6.6.148-rc1 review

@@ -704,18 +704,6 @@ static int mt6797_afe_runtime_resume(str
 	return 0;
 }
 
-static int mt6797_afe_component_probe(struct snd_soc_component *component)
-{
-	return mtk_afe_add_sub_dai_control(component);
-}
-
-static const struct snd_soc_component_driver mt6797_afe_component = {
-	.name		= AFE_PCM_NAME,
-	.probe		= mt6797_afe_component_probe,
-	.pointer	= mtk_afe_pcm_pointer,
-	.pcm_construct	= mtk_afe_pcm_new,
-};
-
 static int mt6797_dai_memif_register(struct mtk_base_afe *afe)
 {
 	struct mtk_base_afe_dai *dai;

@@ -852,7 +840,7 @@ static int mt6797_afe_pcm_dev_probe(stru
 	pm_runtime_get_sync(&pdev->dev);
 
 	/* register component */
-	ret = devm_snd_soc_register_component(dev, &mt6797_afe_component,
+	ret = devm_snd_soc_register_component(dev, &mtk_afe_pcm_platform,
 					      NULL, 0);
 	if (ret) {
 		dev_warn(dev, "err_platform\n");

@@ -429,18 +429,6 @@ static int mt7986_afe_runtime_resume(str
 	return 0;
 }
 
-static int mt7986_afe_component_probe(struct snd_soc_component *component)
-{
-	return mtk_afe_add_sub_dai_control(component);
-}
-
-static const struct snd_soc_component_driver mt7986_afe_component = {
-	.name = AFE_PCM_NAME,
-	.probe = mt7986_afe_component_probe,
-	.pointer	= mtk_afe_pcm_pointer,
-	.pcm_construct	= mtk_afe_pcm_new,
-};
-
 static int mt7986_dai_memif_register(struct mtk_base_afe *afe)
 {
 	struct mtk_base_afe_dai *dai;

@@ -573,7 +561,7 @@ static int mt7986_afe_pcm_dev_probe(stru
 
 	/* register component */
 	ret = devm_snd_soc_register_component(&pdev->dev,
-					      &mt7986_afe_component,
+					      &mtk_afe_pcm_platform,
 					      NULL, 0);
 	if (ret)
 		return dev_err_probe(dev, ret, "Cannot register AFE component\n");

@@ -1043,18 +1043,6 @@ skip_regmap:
 	return 0;
 }
 
-static int mt8183_afe_component_probe(struct snd_soc_component *component)
-{
-	return mtk_afe_add_sub_dai_control(component);
-}
-
-static const struct snd_soc_component_driver mt8183_afe_component = {
-	.name		= AFE_PCM_NAME,
-	.probe		= mt8183_afe_component_probe,
-	.pointer	= mtk_afe_pcm_pointer,
-	.pcm_construct	= mtk_afe_pcm_new,
-};
-
 static int mt8183_dai_memif_register(struct mtk_base_afe *afe)
 {
 	struct mtk_base_afe_dai *dai;

@@ -1250,7 +1238,7 @@ static int mt8183_afe_pcm_dev_probe(stru
 
 	/* register component */
 	ret = devm_snd_soc_register_component(&pdev->dev,
-					      &mt8183_afe_component,
+					      &mtk_afe_pcm_platform,
 					      NULL, 0);
 	if (ret) {
 		dev_warn(dev, "err_platform\n");

@@ -3027,25 +3027,6 @@ skip_regmap:
 	return 0;
 }
 
-static int mt8188_afe_component_probe(struct snd_soc_component *component)
-{
-	struct mtk_base_afe *afe = snd_soc_component_get_drvdata(component);
-	int ret;
-
-	snd_soc_component_init_regmap(component, afe->regmap);
-
-	ret = mtk_afe_add_sub_dai_control(component);
-
-	return ret;
-}
-
-static const struct snd_soc_component_driver mt8188_afe_component = {
-	.name = AFE_PCM_NAME,
-	.pointer       = mtk_afe_pcm_pointer,
-	.pcm_construct = mtk_afe_pcm_new,
-	.probe         = mt8188_afe_component_probe,
-};
-
 static int init_memif_priv_data(struct mtk_base_afe *afe)
 {
 	struct mt8188_afe_private *afe_priv = afe->platform_priv;

@@ -3347,7 +3328,7 @@ static int mt8188_afe_pcm_dev_probe(stru
 	}
 
 	/* register component */
-	ret = devm_snd_soc_register_component(dev, &mt8188_afe_component,
+	ret = devm_snd_soc_register_component(dev, &mtk_afe_pcm_platform,
 					      afe->dai_drivers, afe->num_dai_drivers);
 	if (ret) {
 		dev_warn(dev, "err_platform\n");

@@ -2126,22 +2126,6 @@ skip_regmap:
 	return 0;
 }
 
-static int mt8192_afe_component_probe(struct snd_soc_component *component)
-{
-	return mtk_afe_add_sub_dai_control(component);
-}
-
-static const struct snd_soc_component_driver mt8192_afe_component = {
-	.name = AFE_PCM_NAME,
-	.probe = mt8192_afe_component_probe,
-	.pointer = mtk_afe_pcm_pointer,
-	.pcm_construct = mtk_afe_pcm_new,
-};
-
-static const struct snd_soc_component_driver mt8192_afe_pcm_component = {
-	.name = "mt8192-afe-pcm-dai",
-};
-
 static int mt8192_dai_memif_register(struct mtk_base_afe *afe)
 {
 	struct mtk_base_afe_dai *dai;

@@ -2318,16 +2302,11 @@ static int mt8192_afe_pcm_dev_probe(stru
 
 	/* register platform */
 	ret = devm_snd_soc_register_component(&pdev->dev,
-					      &mt8192_afe_component, NULL, 0);
-	if (ret)
-		return dev_err_probe(dev, ret, "Couldn't register AFE component\n");
-
-	ret = devm_snd_soc_register_component(&pdev->dev,
-					      &mt8192_afe_pcm_component,
+					      &mtk_afe_pcm_platform,
 					      afe->dai_drivers,
 					      afe->num_dai_drivers);
 	if (ret)
-		return dev_err_probe(dev, ret, "Couldn't register AFE-PCM component\n");
+		return dev_err_probe(dev, ret, "Couldn't register AFE component\n");
 
 	return 0;
 }

@@ -2944,25 +2944,6 @@ skip_regmap:
 	return 0;
 }
 
-static int mt8195_afe_component_probe(struct snd_soc_component *component)
-{
-	struct mtk_base_afe *afe = snd_soc_component_get_drvdata(component);
-	int ret = 0;
-
-	snd_soc_component_init_regmap(component, afe->regmap);
-
-	ret = mtk_afe_add_sub_dai_control(component);
-
-	return ret;
-}
-
-static const struct snd_soc_component_driver mt8195_afe_component = {
-	.name = AFE_PCM_NAME,
-	.pointer = mtk_afe_pcm_pointer,
-	.pcm_construct = mtk_afe_pcm_new,
-	.probe = mt8195_afe_component_probe,
-};
-
 static int init_memif_priv_data(struct mtk_base_afe *afe)
 {
 	struct mt8195_afe_private *afe_priv = afe->platform_priv;

@@ -3164,7 +3145,7 @@ static int mt8195_afe_pcm_dev_probe(stru
 	}
 
 	/* register component */
-	ret = devm_snd_soc_register_component(dev, &mt8195_afe_component,
+	ret = devm_snd_soc_register_component(dev, &mtk_afe_pcm_platform,
 					      afe->dai_drivers, afe->num_dai_drivers);
 	if (ret) {
 		dev_warn(dev, "err_platform\n");

@@ -2164,27 +2164,26 @@ static int mt8192_afe_pcm_dev_probe(stru
 {
 	struct mtk_base_afe *afe;
 	struct mt8192_afe_private *afe_priv;
-	struct device *dev;
+	struct device *dev = &pdev->dev;
 	struct reset_control *rstc;
 	int i, ret, irq_id;
 
-	ret = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(34));
+	ret = dma_set_mask_and_coherent(dev, DMA_BIT_MASK(34));
 	if (ret)
 		return ret;
 
-	afe = devm_kzalloc(&pdev->dev, sizeof(*afe), GFP_KERNEL);
+	afe = devm_kzalloc(dev, sizeof(*afe), GFP_KERNEL);
 	if (!afe)
 		return -ENOMEM;
 	platform_set_drvdata(pdev, afe);
 
-	afe->platform_priv = devm_kzalloc(&pdev->dev, sizeof(*afe_priv),
+	afe->platform_priv = devm_kzalloc(dev, sizeof(*afe_priv),
 					  GFP_KERNEL);
 	if (!afe->platform_priv)
 		return -ENOMEM;
 	afe_priv = afe->platform_priv;
 
-	afe->dev = &pdev->dev;
-	dev = afe->dev;
+	afe->dev = dev;
 
 	ret = of_reserved_mem_device_init(dev);
 	if (ret) {

@@ -2212,7 +2211,7 @@ static int mt8192_afe_pcm_dev_probe(stru
 	if (ret)
 		return dev_err_probe(dev, ret, "failed to trigger audio reset\n");
 
-	ret = devm_pm_runtime_enable(&pdev->dev);
+	ret = devm_pm_runtime_enable(dev);
 	if (ret)
 		return ret;
 

@@ -2228,13 +2227,13 @@ static int mt8192_afe_pcm_dev_probe(stru
 
 	/* enable clock for regcache get default value from hw */
 	afe_priv->pm_runtime_bypass_reg_ctl = true;
-	pm_runtime_get_sync(&pdev->dev);
+	pm_runtime_get_sync(dev);
 
 	ret = regmap_reinit_cache(afe->regmap, &mt8192_afe_regmap_config);
 	if (ret)
 		return dev_err_probe(dev, ret, "regmap_reinit_cache fail\n");
 
-	pm_runtime_put_sync(&pdev->dev);
+	pm_runtime_put_sync(dev);
 	afe_priv->pm_runtime_bypass_reg_ctl = false;
 
 	regcache_cache_only(afe->regmap, true);

@@ -2301,7 +2300,7 @@ static int mt8192_afe_pcm_dev_probe(stru
 	afe->runtime_suspend = mt8192_afe_runtime_suspend;
 
 	/* register platform */
-	ret = devm_snd_soc_register_component(&pdev->dev,
+	ret = devm_snd_soc_register_component(dev,
 					      &mtk_afe_pcm_platform,
 					      afe->dai_drivers,
 					      afe->num_dai_drivers);

@@ -2227,15 +2227,19 @@ static int mt8192_afe_pcm_dev_probe(stru
 
 	/* enable clock for regcache get default value from hw */
 	afe_priv->pm_runtime_bypass_reg_ctl = true;
-	pm_runtime_get_sync(dev);
+	ret = pm_runtime_resume_and_get(dev);
+	if (ret) {
+		afe_priv->pm_runtime_bypass_reg_ctl = false;
+		return dev_err_probe(dev, ret, "failed to resume device\n");
+	}
 
 	ret = regmap_reinit_cache(afe->regmap, &mt8192_afe_regmap_config);
-	if (ret)
-		return dev_err_probe(dev, ret, "regmap_reinit_cache fail\n");
-
 	pm_runtime_put_sync(dev);
 	afe_priv->pm_runtime_bypass_reg_ctl = false;
 
+	if (ret)
+		return dev_err_probe(dev, ret, "regmap_reinit_cache fail\n");
+
 	regcache_cache_only(afe->regmap, true);
 	regcache_mark_dirty(afe->regmap);
 

@@ -1149,17 +1149,21 @@ static int mt8183_afe_pcm_dev_probe(stru
 
 	/* enable clock for regcache get default value from hw */
 	afe_priv->pm_runtime_bypass_reg_ctl = true;
-	pm_runtime_get_sync(&pdev->dev);
+	ret = pm_runtime_resume_and_get(dev);
+	if (ret) {
+		afe_priv->pm_runtime_bypass_reg_ctl = false;
+		goto err_pm_disable;
+	}
 
 	ret = regmap_reinit_cache(afe->regmap, &mt8183_afe_regmap_config);
+	pm_runtime_put_sync(dev);
+	afe_priv->pm_runtime_bypass_reg_ctl = false;
+
 	if (ret) {
 		dev_err(dev, "regmap_reinit_cache fail, ret %d\n", ret);
 		goto err_pm_disable;
 	}
 
-	pm_runtime_put_sync(&pdev->dev);
-	afe_priv->pm_runtime_bypass_reg_ctl = false;
-
 	regcache_cache_only(afe->regmap, true);
 	regcache_mark_dirty(afe->regmap);
 

@@ -505,6 +505,7 @@ out:
  * @data:	Key data to be matched against existing elements
  * @genmask:	If set, check that element is active in given genmask
  * @tstamp:	timestamp to check for expired elements
+ * @gfp:	the type of memory to allocate (see kmalloc).
  *
  * This is essentially the same as the lookup function, except that it matches
  * key data against the uncommitted copy and doesn't use preallocated maps for

@@ -515,7 +516,7 @@ out:
 static struct nft_pipapo_elem *pipapo_get(const struct net *net,
 					  const struct nft_set *set,
 					  const u8 *data, u8 genmask,
-					  u64 tstamp)
+					  u64 tstamp, gfp_t gfp)
 {
 	struct nft_pipapo_elem *ret = ERR_PTR(-ENOENT);
 	struct nft_pipapo *priv = nft_set_priv(set);

@@ -528,13 +529,13 @@ static struct nft_pipapo_elem *pipapo_ge
 	if (m->bsize_max == 0)
 		return ret;
 
-	res_map = kmalloc_array(m->bsize_max, sizeof(*res_map), GFP_ATOMIC);
+	res_map = kmalloc_array(m->bsize_max, sizeof(*res_map), gfp);
 	if (!res_map) {
 		ret = ERR_PTR(-ENOMEM);
 		goto out;
 	}
 
-	fill_map = kcalloc(m->bsize_max, sizeof(*res_map), GFP_ATOMIC);
+	fill_map = kcalloc(m->bsize_max, sizeof(*res_map), gfp);
 	if (!fill_map) {
 		ret = ERR_PTR(-ENOMEM);
 		goto out;

