[PATCH net-next v5 0/3] ethtool: Dynamic RSS context indirection table resizing

[PATCH net-next v5 0/3] ethtool: Dynamic RSS context indirection table resizing

[Björn Töpel]

Hi!

[ Michael and Pavan, I removed your Tested-by:/Reviewed-by: although
the changes were pretty small. Please re-add if you're still OK with
it. ]

Some NICs (e.g. bnxt) change their RSS indirection table size based on
the queue count, because the hardware table is a shared resource. The
ethtool core locks ctx->indir_size at context creation, so drivers
have to reject channel changes when RSS contexts exist.

This series adds resize helpers and wires them up in bnxt.

Patch 1 adds core helpers:
  ethtool_rxfh_can_resize() - read-only validation
  ethtool_rxfh_resize() - fold/unfold a raw table in place
  ethtool_rxfh_ctxs_can_resize() - validate all non-default contexts
  ethtool_rxfh_ctxs_resize() - resize all non-default contexts,
    with locking and RSS_NTF notifications

Both ethtool_rxfh_can_resize() and ethtool_rxfh_resize() now take a
user_size parameter: the number of indirection table entries the user
originally provided. When shrinking, the table will not fold below
this floor, preserving user intent.

The user_size is tracked in ctx->indir_user_size for non-default RSS
contexts and in dev->ethtool->rss_indir_user_size for context 0. It
is set when the indirection table is configured via netlink or ioctl,
and cleared to zero on reset-to-default.

Patch 2 uses them in bnxt_set_channels(). Validation runs before
bnxt_close_nic(); actual resize is deferred until after. RSS table
size only changes on P5 chips with older firmware.

Patch 3 adds HW tests in rss_drv.py (devices without dynamic table
sizing are skipped):
  resize_periodic - fold/unfold with a non-default [3,2,1,0]
    sub-table (user_size=4), verifying exact content preservation
    (main + ctx)
  resize_below_user_size_reject - periodic sub-table with user_size
    between big and small device table sizes; verifies that shrinking
    below user_size is rejected even when the table is periodic
    (main + ctx)
  resize_nonperiodic_reject - non-periodic table blocks channel
    reduction, with an extra periodic context to exercise
    multi-context validation (main + ctx)
  resize_nonperiodic_no_corruption - failed resize leaves table
    contents and channel count unchanged (main + ctx)

Running the tests:

  # On real hardware
  sudo NETIF=eth0 ./rss_drv.py

Changes v4 -> v5:

 - Track user-provided indirection table size (user_size) as a resize
   floor. Added indir_user_size to ethtool_rxfh_context and
   rss_indir_user_size to ethtool_netdev_state. ethtool_rxfh_can_resize()
   and ethtool_rxfh_resize() now take a user_size parameter and reject
   shrinking below it. (Jakub)
 - Propagated user_size out of rss_set_prep_indir() and stored it on
   successful set in both netlink and ioctl paths.
 - resize_periodic test now sends a 4-entry sub-table (user_size=4)
   instead of replicating to full device table size
 - Added resize_below_user_size_reject test to verify user_size floor.
 - Removed "Open items" section - user_size tracking is now implemented.

Changes v3 -> v4:

 - Rebased onto net-next
 - Added Reviewed-by: from Michael
 - Added missing Cc: to make the pwbots happier

Changes v2 -> v3:

 - Changed ethtool_rxfh_can_resize() to return bool instead of int;
   true means resize is possible, false means it is not. Inverted
   callers accordingly. (Jakub)
 - Added Tested-by from Pavan

Changes v1 -> v2:

 - Dropped netdevsim support and netdevsim selftest (Jakub)
 - Split ethtool_rxfh_contexts_resize_all() into separate validate
   (ethtool_rxfh_ctxs_can_resize) and apply (ethtool_rxfh_ctxs_resize)
   so drivers can validate before closing the device (Jakub)
 - Shortened helper names (Jakub)
 - Replaced scoped_guard(mutex) with explicit mutex_lock/unlock
   (Jakub)
 - Removed defensive zero-size check, bare expressions instead of != 0
   comparisons, ! instead of == 0 (Jakub)
 - In bnxt, moved bnxt_check_rings() before RSS validation and
   deferred actual resize to after bnxt_close_nic() (Jakub, Michael)
 - Added comment that RSS table size only changes on P5 chips with
   older firmware (Michael)
 - Use non-default [3,2,1,0]xN pattern set via netlink to distinguish
   correct fold from driver resetting to defaults (Jakub)
 - Check exact indirection table pattern, not just set(indir) (Jakub)
 - Use ksft_raises() instead of try/except/else (Jakub)
 - Removed queue_count=8 from NetDrvEnv (Jakub)
 - Added ksft_variants to resize_nonperiodic_reject for ctx coverage
 - Added extra periodic context in reject test for multi-context
   validation coverage
 - Added resize_nonperiodic_no_corruption test


Björn Töpel (3):
  ethtool: Add RSS indirection table resize helpers
  bnxt_en: Resize RSS contexts on channel count change
  selftests: rss_drv: Add RSS indirection table resize tests

 .../net/ethernet/broadcom/bnxt/bnxt_ethtool.c |  38 ++-
 include/linux/ethtool.h                       |  11 +
 net/ethtool/common.c                          | 150 +++++++++++
 net/ethtool/ioctl.c                           |  15 +-
 net/ethtool/rss.c                             |  24 +-
 .../selftests/drivers/net/hw/rss_drv.py       | 233 +++++++++++++++++-
 6 files changed, 450 insertions(+), 21 deletions(-)


base-commit: 348baefbb635cbb448e154f38c93657d4cf23936
-- 
2.53.0

[PATCH net-next v5 1/3] ethtool: Add RSS indirection table resize helpers

[Björn Töpel]

The core locks ctx->indir_size when an RSS context is created. Some
NICs (e.g. bnxt) change their indirection table size based on the
channel count, because the hardware table is a shared resource. This
forces drivers to reject channel changes when RSS contexts exist.

