| Message ID | 97db46739ad095e0ed50f0dbd90e1b506c2991de.1695838185.git.ecree.xilinx@gmail.com (mailing list archive) |
|---|---|
| State | Changes Requested |
| Delegated to: | Netdev Maintainers |
| Headers | show |
| Series | ethtool: track custom RSS contexts in the core | expand |
On 9/27/2023 11:13 AM, edward.cree@amd.com wrote: > From: Edward Cree <ecree.xilinx@gmail.com> > > Since drivers are still choosing the context IDs, we have to force the > XArray to use the ID they've chosen rather than picking one ourselves, > and handle the case where they give us an ID that's already in use. > > Signed-off-by: Edward Cree <ecree.xilinx@gmail.com> > --- > include/linux/ethtool.h | 14 ++++++++ > net/ethtool/ioctl.c | 73 +++++++++++++++++++++++++++++++++++++++-- > 2 files changed, 85 insertions(+), 2 deletions(-) > > diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h > index bb11cb2f477d..229a23571008 100644 > --- a/include/linux/ethtool.h > +++ b/include/linux/ethtool.h > @@ -194,6 +194,17 @@ static inline u8 *ethtool_rxfh_context_key(struct ethtool_rxfh_context *ctx) > return (u8 *)(ethtool_rxfh_context_indir(ctx) + ctx->indir_size); > } > > +static inline size_t ethtool_rxfh_context_size(u32 indir_size, u32 key_size, > + u16 priv_size) > +{ > + size_t indir_bytes = array_size(indir_size, sizeof(u32)); > + size_t flex_len; > + > + flex_len = size_add(size_add(indir_bytes, key_size), > + ALIGN(priv_size, sizeof(u32))); > + return struct_size((struct ethtool_rxfh_context *)0, data, flex_len); > +} > + Ah, we already have to use a flexible area to handle the indirection table, and this way the core is the only one responsible for allocating and freeing the memory. Reviewed-by: Jacob Keller <jacob.e.keller@intel.com> > /* declare a link mode bitmap */ > #define __ETHTOOL_DECLARE_LINK_MODE_MASK(name) \ > DECLARE_BITMAP(name, __ETHTOOL_LINK_MODE_MASK_NBITS) > @@ -731,6 +742,8 @@ struct ethtool_mm_stats { > * will remain unchanged. > * Returns a negative error code or zero. An error code must be returned > * if at least one unsupported change was requested. > + * @rxfh_priv_size: size of the driver private data area the core should > + * allocate for an RSS context. > * @get_rxfh_context: Get the contents of the RX flow hash indirection table, > * hash key, and/or hash function assiciated to the given rss context. > * Returns a negative error code or zero. > @@ -824,6 +837,7 @@ struct ethtool_ops { > u32 cap_link_lanes_supported:1; > u32 supported_coalesce_params; > u32 supported_ring_params; > + u16 rxfh_priv_size; > void (*get_drvinfo)(struct net_device *, struct ethtool_drvinfo *); > int (*get_regs_len)(struct net_device *); > void (*get_regs)(struct net_device *, struct ethtool_regs *, void *); > diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c > index de78b24fffc9..1d13bc8fbb75 100644 > --- a/net/ethtool/ioctl.c > +++ b/net/ethtool/ioctl.c > @@ -1249,6 +1249,7 @@ static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev, > { > int ret; > const struct ethtool_ops *ops = dev->ethtool_ops; > + struct ethtool_rxfh_context *ctx = NULL; > struct ethtool_rxnfc rx_rings; > struct ethtool_rxfh rxfh; > u32 dev_indir_size = 0, dev_key_size = 0, i; > @@ -1256,7 +1257,7 @@ static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev, > u8 *hkey = NULL; > u8 *rss_config; > u32 rss_cfg_offset = offsetof(struct ethtool_rxfh, rss_config[0]); > - bool delete = false; > + bool create = false, delete = false; > > if (!ops->get_rxnfc || !ops->set_rxfh) > return -EOPNOTSUPP; > @@ -1275,6 +1276,7 @@ static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev, > /* Most drivers don't handle rss_context, check it's 0 as well */ > if (rxfh.rss_context && !ops->set_rxfh_context) > return -EOPNOTSUPP; > + create = rxfh.rss_context == ETH_RXFH_CONTEXT_ALLOC; > > /* If either indir, hash key or function is valid, proceed further. > * Must request at least one change: indir size, hash key or function. > @@ -1332,13 +1334,42 @@ static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev, > } > } > > + if (create) { > + if (delete) { > + ret = -EINVAL; > + goto out; > + } > + ctx = kzalloc(ethtool_rxfh_context_size(dev_indir_size, > + dev_key_size, > + ops->rxfh_priv_size), > + GFP_KERNEL_ACCOUNT); > + if (!ctx) { > + ret = -ENOMEM; > + goto out; > + } > + ctx->indir_size = dev_indir_size; > + ctx->key_size = dev_key_size; > + ctx->hfunc = rxfh.hfunc; > + ctx->priv_size = ops->rxfh_priv_size; > + } else if (rxfh.rss_context) { > + ctx = xa_load(&dev->ethtool->rss_ctx, rxfh.rss_context); > + if (!ctx) { > + ret = -ENOENT; > + goto out; > + } > + } > + > if (rxfh.rss_context) > ret = ops->set_rxfh_context(dev, indir, hkey, rxfh.hfunc, > &rxfh.rss_context, delete); > else > ret = ops->set_rxfh(dev, indir, hkey, rxfh.hfunc); > - if (ret) > + if (ret) { > + if (create) > + /* failed to create, free our new tracking entry */ > + kfree(ctx); > goto out; > + } > > if (copy_to_user(useraddr + offsetof(struct ethtool_rxfh, rss_context), > &rxfh.rss_context, sizeof(rxfh.rss_context))) > @@ -1351,6 +1382,44 @@ static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev, > else if (rxfh.indir_size != ETH_RXFH_INDIR_NO_CHANGE) > dev->priv_flags |= IFF_RXFH_CONFIGURED; > } > + /* Update rss_ctx tracking */ > + if (create) { > + /* Ideally this should happen before calling the driver, > + * so that we can fail more cleanly; but we don't have the > + * context ID until the driver picks it, so we have to > + * wait until after. > + */ > + if (WARN_ON(xa_load(&dev->ethtool->rss_ctx, rxfh.rss_context))) { > + /* context ID reused, our tracking is screwed */ > + kfree(ctx); > + goto out; > + } > + /* Allocate the exact ID the driver gave us */ > + if (xa_is_err(xa_store(&dev->ethtool->rss_ctx, rxfh.rss_context, > + ctx, GFP_KERNEL))) { > + kfree(ctx); > + goto out; > + } > + ctx->indir_no_change = rxfh.indir_size == ETH_RXFH_INDIR_NO_CHANGE; > + ctx->key_no_change = !rxfh.key_size; > + } > + if (delete) { > + WARN_ON(xa_erase(&dev->ethtool->rss_ctx, rxfh.rss_context) != ctx); > + kfree(ctx); > + } else if (ctx) { > + if (indir) { > + for (i = 0; i < dev_indir_size; i++) > + ethtool_rxfh_context_indir(ctx)[i] = indir[i]; > + ctx->indir_no_change = 0; > + } > + if (hkey) { > + memcpy(ethtool_rxfh_context_key(ctx), hkey, > + dev_key_size); > + ctx->key_no_change = 0; > + } > + if (rxfh.hfunc != ETH_RSS_HASH_NO_CHANGE) > + ctx->hfunc = rxfh.hfunc; > + } > > out: > kfree(rss_config); >
On Wed, Sep 27, 2023 at 07:13:34PM +0100, edward.cree@amd.com wrote: > From: Edward Cree <ecree.xilinx@gmail.com> > > Since drivers are still choosing the context IDs, we have to force the > XArray to use the ID they've chosen rather than picking one ourselves, > and handle the case where they give us an ID that's already in use. > > Signed-off-by: Edward Cree <ecree.xilinx@gmail.com> > --- > include/linux/ethtool.h | 14 ++++++++ > net/ethtool/ioctl.c | 73 +++++++++++++++++++++++++++++++++++++++-- > 2 files changed, 85 insertions(+), 2 deletions(-) > > diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h > index bb11cb2f477d..229a23571008 100644 > --- a/include/linux/ethtool.h > +++ b/include/linux/ethtool.h > @@ -194,6 +194,17 @@ static inline u8 *ethtool_rxfh_context_key(struct ethtool_rxfh_context *ctx) > return (u8 *)(ethtool_rxfh_context_indir(ctx) + ctx->indir_size); > } > > +static inline size_t ethtool_rxfh_context_size(u32 indir_size, u32 key_size, > + u16 priv_size) > +{ > + size_t indir_bytes = array_size(indir_size, sizeof(u32)); > + size_t flex_len; > + > + flex_len = size_add(size_add(indir_bytes, key_size), > + ALIGN(priv_size, sizeof(u32))); > + return struct_size((struct ethtool_rxfh_context *)0, data, flex_len); > +} > + > /* declare a link mode bitmap */ > #define __ETHTOOL_DECLARE_LINK_MODE_MASK(name) \ > DECLARE_BITMAP(name, __ETHTOOL_LINK_MODE_MASK_NBITS) > @@ -731,6 +742,8 @@ struct ethtool_mm_stats { > * will remain unchanged. > * Returns a negative error code or zero. An error code must be returned > * if at least one unsupported change was requested. > + * @rxfh_priv_size: size of the driver private data area the core should > + * allocate for an RSS context. An odd place to push the documentation. Please keep the ordering the same as the struct has below. Martin > * @get_rxfh_context: Get the contents of the RX flow hash indirection table, > * hash key, and/or hash function assiciated to the given rss context. > * Returns a negative error code or zero. > @@ -824,6 +837,7 @@ struct ethtool_ops { > u32 cap_link_lanes_supported:1; > u32 supported_coalesce_params; > u32 supported_ring_params; > + u16 rxfh_priv_size; > void (*get_drvinfo)(struct net_device *, struct ethtool_drvinfo *); > int (*get_regs_len)(struct net_device *); > void (*get_regs)(struct net_device *, struct ethtool_regs *, void *); > diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c > index de78b24fffc9..1d13bc8fbb75 100644 > --- a/net/ethtool/ioctl.c > +++ b/net/ethtool/ioctl.c > @@ -1249,6 +1249,7 @@ static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev, > { > int ret; > const struct ethtool_ops *ops = dev->ethtool_ops; > + struct ethtool_rxfh_context *ctx = NULL; > struct ethtool_rxnfc rx_rings; > struct ethtool_rxfh rxfh; > u32 dev_indir_size = 0, dev_key_size = 0, i; > @@ -1256,7 +1257,7 @@ static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev, > u8 *hkey = NULL; > u8 *rss_config; > u32 rss_cfg_offset = offsetof(struct ethtool_rxfh, rss_config[0]); > - bool delete = false; > + bool create = false, delete = false; > > if (!ops->get_rxnfc || !ops->set_rxfh) > return -EOPNOTSUPP; > @@ -1275,6 +1276,7 @@ static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev, > /* Most drivers don't handle rss_context, check it's 0 as well */ > if (rxfh.rss_context && !ops->set_rxfh_context) > return -EOPNOTSUPP; > + create = rxfh.rss_context == ETH_RXFH_CONTEXT_ALLOC; > > /* If either indir, hash key or function is valid, proceed further. > * Must request at least one change: indir size, hash key or function. > @@ -1332,13 +1334,42 @@ static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev, > } > } > > + if (create) { > + if (delete) { > + ret = -EINVAL; > + goto out; > + } > + ctx = kzalloc(ethtool_rxfh_context_size(dev_indir_size, > + dev_key_size, > + ops->rxfh_priv_size), > + GFP_KERNEL_ACCOUNT); > + if (!ctx) { > + ret = -ENOMEM; > + goto out; > + } > + ctx->indir_size = dev_indir_size; > + ctx->key_size = dev_key_size; > + ctx->hfunc = rxfh.hfunc; > + ctx->priv_size = ops->rxfh_priv_size; > + } else if (rxfh.rss_context) { > + ctx = xa_load(&dev->ethtool->rss_ctx, rxfh.rss_context); > + if (!ctx) { > + ret = -ENOENT; > + goto out; > + } > + } > + > if (rxfh.rss_context) > ret = ops->set_rxfh_context(dev, indir, hkey, rxfh.hfunc, > &rxfh.rss_context, delete); > else > ret = ops->set_rxfh(dev, indir, hkey, rxfh.hfunc); > - if (ret) > + if (ret) { > + if (create) > + /* failed to create, free our new tracking entry */ > + kfree(ctx); > goto out; > + } > > if (copy_to_user(useraddr + offsetof(struct ethtool_rxfh, rss_context), > &rxfh.rss_context, sizeof(rxfh.rss_context))) > @@ -1351,6 +1382,44 @@ static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev, > else if (rxfh.indir_size != ETH_RXFH_INDIR_NO_CHANGE) > dev->priv_flags |= IFF_RXFH_CONFIGURED; > } > + /* Update rss_ctx tracking */ > + if (create) { > + /* Ideally this should happen before calling the driver, > + * so that we can fail more cleanly; but we don't have the > + * context ID until the driver picks it, so we have to > + * wait until after. > + */ > + if (WARN_ON(xa_load(&dev->ethtool->rss_ctx, rxfh.rss_context))) { > + /* context ID reused, our tracking is screwed */ > + kfree(ctx); > + goto out; > + } > + /* Allocate the exact ID the driver gave us */ > + if (xa_is_err(xa_store(&dev->ethtool->rss_ctx, rxfh.rss_context, > + ctx, GFP_KERNEL))) { > + kfree(ctx); > + goto out; > + } > + ctx->indir_no_change = rxfh.indir_size == ETH_RXFH_INDIR_NO_CHANGE; > + ctx->key_no_change = !rxfh.key_size; > + } > + if (delete) { > + WARN_ON(xa_erase(&dev->ethtool->rss_ctx, rxfh.rss_context) != ctx); > + kfree(ctx); > + } else if (ctx) { > + if (indir) { > + for (i = 0; i < dev_indir_size; i++) > + ethtool_rxfh_context_indir(ctx)[i] = indir[i]; > + ctx->indir_no_change = 0; > + } > + if (hkey) { > + memcpy(ethtool_rxfh_context_key(ctx), hkey, > + dev_key_size); > + ctx->key_no_change = 0; > + } > + if (rxfh.hfunc != ETH_RSS_HASH_NO_CHANGE) > + ctx->hfunc = rxfh.hfunc; > + } > > out: > kfree(rss_config);
diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h index bb11cb2f477d..229a23571008 100644 --- a/include/linux/ethtool.h +++ b/include/linux/ethtool.h @@ -194,6 +194,17 @@ static inline u8 *ethtool_rxfh_context_key(struct ethtool_rxfh_context *ctx) return (u8 *)(ethtool_rxfh_context_indir(ctx) + ctx->indir_size); } +static inline size_t ethtool_rxfh_context_size(u32 indir_size, u32 key_size, + u16 priv_size) +{ + size_t indir_bytes = array_size(indir_size, sizeof(u32)); + size_t flex_len; + + flex_len = size_add(size_add(indir_bytes, key_size), + ALIGN(priv_size, sizeof(u32))); + return struct_size((struct ethtool_rxfh_context *)0, data, flex_len); +} + /* declare a link mode bitmap */ #define __ETHTOOL_DECLARE_LINK_MODE_MASK(name) \ DECLARE_BITMAP(name, __ETHTOOL_LINK_MODE_MASK_NBITS) @@ -731,6 +742,8 @@ struct ethtool_mm_stats { * will remain unchanged. * Returns a negative error code or zero. An error code must be returned * if at least one unsupported change was requested. + * @rxfh_priv_size: size of the driver private data area the core should + * allocate for an RSS context. * @get_rxfh_context: Get the contents of the RX flow hash indirection table, * hash key, and/or hash function assiciated to the given rss context. * Returns a negative error code or zero. @@ -824,6 +837,7 @@ struct ethtool_ops { u32 cap_link_lanes_supported:1; u32 supported_coalesce_params; u32 supported_ring_params; + u16 rxfh_priv_size; void (*get_drvinfo)(struct net_device *, struct ethtool_drvinfo *); int (*get_regs_len)(struct net_device *); void (*get_regs)(struct