@@ -609,7 +610,7 @@ static void *nft_pipapo_get(const struct
 			    const struct nft_set_elem *elem, unsigned int flags)
 {
 	return pipapo_get(net, set, (const u8 *)elem->key.val.data,
-			 nft_genmask_cur(net), get_jiffies_64());
+			 nft_genmask_cur(net), get_jiffies_64(), GFP_ATOMIC);
 }
 
 

@@ -1246,7 +1247,7 @@ static int nft_pipapo_insert(const struc
 	else
 		end = start;
 
-	dup = pipapo_get(net, set, start, genmask, tstamp);
+	dup = pipapo_get(net, set, start, genmask, tstamp, GFP_KERNEL);
 	if (!IS_ERR(dup)) {
 		/* Check if we already have the same exact entry */
 		const struct nft_data *dup_key, *dup_end;

@@ -1268,7 +1269,8 @@ static int nft_pipapo_insert(const struc
 
 	if (PTR_ERR(dup) == -ENOENT) {
 		/* Look for partially overlapping entries */
-		dup = pipapo_get(net, set, end, nft_genmask_next(net), tstamp);
+		dup = pipapo_get(net, set, end, nft_genmask_next(net), tstamp,
+				 GFP_KERNEL);
 	}
 
 	if (PTR_ERR(dup) != -ENOENT) {

@@ -1869,7 +1871,8 @@ static void *pipapo_deactivate(const str
 {
 	struct nft_pipapo_elem *e;
 
-	e = pipapo_get(net, set, data, nft_genmask_next(net), nft_net_tstamp(net));
+	e = pipapo_get(net, set, data, nft_genmask_next(net),
+		       nft_net_tstamp(net), GFP_KERNEL);
 	if (IS_ERR(e))
 		return NULL;
 

@@ -1217,6 +1217,17 @@ static int pipapo_realloc_scratch(struct
 	return 0;
 }
 
+static bool nft_pipapo_transaction_mutex_held(const struct nft_set *set)
+{
+#ifdef CONFIG_PROVE_LOCKING
+	const struct net *net = read_pnet(&set->net);
+
+	return lockdep_is_held(&nft_pernet(net)->commit_mutex);
+#else
+	return true;
+#endif
+}
+
 /**
  * nft_pipapo_insert() - Validate and insert ranged elements
  * @net:	Network namespace

@@ -1802,17 +1813,6 @@ static void nft_pipapo_commit(struct nft
 	pipapo_gc_queue(set);
 }
 
-static bool nft_pipapo_transaction_mutex_held(const struct nft_set *set)
-{
-#ifdef CONFIG_PROVE_LOCKING
-	const struct net *net = read_pnet(&set->net);
-
-	return lockdep_is_held(&nft_pernet(net)->commit_mutex);
-#else
-	return true;
-#endif
-}
-
 static void nft_pipapo_abort(const struct nft_set *set)
 {
 	struct nft_pipapo *priv = nft_set_priv(set);

@@ -1358,7 +1358,7 @@ static int nft_pipapo_insert(const struc
  * pipapo_clone() - Clone matching data to create new working copy
  * @old:	Existing matching data
  *
- * Return: copy of matching data passed as 'old', error pointer on failure
+ * Return: copy of matching data passed as 'old' or NULL.
  */
 static struct nft_pipapo_match *pipapo_clone(struct nft_pipapo_match *old)
 {

@@ -1368,7 +1368,7 @@ static struct nft_pipapo_match *pipapo_c
 
 	new = kmalloc(struct_size(new, f, old->field_count), GFP_KERNEL_ACCOUNT);
 	if (!new)
-		return ERR_PTR(-ENOMEM);
+		return NULL;
 
 	new->field_count = old->field_count;
 	new->bsize_max = old->bsize_max;

@@ -1444,7 +1444,7 @@ out_scratch:
 	free_percpu(new->scratch);
 	kfree(new);
 
-	return ERR_PTR(-ENOMEM);
+	return NULL;
 }
 
 /**

@@ -1798,7 +1798,7 @@ static void nft_pipapo_commit(struct nft
 		return;
 
 	new_clone = pipapo_clone(priv->clone);
-	if (IS_ERR(new_clone))
+	if (!new_clone)
 		return;
 
 	priv->dirty = false;

@@ -1824,7 +1824,7 @@ static void nft_pipapo_abort(const struc
 	m = rcu_dereference_protected(priv->match, nft_pipapo_transaction_mutex_held(set));
 
 	new_clone = pipapo_clone(m);
-	if (IS_ERR(new_clone))
+	if (!new_clone)
 		return;
 
 	priv->dirty = false;

@@ -2265,8 +2265,8 @@ static int nft_pipapo_init(const struct
 
 	/* Create an initial clone of matching data for next insertion */
 	priv->clone = pipapo_clone(m);
-	if (IS_ERR(priv->clone)) {
-		err = PTR_ERR(priv->clone);
+	if (!priv->clone) {
+		err = -ENOMEM;
 		goto out_free;
 	}
 

@@ -2109,35 +2109,23 @@ static void nft_pipapo_remove(const stru
 }
 
 /**
- * nft_pipapo_walk() - Walk over elements
+ * nft_pipapo_do_walk() - Walk over elements in m
  * @ctx:	nftables API context
  * @set:	nftables API set representation
+ * @m:		matching data pointing to key mapping array
  * @iter:	Iterator
  *
  * As elements are referenced in the mapping array for the last field, directly
  * scan that array: there's no need to follow rule mappings from the first
- * field.
+ * field. @m is protected either by RCU read lock or by transaction mutex.
  */
-static void nft_pipapo_walk(const struct nft_ctx *ctx, struct nft_set *set,
-			    struct nft_set_iter *iter)
+static void nft_pipapo_do_walk(const struct nft_ctx *ctx, struct nft_set *set,
+			       const struct nft_pipapo_match *m,
+			       struct nft_set_iter *iter)
 {
-	struct nft_pipapo *priv = nft_set_priv(set);
-	const struct nft_pipapo_match *m;
 	const struct nft_pipapo_field *f;
 	int i, r;
 
-	WARN_ON_ONCE(iter->type != NFT_ITER_READ &&
-		     iter->type != NFT_ITER_UPDATE);
-
-	rcu_read_lock();
-	if (iter->type == NFT_ITER_READ)
-		m = rcu_dereference(priv->match);
-	else
-		m = priv->clone;
-
-	if (unlikely(!m))
-		goto out;
-
 	for (i = 0, f = m->f; i < m->field_count - 1; i++, f++)
 		;
 

@@ -2157,14 +2145,44 @@ static void nft_pipapo_walk(const struct
 
 		iter->err = iter->fn(ctx, set, iter, &elem);
 		if (iter->err < 0)
-			goto out;
+			return;
 
 cont:
 		iter->count++;
 	}
+}
 
-out:
-	rcu_read_unlock();
+/**
+ * nft_pipapo_walk() - Walk over elements
+ * @ctx:	nftables API context
+ * @set:	nftables API set representation
+ * @iter:	Iterator
+ *
+ * Test if destructive action is needed or not, clone active backend if needed
+ * and call the real function to work on the data.
+ */
+static void nft_pipapo_walk(const struct nft_ctx *ctx, struct nft_set *set,
+			    struct nft_set_iter *iter)
+{
+	struct nft_pipapo *priv = nft_set_priv(set);
+	const struct nft_pipapo_match *m;
+
+	switch (iter->type) {
+	case NFT_ITER_UPDATE:
+		m = priv->clone;
+		nft_pipapo_do_walk(ctx, set, m, iter);
+		break;
+	case NFT_ITER_READ:
+		rcu_read_lock();
+		m = rcu_dereference(priv->match);
+		nft_pipapo_do_walk(ctx, set, m, iter);
+		rcu_read_unlock();
+		break;
+	default:
+		iter->err = -EINVAL;
+		WARN_ON_ONCE(1);
+		break;
+	}
 }
 
 /**

@@ -1854,52 +1854,31 @@ static void nft_pipapo_activate(const st
 }
 
 /**
- * pipapo_deactivate() - Check that element is in set, mark as inactive
+ * nft_pipapo_deactivate() - Search for element and make it inactive
  * @net:	Network namespace
  * @set:	nftables API set representation
- * @data:	Input key data
- * @ext:	nftables API extension pointer, used to check for end element
- *
- * This is a convenience function that can be called from both
- * nft_pipapo_deactivate() and nft_pipapo_flush(), as they are in fact the same
- * operation.
+ * @elem:	nftables API element representation containing key data
  *
  * Return: deactivated element if found, NULL otherwise.
  */
-static void *pipapo_deactivate(const struct net *net, const struct nft_set *set,
-			       const u8 *data, const struct nft_set_ext *ext)
+static void *nft_pipapo_deactivate(const struct net *net,
+				   const struct nft_set *set,
+				   const struct nft_set_elem *elem)
 {
 	struct nft_pipapo_elem *e;
 
-	e = pipapo_get(net, set, data, nft_genmask_next(net),
-		       nft_net_tstamp(net), GFP_KERNEL);
+	e = pipapo_get(net, set, (const u8 *)elem->key.val.data,
+		       nft_genmask_next(net), nft_net_tstamp(net), GFP_KERNEL);
 	if (IS_ERR(e))
 		return NULL;
 
 	nft_set_elem_change_active(net, set, &e->ext);
 
-	return e;
-}
-
-/**
- * nft_pipapo_deactivate() - Call pipapo_deactivate() to make element inactive
- * @net:	Network namespace
- * @set:	nftables API set representation
- * @elem:	nftables API element representation containing key data
- *
- * Return: deactivated element if found, NULL otherwise.
- */
-static void *nft_pipapo_deactivate(const struct net *net,
-				   const struct nft_set *set,
-				   const struct nft_set_elem *elem)
-{
-	const struct nft_set_ext *ext = nft_set_elem_ext(set, elem->priv);
-
-	return pipapo_deactivate(net, set, (const u8 *)elem->key.val.data, ext);
+	return &e->priv;
 }
 
 /**
- * nft_pipapo_flush() - Call pipapo_deactivate() to make element inactive
+ * nft_pipapo_flush() - make element inactive
  * @net:	Network namespace
  * @set:	nftables API set representation
  * @elem:	nftables API element representation containing key data

@@ -502,6 +502,7 @@ out:
  * pipapo_get() - Get matching element reference given key data
  * @net:	Network namespace
  * @set:	nftables API set representation
+ * @m:		storage containing active/existing elements
  * @data:	Key data to be matched against existing elements
  * @genmask:	If set, check that element is active in given genmask
  * @tstamp:	timestamp to check for expired elements

@@ -515,17 +516,15 @@ out:
  */
 static struct nft_pipapo_elem *pipapo_get(const struct net *net,
 					  const struct nft_set *set,
+					  const struct nft_pipapo_match *m,
 					  const u8 *data, u8 genmask,
 					  u64 tstamp, gfp_t gfp)
 {
 	struct nft_pipapo_elem *ret = ERR_PTR(-ENOENT);
-	struct nft_pipapo *priv = nft_set_priv(set);
 	unsigned long *res_map, *fill_map = NULL;
-	const struct nft_pipapo_match *m;
 	const struct nft_pipapo_field *f;
 	int i;
 
-	m = priv->clone;
 	if (m->bsize_max == 0)
 		return ret;
 

@@ -609,7 +608,10 @@ out:
 static void *nft_pipapo_get(const struct net *net, const struct nft_set *set,
 			    const struct nft_set_elem *elem, unsigned int flags)
 {
-	return pipapo_get(net, set, (const u8 *)elem->key.val.data,
+	struct nft_pipapo *priv = nft_set_priv(set);
+	struct nft_pipapo_match *m = rcu_dereference(priv->match);
+
+	return pipapo_get(net, set, m, (const u8 *)elem->key.val.data,
 			 nft_genmask_cur(net), get_jiffies_64(), GFP_ATOMIC);
 }
 

@@ -1258,7 +1260,7 @@ static int nft_pipapo_insert(const struc
 	else
 		end = start;
 
-	dup = pipapo_get(net, set, start, genmask, tstamp, GFP_KERNEL);
+	dup = pipapo_get(net, set, m, start, genmask, tstamp, GFP_KERNEL);
 	if (!IS_ERR(dup)) {
 		/* Check if we already have the same exact entry */
 		const struct nft_data *dup_key, *dup_end;

@@ -1280,7 +1282,7 @@ static int nft_pipapo_insert(const struc
 
 	if (PTR_ERR(dup) == -ENOENT) {
 		/* Look for partially overlapping entries */
-		dup = pipapo_get(net, set, end, nft_genmask_next(net), tstamp,
+		dup = pipapo_get(net, set, m, end, nft_genmask_next(net), tstamp,
 				 GFP_KERNEL);
 	}
 

@@ -1865,16 +1867,18 @@ static void *nft_pipapo_deactivate(const
 				   const struct nft_set *set,
 				   const struct nft_set_elem *elem)
 {
+	const struct nft_pipapo *priv = nft_set_priv(set);
+	struct nft_pipapo_match *m = priv->clone;
 	struct nft_pipapo_elem *e;
 
-	e = pipapo_get(net, set, (const u8 *)elem->key.val.data,
+	e = pipapo_get(net, set, m, (const u8 *)elem->key.val.data,
 		       nft_genmask_next(net), nft_net_tstamp(net), GFP_KERNEL);
 	if (IS_ERR(e))
 		return NULL;
 
 	nft_set_elem_change_active(net, set, &e->ext);
 
-	return &e->priv;
+	return e;
 }
 
 /**

@@ -1900,8 +1904,9 @@ static bool nft_pipapo_flush(const struc
 {
 	struct nft_pipapo_elem *e = elem;
 
-	return pipapo_deactivate(net, set, (const u8 *)nft_set_ext_key(&e->ext),
-				 &e->ext);
+	nft_set_elem_change_active(net, set, &e->ext);
+
+	return true;
 }
 
 /**

@@ -1230,6 +1230,29 @@ static bool nft_pipapo_transaction_mutex
 #endif
 }
 
+static struct nft_pipapo_match *pipapo_clone(struct nft_pipapo_match *old);
+
+/**
+ * pipapo_maybe_clone() - Build clone for pending data changes, if not existing
+ * @set:	nftables API set representation
+ *
+ * Return: newly created or existing clone, if any. NULL on allocation failure
+ */
+static struct nft_pipapo_match *pipapo_maybe_clone(const struct nft_set *set)
+{
+	struct nft_pipapo *priv = nft_set_priv(set);
+	struct nft_pipapo_match *m;
+
+	if (priv->clone)
+		return priv->clone;
+
+	m = rcu_dereference_protected(priv->match,
+				      nft_pipapo_transaction_mutex_held(set));
+	priv->clone = pipapo_clone(m);
+
+	return priv->clone;
+}
+
 /**
  * nft_pipapo_insert() - Validate and insert ranged elements
  * @net:	Network namespace