Add helpers to resize indirection tables:

ethtool_rxfh_can_resize() checks whether a table can be resized
without modifying it.

ethtool_rxfh_resize() resizes a raw u32 table in place. Folding
(shrink) requires the table to be periodic at the new size;
non-periodic tables are rejected. Unfolding (grow) replicates the
existing pattern. Sizes must be multiples of each other.

Both ethtool_rxfh_can_resize() and ethtool_rxfh_resize() take a
user_size parameter: the number of indirection table entries the user
originally provided. When shrinking, the table will not fold below
this floor.

ethtool_rxfh_ctxs_can_resize() validates all non-default RSS contexts
can be resized. ethtool_rxfh_ctxs_resize() applies the resize and
sends ETHTOOL_MSG_RSS_NTF per resized context after releasing
rss_lock.

No reallocation is needed because ethtool_rxfh_ctx_alloc() reserves
space for rxfh_indir_space entries, and key_off is based on that
maximum.

Signed-off-by: Björn Töpel <bjorn@kernel.org>
---
 include/linux/ethtool.h |  11 +++
 net/ethtool/common.c    | 150 ++++++++++++++++++++++++++++++++++++++++
 net/ethtool/ioctl.c     |  15 ++--
 net/ethtool/rss.c       |  24 +++++--
 4 files changed, 190 insertions(+), 10 deletions(-)

diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index 83c375840835..6283a8cf3004 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -182,10 +182,13 @@ static inline u32 ethtool_rxfh_indir_default(u32 index, u32 n_rx_rings)
  *	of %RXH_XFRM_*.
  * @indir_configured: indir has been specified (at create time or subsequently)
  * @key_configured: hkey has been specified (at create time or subsequently)
+ * @indir_user_size: number of entries the user provided for the indirection
+ *	table.
  */
 struct ethtool_rxfh_context {
 	u32 indir_size;
 	u32 key_size;
+	u32 indir_user_size;
 	u16 priv_size;
 	u8 hfunc;
 	u8 input_xfrm;
@@ -214,6 +217,11 @@ static inline u8 *ethtool_rxfh_context_key(struct ethtool_rxfh_context *ctx)
 }
 
 void ethtool_rxfh_context_lost(struct net_device *dev, u32 context_id);
+bool ethtool_rxfh_can_resize(const u32 *tbl, u32 old_size, u32 new_size,
+			     u32 user_size);
+int ethtool_rxfh_resize(u32 *tbl, u32 old_size, u32 new_size, u32 user_size);
+int ethtool_rxfh_ctxs_can_resize(struct net_device *dev, u32 new_indir_size);
+void ethtool_rxfh_ctxs_resize(struct net_device *dev, u32 new_indir_size);
 
 struct link_mode_info {
 	int	speed;
@@ -1333,12 +1341,15 @@ int ethtool_virtdev_set_link_ksettings(struct net_device *dev,
  * @rss_ctx:		XArray of custom RSS contexts
  * @rss_lock:		Protects entries in @rss_ctx.  May be taken from
  *			within RTNL.
+ * @rss_indir_user_size: number of indirection table entries the user
+ *			provided for the default (context 0) RSS table.
  * @wol_enabled:	Wake-on-LAN is enabled
  * @module_fw_flash_in_progress: Module firmware flashing is in progress.
  */
 struct ethtool_netdev_state {
 	struct xarray		rss_ctx;
 	struct mutex		rss_lock;
+	u32			rss_indir_user_size;
 	unsigned		wol_enabled:1;
 	unsigned		module_fw_flash_in_progress:1;
 };
diff --git a/net/ethtool/common.c b/net/ethtool/common.c
index e252cf20c22f..0f8b5c84aba6 100644
--- a/net/ethtool/common.c
+++ b/net/ethtool/common.c
@@ -1204,6 +1204,156 @@ void ethtool_rxfh_context_lost(struct net_device *dev, u32 context_id)
 }
 EXPORT_SYMBOL(ethtool_rxfh_context_lost);
 
