Add XSKMAP support (iovisor#2730)

* add XSKMAP support
* enable lookup for XSKMAP available from kernel 5.3
* add section for XSKMAP in reference guide

docs/reference_guide.md

@@ -44,16 +44,17 @@ This guide is incomplete. If something feels missing, check the bcc and kernel s
  - [9. BPF_PROG_ARRAY](#9-bpf_prog_array)
  - [10. BPF_DEVMAP](#10-bpf_devmap)
  - [11. BPF_CPUMAP](#11-bpf_cpumap)
- - [12. map.lookup()](#12-maplookup)
- - [13. map.lookup_or_try_init()](#13-maplookup_or_try_init)
- - [14. map.delete()](#14-mapdelete)
- - [15. map.update()](#15-mapupdate)
- - [16. map.insert()](#16-mapinsert)
- - [17. map.increment()](#17-mapincrement)
- - [18. map.get_stackid()](#18-mapget_stackid)
- - [19. map.perf_read()](#19-mapperf_read)
- - [20. map.call()](#20-mapcall)
- - [21. map.redirect_map()](#21-mapredirect_map)
+ - [12. BPF_XSKMAP](#12-bpf_xskmap)
+ - [13. map.lookup()](#13-maplookup)
+ - [14. map.lookup_or_try_init()](#14-maplookup_or_try_init)
+ - [15. map.delete()](#15-mapdelete)
+ - [16. map.update()](#16-mapupdate)
+ - [17. map.insert()](#17-mapinsert)
+ - [18. map.increment()](#18-mapincrement)
+ - [19. map.get_stackid()](#19-mapget_stackid)
+ - [20. map.perf_read()](#20-mapperf_read)
+ - [21. map.call()](#21-mapcall)
+ - [22. map.redirect_map()](#22-mapredirect_map)
  - [Licensing](#licensing)
 
 - [bcc Python](#bcc-python)
@@ -759,7 +760,23 @@ Methods (covered later): map.redirect_map().
 Examples in situ:
 [search /examples](https://github.com/iovisor/bcc/search?q=BPF_CPUMAP+path%3Aexamples&type=Code),
 
-### 12. BPF_ARRAY_OF_MAPS
+### 12. BPF_XSKMAP
+
+Syntax: ```BPF_XSKMAP(name, size)```
+
+This creates a xsk map named ```name``` with ```size``` entries. Each entry represents one NIC's queue id. This map is only used in XDP to redirect packet to an AF_XDP socket. If the AF_XDP socket is binded to a queue which is different than the current packet's queue id, the packet will be dropped. For kernel v5.3 and latter, `lookup` method is available and can be used to check whether and AF_XDP socket is available for the current packet's queue id. More details at [AF_XDP](https://www.kernel.org/doc/html/latest/networking/af_xdp.html).
+
+For example:
+```C
+BPF_XSKMAP(xsks_map, 8);
+```
+
+Methods (covered later): map.redirect_map(). map.lookup()
+
+Examples in situ:
+[search /examples](https://github.com/iovisor/bcc/search?q=BPF_XSKMAP+path%3Aexamples&type=Code),
+
+### 13. BPF_ARRAY_OF_MAPS
 
 Syntax: ```BPF_ARRAY_OF_MAPS(name, inner_map_name, size)```
 
@@ -772,7 +789,7 @@ BPF_TABLE("hash", int, int, ex2, 1024);
 BPF_ARRAY_OF_MAPS(maps_array, "ex1", 10);
 ```
 
-### 13. BPF_HASH_OF_MAPS
+### 14. BPF_HASH_OF_MAPS
 
 Syntax: ```BPF_HASH_OF_MAPS(name, inner_map_name, size)```
 
@@ -785,7 +802,7 @@ BPF_ARRAY(ex2, int, 1024);
 BPF_HASH_OF_MAPS(maps_hash, "ex1", 10);
 ```
 
-### 14. map.lookup()
+### 15. map.lookup()
 
 Syntax: ```*val map.lookup(&key)```
 
@@ -795,7 +812,7 @@ Examples in situ:
 [search /examples](https://github.com/iovisor/bcc/search?q=lookup+path%3Aexamples&type=Code),
 [search /tools](https://github.com/iovisor/bcc/search?q=lookup+path%3Atools&type=Code)
 
-### 15. map.lookup_or_try_init()
+### 16. map.lookup_or_try_init()
 
 Syntax: ```*val map.lookup_or_try_init(&key, &zero)```
 
@@ -808,7 +825,7 @@ Examples in situ:
 Note: The old map.lookup_or_init() may cause return from the function, so lookup_or_try_init() is recommended as it
 does not have this side effect.
 
