** Description changed:
SRU Justification
[Impact]
- * Microsoft requested additional MANA patches for 5.15 to bring
- important fixes from 6.8 and newer.
+ * Microsoft requested additional MANA patches for 5.15 to bring important
fixes from 6.8 and newer.
+ * Requires previous commit ("net: mana: Fix oversized sge0 for GSO packets")
to be reverted and re-applied on top, as the remainder of the patchset would
cause incorrect code addition during cherry-pick
[Fix]
* 11 clean cherry-picks from upstream
72fc81e5bbce: "scsi: storvsc: Set correct data length for sending SCSI
command without payload"
368809a586bd: "net: mana: Add get_link and get_link_ksettings in ethtool"
9480d2fac80b: "net: mana: Cleanup "mana" debugfs dir after cleanup of all
children"
232ffe595480: "net: netvsc: Update default VMBus channels"
51c617891b64: "net: mana: Fix the extra HZ in mana_hwc_send_request"
77b356028961: "net: mana: Configure hwc timeout from hardware"
77be12274124: "net: mana: Fix perf regression: remove rx_cqes, tx_cqes
counters"
87b01cb815c7: "net: mana: Fix possible double free in error handling path"
12d56dcd1fec: "net: mana: Increase the DEF_RX_BUFFERS_PER_QUEUE to 1024"
ac30ab1488f8: "RDMA/mana_ib: query device capabilities"
2948a31f3f4f: "RDMA/mana_ib: register RDMA device with GDMA"
- * 4 backports from upstream
+ * 5 backports from upstream
+ a43e8e9ffa0d: "net: mana: Fix oversized sge0 for GSO packets"
8942bba12fbb: "net: mana: Enable debugfs files for MANA device"
3d133ab4a367: "net: mana: Add gdma stats to ethtool output for mana"
7f4df1601556: "net: mana: Add new MANA VF performance counters for easier
troubleshooting"
6ce8eb462c1b: "net: mana: Implement get_ringparam/set_ringparam for mana"
+
+ * 1 revert (from jammy:linux-azure)
+ ce02deecee38: Revert "net: mana: Fix oversized sge0 for GSO packets"
[Test Plan]
* Compile tested
* Boot tested on Standard_A2_v2 instance
* Microsoft tested successfully
[Regression potential]
* Several of the backports took a bit of effort. For example, 5.15 doesn't
have 7462494408cd: "ethtool: extend ringparam setting/getting API with
rx_buf_len" so the mana_ethtool changes requires function parameter
modifications to match 5.15. Additional, 5.15 doesn't have 504148fedb85: "net:
add skb_[inner_]tcp_all_headers helpers", so previous arithmetic had to be
re-implemented. There were a couple of other places where previous functions
were maintained rather than bringing in unknown context from the commit.
* Most changes were to MANA related source code, so leaning heavily on
Microsoft to test there, but it is localized. MANA changes could lead to
network instability.
* Other changes to scsi and hyperv were small and clean cherry-picks from
upstream, so expect little regression risk there.
[Other info]
* SF#00408239
* SF#00406456
* SF#00407579
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux-azure in Ubuntu.
https://bugs.launchpad.net/bugs/2103651
Title:
Additional Mana Patch Requests for 5.15 Kernel
Status in linux-azure package in Ubuntu:
New
Status in linux-azure source package in Jammy:
In Progress
Bug description:
SRU Justification
[Impact]
* Microsoft requested additional MANA patches for 5.15 to bring important
fixes from 6.8 and newer.
* Requires previous commit ("net: mana: Fix oversized sge0 for GSO packets")
to be reverted and re-applied on top, as the remainder of the patchset would
cause incorrect code addition during cherry-pick
[Fix]
* 11 clean cherry-picks from upstream
72fc81e5bbce: "scsi: storvsc: Set correct data length for sending SCSI
command without payload"
368809a586bd: "net: mana: Add get_link and get_link_ksettings in ethtool"
9480d2fac80b: "net: mana: Cleanup "mana" debugfs dir after cleanup of all
children"
232ffe595480: "net: netvsc: Update default VMBus channels"
51c617891b64: "net: mana: Fix the extra HZ in mana_hwc_send_request"
77b356028961: "net: mana: Configure hwc timeout from hardware"
77be12274124: "net: mana: Fix perf regression: remove rx_cqes, tx_cqes
counters"
87b01cb815c7: "net: mana: Fix possible double free in error handling path"
12d56dcd1fec: "net: mana: Increase the DEF_RX_BUFFERS_PER_QUEUE to 1024"
ac30ab1488f8: "RDMA/mana_ib: query device capabilities"
2948a31f3f4f: "RDMA/mana_ib: register RDMA device with GDMA"
* 5 backports from upstream
a43e8e9ffa0d: "net: mana: Fix oversized sge0 for GSO packets"
8942bba12fbb: "net: mana: Enable debugfs files for MANA device"
3d133ab4a367: "net: mana: Add gdma stats to ethtool output for mana"
7f4df1601556: "net: mana: Add new MANA VF performance counters for easier
troubleshooting"
6ce8eb462c1b: "net: mana: Implement get_ringparam/set_ringparam for mana"
* 1 revert (from jammy:linux-azure)
ce02deecee38: Revert "net: mana: Fix oversized sge0 for GSO packets"
[Test Plan]
* Compile tested
* Boot tested on Standard_A2_v2 instance
* Microsoft tested successfully
[Regression potential]
* Several of the backports took a bit of effort. For example, 5.15 doesn't
have 7462494408cd: "ethtool: extend ringparam setting/getting API with
rx_buf_len" so the mana_ethtool changes requires function parameter
modifications to match 5.15. Additional, 5.15 doesn't have 504148fedb85: "net:
add skb_[inner_]tcp_all_headers helpers", so previous arithmetic had to be
re-implemented. There were a couple of other places where previous functions
were maintained rather than bringing in unknown context from the commit.
* Most changes were to MANA related source code, so leaning heavily on
Microsoft to test there, but it is localized. MANA changes could lead to
network instability.
* Other changes to scsi and hyperv were small and clean cherry-picks from
upstream, so expect little regression risk there.
[Other info]
* SF#00408239
* SF#00406456
* SF#00407579
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux-azure/+bug/2103651/+subscriptions
--
Mailing list: https://launchpad.net/~kernel-packages
Post to : kernel-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kernel-packages
More help : https://help.launchpad.net/ListHelp