+static bool rxfh_indir_is_periodic(const u32 *tbl, u32 old_size, u32 new_size)
+{
+	u32 i;
+
+	for (i = new_size; i < old_size; i++)
+		if (tbl[i] != tbl[i % new_size])
+			return false;
+	return true;
+}
+
+/**
+ * ethtool_rxfh_can_resize - Check if an indirection table can be resized
+ * @tbl: indirection table
+ * @old_size: current number of entries in the table
+ * @new_size: desired number of entries
+ * @user_size: number of entries the user configured (lower bound),
+ *
+ * Validate that @tbl can be resized from @old_size to @new_size without
+ * data loss. When shrinking, @new_size must not be smaller than @user_size.
+ * Read-only; does not modify the table.
+ *
+ * Return: true if resize is possible, false otherwise.
+ */
+bool ethtool_rxfh_can_resize(const u32 *tbl, u32 old_size, u32 new_size,
+			     u32 user_size)
+{
+	if (new_size == old_size)
+		return true;
+
+	if (new_size < old_size) {
+		if (user_size && new_size < user_size)
+			return false;
+		if (old_size % new_size)
+			return false;
+		if (!rxfh_indir_is_periodic(tbl, old_size, new_size))
+			return false;
+		return true;
+	}
+
+	if (new_size % old_size)
+		return false;
+	return true;
+}
+EXPORT_SYMBOL(ethtool_rxfh_can_resize);
+
+/* Resize without validation; caller must have called can_resize first */
+static void __ethtool_rxfh_resize(u32 *tbl, u32 old_size, u32 new_size)
+{
+	u32 i;
+
+	/* Grow: replicate existing pattern; shrink is a no-op on the data */
+	for (i = old_size; i < new_size; i++)
+		tbl[i] = tbl[i % old_size];
+}
+
+/**
+ * ethtool_rxfh_resize - Fold or unfold an indirection table
+ * @tbl: indirection table (must have room for max(old_size, new_size) entries)
+ * @old_size: current number of entries in the table
+ * @new_size: desired number of entries
+ * @user_size: number of entries the user configured (lower bound),
+ *
+ * Resize an RSS indirection table in place. When folding (shrinking),
+ * the table must be periodic with period @new_size and @new_size must
+ * not be smaller than @user_size.
+ * When unfolding (growing), the existing pattern is replicated. Both
+ * directions require the sizes to be multiples of each other.
+ *
+ * Return: 0 on success, -%EINVAL on failure (no mutation on failure).
+ */
+int ethtool_rxfh_resize(u32 *tbl, u32 old_size, u32 new_size, u32 user_size)
+{
+	if (!ethtool_rxfh_can_resize(tbl, old_size, new_size, user_size))
+		return -EINVAL;
+
+	__ethtool_rxfh_resize(tbl, old_size, new_size);
+	return 0;
+}
+EXPORT_SYMBOL(ethtool_rxfh_resize);
+
+/**
+ * ethtool_rxfh_ctxs_can_resize - Validate resize for all RSS contexts
+ * @dev: network device
+ * @new_indir_size: new indirection table size
+ *
+ * Validate that the indirection tables of all non-default RSS contexts
+ * can be resized to @new_indir_size. Read-only; does not modify any
+ * context. Intended to be paired with ethtool_rxfh_ctxs_resize().
+ *
+ * Return: 0 if all contexts can be resized, negative errno on failure.
+ */
+int ethtool_rxfh_ctxs_can_resize(struct net_device *dev,
+				 u32 new_indir_size)
+{
+	struct ethtool_rxfh_context *ctx;
+	unsigned long context;
+	int ret = 0;
+
+	if (!dev->ethtool_ops->rxfh_indir_space ||
+	    new_indir_size > dev->ethtool_ops->rxfh_indir_space)
+		return -EINVAL;
+
+	mutex_lock(&dev->ethtool->rss_lock);
+	xa_for_each(&dev->ethtool->rss_ctx, context, ctx) {
+		u32 *indir = ethtool_rxfh_context_indir(ctx);
+
+		if (!ethtool_rxfh_can_resize(indir, ctx->indir_size,
+					     new_indir_size,
+					     ctx->indir_user_size)) {
+			ret = -EINVAL;
+			goto unlock;
+		}
+	}
+unlock:
+	mutex_unlock(&dev->ethtool->rss_lock);
+	return ret;
+}
+EXPORT_SYMBOL(ethtool_rxfh_ctxs_can_resize);
+
+/**
+ * ethtool_rxfh_ctxs_resize - Resize all RSS context indirection tables
+ * @dev: network device
+ * @new_indir_size: new indirection table size
+ *
+ * Resize the indirection table of every non-default RSS context to
+ * @new_indir_size. Caller must have validated with
+ * ethtool_rxfh_ctxs_can_resize() first. An %ETHTOOL_MSG_RSS_NTF is
+ * sent for each resized context.
+ *
+ * Notifications are sent outside the RSS lock to avoid holding the
+ * mutex during notification delivery.
+ */
+void ethtool_rxfh_ctxs_resize(struct net_device *dev, u32 new_indir_size)
+{
+	struct ethtool_rxfh_context *ctx;
+	unsigned long context;
+
+	mutex_lock(&dev->ethtool->rss_lock);
+	xa_for_each(&dev->ethtool->rss_ctx, context, ctx) {
+		__ethtool_rxfh_resize(ethtool_rxfh_context_indir(ctx),
+				      ctx->indir_size, new_indir_size);
+		ctx->indir_size = new_indir_size;
+	}
+	mutex_unlock(&dev->ethtool->rss_lock);
+
+	xa_for_each(&dev->ethtool->rss_ctx, context, ctx)
+		ethtool_rss_notify(dev, ETHTOOL_MSG_RSS_NTF, context);
+}
+EXPORT_SYMBOL(ethtool_rxfh_ctxs_resize);
+
 enum ethtool_link_medium ethtool_str_to_medium(const char *str)
 {
 	int i;
diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c
index ff4b4780d6af..c60b92bf09a4 100644
--- a/net/ethtool/ioctl.c
+++ b/net/ethtool/ioctl.c
@@ -1403,10 +1403,13 @@ static noinline_for_stack int ethtool_set_rxfh_indir(struct net_device *dev,
 		goto out_unlock;
 