net_device *, struct ethtool_regs *, void *); diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c index de78b24fffc9..1d13bc8fbb75 100644 --- a/net/ethtool/ioctl.c +++ b/net/ethtool/ioctl.c @@ -1249,6 +1249,7 @@ static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev, { int ret; const struct ethtool_ops *ops = dev->ethtool_ops; + struct ethtool_rxfh_context *ctx = NULL; struct ethtool_rxnfc rx_rings; struct ethtool_rxfh rxfh; u32 dev_indir_size = 0, dev_key_size = 0, i; @@ -1256,7 +1257,7 @@ static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev, u8 *hkey = NULL; u8 *rss_config; u32 rss_cfg_offset = offsetof(struct ethtool_rxfh, rss_config[0]); - bool delete = false; + bool create = false, delete = false; if (!ops->get_rxnfc || !ops->set_rxfh) return -EOPNOTSUPP; @@ -1275,6 +1276,7 @@ static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev, /* Most drivers don't handle rss_context, check it's 0 as well */ if (rxfh.rss_context && !ops->set_rxfh_context) return -EOPNOTSUPP; + create = rxfh.rss_context == ETH_RXFH_CONTEXT_ALLOC; /* If either indir, hash key or function is valid, proceed further. * Must request at least one change: indir size, hash key or function. @@ -1332,13 +1334,42 @@ static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev, } } + if (create) { + if (delete) { + ret = -EINVAL; + goto out; + } + ctx = kzalloc(ethtool_rxfh_context_size(dev_indir_size, + dev_key_size, + ops->rxfh_priv_size), + GFP_KERNEL_ACCOUNT); + if (!ctx) { + ret = -ENOMEM; + goto out; + } + ctx->indir_size = dev_indir_size; + ctx->key_size = dev_key_size; + ctx->hfunc = rxfh.hfunc; + ctx->priv_size = ops->rxfh_priv_size; + } else if (rxfh.rss_context) { + ctx = xa_load(&dev->ethtool->rss_ctx, rxfh.rss_context); + if (!ctx) { + ret = -ENOENT; + goto out; + } + } + if (rxfh.rss_context) ret = ops->set_rxfh_context(dev, indir, hkey, rxfh.hfunc, &rxfh.rss_context, delete); else ret = ops->set_rxfh(dev, indir, hkey, rxfh.hfunc); - if (ret) + if (ret) { + if (create) + /* failed to create, free our new tracking entry */ + kfree(ctx); goto out; + } if (copy_to_user(useraddr + offsetof(struct ethtool_rxfh, rss_context), &rxfh.rss_context, sizeof(rxfh.rss_context))) @@ -1351,6 +1382,44 @@ static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev, else if (rxfh.indir_size != ETH_RXFH_INDIR_NO_CHANGE) dev->priv_flags |= IFF_RXFH_CONFIGURED; } + /* Update rss_ctx tracking */ + if (create) { + /* Ideally this should happen before calling the driver, + * so that we can fail more cleanly; but we don't have the + * context ID until the driver picks it, so we have to + * wait until after. + */ + if (WARN_ON(xa_load(&dev->ethtool->rss_ctx, rxfh.rss_context))) { + /* context ID reused, our tracking is screwed */ + kfree(ctx); + goto out; + } + /* Allocate the exact ID the driver gave us */ + if (xa_is_err(xa_store(&dev->ethtool->rss_ctx, rxfh.rss_context, + ctx, GFP_KERNEL))) { + kfree(ctx); + goto out; + } + ctx->indir_no_change = rxfh.indir_size == ETH_RXFH_INDIR_NO_CHANGE; + ctx->key_no_change = !rxfh.key_size; + } + if (delete) { + WARN_ON(xa_erase(&dev->ethtool->rss_ctx, rxfh.rss_context) != ctx); + kfree(ctx); + } else if (ctx) { + if (indir) { + for (i = 0; i < dev_indir_size; i++) + ethtool_rxfh_context_indir(ctx)[i] = indir[i]; + ctx->indir_no_change = 0; + } + if (hkey) { + memcpy(ethtool_rxfh_context_key(ctx), hkey, + dev_key_size); + ctx->key_no_change = 0; + } + if (rxfh.hfunc != ETH_RSS_HASH_NO_CHANGE) + ctx->hfunc = rxfh.hfunc; + } out: kfree(rss_config);