@@ -1247,14 +1270,17 @@ static int nft_pipapo_insert(const struc
 	union nft_pipapo_map_bucket rulemap[NFT_PIPAPO_MAX_FIELDS];
 	const u8 *start = (const u8 *)elem->key.val.data, *end;
 	struct nft_pipapo_elem *e = elem->priv, *dup;
+	struct nft_pipapo_match *m = pipapo_maybe_clone(set);
 	struct nft_pipapo *priv = nft_set_priv(set);
-	struct nft_pipapo_match *m = priv->clone;
 	u8 genmask = nft_genmask_next(net);
 	u64 tstamp = nft_net_tstamp(net);
 	struct nft_pipapo_field *f;
 	const u8 *start_p, *end_p;
 	int i, bsize_max, err = 0;
 
+	if (!m)
+		return -ENOMEM;
+
 	if (nft_set_ext_exists(ext, NFT_SET_EXT_KEY_END))
 		end = (const u8 *)nft_set_ext_key_end(ext)->data;
 	else

@@ -1791,7 +1817,10 @@ static void pipapo_reclaim_match(struct
 static void nft_pipapo_commit(struct nft_set *set)
 {
 	struct nft_pipapo *priv = nft_set_priv(set);
-	struct nft_pipapo_match *new_clone, *old;
+	struct nft_pipapo_match *old;
+
+	if (!priv->clone)
+		return;
 
 	if (time_after_eq(jiffies, priv->last_gc + nft_set_gc_interval(set)))
 		pipapo_gc_scan(set, priv->clone);

@@ -1799,40 +1828,29 @@ static void nft_pipapo_commit(struct nft
 	if (!priv->dirty)
 		return;
 
-	new_clone = pipapo_clone(priv->clone);
-	if (!new_clone)
-		return;
-
+	old = rcu_replace_pointer(priv->match, priv->clone,
+				  nft_pipapo_transaction_mutex_held(set));
+	priv->clone = NULL;
 	priv->dirty = false;
 
-	old = rcu_access_pointer(priv->match);
-	rcu_assign_pointer(priv->match, priv->clone);
 	if (old)
 		call_rcu(&old->rcu, pipapo_reclaim_match);
 
-	priv->clone = new_clone;
-
 	pipapo_gc_queue(set);
 }
 
 static void nft_pipapo_abort(const struct nft_set *set)
 {
 	struct nft_pipapo *priv = nft_set_priv(set);
-	struct nft_pipapo_match *new_clone, *m;
 
 	if (!priv->dirty)
 		return;
 
-	m = rcu_dereference_protected(priv->match, nft_pipapo_transaction_mutex_held(set));
-
-	new_clone = pipapo_clone(m);
-	if (!new_clone)
+	if (!priv->clone)
 		return;
-
 	priv->dirty = false;
-
 	pipapo_free_match(priv->clone);
-	priv->clone = new_clone;
+	priv->clone = NULL;
 }
 
 /**

@@ -1867,10 +1885,15 @@ static void *nft_pipapo_deactivate(const
 				   const struct nft_set *set,
 				   const struct nft_set_elem *elem)
 {
-	const struct nft_pipapo *priv = nft_set_priv(set);
-	struct nft_pipapo_match *m = priv->clone;
+	struct nft_pipapo_match *m = pipapo_maybe_clone(set);
 	struct nft_pipapo_elem *e;
 
+	/* removal must occur on priv->clone, if we are low on memory
+	 * we have no choice and must fail the removal request.
+	 */
+	if (!m)
+		return NULL;
+
 	e = pipapo_get(net, set, m, (const u8 *)elem->key.val.data,
 		       nft_genmask_next(net), nft_net_tstamp(net), GFP_KERNEL);
 	if (IS_ERR(e))

@@ -2153,7 +2176,12 @@ static void nft_pipapo_walk(const struct
 
 	switch (iter->type) {
 	case NFT_ITER_UPDATE:
-		m = priv->clone;
+		m = pipapo_maybe_clone(set);
+		if (!m) {
+			iter->err = -ENOMEM;
+			return;
+		}
+
 		nft_pipapo_do_walk(ctx, set, m, iter);
 		break;
 	case NFT_ITER_READ:

@@ -342,6 +342,8 @@
 #include "nft_set_pipapo_avx2.h"
 #include "nft_set_pipapo.h"
 
+static void nft_pipapo_abort(const struct nft_set *set);
+
 /**
  * pipapo_refill() - For each set bit, set bits from selected mapping table item
  * @map:	Bitmap to be scanned for set bits

@@ -1278,7 +1280,7 @@ static int nft_pipapo_insert(const struc
 	const u8 *start_p, *end_p;
 	int i, bsize_max, err = 0;
 
-	if (!m)
+	if (!m || m->state == NFT_PIPAPO_CLONE_ERR)
 		return -ENOMEM;
 
 	if (nft_set_ext_exists(ext, NFT_SET_EXT_KEY_END))

@@ -1351,8 +1353,10 @@ static int nft_pipapo_insert(const struc
 		else
 			ret = pipapo_expand(f, start, end, f->groups * f->bb);
 
-		if (ret < 0)
-			return ret;
+		if (ret < 0) {
+			err = ret;
+			goto abort;
+		}
 
 		if (f->bsize > bsize_max)
 			bsize_max = f->bsize;

@@ -1368,7 +1372,7 @@ static int nft_pipapo_insert(const struc
 
 		err = pipapo_realloc_scratch(m, bsize_max);
 		if (err)
-			return err;
+			goto abort;
 
 		m->bsize_max = bsize_max;
 	} else {

@@ -1379,7 +1383,26 @@ static int nft_pipapo_insert(const struc
 
 	pipapo_map(m, rulemap, e);
 
+	m->state = NFT_PIPAPO_CLONE_MOD;
 	return 0;
+abort:
+	DEBUG_NET_WARN_ON_ONCE(m->state == NFT_PIPAPO_CLONE_ERR);
+
+	/* Two rollback cases:
+	 * 1) no previous changes.  nft_pipapo_abort is not
+	 * guaranteed to be invoked (there might be no further
+	 * add/delete requests coming after this).
+	 *
+	 * 2) we had previous changes: there are transaction
+	 * records pointing to this set.  Leave the rollback to
+	 * the transaction handling.
+	 */
+	if (m->state == NFT_PIPAPO_CLONE_NEW)
+		nft_pipapo_abort(set); /* releases m */
+	else
+		m->state = NFT_PIPAPO_CLONE_ERR;
+
+	return err;
 }
 
 /**

@@ -1455,6 +1478,7 @@ static struct nft_pipapo_match *pipapo_c
 		dst++;
 	}
 
+	new->state = NFT_PIPAPO_CLONE_NEW;
 	return new;
 
 out_mt:

@@ -1891,7 +1915,7 @@ static void *nft_pipapo_deactivate(const
 	/* removal must occur on priv->clone, if we are low on memory
 	 * we have no choice and must fail the removal request.
 	 */
-	if (!m)
+	if (!m || m->state == NFT_PIPAPO_CLONE_ERR)
 		return NULL;
 
 	e = pipapo_get(net, set, m, (const u8 *)elem->key.val.data,

@@ -142,9 +142,16 @@ struct nft_pipapo_scratch {
 	unsigned long map[];
 };
 
+enum nft_pipapo_clone_state {
+	NFT_PIPAPO_CLONE_NEW,
+	NFT_PIPAPO_CLONE_MOD,
+	NFT_PIPAPO_CLONE_ERR,
+};
+
 /**
  * struct nft_pipapo_match - Data used for lookup and matching
  * @field_count		Amount of fields in set
+ * @state:		add/delete state; used from control plane
  * @scratch:		Preallocated per-CPU maps for partial matching results
  * @bsize_max:		Maximum lookup table bucket size of all fields, in longs
  * @rcu			Matching data is swapped on commits

@@ -152,6 +159,7 @@ struct nft_pipapo_scratch {
  */
 struct nft_pipapo_match {
 	int field_count;
+	enum nft_pipapo_clone_state state:8;
 	struct nft_pipapo_scratch * __percpu *scratch;
 	size_t bsize_max;
 	struct rcu_head rcu;

@@ -61,6 +61,7 @@ struct lsm_blob_sizes {
 	int	lbs_file;
 	int lbs_backing_file;
 	int	lbs_inode;
+	int	lbs_sock;
 	int	lbs_superblock;
 	int	lbs_ipc;
 	int	lbs_msg_msg;

@@ -51,8 +51,12 @@ struct aa_sk_ctx {
 	struct aa_label *peer;
 };
 
-#define SK_CTX(X) ((X)->sk_security)
 #define SOCK_ctx(X) SOCK_INODE(X)->i_security
+static inline struct aa_sk_ctx *aa_sock(const struct sock *sk)
+{
+	return sk->sk_security + apparmor_blob_sizes.lbs_sock;
+}
+
 #define DEFINE_AUDIT_NET(NAME, OP, SK, F, T, P)				  \
 	struct lsm_network_audit NAME ## _net = { .sk = (SK),		  \
 						  .family = (F)};	  \

@@ -851,32 +851,14 @@ static int apparmor_task_kill(struct tas
 }
 
 /**
- * apparmor_sk_alloc_security - allocate and attach the sk_security field
- */
-static int apparmor_sk_alloc_security(struct sock *sk, int family, gfp_t flags)
-{
-	struct aa_sk_ctx *ctx;
-
-	ctx = kzalloc(sizeof(*ctx), flags);
-	if (!ctx)
-		return -ENOMEM;
-
-	SK_CTX(sk) = ctx;
-
-	return 0;
-}
-
-/**
  * apparmor_sk_free_security - free the sk_security field
  */
 static void apparmor_sk_free_security(struct sock *sk)
 {
-	struct aa_sk_ctx *ctx = SK_CTX(sk);
+	struct aa_sk_ctx *ctx = aa_sock(sk);
 
-	SK_CTX(sk) = NULL;
 	aa_put_label(ctx->label);
 	aa_put_label(ctx->peer);
-	kfree(ctx);
 }
 
 /**

@@ -885,8 +867,8 @@ static void apparmor_sk_free_security(st
 static void apparmor_sk_clone_security(const struct sock *sk,
 				       struct sock *newsk)
 {
-	struct aa_sk_ctx *ctx = SK_CTX(sk);
-	struct aa_sk_ctx *new = SK_CTX(newsk);
+	struct aa_sk_ctx *ctx = aa_sock(sk);
+	struct aa_sk_ctx *new = aa_sock(newsk);
 
 	if (new->label)
 		aa_put_label(new->label);

@@ -940,7 +922,7 @@ static int apparmor_socket_post_create(s
 		label = aa_get_current_label();
 
 	if (sock->sk) {
-		struct aa_sk_ctx *ctx = SK_CTX(sock->sk);
+		struct aa_sk_ctx *ctx = aa_sock(sock->sk);
 
 		aa_put_label(ctx->label);
 		ctx->label = aa_get_label(label);

@@ -1139,7 +1121,7 @@ static int apparmor_socket_shutdown(stru
  */
 static int apparmor_socket_sock_rcv_skb(struct sock *sk, struct sk_buff *skb)
 {
-	struct aa_sk_ctx *ctx = SK_CTX(sk);
+	struct aa_sk_ctx *ctx = aa_sock(sk);
 
 	if (!skb->secmark)
 		return 0;

@@ -1159,7 +1141,7 @@ static int apparmor_socket_sock_rcv_skb(
 
 static struct aa_label *sk_peer_label(struct sock *sk)
 {
-	struct aa_sk_ctx *ctx = SK_CTX(sk);
+	struct aa_sk_ctx *ctx = aa_sock(sk);
 
 	if (ctx->peer)
 		return ctx->peer;

@@ -1240,7 +1222,7 @@ static int apparmor_socket_getpeersec_dg
  */
 static void apparmor_sock_graft(struct sock *sk, struct socket *parent)
 {
-	struct aa_sk_ctx *ctx = SK_CTX(sk);
+	struct aa_sk_ctx *ctx = aa_sock(sk);
 
 	if (!ctx->label)
 		ctx->label = aa_get_current_label();

@@ -1250,7 +1232,7 @@ static void apparmor_sock_graft(struct s
 static int apparmor_inet_conn_request(const struct sock *sk, struct sk_buff *skb,
 				      struct request_sock *req)
 {
-	struct aa_sk_ctx *ctx = SK_CTX(sk);
+	struct aa_sk_ctx *ctx = aa_sock(sk);
 
 	if (!skb->secmark)
 		return 0;

@@ -1267,6 +1249,7 @@ struct lsm_blob_sizes apparmor_blob_size
 	.lbs_cred = sizeof(struct aa_label *),
 	.lbs_file = sizeof(struct aa_file_ctx),
 	.lbs_task = sizeof(struct aa_task_ctx),
+	.lbs_sock = sizeof(struct aa_sk_ctx),
 };
 
 static struct security_hook_list apparmor_hooks[] __ro_after_init = {

@@ -1305,7 +1288,6 @@ static struct security_hook_list apparmo
 	LSM_HOOK_INIT(getprocattr, apparmor_getprocattr),
 	LSM_HOOK_INIT(setprocattr, apparmor_setprocattr),
 
-	LSM_HOOK_INIT(sk_alloc_security, apparmor_sk_alloc_security),
 	LSM_HOOK_INIT(sk_free_security, apparmor_sk_free_security),
 	LSM_HOOK_INIT(sk_clone_security, apparmor_sk_clone_security),
 

@@ -1864,7 +1846,7 @@ static unsigned int apparmor_ip_postrout
 	if (sk == NULL)
 		return NF_ACCEPT;
 
-	ctx = SK_CTX(sk);
+	ctx = aa_sock(sk);
 	if (!apparmor_secmark_check(ctx->label, OP_SENDMSG, AA_MAY_SEND,
 				    skb->secmark, sk))
 		return NF_ACCEPT;