 	/* indicate whether rxfh was set to default */
-	if (user_size == 0)
+	if (user_size == 0) {
 		dev->priv_flags &= ~IFF_RXFH_CONFIGURED;
-	else
+		dev->ethtool->rss_indir_user_size = 0;
+	} else {
 		dev->priv_flags |= IFF_RXFH_CONFIGURED;
+		dev->ethtool->rss_indir_user_size = rxfh_dev.indir_size;
+	}
 
 out_unlock:
 	mutex_unlock(&dev->ethtool->rss_lock);
@@ -1720,10 +1723,13 @@ static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev,
 
 	if (!rxfh_dev.rss_context) {
 		/* indicate whether rxfh was set to default */
-		if (rxfh.indir_size == 0)
+		if (rxfh.indir_size == 0) {
 			dev->priv_flags &= ~IFF_RXFH_CONFIGURED;
-		else if (rxfh.indir_size != ETH_RXFH_INDIR_NO_CHANGE)
+			dev->ethtool->rss_indir_user_size = 0;
+		} else if (rxfh.indir_size != ETH_RXFH_INDIR_NO_CHANGE) {
 			dev->priv_flags |= IFF_RXFH_CONFIGURED;
+			dev->ethtool->rss_indir_user_size = dev_indir_size;
+		}
 	}
 	/* Update rss_ctx tracking */
 	if (rxfh_dev.rss_delete) {
@@ -1736,6 +1742,7 @@ static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev,
 			ctx->indir_configured =
 				rxfh.indir_size &&
 				rxfh.indir_size != ETH_RXFH_INDIR_NO_CHANGE;
+			ctx->indir_user_size = dev_indir_size;
 		}
 		if (rxfh_dev.key) {
 			memcpy(ethtool_rxfh_context_key(ctx), rxfh_dev.key,
diff --git a/net/ethtool/rss.c b/net/ethtool/rss.c
index da5934cceb07..93a0f1de4cee 100644
--- a/net/ethtool/rss.c
+++ b/net/ethtool/rss.c
@@ -618,7 +618,7 @@ ethnl_rss_set_validate(struct ethnl_req_info *req_info, struct genl_info *info)
 static int
 rss_set_prep_indir(struct net_device *dev, struct genl_info *info,
 		   struct rss_reply_data *data, struct ethtool_rxfh_param *rxfh,
-		   bool *reset, bool *mod)
+		   bool *reset, bool *mod, u32 *user_sizep)
 {
 	struct netlink_ext_ack *extack = info->extack;
 	struct nlattr **tb = info->attrs;
@@ -685,6 +685,7 @@ rss_set_prep_indir(struct net_device *dev, struct genl_info *info,
 	}
 
 	*mod |= memcmp(rxfh->indir, data->indir_table, data->indir_size);
+	*user_sizep = user_size;
 
 	return 0;
 
@@ -833,6 +834,7 @@ ethnl_rss_set(struct ethnl_req_info *req_info, struct genl_info *info)
 	struct nlattr **tb = info->attrs;
 	struct rss_reply_data data = {};
 	const struct ethtool_ops *ops;
+	u32 indir_user_size = 0;
 	int ret;
 
 	ops = dev->ethtool_ops;
@@ -844,7 +846,8 @@ ethnl_rss_set(struct ethnl_req_info *req_info, struct genl_info *info)
 
 	rxfh.rss_context = request->rss_context;
 
-	ret = rss_set_prep_indir(dev, info, &data, &rxfh, &indir_reset, &mod);
+	ret = rss_set_prep_indir(dev, info, &data, &rxfh, &indir_reset, &mod,
+				 &indir_user_size);
 	if (ret)
 		goto exit_clean_data;
 	indir_mod = !!tb[ETHTOOL_A_RSS_INDIR];
@@ -889,12 +892,17 @@ ethnl_rss_set(struct ethnl_req_info *req_info, struct genl_info *info)
 	if (ret)
 		goto exit_unlock;
 
-	if (ctx)
+	if (ctx) {
 		rss_set_ctx_update(ctx, tb, &data, &rxfh);
-	else if (indir_reset)
+		if (indir_user_size)
+			ctx->indir_user_size = indir_user_size;
+	} else if (indir_reset) {
 		dev->priv_flags &= ~IFF_RXFH_CONFIGURED;
-	else if (indir_mod)
+		dev->ethtool->rss_indir_user_size = 0;
+	} else if (indir_mod) {
 		dev->priv_flags |= IFF_RXFH_CONFIGURED;
+		dev->ethtool->rss_indir_user_size = indir_user_size;
+	}
 
 exit_unlock:
 	mutex_unlock(&dev->ethtool->rss_lock);
@@ -998,6 +1006,7 @@ int ethnl_rss_create_doit(struct sk_buff *skb, struct genl_info *info)
 	struct rss_reply_data data = {};
 	const struct ethtool_ops *ops;
 	struct rss_req_info req = {};
+	u32 indir_user_size = 0;
 	struct net_device *dev;
 	struct sk_buff *rsp;
 	void *hdr;
@@ -1034,7 +1043,8 @@ int ethnl_rss_create_doit(struct sk_buff *skb, struct genl_info *info)
 	if (ret)
 		goto exit_ops;
 
-	ret = rss_set_prep_indir(dev, info, &data, &rxfh, &indir_dflt, &mod);
+	ret = rss_set_prep_indir(dev, info, &data, &rxfh, &indir_dflt, &mod,
+				 &indir_user_size);
 	if (ret)
 		goto exit_clean_data;
 
@@ -1080,6 +1090,8 @@ int ethnl_rss_create_doit(struct sk_buff *skb, struct genl_info *info)
 
 	/* Store the config from rxfh to Xarray.. */
 	rss_set_ctx_update(ctx, tb, &data, &rxfh);
+	if (indir_user_size)
+		ctx->indir_user_size = indir_user_size;
 	/* .. copy from Xarray to data. */
 	__rss_prepare_ctx(dev, &data, ctx);
 
-- 
2.53.0

[PATCH net-next v5 2/3] bnxt_en: Resize RSS contexts on channel count change

[Björn Töpel]

bnxt_set_channels() rejects channel changes that alter the RSS table
size when IFF_RXFH_CONFIGURED is set, because non-default context
sizes were locked at creation.

