    inet 10.19.177.19/21 brd 10.19.183.255 scope global dynamic noprefixroute enp0s30f4
       valid_lft 17012sec preferred_lft 17012sec
    inet6 2620:52:0:13b0:9a4f:eeff:fe16:119b/64 scope global dynamic noprefixroute 
       valid_lft 2591880sec preferred_lft 604680sec
    inet6 fe80::9a4f:eeff:fe16:119b/64 scope link noprefixroute 
       valid_lft forever preferred_lft forever

No errors are seen, setting to Verified
Comment 15 errata-xmlrpc 2022-05-17 15:44:56 UTC 
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory (new packages: kernel), and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHBA-2022:3907
Note You need to log in before you can comment on or make changes to this bug.