@@ -152,7 +152,7 @@ static int aa_label_sk_perm(const struct
 			    const char *op, u32 request,
 			    struct sock *sk)
 {
-	struct aa_sk_ctx *ctx = SK_CTX(sk);
+	struct aa_sk_ctx *ctx = aa_sock(sk);
 	int error = 0;
 
 	AA_BUG(!label);

@@ -30,6 +30,7 @@
 #include <linux/string.h>
 #include <linux/msg.h>
 #include <net/flow.h>
+#include <net/sock.h>
 
 /* How many LSMs were built into the kernel? */
 #define LSM_COUNT (__end_lsm_info - __start_lsm_info)

@@ -212,6 +213,7 @@ static void __init lsm_set_blob_sizes(st
 	lsm_set_blob_size(&needed->lbs_inode, &blob_sizes.lbs_inode);
 	lsm_set_blob_size(&needed->lbs_ipc, &blob_sizes.lbs_ipc);
 	lsm_set_blob_size(&needed->lbs_msg_msg, &blob_sizes.lbs_msg_msg);
+	lsm_set_blob_size(&needed->lbs_sock, &blob_sizes.lbs_sock);
 	lsm_set_blob_size(&needed->lbs_superblock, &blob_sizes.lbs_superblock);
 	lsm_set_blob_size(&needed->lbs_task, &blob_sizes.lbs_task);
 	lsm_set_blob_size(&needed->lbs_xattr_count,

@@ -381,6 +383,7 @@ static void __init ordered_lsm_init(void
 	init_debug("inode blob size      = %d\n", blob_sizes.lbs_inode);
 	init_debug("ipc blob size        = %d\n", blob_sizes.lbs_ipc);
 	init_debug("msg_msg blob size    = %d\n", blob_sizes.lbs_msg_msg);
+	init_debug("sock blob size       = %d\n", blob_sizes.lbs_sock);
 	init_debug("superblock blob size = %d\n", blob_sizes.lbs_superblock);
 	init_debug("task blob size       = %d\n", blob_sizes.lbs_task);
 	init_debug("xattr slots          = %d\n", blob_sizes.lbs_xattr_count);

@@ -4560,6 +4563,28 @@ int security_socket_getpeersec_dgram(str
 EXPORT_SYMBOL(security_socket_getpeersec_dgram);
 
 /**
+ * lsm_sock_alloc - allocate a composite sock blob
+ * @sock: the sock that needs a blob
+ * @priority: allocation mode
+ *
+ * Allocate the sock blob for all the modules
+ *
+ * Returns 0, or -ENOMEM if memory can't be allocated.
+ */
+static int lsm_sock_alloc(struct sock *sock, gfp_t priority)
+{
+	if (blob_sizes.lbs_sock == 0) {
+		sock->sk_security = NULL;
+		return 0;
+	}
+
+	sock->sk_security = kzalloc(blob_sizes.lbs_sock, priority);
+	if (sock->sk_security == NULL)
+		return -ENOMEM;
+	return 0;
+}
+
+/**
  * security_sk_alloc() - Allocate and initialize a sock's LSM blob
  * @sk: sock
  * @family: protocol family

@@ -4572,7 +4597,14 @@ EXPORT_SYMBOL(security_socket_getpeersec
  */
 int security_sk_alloc(struct sock *sk, int family, gfp_t priority)
 {
-	return call_int_hook(sk_alloc_security, 0, sk, family, priority);
+	int rc = lsm_sock_alloc(sk, priority);
+
+	if (unlikely(rc))
+		return rc;
+	rc = call_int_hook(sk_alloc_security, 0, sk, family, priority);
+	if (unlikely(rc))
+		security_sk_free(sk);
+	return rc;
 }
 
 /**

@@ -4584,6 +4616,8 @@ int security_sk_alloc(struct sock *sk, i
 void security_sk_free(struct sock *sk)
 {
 	call_void_hook(sk_free_security, sk);
+	kfree(sk->sk_security);
+	sk->sk_security = NULL;
 }
 
 /**

265 lines not on this page · raw message

@@ -205,4 +205,9 @@ static inline struct superblock_security
 	return superblock->s_security + selinux_blob_sizes.lbs_superblock;
 }
 
+static inline struct sk_security_struct *selinux_sock(const struct sock *sock)
+{
+	return sock->sk_security + selinux_blob_sizes.lbs_sock;
+}
+
 #endif /* _SELINUX_OBJSEC_H_ */

@@ -17,6 +17,7 @@
 #include <linux/gfp.h>
 #include <linux/ip.h>
 #include <linux/ipv6.h>
+#include <linux/lsm_hooks.h>
 #include <net/sock.h>
 #include <net/netlabel.h>
 #include <net/ip.h>

@@ -68,7 +69,7 @@ static int selinux_netlbl_sidlookup_cach
 static struct netlbl_lsm_secattr *selinux_netlbl_sock_genattr(struct sock *sk)
 {
 	int rc;
-	struct sk_security_struct *sksec = sk->sk_security;
+	struct sk_security_struct *sksec = selinux_sock(sk);
 	struct netlbl_lsm_secattr *secattr;
 
 	if (sksec->nlbl_secattr != NULL)

@@ -100,7 +101,7 @@ static struct netlbl_lsm_secattr *selinu
 							const struct sock *sk,
 							u32 sid)
 {
-	struct sk_security_struct *sksec = sk->sk_security;
+	struct sk_security_struct *sksec = selinux_sock(sk);
 	struct netlbl_lsm_secattr *secattr = sksec->nlbl_secattr;
 
 	if (secattr == NULL)

@@ -240,7 +241,7 @@ int selinux_netlbl_skbuff_setsid(struct
 	 * being labeled by it's parent socket, if it is just exit */
 	sk = skb_to_full_sk(skb);
 	if (sk != NULL) {
-		struct sk_security_struct *sksec = sk->sk_security;
+		struct sk_security_struct *sksec = selinux_sock(sk);
 
 		if (sksec->nlbl_state != NLBL_REQSKB)
 			return 0;

@@ -277,7 +278,7 @@ int selinux_netlbl_sctp_assoc_request(st
 {
 	int rc;
 	struct netlbl_lsm_secattr secattr;
-	struct sk_security_struct *sksec = asoc->base.sk->sk_security;
+	struct sk_security_struct *sksec = selinux_sock(asoc->base.sk);
 	struct sockaddr_in addr4;
 	struct sockaddr_in6 addr6;
 

@@ -356,7 +357,7 @@ inet_conn_request_return:
  */
 void selinux_netlbl_inet_csk_clone(struct sock *sk, u16 family)
 {
-	struct sk_security_struct *sksec = sk->sk_security;
+	struct sk_security_struct *sksec = selinux_sock(sk);
 
 	if (family == PF_INET)
 		sksec->nlbl_state = NLBL_LABELED;

@@ -374,8 +375,8 @@ void selinux_netlbl_inet_csk_clone(struc
  */
 void selinux_netlbl_sctp_sk_clone(struct sock *sk, struct sock *newsk)
 {
-	struct sk_security_struct *sksec = sk->sk_security;
-	struct sk_security_struct *newsksec = newsk->sk_security;
+	struct sk_security_struct *sksec = selinux_sock(sk);
+	struct sk_security_struct *newsksec = selinux_sock(newsk);
 
 	newsksec->nlbl_state = sksec->nlbl_state;
 }

@@ -393,7 +394,7 @@ void selinux_netlbl_sctp_sk_clone(struct
 int selinux_netlbl_socket_post_create(struct sock *sk, u16 family)
 {
 	int rc;
-	struct sk_security_struct *sksec = sk->sk_security;
+	struct sk_security_struct *sksec = selinux_sock(sk);
 	struct netlbl_lsm_secattr *secattr;
 
 	if (family != PF_INET && family != PF_INET6)

@@ -507,7 +508,7 @@ int selinux_netlbl_socket_setsockopt(str
 {
 	int rc = 0;
 	struct sock *sk = sock->sk;
-	struct sk_security_struct *sksec = sk->sk_security;
+	struct sk_security_struct *sksec = selinux_sock(sk);
 	struct netlbl_lsm_secattr secattr;
 
 	if (selinux_netlbl_option(level, optname) &&

@@ -545,7 +546,7 @@ static int selinux_netlbl_socket_connect
 						struct sockaddr *addr)
 {
 	int rc;
-	struct sk_security_struct *sksec = sk->sk_security;
+	struct sk_security_struct *sksec = selinux_sock(sk);
 	struct netlbl_lsm_secattr *secattr;
 
 	/* connected sockets are allowed to disconnect when the address family

@@ -584,7 +585,7 @@ static int selinux_netlbl_socket_connect
 int selinux_netlbl_socket_connect_locked(struct sock *sk,
 					 struct sockaddr *addr)
 {
-	struct sk_security_struct *sksec = sk->sk_security;
+	struct sk_security_struct *sksec = selinux_sock(sk);
 
 	if (sksec->nlbl_state != NLBL_REQSKB &&
 	    sksec->nlbl_state != NLBL_CONNLABELED)

@@ -361,6 +361,11 @@ static inline struct superblock_smack *s
 	return superblock->s_security + smack_blob_sizes.lbs_superblock;
 }
 
+static inline struct socket_smack *smack_sock(const struct sock *sock)
+{
+	return sock->sk_security + smack_blob_sizes.lbs_sock;
+}
+
 /*
  * Is the directory transmuting?
  */

243 lines not on this page · raw message

@@ -26,8 +26,8 @@ static unsigned int smack_ip_output(void
 	struct socket_smack *ssp;
 	struct smack_known *skp;
 
-	if (sk && sk->sk_security) {
-		ssp = sk->sk_security;
+	if (sk) {
+		ssp = smack_sock(sk);
 		skp = ssp->smk_out;
 		skb->secmark = skp->smk_secid;
 	}

@@ -4777,9 +4777,8 @@ static int selinux_socket_socketpair(str
    Need to determine whether we should perform a name_bind
    permission check between the socket and the port number. */
 
-static int selinux_socket_bind(struct socket *sock, struct sockaddr *address, int addrlen)
+static int __selinux_socket_bind(struct sock *sk, struct sockaddr *address, int addrlen)
 {
-	struct sock *sk = sock->sk;
 	struct sk_security_struct *sksec = selinux_sock(sk);
 	u16 family;
 	int err;

@@ -4913,13 +4912,17 @@ err_af:
 	return -EAFNOSUPPORT;
 }
 
+static int selinux_socket_bind(struct socket *sock, struct sockaddr *address, int addrlen)
+{
+	return __selinux_socket_bind(sock->sk, address, addrlen);
+}
+
 /* This supports connect(2) and SCTP connect services such as sctp_connectx(3)
  * and sctp_sendmsg(3) as described in Documentation/security/SCTP.rst
  */
-static int selinux_socket_connect_helper(struct socket *sock,
+static int selinux_socket_connect_helper(struct sock *sk,
 					 struct sockaddr *address, int addrlen)
 {
-	struct sock *sk = sock->sk;
 	struct sk_security_struct *sksec = selinux_sock(sk);
 	int err;
 

@@ -5012,7 +5015,7 @@ static int selinux_socket_connect(struct
 	int err;
 	struct sock *sk = sock->sk;
 
-	err = selinux_socket_connect_helper(sock, address, addrlen);
+	err = selinux_socket_connect_helper(sk, address, addrlen);
 	if (err)
 		return err;
 

@@ -5512,13 +5515,11 @@ static int selinux_sctp_bind_connect(str
 	int len, err = 0, walk_size = 0;
 	void *addr_buf;
 	struct sockaddr *addr;
-	struct socket *sock;
 
 	if (!selinux_policycap_extsockclass())
 		return 0;
 
 	/* Process one or more addresses that may be IPv4 or IPv6 */
-	sock = sk->sk_socket;
 	addr_buf = address;
 
 	while (walk_size < addrlen) {

@@ -5547,14 +5548,14 @@ static int selinux_sctp_bind_connect(str
 		case SCTP_PRIMARY_ADDR:
 		case SCTP_SET_PEER_PRIMARY_ADDR:
 		case SCTP_SOCKOPT_BINDX_ADD:
-			err = selinux_socket_bind(sock, addr, len);
+			err = __selinux_socket_bind(sk, addr, len);
 			break;
 		/* Connect checks */
 		case SCTP_SOCKOPT_CONNECTX:
 		case SCTP_PARAM_SET_PRIMARY:
 		case SCTP_PARAM_ADD_IP:
 		case SCTP_SENDMSG_CONNECT:
-			err = selinux_socket_connect_helper(sock, addr, len);
+			err = selinux_socket_connect_helper(sk, addr, len);
 			if (err)
 				return err;
 

@@ -948,9 +948,8 @@ static int set_expected_rtp_rtcp(struct
 		saddr = &ct->tuplehash[!dir].tuple.src.u3;
 	} else if (sip_external_media) {
 		struct net_device *dev = skb_dst(skb)->dev;
-		struct net *net = dev_net(dev);
-		struct flowi fl;
 		struct dst_entry *dst = NULL;
+		struct flowi fl;
 
 		memset(&fl, 0, sizeof(fl));
 

@@ -947,7 +947,6 @@ static int set_expected_rtp_rtcp(struct
 			return NF_ACCEPT;
 		saddr = &ct->tuplehash[!dir].tuple.src.u3;
 	} else if (sip_external_media) {
-		struct net_device *dev = skb_dst(skb)->dev;
 		struct dst_entry *dst = NULL;
 		struct flowi fl;
 

@@ -969,7 +968,11 @@ static int set_expected_rtp_rtcp(struct
 		 * through the same interface as the signalling peer.
 		 */
 		if (dst) {
-			bool external_media = (dst->dev == dev);
+			const struct dst_entry *this_dst = skb_dst(skb);
+			bool external_media = false;
+
+			if (this_dst && dst->dev == this_dst->dev)
+				external_media = true;
 
 			dst_release(dst);
 			if (external_media)

@@ -1851,11 +1851,6 @@ static inline u64 nft_net_tstamp(const s
 #define __NFT_REDUCE_READONLY	1UL
 #define NFT_REDUCE_READONLY	(void *)__NFT_REDUCE_READONLY
 
-static inline bool nft_reduce_is_readonly(const struct nft_expr *expr)
-{
-	return expr->ops->reduce == NFT_REDUCE_READONLY;
-}
-
 void nft_reg_track_update(struct nft_regs_track *track,
 			  const struct nft_expr *expr, u8 dreg, u8 len);
 void nft_reg_track_cancel(struct nft_regs_track *track, u8 dreg, u8 len);