Replace the rejection with the new resize helpers. All validation runs
before the device is closed; actual resize is deferred until after
bnxt_close_nic():

 1. ethtool_rxfh_can_resize() checks context 0, passing
    dev->ethtool->rss_indir_user_size as the user_size floor.
 2. ethtool_rxfh_ctxs_can_resize() validates all non-default contexts.
 3. After bnxt_close_nic(), ethtool_rxfh_resize() applies context 0
    changes, and ethtool_rxfh_ctxs_resize() resizes non-default
    contexts.

RSS table size only changes on P5 chips with older firmware; newer
firmware always uses the largest table size.

When context 0 uses defaults (!IFF_RXFH_CONFIGURED), steps 1 and 3 are
skipped; the driver regenerates the table via
bnxt_set_dflt_rss_indir_tbl().

Signed-off-by: Björn Töpel <bjorn@kernel.org>
---
 .../net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 38 +++++++++++++++----
 1 file changed, 31 insertions(+), 7 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
index 48e8e3be70d3..469f4720c1d7 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
@@ -942,6 +942,7 @@ static int bnxt_set_channels(struct net_device *dev,
 {
 	struct bnxt *bp = netdev_priv(dev);
 	int req_tx_rings, req_rx_rings, tcs;
+	u32 new_tbl_size = 0, old_tbl_size;
 	bool sh = false;
 	int tx_xdp = 0;
 	int rc = 0;
@@ -977,19 +978,34 @@ static int bnxt_set_channels(struct net_device *dev,
 		tx_xdp = req_rx_rings;
 	}
 
-	if (bnxt_get_nr_rss_ctxs(bp, req_rx_rings) !=
-	    bnxt_get_nr_rss_ctxs(bp, bp->rx_nr_rings) &&
-	    (netif_is_rxfh_configured(dev) || bp->num_rss_ctx)) {
-		netdev_warn(dev, "RSS table size change required, RSS table entries must be default (with no additional RSS contexts present) to proceed\n");
-		return -EINVAL;
-	}
-
 	rc = bnxt_check_rings(bp, req_tx_rings, req_rx_rings, sh, tcs, tx_xdp);
 	if (rc) {
 		netdev_warn(dev, "Unable to allocate the requested rings\n");
 		return rc;
 	}
 
+	/* RSS table size only changes on P5 chips with older firmware;
+	 * newer firmware always uses the largest table size.
+	 */
+	if (bnxt_get_nr_rss_ctxs(bp, req_rx_rings) !=
+	    bnxt_get_nr_rss_ctxs(bp, bp->rx_nr_rings)) {
+		new_tbl_size = bnxt_get_nr_rss_ctxs(bp, req_rx_rings) *
+			       BNXT_RSS_TABLE_ENTRIES_P5;
+		old_tbl_size = bnxt_get_rxfh_indir_size(dev);
+
+		if (netif_is_rxfh_configured(dev) &&
+		    !ethtool_rxfh_can_resize(bp->rss_indir_tbl,
+					old_tbl_size, new_tbl_size,
+					dev->ethtool->rss_indir_user_size)) {
+			netdev_warn(dev, "RSS table resize not possible\n");
+			return -EINVAL;
+		}
+
+		rc = ethtool_rxfh_ctxs_can_resize(dev, new_tbl_size);
+		if (rc)
+			return rc;
+	}
+
 	if (netif_running(dev)) {
 		if (BNXT_PF(bp)) {
 			/* TODO CHIMP_FW: Send message to all VF's
@@ -999,6 +1015,14 @@ static int bnxt_set_channels(struct net_device *dev,
 		bnxt_close_nic(bp, true, false);
 	}
 
+	if (new_tbl_size) {
+		if (netif_is_rxfh_configured(dev))
+			ethtool_rxfh_resize(bp->rss_indir_tbl,
+					    old_tbl_size, new_tbl_size,
+					    dev->ethtool->rss_indir_user_size);
+		ethtool_rxfh_ctxs_resize(dev, new_tbl_size);
+	}
+
 	if (sh) {
 		bp->flags |= BNXT_FLAG_SHARED_RINGS;
 		bp->rx_nr_rings = channel->combined_count;
-- 
2.53.0

[PATCH net-next v5 3/3] selftests: rss_drv: Add RSS indirection table resize tests

[Björn Töpel]

Add resize tests to rss_drv.py. Devices without dynamic table sizing
are skipped via _require_dynamic_indir_size().

resize_periodic: set a periodic 4-entry table, shrink channels to
fold, grow back to unfold. Check the exact pattern is preserved. Has
main and non-default context variants.

resize_below_user_size_reject: send a periodic table with user_size
between the big and small device table sizes. Verify that shrinking
below user_size is rejected even though the table is periodic. Has
main and non-default context variants.

resize_nonperiodic_reject: set a non-periodic table (equal N), verify
that channel reduction is rejected.

resize_nonperiodic_no_corruption: verify a failed resize leaves both
the indirection table contents and the channel count unchanged.