-### 16. map.delete()
+### 17. map.delete()
 
 Syntax: ```map.delete(&key)```
 
@@ -818,7 +835,7 @@ Examples in situ:
 [search /examples](https://github.com/iovisor/bcc/search?q=delete+path%3Aexamples&type=Code),
 [search /tools](https://github.com/iovisor/bcc/search?q=delete+path%3Atools&type=Code)
 
-### 17. map.update()
+### 18. map.update()
 
 Syntax: ```map.update(&key, &val)```
 
@@ -828,7 +845,7 @@ Examples in situ:
 [search /examples](https://github.com/iovisor/bcc/search?q=update+path%3Aexamples&type=Code),
 [search /tools](https://github.com/iovisor/bcc/search?q=update+path%3Atools&type=Code)
 
-### 18. map.insert()
+### 19. map.insert()
 
 Syntax: ```map.insert(&key, &val)```
 
@@ -838,7 +855,7 @@ Examples in situ:
 [search /examples](https://github.com/iovisor/bcc/search?q=insert+path%3Aexamples&type=Code),
 [search /tools](https://github.com/iovisor/bcc/search?q=insert+path%3Atools&type=Code)
 
-### 19. map.increment()
+### 20. map.increment()
 
 Syntax: ```map.increment(key[, increment_amount])```
 
@@ -848,7 +865,7 @@ Examples in situ:
 [search /examples](https://github.com/iovisor/bcc/search?q=increment+path%3Aexamples&type=Code),
 [search /tools](https://github.com/iovisor/bcc/search?q=increment+path%3Atools&type=Code)
 
-### 20. map.get_stackid()
+### 21. map.get_stackid()
 
 Syntax: ```int map.get_stackid(void *ctx, u64 flags)```
 
@@ -858,7 +875,7 @@ Examples in situ:
 [search /examples](https://github.com/iovisor/bcc/search?q=get_stackid+path%3Aexamples&type=Code),
 [search /tools](https://github.com/iovisor/bcc/search?q=get_stackid+path%3Atools&type=Code)
 
-### 21. map.perf_read()
+### 22. map.perf_read()
 
 Syntax: ```u64 map.perf_read(u32 cpu)```
 
@@ -867,7 +884,7 @@ This returns the hardware performance counter as configured in [5. BPF_PERF_ARRA
 Examples in situ:
 [search /tests](https://github.com/iovisor/bcc/search?q=perf_read+path%3Atests&type=Code)
 
-### 22. map.call()
+### 23. map.call()
 
 Syntax: ```void map.call(void *ctx, int index)```
 
@@ -906,11 +923,11 @@ Examples in situ:
 [search /examples](https://github.com/iovisor/bcc/search?l=C&q=call+path%3Aexamples&type=Code),
 [search /tests](https://github.com/iovisor/bcc/search?l=C&q=call+path%3Atests&type=Code)
 
-### 23. map.redirect_map()
+### 24. map.redirect_map()
 
 Syntax: ```int map.redirect_map(int index, int flags)```
 
-This redirects the incoming packets based on the ```index``` entry. If the map is [10. BPF_DEVMAP](#10-bpf_devmap), the packet will be sent to the transmit queue of the network interface that the entry points to. If the map is [11. BPF_CPUMAP](#11-bpf_cpumap), the packet will be sent to the ring buffer of the ```index``` CPU and be processed by the CPU later.
+This redirects the incoming packets based on the ```index``` entry. If the map is [10. BPF_DEVMAP](#10-bpf_devmap), the packet will be sent to the transmit queue of the network interface that the entry points to. If the map is [11. BPF_CPUMAP](#11-bpf_cpumap), the packet will be sent to the ring buffer of the ```index``` CPU and be processed by the CPU later. If the map is [12. BPF_XSKMAP](#12-bpf_xskmap), the packet will be sent to the AF_XDP socket attached to the queue.
 
 If the packet is redirected successfully, the function will return XDP_REDIRECT. Otherwise, it will return XDP_ABORTED to discard the packet.
 

src/cc/api/BPF.cc

@@ -677,6 +677,13 @@ BPFDevmapTable BPF::get_devmap_table(const std::string& name) {
  return BPFDevmapTable({});
 }
 
+BPFXskmapTable BPF::get_xskmap_table(const std::string& name) {
+ TableStorage::iterator it;
+ if (bpf_module_->table_storage().Find(Path({bpf_module_->id(), name}), it))
+ return BPFXskmapTable(it->second);
+ return BPFXskmapTable({});
+}
+
 BPFStackTable BPF::get_stack_table(const std::string& name, bool use_debug_file,
  bool check_debug_file_crc) {
  TableStorage::iterator it;

src/cc/api/BPF.h

@@ -167,6 +167,8 @@ class BPF {
 
  BPFDevmapTable get_devmap_table(const std::string& name);
 