@@ -123,17 +123,6 @@ struct nft_regs {
 	};
 };
 
-struct nft_regs_track {
-	struct {
-		const struct nft_expr		*selector;
-		const struct nft_expr		*bitwise;
-		u8				num_reg;
-	} regs[NFT_REG32_NUM];
-
-	const struct nft_expr			*cur;
-	const struct nft_expr			*last;
-};
-
 /* Store/load an u8, u16 or u64 integer to/from the u32 data register.
  *
  * Note, when using concatenations, register allocation happens at 32-bit

@@ -420,8 +409,6 @@ int nft_expr_clone(struct nft_expr *dst,
 void nft_expr_destroy(const struct nft_ctx *ctx, struct nft_expr *expr);
 int nft_expr_dump(struct sk_buff *skb, unsigned int attr,
 		  const struct nft_expr *expr, bool reset);
-bool nft_expr_reduce_bitwise(struct nft_regs_track *track,
-			     const struct nft_expr *expr);
 
 struct nft_set_ext;
 

@@ -933,7 +920,6 @@ struct nft_offload_ctx;
  *	@destroy_clone: destruction clone function
  *	@dump: function to dump parameters
  *	@validate: validate expression, called during loop detection
- *	@reduce: reduce expression
  *	@gc: garbage collection expression
  *	@offload: hardware offload expression
  *	@offload_action: function to report true/false to allocate one slot or not in the flow

@@ -967,8 +953,6 @@ struct nft_expr_ops {
 						bool reset);
 	int				(*validate)(const struct nft_ctx *ctx,
 						    const struct nft_expr *expr);
-	bool				(*reduce)(struct nft_regs_track *track,
-						  const struct nft_expr *expr);
 	bool				(*gc)(struct net *net,
 					      const struct nft_expr *expr);
 	int				(*offload)(struct nft_offload_ctx *ctx,

@@ -1848,20 +1832,4 @@ static inline u64 nft_net_tstamp(const s
 	return nft_pernet(net)->tstamp;
 }
 
-#define __NFT_REDUCE_READONLY	1UL
-#define NFT_REDUCE_READONLY	(void *)__NFT_REDUCE_READONLY
-
-void nft_reg_track_update(struct nft_regs_track *track,
-			  const struct nft_expr *expr, u8 dreg, u8 len);
-void nft_reg_track_cancel(struct nft_regs_track *track, u8 dreg, u8 len);
-void __nft_reg_track_cancel(struct nft_regs_track *track, u8 dreg);
-
-static inline bool nft_reg_track_cmp(struct nft_regs_track *track,
-				     const struct nft_expr *expr, u8 dreg)
-{
-	return track->regs[dreg].selector &&
-	       track->regs[dreg].selector->ops == expr->ops &&
-	       track->regs[dreg].num_reg == 0;
-}
-
 #endif /* _NET_NF_TABLES_H */

@@ -36,6 +36,4 @@ void nft_fib6_eval(const struct nft_expr
 void nft_fib_store_result(void *reg, const struct nft_fib *priv,
 			  const struct net_device *dev);
 
-bool nft_fib_reduce(struct nft_regs_track *track,
-		    const struct nft_expr *expr);
 #endif

@@ -43,9 +43,6 @@ void nft_meta_set_destroy(const struct n
 int nft_meta_set_validate(const struct nft_ctx *ctx,
 			  const struct nft_expr *expr);
 
-bool nft_meta_get_reduce(struct nft_regs_track *track,
-			 const struct nft_expr *expr);
-
 struct nft_inner_tun_ctx;
 void nft_meta_inner_eval(const struct nft_expr *expr,
 			 struct nft_regs *regs, const struct nft_pktinfo *pkt,

@@ -104,7 +104,6 @@ static const struct nft_expr_ops nft_met
 	.eval		= nft_meta_bridge_get_eval,
 	.init		= nft_meta_bridge_get_init,
 	.dump		= nft_meta_get_dump,
-	.reduce		= nft_meta_get_reduce,
 };
 
 static void nft_meta_bridge_set_eval(const struct nft_expr *expr,

@@ -151,24 +150,6 @@ static int nft_meta_bridge_set_init(cons
 	return 0;
 }
 
-static bool nft_meta_bridge_set_reduce(struct nft_regs_track *track,
-				       const struct nft_expr *expr)
-{
-	int i;
-
-	for (i = 0; i < NFT_REG32_NUM; i++) {
-		if (!track->regs[i].selector)
-			continue;
-
-		if (track->regs[i].selector->ops != &nft_meta_bridge_get_ops)
-			continue;
-
-		__nft_reg_track_cancel(track, i);
-	}
-
-	return false;
-}
-
 static int nft_meta_bridge_set_validate(const struct nft_ctx *ctx,
 					const struct nft_expr *expr)
 {

@@ -193,7 +174,6 @@ static const struct nft_expr_ops nft_met
 	.init		= nft_meta_bridge_set_init,
 	.destroy	= nft_meta_set_destroy,
 	.dump		= nft_meta_set_dump,
-	.reduce		= nft_meta_bridge_set_reduce,
 	.validate	= nft_meta_bridge_set_validate,
 };
 

@@ -184,7 +184,6 @@ static const struct nft_expr_ops nft_rej
 	.init		= nft_reject_init,
 	.dump		= nft_reject_dump,
 	.validate	= nft_reject_bridge_validate,
-	.reduce		= NFT_REDUCE_READONLY,
 };
 
 static struct nft_expr_type nft_reject_bridge_type __read_mostly = {

@@ -76,7 +76,6 @@ static const struct nft_expr_ops nft_dup
 	.eval		= nft_dup_ipv4_eval,
 	.init		= nft_dup_ipv4_init,
 	.dump		= nft_dup_ipv4_dump,
-	.reduce		= NFT_REDUCE_READONLY,
 };
 
 static const struct nla_policy nft_dup_ipv4_policy[NFTA_DUP_MAX + 1] = {

@@ -157,7 +157,6 @@ static const struct nft_expr_ops nft_fib
 	.init		= nft_fib_init,
 	.dump		= nft_fib_dump,
 	.validate	= nft_fib_validate,
-	.reduce		= nft_fib_reduce,
 };
 
 static const struct nft_expr_ops nft_fib4_ops = {

@@ -167,7 +166,6 @@ static const struct nft_expr_ops nft_fib
 	.init		= nft_fib_init,
 	.dump		= nft_fib_dump,
 	.validate	= nft_fib_validate,
-	.reduce		= nft_fib_reduce,
 };
 
 static const struct nft_expr_ops *

@@ -45,7 +45,6 @@ static const struct nft_expr_ops nft_rej
 	.init		= nft_reject_init,
 	.dump		= nft_reject_dump,
 	.validate	= nft_reject_validate,
-	.reduce		= NFT_REDUCE_READONLY,
 };
 
 static struct nft_expr_type nft_reject_ipv4_type __read_mostly = {

@@ -74,7 +74,6 @@ static const struct nft_expr_ops nft_dup
 	.eval		= nft_dup_ipv6_eval,
 	.init		= nft_dup_ipv6_init,
 	.dump		= nft_dup_ipv6_dump,
-	.reduce		= NFT_REDUCE_READONLY,
 };
 
 static const struct nla_policy nft_dup_ipv6_policy[NFTA_DUP_MAX + 1] = {

@@ -226,7 +226,6 @@ static const struct nft_expr_ops nft_fib
 	.init		= nft_fib_init,
 	.dump		= nft_fib_dump,
 	.validate	= nft_fib_validate,
-	.reduce		= nft_fib_reduce,
 };
 
 static const struct nft_expr_ops nft_fib6_ops = {

@@ -236,7 +235,6 @@ static const struct nft_expr_ops nft_fib
 	.init		= nft_fib_init,
 	.dump		= nft_fib_dump,
 	.validate	= nft_fib_validate,
-	.reduce		= nft_fib_reduce,
 };
 
 static const struct nft_expr_ops *

@@ -46,7 +46,6 @@ static const struct nft_expr_ops nft_rej
 	.init		= nft_reject_init,
 	.dump		= nft_reject_dump,
 	.validate	= nft_reject_validate,
-	.reduce		= NFT_REDUCE_READONLY,
 };
 
 static struct nft_expr_type nft_reject_ipv6_type __read_mostly = {

89 lines not on this page · raw message

131 lines not on this page · raw message

@@ -177,23 +177,12 @@ nla_put_failure:
 	return -1;
 }
 
-static bool nft_byteorder_reduce(struct nft_regs_track *track,
-				 const struct nft_expr *expr)
-{
-	struct nft_byteorder *priv = nft_expr_priv(expr);
-
-	nft_reg_track_cancel(track, priv->dreg, priv->len);
-
-	return false;
-}
-
 static const struct nft_expr_ops nft_byteorder_ops = {
 	.type		= &nft_byteorder_type,
 	.size		= NFT_EXPR_SIZE(sizeof(struct nft_byteorder)),
 	.eval		= nft_byteorder_eval,
 	.init		= nft_byteorder_init,
 	.dump		= nft_byteorder_dump,
-	.reduce		= nft_byteorder_reduce,
 };
 
 struct nft_expr_type nft_byteorder_type __read_mostly = {

@@ -190,7 +190,6 @@ static const struct nft_expr_ops nft_cmp
 	.eval		= nft_cmp_eval,
 	.init		= nft_cmp_init,
 	.dump		= nft_cmp_dump,
-	.reduce		= NFT_REDUCE_READONLY,
 	.offload	= nft_cmp_offload,
 };
 

@@ -282,7 +281,6 @@ const struct nft_expr_ops nft_cmp_fast_o
 	.eval		= NULL,	/* inlined */
 	.init		= nft_cmp_fast_init,
 	.dump		= nft_cmp_fast_dump,
-	.reduce		= NFT_REDUCE_READONLY,
 	.offload	= nft_cmp_fast_offload,
 };
 

@@ -376,7 +374,6 @@ const struct nft_expr_ops nft_cmp16_fast
 	.eval		= NULL,	/* inlined */
 	.init		= nft_cmp16_fast_init,
 	.dump		= nft_cmp16_fast_dump,
-	.reduce		= NFT_REDUCE_READONLY,
 	.offload	= nft_cmp16_fast_offload,
 };
 

@@ -794,14 +794,6 @@ static const struct nfnetlink_subsystem
 
 static struct nft_expr_type nft_match_type;
 
-static bool nft_match_reduce(struct nft_regs_track *track,
-			     const struct nft_expr *expr)
-{
-	const struct xt_match *match = expr->ops->data;
-
-	return strcmp(match->name, "comment") == 0;
-}
-
 static const struct nft_expr_ops *
 nft_match_select_ops(const struct nft_ctx *ctx,
 		     const struct nlattr * const tb[])

@@ -844,7 +836,6 @@ nft_match_select_ops(const struct nft_ct
 	ops->dump = nft_match_dump;
 	ops->validate = nft_match_validate;
 	ops->data = match;
-	ops->reduce = nft_match_reduce;
 
 	matchsize = NFT_EXPR_SIZE(XT_ALIGN(match->matchsize));
 	if (matchsize > NFT_MATCH_LARGE_THRESH) {

@@ -933,7 +924,6 @@ nft_target_select_ops(const struct nft_c
 	ops->destroy = nft_target_destroy;
 	ops->dump = nft_target_dump;
 	ops->data = target;
-	ops->reduce = NFT_REDUCE_READONLY;
 
 	if (family == NFPROTO_BRIDGE) {
 		ops->eval = nft_target_eval_bridge;

@@ -247,7 +247,6 @@ static const struct nft_expr_ops nft_con
 	.destroy_clone	= nft_connlimit_destroy_clone,
 	.dump		= nft_connlimit_dump,
 	.gc		= nft_connlimit_gc,
-	.reduce		= NFT_REDUCE_READONLY,
 };
 
 static struct nft_expr_type nft_connlimit_type __read_mostly = {

@@ -301,7 +301,6 @@ static const struct nft_expr_ops nft_cou
 	.destroy_clone	= nft_counter_destroy,
 	.dump		= nft_counter_dump,
 	.clone		= nft_counter_clone,
-	.reduce		= NFT_REDUCE_READONLY,
 	.offload	= nft_counter_offload,
 	.offload_stats	= nft_counter_offload_stats,
 };

84 lines not on this page · raw message

@@ -80,7 +80,6 @@ static const struct nft_expr_ops nft_dup
 	.eval		= nft_dup_netdev_eval,
 	.init		= nft_dup_netdev_init,
 	.dump		= nft_dup_netdev_dump,
-	.reduce		= NFT_REDUCE_READONLY,
 	.offload	= nft_dup_netdev_offload,
 	.offload_action	= nft_dup_netdev_offload_action,
 };

@@ -430,7 +430,6 @@ static const struct nft_expr_ops nft_dyn
 	.activate	= nft_dynset_activate,
 	.deactivate	= nft_dynset_deactivate,
 	.dump		= nft_dynset_dump,
-	.reduce		= NFT_REDUCE_READONLY,
 };
 
 struct nft_expr_type nft_dynset_type __read_mostly = {

@@ -701,40 +701,12 @@ static int nft_exthdr_dump_strip(struct
 	return nft_exthdr_dump_common(skb, priv);
 }
 