Signed-off-by: Björn Töpel <bjorn@kernel.org>
---
 .../selftests/drivers/net/hw/rss_drv.py       | 233 +++++++++++++++++-
 1 file changed, 229 insertions(+), 4 deletions(-)

diff --git a/tools/testing/selftests/drivers/net/hw/rss_drv.py b/tools/testing/selftests/drivers/net/hw/rss_drv.py
index 2d1a33189076..bd59dace6e15 100755
--- a/tools/testing/selftests/drivers/net/hw/rss_drv.py
+++ b/tools/testing/selftests/drivers/net/hw/rss_drv.py
@@ -5,9 +5,9 @@
 Driver-related behavior tests for RSS.
 """
 
-from lib.py import ksft_run, ksft_exit, ksft_ge
-from lib.py import ksft_variants, KsftNamedVariant, KsftSkipEx
-from lib.py import defer, ethtool
+from lib.py import ksft_run, ksft_exit, ksft_eq, ksft_ge
+from lib.py import ksft_variants, KsftNamedVariant, KsftSkipEx, ksft_raises
+from lib.py import defer, ethtool, CmdExitFailure
 from lib.py import EthtoolFamily, NlError
 from lib.py import NetDrvEnv
 
@@ -45,6 +45,18 @@ def _maybe_create_context(cfg, create_context):
     return ctx_id
 
 
+def _require_dynamic_indir_size(cfg, ch_max):
+    """Skip if the device does not dynamically size its indirection table."""
+    ethtool(f"-X {cfg.ifname} default")
+    ethtool(f"-L {cfg.ifname} combined 2")
+    small = len(_get_rss(cfg)['rss-indirection-table'])
+    ethtool(f"-L {cfg.ifname} combined {ch_max}")
+    large = len(_get_rss(cfg)['rss-indirection-table'])
+
+    if small == large:
+        raise KsftSkipEx("Device does not dynamically size indirection table")
+
+
 @ksft_variants([
     KsftNamedVariant("main", False),
     KsftNamedVariant("ctx", True),
@@ -76,11 +88,224 @@ def indir_size_4x(cfg, create_context):
         _test_rss_indir_size(cfg, test_max, context=ctx_id)
 