-static bool nft_exthdr_reduce(struct nft_regs_track *track,
-			       const struct nft_expr *expr)
-{
-	const struct nft_exthdr *priv = nft_expr_priv(expr);
-	const struct nft_exthdr *exthdr;
-
-	if (!nft_reg_track_cmp(track, expr, priv->dreg)) {
-		nft_reg_track_update(track, expr, priv->dreg, priv->len);
-		return false;
-	}
-
-	exthdr = nft_expr_priv(track->regs[priv->dreg].selector);
-	if (priv->type != exthdr->type ||
-	    priv->op != exthdr->op ||
-	    priv->flags != exthdr->flags ||
-	    priv->offset != exthdr->offset ||
-	    priv->len != exthdr->len) {
-		nft_reg_track_update(track, expr, priv->dreg, priv->len);
-		return false;
-	}
-
-	if (!track->regs[priv->dreg].bitwise)
-		return true;
-
-	return nft_expr_reduce_bitwise(track, expr);
-}
-
 static const struct nft_expr_ops nft_exthdr_ipv6_ops = {
 	.type		= &nft_exthdr_type,
 	.size		= NFT_EXPR_SIZE(sizeof(struct nft_exthdr)),
 	.eval		= nft_exthdr_ipv6_eval,
 	.init		= nft_exthdr_init,
 	.dump		= nft_exthdr_dump,
-	.reduce		= nft_exthdr_reduce,
 };
 
 static const struct nft_expr_ops nft_exthdr_ipv4_ops = {

@@ -743,7 +715,6 @@ static const struct nft_expr_ops nft_ext
 	.eval		= nft_exthdr_ipv4_eval,
 	.init		= nft_exthdr_ipv4_init,
 	.dump		= nft_exthdr_dump,
-	.reduce		= nft_exthdr_reduce,
 };
 
 static const struct nft_expr_ops nft_exthdr_tcp_ops = {

@@ -752,7 +723,6 @@ static const struct nft_expr_ops nft_ext
 	.eval		= nft_exthdr_tcp_eval,
 	.init		= nft_exthdr_init,
 	.dump		= nft_exthdr_dump,
-	.reduce		= nft_exthdr_reduce,
 };
 
 static const struct nft_expr_ops nft_exthdr_tcp_set_ops = {

@@ -761,7 +731,6 @@ static const struct nft_expr_ops nft_ext
 	.eval		= nft_exthdr_tcp_set_eval,
 	.init		= nft_exthdr_tcp_set_init,
 	.dump		= nft_exthdr_dump_set,
-	.reduce		= NFT_REDUCE_READONLY,
 };
 
 static const struct nft_expr_ops nft_exthdr_tcp_strip_ops = {

@@ -770,7 +739,6 @@ static const struct nft_expr_ops nft_ext
 	.eval		= nft_exthdr_tcp_strip_eval,
 	.init		= nft_exthdr_tcp_strip_init,
 	.dump		= nft_exthdr_dump_strip,
-	.reduce		= NFT_REDUCE_READONLY,
 };
 
 static const struct nft_expr_ops nft_exthdr_sctp_ops = {

@@ -779,7 +747,6 @@ static const struct nft_expr_ops nft_ext
 	.eval		= nft_exthdr_sctp_eval,
 	.init		= nft_exthdr_init,
 	.dump		= nft_exthdr_dump,
-	.reduce		= nft_exthdr_reduce,
 };
 
 static const struct nft_expr_ops nft_exthdr_dccp_ops = {

@@ -788,7 +755,6 @@ static const struct nft_expr_ops nft_ext
 	.eval		= nft_exthdr_dccp_eval,
 	.init		= nft_exthdr_dccp_init,
 	.dump		= nft_exthdr_dump,
-	.reduce		= nft_exthdr_reduce,
 };
 
 static const struct nft_expr_ops *

47 lines not on this page · raw message

@@ -49,7 +49,6 @@ static const struct nft_expr_ops nft_fib
 	.init		= nft_fib_init,
 	.dump		= nft_fib_dump,
 	.validate	= nft_fib_validate,
-	.reduce		= nft_fib_reduce,
 };
 
 static struct nft_expr_type nft_fib_inet_type __read_mostly = {

@@ -58,7 +58,6 @@ static const struct nft_expr_ops nft_fib
 	.init		= nft_fib_init,
 	.dump		= nft_fib_dump,
 	.validate	= nft_fib_validate,
-	.reduce		= nft_fib_reduce,
 };
 
 static struct nft_expr_type nft_fib_netdev_type __read_mostly = {

@@ -481,7 +481,6 @@ static const struct nft_expr_ops nft_flo
 	.destroy	= nft_flow_offload_destroy,
 	.validate	= nft_flow_offload_validate,
 	.dump		= nft_flow_offload_dump,
-	.reduce		= NFT_REDUCE_READONLY,
 };
 
 static struct nft_expr_type nft_flow_offload_type __read_mostly = {

@@ -228,7 +228,6 @@ static const struct nft_expr_ops nft_fwd
 	.init		= nft_fwd_neigh_init,
 	.dump		= nft_fwd_neigh_dump,
 	.validate	= nft_fwd_validate,
-	.reduce		= NFT_REDUCE_READONLY,
 };
 
 static const struct nft_expr_ops nft_fwd_netdev_ops = {

@@ -238,7 +237,6 @@ static const struct nft_expr_ops nft_fwd
 	.init		= nft_fwd_netdev_init,
 	.dump		= nft_fwd_netdev_dump,
 	.validate	= nft_fwd_validate,
-	.reduce		= NFT_REDUCE_READONLY,
 	.offload	= nft_fwd_netdev_offload,
 	.offload_action	= nft_fwd_netdev_offload_action,
 };

@@ -165,16 +165,6 @@ nla_put_failure:
 	return -1;
 }
 
-static bool nft_jhash_reduce(struct nft_regs_track *track,
-			     const struct nft_expr *expr)
-{
-	const struct nft_jhash *priv = nft_expr_priv(expr);
-
-	nft_reg_track_cancel(track, priv->dreg, sizeof(u32));
-
-	return false;
-}
-
 static int nft_symhash_dump(struct sk_buff *skb,
 			    const struct nft_expr *expr, bool reset)
 {

@@ -195,30 +185,6 @@ nla_put_failure:
 	return -1;
 }
 
-static bool nft_symhash_reduce(struct nft_regs_track *track,
-			       const struct nft_expr *expr)
-{
-	struct nft_symhash *priv = nft_expr_priv(expr);
-	struct nft_symhash *symhash;
-
-	if (!nft_reg_track_cmp(track, expr, priv->dreg)) {
-		nft_reg_track_update(track, expr, priv->dreg, sizeof(u32));
-		return false;
-	}
-
-	symhash = nft_expr_priv(track->regs[priv->dreg].selector);
-	if (priv->offset != symhash->offset ||
-	    priv->modulus != symhash->modulus) {
-		nft_reg_track_update(track, expr, priv->dreg, sizeof(u32));
-		return false;
-	}
-
-	if (!track->regs[priv->dreg].bitwise)
-		return true;
-
-	return false;
-}
-
 static struct nft_expr_type nft_hash_type;
 static const struct nft_expr_ops nft_jhash_ops = {
 	.type		= &nft_hash_type,

@@ -226,7 +192,6 @@ static const struct nft_expr_ops nft_jha
 	.eval		= nft_jhash_eval,
 	.init		= nft_jhash_init,
 	.dump		= nft_jhash_dump,
-	.reduce		= nft_jhash_reduce,
 };
 
 static const struct nft_expr_ops nft_symhash_ops = {

@@ -235,7 +200,6 @@ static const struct nft_expr_ops nft_sym
 	.eval		= nft_symhash_eval,
 	.init		= nft_symhash_init,
 	.dump		= nft_symhash_dump,
-	.reduce		= nft_symhash_reduce,
 };
 
 static const struct nft_expr_ops *

@@ -320,17 +320,6 @@ static bool nft_immediate_offload_action
 	return false;
 }
 
-static bool nft_immediate_reduce(struct nft_regs_track *track,
-				 const struct nft_expr *expr)
-{
-	const struct nft_immediate_expr *priv = nft_expr_priv(expr);
-
-	if (priv->dreg != NFT_REG_VERDICT)
-		nft_reg_track_cancel(track, priv->dreg, priv->dlen);
-
-	return false;
-}
-
 static const struct nft_expr_ops nft_imm_ops = {
 	.type		= &nft_imm_type,
 	.size		= NFT_EXPR_SIZE(sizeof(struct nft_immediate_expr)),

@@ -341,7 +330,6 @@ static const struct nft_expr_ops nft_imm
 	.destroy	= nft_immediate_destroy,
 	.dump		= nft_immediate_dump,
 	.validate	= nft_immediate_validate,
-	.reduce		= nft_immediate_reduce,
 	.offload	= nft_immediate_offload,
 	.offload_action	= nft_immediate_offload_action,
 };

@@ -125,7 +125,6 @@ static const struct nft_expr_ops nft_las
 	.destroy	= nft_last_destroy,
 	.clone		= nft_last_clone,
 	.dump		= nft_last_dump,
-	.reduce		= NFT_REDUCE_READONLY,
 };
 
 struct nft_expr_type nft_last_type __read_mostly = {

@@ -243,7 +243,6 @@ static const struct nft_expr_ops nft_lim
 	.destroy	= nft_limit_pkts_destroy,
 	.clone		= nft_limit_pkts_clone,
 	.dump		= nft_limit_pkts_dump,
-	.reduce		= NFT_REDUCE_READONLY,
 };
 
 static void nft_limit_bytes_eval(const struct nft_expr *expr,

@@ -299,7 +298,6 @@ static const struct nft_expr_ops nft_lim
 	.dump		= nft_limit_bytes_dump,
 	.clone		= nft_limit_bytes_clone,
 	.destroy	= nft_limit_bytes_destroy,
-	.reduce		= NFT_REDUCE_READONLY,
 };
 
 static const struct nft_expr_ops *

@@ -235,7 +235,6 @@ static const struct nft_expr_ops nft_log
 	.init		= nft_log_init,
 	.destroy	= nft_log_destroy,
 	.dump		= nft_log_dump,
-	.reduce		= NFT_REDUCE_READONLY,
 };
 
 static struct nft_expr_type nft_log_type __read_mostly = {

@@ -232,17 +232,6 @@ static int nft_lookup_validate(const str
 	return 0;
 }
 
-static bool nft_lookup_reduce(struct nft_regs_track *track,
-			      const struct nft_expr *expr)
-{
-	const struct nft_lookup *priv = nft_expr_priv(expr);
-
-	if (priv->set->flags & NFT_SET_MAP)
-		nft_reg_track_cancel(track, priv->dreg, priv->set->dlen);
-
-	return false;
-}
-
 static const struct nft_expr_ops nft_lookup_ops = {
 	.type		= &nft_lookup_type,
 	.size		= NFT_EXPR_SIZE(sizeof(struct nft_lookup)),

@@ -253,7 +242,6 @@ static const struct nft_expr_ops nft_loo
 	.destroy	= nft_lookup_destroy,
 	.dump		= nft_lookup_dump,
 	.validate	= nft_lookup_validate,
-	.reduce		= nft_lookup_reduce,
 };
 
 struct nft_expr_type nft_lookup_type __read_mostly = {

@@ -143,7 +143,6 @@ static const struct nft_expr_ops nft_mas
 	.destroy	= nft_masq_ipv4_destroy,
 	.dump		= nft_masq_dump,
 	.validate	= nft_masq_validate,
-	.reduce		= NFT_REDUCE_READONLY,
 };
 
 static struct nft_expr_type nft_masq_ipv4_type __read_mostly = {

@@ -171,7 +170,6 @@ static const struct nft_expr_ops nft_mas
 	.destroy	= nft_masq_ipv6_destroy,
 	.dump		= nft_masq_dump,
 	.validate	= nft_masq_validate,
-	.reduce		= NFT_REDUCE_READONLY,
 };
 
 static struct nft_expr_type nft_masq_ipv6_type __read_mostly = {

@@ -213,7 +211,6 @@ static const struct nft_expr_ops nft_mas
 	.destroy	= nft_masq_inet_destroy,
 	.dump		= nft_masq_dump,
 	.validate	= nft_masq_validate,
-	.reduce		= NFT_REDUCE_READONLY,
 };
 
 static struct nft_expr_type nft_masq_inet_type __read_mostly = {

67 lines not on this page · raw message

@@ -320,7 +320,6 @@ static const struct nft_expr_ops nft_nat
 	.destroy        = nft_nat_destroy,
 	.dump           = nft_nat_dump,
 	.validate	= nft_nat_validate,
-	.reduce		= NFT_REDUCE_READONLY,
 };
 
 static struct nft_expr_type nft_nat_type __read_mostly = {

@@ -351,7 +350,6 @@ static const struct nft_expr_ops nft_nat
 	.destroy        = nft_nat_destroy,
 	.dump           = nft_nat_dump,
 	.validate	= nft_nat_validate,
-	.reduce		= NFT_REDUCE_READONLY,
 };
 
 static struct nft_expr_type nft_inet_nat_type __read_mostly = {

@@ -84,16 +84,6 @@ err:
 	return err;
 }
 
-static bool nft_ng_inc_reduce(struct nft_regs_track *track,
-				 const struct nft_expr *expr)
-{
-	const struct nft_ng_inc *priv = nft_expr_priv(expr);
-
-	nft_reg_track_cancel(track, priv->dreg, NFT_REG32_SIZE);
-
-	return false;
-}
-
 static int nft_ng_dump(struct sk_buff *skb, enum nft_registers dreg,
 		       u32 modulus, enum nft_ng_types type, u32 offset)
 {

@@ -178,16 +168,6 @@ static int nft_ng_random_dump(struct sk_
 			   priv->offset);
 }
 
-static bool nft_ng_random_reduce(struct nft_regs_track *track,
-				 const struct nft_expr *expr)
-{
-	const struct nft_ng_random *priv = nft_expr_priv(expr);
-
-	nft_reg_track_cancel(track, priv->dreg, NFT_REG32_SIZE);
-
-	return false;
-}
-
 static struct nft_expr_type nft_ng_type;
 static const struct nft_expr_ops nft_ng_inc_ops = {
 	.type		= &nft_ng_type,

@@ -196,7 +176,6 @@ static const struct nft_expr_ops nft_ng_
 	.init		= nft_ng_inc_init,
 	.destroy	= nft_ng_inc_destroy,
 	.dump		= nft_ng_inc_dump,
-	.reduce		= nft_ng_inc_reduce,
 };
 
 static const struct nft_expr_ops nft_ng_random_ops = {

@@ -205,7 +184,6 @@ static const struct nft_expr_ops nft_ng_
 	.eval		= nft_ng_random_eval,
 	.init		= nft_ng_random_init,
 	.dump		= nft_ng_random_dump,
-	.reduce		= nft_ng_random_reduce,
 };
 
 static const struct nft_expr_ops *

@@ -123,7 +123,6 @@ static const struct nft_expr_ops nft_obj
 	.deactivate	= nft_objref_deactivate,
 	.dump		= nft_objref_dump,
 	.validate	= nft_objref_validate,
-	.reduce		= NFT_REDUCE_READONLY,
 };
 
 struct nft_objref_map {

@@ -246,7 +245,6 @@ static const struct nft_expr_ops nft_obj
 	.destroy	= nft_objref_map_destroy,
 	.dump		= nft_objref_map_dump,
 	.validate	= nft_objref_map_validate,
-	.reduce		= NFT_REDUCE_READONLY,
 };
 
 static const struct nft_expr_ops *

@@ -136,30 +136,6 @@ static int nft_osf_validate(const struct
 	return nft_chain_validate_hooks(ctx->chain, hooks);
 }
 
-static bool nft_osf_reduce(struct nft_regs_track *track,
-			   const struct nft_expr *expr)
-{
-	struct nft_osf *priv = nft_expr_priv(expr);
-	struct nft_osf *osf;
-
-	if (!nft_reg_track_cmp(track, expr, priv->dreg)) {
-		nft_reg_track_update(track, expr, priv->dreg, NFT_OSF_MAXGENRELEN);
-		return false;
-	}
-
-	osf = nft_expr_priv(track->regs[priv->dreg].selector);
-	if (priv->flags != osf->flags ||
-	    priv->ttl != osf->ttl) {
-		nft_reg_track_update(track, expr, priv->dreg, NFT_OSF_MAXGENRELEN);
-		return false;
-	}
-
-	if (!track->regs[priv->dreg].bitwise)
-		return true;
-
-	return false;
-}
-
 static struct nft_expr_type nft_osf_type;
 static const struct nft_expr_ops nft_osf_op = {
 	.eval		= nft_osf_eval,

@@ -168,7 +144,6 @@ static const struct nft_expr_ops nft_osf
 	.dump		= nft_osf_dump,
 	.type		= &nft_osf_type,
 	.validate	= nft_osf_validate,
-	.reduce		= nft_osf_reduce,
 };
 
 static struct nft_expr_type nft_osf_type __read_mostly = {

77 lines not on this page · raw message

@@ -191,7 +191,6 @@ static const struct nft_expr_ops nft_que
 	.init		= nft_queue_init,
 	.dump		= nft_queue_dump,
 	.validate	= nft_queue_validate,
-	.reduce		= NFT_REDUCE_READONLY,
 };
 
 static const struct nft_expr_ops nft_queue_sreg_ops = {

@@ -201,7 +200,6 @@ static const struct nft_expr_ops nft_que
 	.init		= nft_queue_sreg_init,
 	.dump		= nft_queue_sreg_dump,
 	.validate	= nft_queue_validate,
-	.reduce		= NFT_REDUCE_READONLY,
 };
 
 static const struct nft_expr_ops *

@@ -265,7 +265,6 @@ static const struct nft_expr_ops nft_quo
 	.destroy	= nft_quota_destroy,
 	.clone		= nft_quota_clone,
 	.dump		= nft_quota_dump,
-	.reduce		= NFT_REDUCE_READONLY,
 };
 
 static struct nft_expr_type nft_quota_type __read_mostly = {

@@ -138,7 +138,6 @@ static const struct nft_expr_ops nft_ran
 	.eval		= nft_range_eval,
 	.init		= nft_range_init,
 	.dump		= nft_range_dump,
-	.reduce		= NFT_REDUCE_READONLY,
 };
 
 struct nft_expr_type nft_range_type __read_mostly = {

@@ -146,7 +146,6 @@ static const struct nft_expr_ops nft_red
 	.destroy	= nft_redir_ipv4_destroy,
 	.dump		= nft_redir_dump,
 	.validate	= nft_redir_validate,
-	.reduce		= NFT_REDUCE_READONLY,
 };
 
 static struct nft_expr_type nft_redir_ipv4_type __read_mostly = {

@@ -174,7 +173,6 @@ static const struct nft_expr_ops nft_red
 	.destroy	= nft_redir_ipv6_destroy,
 	.dump		= nft_redir_dump,
 	.validate	= nft_redir_validate,
-	.reduce		= NFT_REDUCE_READONLY,
 };
 
 static struct nft_expr_type nft_redir_ipv6_type __read_mostly = {

@@ -203,7 +201,6 @@ static const struct nft_expr_ops nft_red
 	.destroy	= nft_redir_inet_destroy,
 	.dump		= nft_redir_dump,
 	.validate	= nft_redir_validate,
-	.reduce		= NFT_REDUCE_READONLY,
 };
 
 static struct nft_expr_type nft_redir_inet_type __read_mostly = {

@@ -79,7 +79,6 @@ static const struct nft_expr_ops nft_rej
 	.init		= nft_reject_init,
 	.dump		= nft_reject_dump,
 	.validate	= nft_reject_inet_validate,
-	.reduce		= NFT_REDUCE_READONLY,
 };
 
 static struct nft_expr_type nft_reject_inet_type __read_mostly = {

@@ -158,7 +158,6 @@ static const struct nft_expr_ops nft_rej
 	.init		= nft_reject_init,
 	.dump		= nft_reject_dump,
 	.validate	= nft_reject_netdev_validate,
-	.reduce		= NFT_REDUCE_READONLY,
 };
 
 static struct nft_expr_type nft_reject_netdev_type __read_mostly = {

@@ -195,7 +195,6 @@ static const struct nft_expr_ops nft_rt_
 	.init		= nft_rt_get_init,
 	.dump		= nft_rt_get_dump,
 	.validate	= nft_rt_validate,
-	.reduce		= NFT_REDUCE_READONLY,
 };
 
 struct nft_expr_type nft_rt_type __read_mostly = {

@@ -249,31 +249,6 @@ static int nft_socket_dump(struct sk_buf
 	return 0;
 }
 
-static bool nft_socket_reduce(struct nft_regs_track *track,
-			      const struct nft_expr *expr)
-{
-	const struct nft_socket *priv = nft_expr_priv(expr);
-	const struct nft_socket *socket;
-
-	if (!nft_reg_track_cmp(track, expr, priv->dreg)) {
-		nft_reg_track_update(track, expr, priv->dreg, priv->len);
-		return false;
-	}
-
-	socket = nft_expr_priv(track->regs[priv->dreg].selector);
-	if (priv->key != socket->key ||
-	    priv->dreg != socket->dreg ||
-	    priv->level != socket->level) {
-		nft_reg_track_update(track, expr, priv->dreg, priv->len);
-		return false;
-	}
-
-	if (!track->regs[priv->dreg].bitwise)
-		return true;
-
-	return nft_expr_reduce_bitwise(track, expr);
-}
-
 static int nft_socket_validate(const struct nft_ctx *ctx,
 			       const struct nft_expr *expr)
 {

@@ -296,7 +271,6 @@ static const struct nft_expr_ops nft_soc
 	.init		= nft_socket_init,
 	.dump		= nft_socket_dump,
 	.validate	= nft_socket_validate,
-	.reduce		= nft_socket_reduce,
 };
 
 static struct nft_expr_type nft_socket_type __read_mostly = {

@@ -290,7 +290,6 @@ static const struct nft_expr_ops nft_syn
 	.dump		= nft_synproxy_dump,
 	.type		= &nft_synproxy_type,
 	.validate	= nft_synproxy_validate,
-	.reduce		= NFT_REDUCE_READONLY,
 };
 
 static struct nft_expr_type nft_synproxy_type __read_mostly = {

@@ -331,7 +331,6 @@ static const struct nft_expr_ops nft_tpr
 	.init		= nft_tproxy_init,
 	.destroy	= nft_tproxy_destroy,
 	.dump		= nft_tproxy_dump,
-	.reduce		= NFT_REDUCE_READONLY,
 	.validate	= nft_tproxy_validate,
 };
 

@@ -124,31 +124,6 @@ nla_put_failure:
 	return -1;
 }
 
-static bool nft_tunnel_get_reduce(struct nft_regs_track *track,
-				  const struct nft_expr *expr)
-{
-	const struct nft_tunnel *priv = nft_expr_priv(expr);
-	const struct nft_tunnel *tunnel;
-
-	if (!nft_reg_track_cmp(track, expr, priv->dreg)) {
-		nft_reg_track_update(track, expr, priv->dreg, priv->len);
-		return false;
-	}
-
-	tunnel = nft_expr_priv(track->regs[priv->dreg].selector);
-	if (priv->key != tunnel->key ||
-	    priv->dreg != tunnel->dreg ||
-	    priv->mode != tunnel->mode) {
-		nft_reg_track_update(track, expr, priv->dreg, priv->len);
-		return false;
-	}
-
-	if (!track->regs[priv->dreg].bitwise)
-		return true;
-
-	return false;
-}
-
 static struct nft_expr_type nft_tunnel_type;
 static const struct nft_expr_ops nft_tunnel_get_ops = {
 	.type		= &nft_tunnel_type,

@@ -156,7 +131,6 @@ static const struct nft_expr_ops nft_tun
 	.eval		= nft_tunnel_get_eval,
 	.init		= nft_tunnel_get_init,
 	.dump		= nft_tunnel_get_dump,
-	.reduce		= nft_tunnel_get_reduce,
 };
 
 static struct nft_expr_type nft_tunnel_type __read_mostly = {

@@ -258,32 +258,6 @@ static int nft_xfrm_validate(const struc
 	return nft_chain_validate_hooks(ctx->chain, hooks);
 }
 
-static bool nft_xfrm_reduce(struct nft_regs_track *track,
-			    const struct nft_expr *expr)
-{
-	const struct nft_xfrm *priv = nft_expr_priv(expr);
-	const struct nft_xfrm *xfrm;
-
-	if (!nft_reg_track_cmp(track, expr, priv->dreg)) {
-		nft_reg_track_update(track, expr, priv->dreg, priv->len);
-		return false;
-	}
-
-	xfrm = nft_expr_priv(track->regs[priv->dreg].selector);
-	if (priv->key != xfrm->key ||
-	    priv->dreg != xfrm->dreg ||
-	    priv->dir != xfrm->dir ||
-	    priv->spnum != xfrm->spnum) {
-		nft_reg_track_update(track, expr, priv->dreg, priv->len);
-		return false;
-	}
-
-	if (!track->regs[priv->dreg].bitwise)
-		return true;
-
-	return nft_expr_reduce_bitwise(track, expr);
-}
-
 static struct nft_expr_type nft_xfrm_type;
 static const struct nft_expr_ops nft_xfrm_get_ops = {
 	.type		= &nft_xfrm_type,

@@ -292,7 +266,6 @@ static const struct nft_expr_ops nft_xfr
 	.init		= nft_xfrm_get_init,
 	.dump		= nft_xfrm_get_dump,
 	.validate	= nft_xfrm_validate,
-	.reduce		= nft_xfrm_reduce,
 };
 
 static struct nft_expr_type nft_xfrm_type __read_mostly = {

@@ -31,6 +31,15 @@ int nft_fib_validate(const struct nft_ct
 	const struct nft_fib *priv = nft_expr_priv(expr);
 	unsigned int hooks;
 
+	switch (ctx->family) {
+	case NFPROTO_IPV4:
+	case NFPROTO_IPV6:
+	case NFPROTO_INET:
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
 	switch (priv->result) {
 	case NFT_FIB_RESULT_OIF:
 	case NFT_FIB_RESULT_OIFNAME:

@@ -50,6 +50,33 @@ static void nft_fib_netdev_eval(const st
 	regs->verdict.code = NFT_BREAK;
 }
 
+static int nft_fib_netdev_validate(const struct nft_ctx *ctx,
+				   const struct nft_expr *expr)
+{
+	const struct nft_fib *priv = nft_expr_priv(expr);
+	unsigned int hooks;
+
+	switch (priv->result) {
+	case NFT_FIB_RESULT_OIF:
+	case NFT_FIB_RESULT_OIFNAME:
+		hooks = (1 << NF_NETDEV_INGRESS);
+		break;
+	case NFT_FIB_RESULT_ADDRTYPE:
+		if (priv->flags & NFTA_FIB_F_IIF)
+			hooks = (1 << NF_NETDEV_INGRESS);
+		else if (priv->flags & NFTA_FIB_F_OIF)
+			hooks = (1 << NF_NETDEV_EGRESS);
+		else
+			hooks = (1 << NF_NETDEV_INGRESS) |
+				(1 << NF_NETDEV_EGRESS);
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return nft_chain_validate_hooks(ctx->chain, hooks);
+}
+
 static struct nft_expr_type nft_fib_netdev_type;
 static const struct nft_expr_ops nft_fib_netdev_ops = {
 	.type		= &nft_fib_netdev_type,

@@ -57,7 +84,7 @@ static const struct nft_expr_ops nft_fib
 	.eval		= nft_fib_netdev_eval,
 	.init		= nft_fib_init,
 	.dump		= nft_fib_dump,
-	.validate	= nft_fib_validate,
+	.validate	= nft_fib_netdev_validate,
 };
 
 static struct nft_expr_type nft_fib_netdev_type __read_mostly = {

@@ -4096,7 +4096,7 @@ out_err:
 
 static __be32 nfsd4_encode_readv(struct nfsd4_compoundres *resp,
 				 struct nfsd4_read *read,
-				 struct file *file, unsigned long maxcount)
+				 unsigned long maxcount)
 {
 	struct xdr_stream *xdr = resp->xdr;
 	unsigned int base = xdr->buf->page_len & ~PAGE_MASK;

@@ -4107,7 +4107,7 @@ static __be32 nfsd4_encode_readv(struct
 	if (xdr_reserve_space_vec(xdr, maxcount) < 0)
 		return nfserr_resource;
 
-	nfserr = nfsd_iter_read(resp->rqstp, read->rd_fhp, file,
+	nfserr = nfsd_iter_read(resp->rqstp, read->rd_fhp, read->rd_nf,
 				read->rd_offset, &maxcount, base,
 				&read->rd_eof);
 	read->rd_length = maxcount;

@@ -4155,7 +4155,7 @@ nfsd4_encode_read(struct nfsd4_compoundr
 	if (file->f_op->splice_read && splice_ok)
 		nfserr = nfsd4_encode_splice_read(resp, read, file, maxcount);
 	else
-		nfserr = nfsd4_encode_readv(resp, read, file, maxcount);
+		nfserr = nfsd4_encode_readv(resp, read, maxcount);
 	if (nfserr) {
 		xdr_truncate_encode(xdr, starting_len);
 		return nfserr;

@@ -4906,7 +4906,7 @@ nfsd4_encode_read_plus_data(struct nfsd4
 	if (file->f_op->splice_read && splice_ok)
 		nfserr = nfsd4_encode_splice_read(resp, read, file, maxcount);
 	else
-		nfserr = nfsd4_encode_readv(resp, read, file, maxcount);
+		nfserr = nfsd4_encode_readv(resp, read, maxcount);
 	if (nfserr)
 		return nfserr;
 