 
+@ksft_variants([
+    KsftNamedVariant("main", False),
+    KsftNamedVariant("ctx", True),
+])
+def resize_periodic(cfg, create_context):
+    """Test that a periodic indirection table survives channel changes.
+
+    Set a non-default periodic table ([3, 2, 1, 0] x N) via netlink,
+    reduce channels to trigger a fold, then increase to trigger an
+    unfold. Using a reversed pattern (instead of [0, 1, 2, 3]) ensures
+    the test can distinguish a correct fold from a driver that silently
+    resets the table to defaults. Verify the exact pattern is preserved
+    and the size tracks the channel count.
+    """
+    channels = cfg.ethnl.channels_get({'header': {'dev-index': cfg.ifindex}})
+    ch_max = channels.get('combined-max', 0)
+    qcnt = channels['combined-count']
+
+    if ch_max < 4:
+        raise KsftSkipEx(f"Not enough queues for the test: max={ch_max}")
+
+    defer(ethtool, f"-L {cfg.ifname} combined {qcnt}")
+
+    _require_dynamic_indir_size(cfg, ch_max)
+
+    ctx_id = _maybe_create_context(cfg, create_context)
+
+    # Set a non-default periodic pattern via netlink.
+    # Send only 4 entries (user_size=4) so the kernel replicates it
+    # to fill the device table. This allows folding down to 4 entries.
+    rss = _get_rss(cfg, context=ctx_id)
+    orig_size = len(rss['rss-indirection-table'])
+    pattern = [3, 2, 1, 0]
+    req = {'header': {'dev-index': cfg.ifindex}, 'indir': pattern}
+    if ctx_id:
+        req['context'] = ctx_id
+    else:
+        defer(ethtool, f"-X {cfg.ifname} default")
+    cfg.ethnl.rss_set(req)
+
+    # Shrink — should fold
+    ethtool(f"-L {cfg.ifname} combined 4")
+    rss = _get_rss(cfg, context=ctx_id)
+    indir = rss['rss-indirection-table']
+
+    ksft_ge(orig_size, len(indir), "Table did not shrink")
+    ksft_eq(indir, [3, 2, 1, 0] * (len(indir) // 4),
+            "Folded table has wrong pattern")
+
+    # Grow back — should unfold
+    ethtool(f"-L {cfg.ifname} combined {ch_max}")
+    rss = _get_rss(cfg, context=ctx_id)
+    indir = rss['rss-indirection-table']
+
+    ksft_eq(len(indir), orig_size, "Table size not restored")
+    ksft_eq(indir, [3, 2, 1, 0] * (len(indir) // 4),
+            "Unfolded table has wrong pattern")
+
+
+@ksft_variants([
+    KsftNamedVariant("main", False),
+    KsftNamedVariant("ctx", True),
+])
+def resize_below_user_size_reject(cfg, create_context):
+    """Test that shrinking below user_size is rejected.
+
+    Send a table via netlink whose size (user_size) sits between
+    the small and large device table sizes. The table is periodic,
+    so folding would normally succeed, but the user_size floor must
+    prevent it.
+    """
+    channels = cfg.ethnl.channels_get({'header': {'dev-index': cfg.ifindex}})
+    ch_max = channels.get('combined-max', 0)
+    qcnt = channels['combined-count']
+
+    if ch_max < 4:
+        raise KsftSkipEx(f"Not enough queues for the test: max={ch_max}")
+
+    defer(ethtool, f"-L {cfg.ifname} combined {qcnt}")
+
+    _require_dynamic_indir_size(cfg, ch_max)
+
+    ctx_id = _maybe_create_context(cfg, create_context)
+
+    # Measure the table size at max channels
+    rss = _get_rss(cfg, context=ctx_id)
+    big_size = len(rss['rss-indirection-table'])
+
+    # Measure the table size at reduced channels
+    ethtool(f"-L {cfg.ifname} combined 4")
+    rss = _get_rss(cfg, context=ctx_id)
+    small_size = len(rss['rss-indirection-table'])
+    ethtool(f"-L {cfg.ifname} combined {ch_max}")
+
+    if small_size >= big_size:
+        raise KsftSkipEx("Table did not shrink at reduced channels")
+
+    # Find a user_size
+    user_size = None
+    for div in [2, 4]:
+        candidate = big_size // div
+        if candidate > small_size and big_size % candidate == 0:
+            user_size = candidate
+            break
+    if user_size is None:
+        raise KsftSkipEx("No suitable user_size between small and big table")
+
+    # Send a periodic sub-table of exactly user_size entries.
+    # Pattern safe for 4 channels.
+    pattern = [0, 1, 2, 3] * (user_size // 4)
+    if len(pattern) != user_size:
+        raise KsftSkipEx(f"user_size ({user_size}) not divisible by 4")
+    req = {'header': {'dev-index': cfg.ifindex}, 'indir': pattern}
+    if ctx_id:
+        req['context'] = ctx_id
+    else:
+        defer(ethtool, f"-X {cfg.ifname} default")
+    cfg.ethnl.rss_set(req)
+
+    # Shrink channels — table would go to small_size < user_size.
+    # The table is periodic so folding would work, but user_size
+    # floor must reject it.
+    with ksft_raises(CmdExitFailure):
+        ethtool(f"-L {cfg.ifname} combined 4")
+
+
+@ksft_variants([
+    KsftNamedVariant("main", False),
+    KsftNamedVariant("ctx", True),
+])
+def resize_nonperiodic_reject(cfg, create_context):
+    """Test that a non-periodic table blocks channel reduction.
+
+    Set equal weight across all queues so the table is not periodic
+    at any smaller size, then verify channel reduction is rejected.
+    An additional context with a periodic table is created to verify
+    that validation catches the non-periodic one even when others
+    are fine.
+    """
+    channels = cfg.ethnl.channels_get({'header': {'dev-index': cfg.ifindex}})
+    ch_max = channels.get('combined-max', 0)
+    qcnt = channels['combined-count']
+
+    if ch_max < 4:
+        raise KsftSkipEx(f"Not enough queues for the test: max={ch_max}")
+
+    defer(ethtool, f"-L {cfg.ifname} combined {qcnt}")
+
+    _require_dynamic_indir_size(cfg, ch_max)
+
+    ctx_id = _maybe_create_context(cfg, create_context)
+    ctx_ref = f"context {ctx_id}" if ctx_id else ""
+
+    # Create an extra context with a periodic (foldable) table so that
+    # the validation must iterate all contexts to find the bad one.
+    extra_ctx = _maybe_create_context(cfg, True)
+    ethtool(f"-X {cfg.ifname} context {extra_ctx} equal 2")
+
+    ethtool(f"-X {cfg.ifname} {ctx_ref} equal {ch_max}")
+    if not create_context:
+        defer(ethtool, f"-X {cfg.ifname} default")
+
+    with ksft_raises(CmdExitFailure):
+        ethtool(f"-L {cfg.ifname} combined 2")
+
+
+@ksft_variants([
+    KsftNamedVariant("main", False),
+    KsftNamedVariant("ctx", True),
+])
+def resize_nonperiodic_no_corruption(cfg, create_context):
+    """Test that a failed resize does not corrupt table or channel count.
+
+    Set a non-periodic table, attempt a channel reduction (which must
+    fail), then verify both the indirection table contents and the
+    channel count are unchanged.
+    """
+    channels = cfg.ethnl.channels_get({'header': {'dev-index': cfg.ifindex}})
+    ch_max = channels.get('combined-max', 0)
+    qcnt = channels['combined-count']
+
+    if ch_max < 4:
+        raise KsftSkipEx(f"Not enough queues for the test: max={ch_max}")
+
+    defer(ethtool, f"-L {cfg.ifname} combined {qcnt}")
+
+    _require_dynamic_indir_size(cfg, ch_max)
+
+    ctx_id = _maybe_create_context(cfg, create_context)
+    ctx_ref = f"context {ctx_id}" if ctx_id else ""
+
+    ethtool(f"-X {cfg.ifname} {ctx_ref} equal {ch_max}")
+    if not create_context:
+        defer(ethtool, f"-X {cfg.ifname} default")
+
+    rss_before = _get_rss(cfg, context=ctx_id)
+
+    with ksft_raises(CmdExitFailure):
+        ethtool(f"-L {cfg.ifname} combined 2")
+
+    rss_after = _get_rss(cfg, context=ctx_id)
+    ksft_eq(rss_after['rss-indirection-table'],
+            rss_before['rss-indirection-table'],
+            "Indirection table corrupted after failed resize")
+
+    channels = cfg.ethnl.channels_get({'header': {'dev-index': cfg.ifindex}})
+    ksft_eq(channels['combined-count'], ch_max,
+            "Channel count changed after failed resize")
+
+
 def main() -> None:
     """ Ksft boiler plate main """
     with NetDrvEnv(__file__) as cfg:
         cfg.ethnl = EthtoolFamily()
-        ksft_run([indir_size_4x], args=(cfg, ))
+        ksft_run([indir_size_4x, resize_periodic,
+                  resize_below_user_size_reject,
+                  resize_nonperiodic_reject,
+                  resize_nonperiodic_no_corruption], args=(cfg, ))
     ksft_exit()
 
 
-- 
2.53.0

Re: [PATCH net-next v5 1/3] ethtool: Add RSS indirection table resize helpers

[Jakub Kicinski]

On Tue, 17 Mar 2026 09:33:34 +0100 Björn Töpel wrote:
>  static int
>  rss_set_prep_indir(struct net_device *dev, struct genl_info *info,
>  		   struct rss_reply_data *data, struct ethtool_rxfh_param *rxfh,
> -		   bool *reset, bool *mod)
> +		   bool *reset, bool *mod, u32 *user_sizep)
>  {
>  	struct netlink_ext_ack *extack = info->extack;
>  	struct nlattr **tb = info->attrs;
> @@ -685,6 +685,7 @@ rss_set_prep_indir(struct net_device *dev, struct genl_info *info,
>  	}
>  
>  	*mod |= memcmp(rxfh->indir, data->indir_table, data->indir_size);
> +	*user_sizep = user_size;
>  
>  	return 0;

Weak preference for return user_size; instead of output arg, 7th, "p"?

Re: [PATCH net-next v5 2/3] bnxt_en: Resize RSS contexts on channel count change

[Jakub Kicinski]

On Tue, 17 Mar 2026 09:33:35 +0100 Björn Töpel wrote:
> +		if (netif_is_rxfh_configured(dev) &&
> +		    !ethtool_rxfh_can_resize(bp->rss_indir_tbl,
> +					old_tbl_size, new_tbl_size,
> +					dev->ethtool->rss_indir_user_size)) {
> +			netdev_warn(dev, "RSS table resize not possible\n");
> +			return -EINVAL;
> +		}

Now that we know the size of the table that user has set 
we can skip the netif_is_rxfh_configured(dev) and depend fully on
ethtool_rxfh_can_resize(... rss_indir_user_size) ?

Otherwise we're not making much use of dev->ethtool->rss_indir_user_size
(the one for the default context / context 0).
Regardless of the table size user provided IFF_RXFH_CONFIGURED will be
set.

Which brings me to my last point - I think you can delete
IFF_RXFH_CONFIGURED and have netif_is_rxfh_configured(dev)
operate on ..user_size ? We should probably have a new helper
for when driver lost control over rings and has to reset indir table
forgetting user config (bnxt mlx5 clearing IFF_RXFH_CONFIGURED).
Helper should clear the state and issue a Netlink notification?

Re: [PATCH net-next v5 1/3] ethtool: Add RSS indirection table resize helpers

[Björn Töpel]

Jakub Kicinski <kuba@kernel.org> writes:

> On Tue, 17 Mar 2026 09:33:34 +0100 Björn Töpel wrote:
>>  static int
>>  rss_set_prep_indir(struct net_device *dev, struct genl_info *info,
>>  		   struct rss_reply_data *data, struct ethtool_rxfh_param *rxfh,
>> -		   bool *reset, bool *mod)
>> +		   bool *reset, bool *mod, u32 *user_sizep)
>>  {
>>  	struct netlink_ext_ack *extack = info->extack;
>>  	struct nlattr **tb = info->attrs;
>> @@ -685,6 +685,7 @@ rss_set_prep_indir(struct net_device *dev, struct genl_info *info,
>>  	}
>>  
>>  	*mod |= memcmp(rxfh->indir, data->indir_table, data->indir_size);
>> +	*user_sizep = user_size;
>>  
>>  	return 0;
>
> Weak preference for return user_size; instead of output arg, 7th, "p"?

Good point! The 7th and p does have a bad smell to it...

Re: [PATCH net-next v5 2/3] bnxt_en: Resize RSS contexts on channel count change

[Björn Töpel]

Jakub Kicinski <kuba@kernel.org> writes:

> On Tue, 17 Mar 2026 09:33:35 +0100 Björn Töpel wrote:
>> +		if (netif_is_rxfh_configured(dev) &&
>> +		    !ethtool_rxfh_can_resize(bp->rss_indir_tbl,
>> +					old_tbl_size, new_tbl_size,
>> +					dev->ethtool->rss_indir_user_size)) {
>> +			netdev_warn(dev, "RSS table resize not possible\n");
>> +			return -EINVAL;
>> +		}
>
> Now that we know the size of the table that user has set 
> we can skip the netif_is_rxfh_configured(dev) and depend fully on
> ethtool_rxfh_can_resize(... rss_indir_user_size) ?
>
> Otherwise we're not making much use of dev->ethtool->rss_indir_user_size
> (the one for the default context / context 0).
> Regardless of the table size user provided IFF_RXFH_CONFIGURED will be
> set.
>
> Which brings me to my last point - I think you can delete
> IFF_RXFH_CONFIGURED and have netif_is_rxfh_configured(dev)
> operate on ..user_size ? We should probably have a new helper
> for when driver lost control over rings and has to reset indir table
> forgetting user config (bnxt mlx5 clearing IFF_RXFH_CONFIGURED).
> Helper should clear the state and issue a Netlink notification?

All good points -- I'll try that! Thanks!