@@ -1061,7 +1061,7 @@ __be32 nfsd_splice_read(struct svc_rqst
  * nfsd_iter_read - Perform a VFS read using an iterator
  * @rqstp: RPC transaction context
  * @fhp: file handle of file to be read
- * @file: opened struct file of file to be read
+ * @nf: opened struct nfsd_file of file to be read
  * @offset: starting byte offset
  * @count: IN: requested number of bytes; OUT: number of bytes read
  * @base: offset in first page of read buffer

@@ -1074,9 +1074,10 @@ __be32 nfsd_splice_read(struct svc_rqst
  * returned.
  */
 __be32 nfsd_iter_read(struct svc_rqst *rqstp, struct svc_fh *fhp,
-		      struct file *file, loff_t offset, unsigned long *count,
+		      struct nfsd_file *nf, loff_t offset, unsigned long *count,
 		      unsigned int base, u32 *eof)
 {
+	struct file *file = nf->nf_file;
 	unsigned long v, total;
 	struct iov_iter iter;
 	loff_t ppos = offset;

@@ -1258,7 +1259,7 @@ __be32 nfsd_read(struct svc_rqst *rqstp,
 	if (file->f_op->splice_read && test_bit(RQ_SPLICE_OK, &rqstp->rq_flags))
 		err = nfsd_splice_read(rqstp, fhp, file, offset, count, eof);
 	else
-		err = nfsd_iter_read(rqstp, fhp, file, offset, count, 0, eof);
+		err = nfsd_iter_read(rqstp, fhp, nf, offset, count, 0, eof);
 
 	nfsd_file_put(nf);
 	trace_nfsd_read_done(rqstp, fhp, offset, *count);

@@ -120,7 +120,7 @@ __be32		nfsd_splice_read(struct svc_rqst
 				unsigned long *count,
 				u32 *eof);
 __be32		nfsd_iter_read(struct svc_rqst *rqstp, struct svc_fh *fhp,
-				struct file *file, loff_t offset,
+				struct nfsd_file *nf, loff_t offset,
 				unsigned long *count, unsigned int base,
 				u32 *eof);
 __be32		nfsd_read(struct svc_rqst *rqstp, struct svc_fh *fhp,

@@ -26,6 +26,9 @@ struct svc_sock {
 	void			(*sk_odata)(struct sock *);
 	void			(*sk_owspace)(struct sock *);
 
+	/* For sends (protected by xpt_mutex) */
+	struct bio_vec		*sk_bvec;
+
 	/* private TCP part */
 	/* On-the-wire fragment header: */
 	__be32			sk_marker;

@@ -68,6 +68,17 @@
 
 #define RPCDBG_FACILITY	RPCDBG_SVCXPRT
 
+/*
+ * For UDP:
+ * 1 for header page
+ * enough pages for RPCSVC_MAXPAYLOAD_UDP
+ * 1 in case payload is not aligned
+ * 1 for tail page
+ */
+enum {
+	SUNRPC_MAX_UDP_SENDPAGES = 1 + RPCSVC_MAXPAYLOAD_UDP / PAGE_SIZE + 1 + 1
+};
+
 /* To-do: to avoid tying up an nfsd thread while waiting for a
  * handshake request, the request could instead be deferred.
  */

@@ -750,15 +761,14 @@ static int svc_udp_sendto(struct svc_rqs
 	if (svc_xprt_is_dead(xprt))
 		goto out_notconn;
 
-	count = xdr_buf_to_bvec(rqstp->rq_bvec,
-				ARRAY_SIZE(rqstp->rq_bvec), xdr);
+	count = xdr_buf_to_bvec(svsk->sk_bvec, SUNRPC_MAX_UDP_SENDPAGES, xdr);
 
-	iov_iter_bvec(&msg.msg_iter, ITER_SOURCE, rqstp->rq_bvec,
+	iov_iter_bvec(&msg.msg_iter, ITER_SOURCE, svsk->sk_bvec,
 		      count, rqstp->rq_res.len);
 	err = sock_sendmsg(svsk->sk_sock, &msg);
 	if (err == -ECONNREFUSED) {
 		/* ICMP error on earlier request. */
-		iov_iter_bvec(&msg.msg_iter, ITER_SOURCE, rqstp->rq_bvec,
+		iov_iter_bvec(&msg.msg_iter, ITER_SOURCE, svsk->sk_bvec,
 			      count, rqstp->rq_res.len);
 		err = sock_sendmsg(svsk->sk_sock, &msg);
 	}

@@ -1257,19 +1267,19 @@ static int svc_tcp_sendmsg(struct svc_so
 	*sentp = 0;
 
 	/* The stream record marker is copied into a temporary page
-	 * fragment buffer so that it can be included in rq_bvec.
+	 * fragment buffer so that it can be included in sk_bvec.
 	 */
 	buf = page_frag_alloc(&svsk->sk_frag_cache, sizeof(marker),
 			      GFP_KERNEL);
 	if (!buf)
 		return -ENOMEM;
 	memcpy(buf, &marker, sizeof(marker));
-	bvec_set_virt(rqstp->rq_bvec, buf, sizeof(marker));
+	bvec_set_virt(svsk->sk_bvec, buf, sizeof(marker));
 
-	count = xdr_buf_to_bvec(rqstp->rq_bvec + 1,
-				ARRAY_SIZE(rqstp->rq_bvec) - 1, &rqstp->rq_res);
+	count = xdr_buf_to_bvec(svsk->sk_bvec + 1, RPCSVC_MAXPAGES,
+				&rqstp->rq_res);
 
-	iov_iter_bvec(&msg.msg_iter, ITER_SOURCE, rqstp->rq_bvec,
+	iov_iter_bvec(&msg.msg_iter, ITER_SOURCE, svsk->sk_bvec,
 		      1 + count, sizeof(marker) + rqstp->rq_res.len);
 	ret = sock_sendmsg(svsk->sk_sock, &msg);
 	page_frag_free(buf);

@@ -1432,6 +1442,13 @@ static struct svc_sock *svc_setup_socket
 	if (!svsk)
 		return ERR_PTR(-ENOMEM);
 
+	svsk->sk_bvec = kcalloc(RPCSVC_MAXPAGES + 1, sizeof(*svsk->sk_bvec),
+				GFP_KERNEL);
+	if (!svsk->sk_bvec) {
+		kfree(svsk);
+		return ERR_PTR(-ENOMEM);
+	}
+
 	inet = sock->sk;
 
 	if (pmap_register) {

@@ -1441,6 +1458,7 @@ static struct svc_sock *svc_setup_socket
 				     inet->sk_protocol,
 				     ntohs(inet_sk(inet)->inet_sport));
 		if (err < 0) {
+			kfree(svsk->sk_bvec);
 			kfree(svsk);
 			return ERR_PTR(err);
 		}

@@ -1660,5 +1678,6 @@ static void svc_sock_free(struct svc_xpr
 	if (pfc->va)
 		__page_frag_cache_drain(virt_to_head_page(pfc->va),
 					pfc->pagecnt_bias);
+	kfree(svsk->sk_bvec);
 	kfree(svsk);
 }

@@ -141,6 +141,21 @@ int	xdr_alloc_bvec(struct xdr_buf *buf,
 void	xdr_free_bvec(struct xdr_buf *buf);
 unsigned int xdr_buf_to_bvec(struct bio_vec *bvec, unsigned int bvec_size,
 			     const struct xdr_buf *xdr);
+int xdr_buf_to_sg(const struct xdr_buf *buf, unsigned int offset,
+		  unsigned int len, struct scatterlist *sg, unsigned int nsg);
+int xdr_buf_to_sg_alloc(const struct xdr_buf *buf, unsigned int offset,
+			unsigned int len, struct scatterlist *sg_head,
+			unsigned int sg_head_nents,
+			struct scatterlist **sg_overflow, gfp_t gfp);
+
+/*
+ * Inline scatterlist entries for xdr_buf_to_sg_alloc().  Sized to cover the
+ * head kvec, tail kvec, and a few page fragments without any heap allocation.
+ */
+enum {
+	XDR_BUF_TO_SG_NENTS	= 8,
+};
+
 
 static inline __be32 *xdr_encode_array(__be32 *p, const void *s, unsigned int len)
 {

205 lines not on this page · raw message

@@ -139,8 +139,8 @@ void	xdr_terminate_string(const struct x
 size_t	xdr_buf_pagecount(const struct xdr_buf *buf);
 int	xdr_alloc_bvec(struct xdr_buf *buf, gfp_t gfp);
 void	xdr_free_bvec(struct xdr_buf *buf);
-unsigned int xdr_buf_to_bvec(struct bio_vec *bvec, unsigned int bvec_size,
-			     const struct xdr_buf *xdr);
+int xdr_buf_to_bvec(struct bio_vec *bvec, unsigned int bvec_size,
+		    const struct xdr_buf *xdr);
 int xdr_buf_to_sg(const struct xdr_buf *buf, unsigned int offset,
 		  unsigned int len, struct scatterlist *sg, unsigned int nsg);
 int xdr_buf_to_sg_alloc(const struct xdr_buf *buf, unsigned int offset,

@@ -748,7 +748,7 @@ static int svc_udp_sendto(struct svc_rqs
 		.msg_flags	= MSG_SPLICE_PAGES,
 		.msg_controllen	= sizeof(buffer),
 	};
-	unsigned int count;
+	int count;
 	int err;
 
 	svc_udp_release_ctxt(xprt, rqstp->rq_xprt_ctxt);

@@ -762,6 +762,10 @@ static int svc_udp_sendto(struct svc_rqs
 		goto out_notconn;
 
 	count = xdr_buf_to_bvec(svsk->sk_bvec, SUNRPC_MAX_UDP_SENDPAGES, xdr);
+	if (count < 0) {
+		err = count;
+		goto out_trace;
+	}
 
 	iov_iter_bvec(&msg.msg_iter, ITER_SOURCE, svsk->sk_bvec,
 		      count, rqstp->rq_res.len);

@@ -773,6 +777,7 @@ static int svc_udp_sendto(struct svc_rqs
 		err = sock_sendmsg(svsk->sk_sock, &msg);
 	}
 
+out_trace:
 	trace_svcsock_udp_send(xprt, err);
 
 	mutex_unlock(&xprt->xpt_mutex);

@@ -1260,7 +1265,7 @@ static int svc_tcp_sendmsg(struct svc_so
 	struct msghdr msg = {
 		.msg_flags	= MSG_SPLICE_PAGES,
 	};
-	unsigned int count;
+	int count;
 	void *buf;
 	int ret;
 

@@ -1278,10 +1283,15 @@ static int svc_tcp_sendmsg(struct svc_so
 
 	count = xdr_buf_to_bvec(svsk->sk_bvec + 1, RPCSVC_MAXPAGES,
 				&rqstp->rq_res);
+	if (count < 0) {
+		ret = count;
+		goto out;
+	}
 
 	iov_iter_bvec(&msg.msg_iter, ITER_SOURCE, svsk->sk_bvec,
 		      1 + count, sizeof(marker) + rqstp->rq_res.len);
 	ret = sock_sendmsg(svsk->sk_sock, &msg);
+out:
 	page_frag_free(buf);
 	if (ret < 0)
 		return ret;

@@ -167,13 +167,14 @@ xdr_free_bvec(struct xdr_buf *buf)
 /**
  * xdr_buf_to_bvec - Copy components of an xdr_buf into a bio_vec array
  * @bvec: bio_vec array to populate
- * @bvec_size: element count of @bio_vec
+ * @bvec_size: element count of @bvec
  * @xdr: xdr_buf to be copied
  *
- * Returns the number of entries consumed in @bvec.
+ * Returns the number of entries consumed in @bvec on success, or
+ * -ESERVERFAULT when @xdr does not fit within @bvec_size entries.
  */
-unsigned int xdr_buf_to_bvec(struct bio_vec *bvec, unsigned int bvec_size,
-			     const struct xdr_buf *xdr)
+int xdr_buf_to_bvec(struct bio_vec *bvec, unsigned int bvec_size,
+		    const struct xdr_buf *xdr)
 {
 	const struct kvec *head = xdr->head;
 	const struct kvec *tail = xdr->tail;

@@ -215,7 +216,7 @@ unsigned int xdr_buf_to_bvec(struct bio_
 
 bvec_overflow:
 	pr_warn_once("%s: bio_vec array overflow\n", __func__);
-	return count;
+	return -ESERVERFAULT;
 }
 
 /**

@@ -649,6 +649,20 @@ static inline void fixup_hashdist(void)
 static inline void fixup_hashdist(void) {}
 #endif /* CONFIG_NUMA */
 
+#ifdef CONFIG_ZONE_DEVICE
+static __meminit void pageblock_migratetype_init_range(unsigned long pfn,
+		unsigned long nr_pages, int migratetype)
+{
+	const unsigned long end = pfn + nr_pages;
+
+	for (pfn = pageblock_align(pfn); pfn < end; pfn += pageblock_nr_pages) {
+		set_pageblock_migratetype(pfn_to_page(pfn), migratetype);
+		if (IS_ALIGNED(pfn, PAGES_PER_SECTION))
+			cond_resched();
+	}
+}
+#endif
+
 #ifdef CONFIG_DEFERRED_STRUCT_PAGE_INIT
 static inline void pgdat_set_deferred_range(pg_data_t *pgdat)
 {

@@ -993,21 +1007,6 @@ static void __ref __init_zone_device_pag
 	page->zone_device_data = NULL;
 
 	/*
-	 * Mark the block movable so that blocks are reserved for
-	 * movable at startup. This will force kernel allocations
-	 * to reserve their blocks rather than leaking throughout
-	 * the address space during boot when many long-lived
-	 * kernel allocations are made.
-	 *
-	 * Please note that MEMINIT_HOTPLUG path doesn't clear memmap
-	 * because this is done early in section_activate()
-	 */
-	if (pageblock_aligned(pfn)) {
-		set_pageblock_migratetype(page, MIGRATE_MOVABLE);
-		cond_resched();
-	}
-
-	/*
 	 * ZONE_DEVICE pages are released directly to the driver page allocator
 	 * which will set the page count to 1 when allocating the page.
 	 */