+ BPFXskmapTable get_xskmap_table(const std::string& name);
+
  BPFStackTable get_stack_table(const std::string& name,
  bool use_debug_file = true,
  bool check_debug_file_crc = true);

src/cc/api/BPFTable.cc

@@ -627,21 +627,21 @@ StatusTuple BPFCgroupArray::remove_value(const int& index) {
  return StatusTuple(0);
 }
 
-BPFDevmapTable::BPFDevmapTable(const TableDesc& desc) 
+BPFDevmapTable::BPFDevmapTable(const TableDesc& desc)
  : BPFTableBase<int, int>(desc) {
  if(desc.type != BPF_MAP_TYPE_DEVMAP)
- throw std::invalid_argument("Table '" + desc.name + 
+ throw std::invalid_argument("Table '" + desc.name +
  "' is not a devmap table");
 }
 
-StatusTuple BPFDevmapTable::update_value(const int& index, 
+StatusTuple BPFDevmapTable::update_value(const int& index,
  const int& value) {
  if (!this->update(const_cast<int*>(&index), const_cast<int*>(&value)))
  return StatusTuple(-1, "Error updating value: %s", std::strerror(errno));
  return StatusTuple(0);
 }
 
-StatusTuple BPFDevmapTable::get_value(const int& index, 
+StatusTuple BPFDevmapTable::get_value(const int& index,
  int& value) {
  if (!this->lookup(const_cast<int*>(&index), &value))
  return StatusTuple(-1, "Error getting value: %s", std::strerror(errno));
@@ -654,6 +654,33 @@ StatusTuple BPFDevmapTable::remove_value(const int& index) {
  return StatusTuple(0);
 }
 
+BPFXskmapTable::BPFXskmapTable(const TableDesc& desc)
+ : BPFTableBase<int, int>(desc) {
+ if(desc.type != BPF_MAP_TYPE_XSKMAP)
+ throw std::invalid_argument("Table '" + desc.name +
+ "' is not a xskmap table");
+}
+
+StatusTuple BPFXskmapTable::update_value(const int& index,
+ const int& value) {
+ if (!this->update(const_cast<int*>(&index), const_cast<int*>(&value)))
+ return StatusTuple(-1, "Error updating value: %s", std::strerror(errno));
+ return StatusTuple(0);
+}
+
+StatusTuple BPFXskmapTable::get_value(const int& index,
+ int& value) {
+ if (!this->lookup(const_cast<int*>(&index), &value))
+ return StatusTuple(-1, "Error getting value: %s", std::strerror(errno));
+ return StatusTuple(0);
+}
+
+StatusTuple BPFXskmapTable::remove_value(const int& index) {
+ if (!this->remove(const_cast<int*>(&index)))
+ return StatusTuple(-1, "Error removing value: %s", std::strerror(errno));
+ return StatusTuple(0);
+}
+
 BPFMapInMapTable::BPFMapInMapTable(const TableDesc& desc)
  : BPFTableBase<int, int>(desc) {
  if(desc.type != BPF_MAP_TYPE_ARRAY_OF_MAPS &&

src/cc/api/BPFTable.h

@@ -395,7 +395,16 @@ class BPFCgroupArray : public BPFTableBase<int, int> {
 class BPFDevmapTable : public BPFTableBase<int, int> {
 public:
  BPFDevmapTable(const TableDesc& desc);
-
+
+ StatusTuple update_value(const int& index, const int& value);
+ StatusTuple get_value(const int& index, int& value);
+ StatusTuple remove_value(const int& index);
+};
+
+class BPFXskmapTable : public BPFTableBase<int, int> {
+public:
+ BPFXskmapTable(const TableDesc& desc);
+
  StatusTuple update_value(const int& index, const int& value);
  StatusTuple get_value(const int& index, int& value);
  StatusTuple remove_value(const int& index);

src/cc/export/helpers.h

@@ -269,6 +269,18 @@ struct _name##_table_t _name = { .max_entries = (_max_entries) }
 #define BPF_CPUMAP(_name, _max_entries) \
  BPF_XDP_REDIRECT_MAP("cpumap", u32, _name, _max_entries)
 
+#define BPF_XSKMAP(_name, _max_entries) \
+struct _name##_table_t { \
+ u32 key; \
+ int leaf; \
+ int * (*lookup) (int *); \
+ /* xdp_act = map.redirect_map(index, flag) */ \
+ u64 (*redirect_map) (int, int); \
+ u32 max_entries; \
+}; \
+__attribute__((section("maps/xskmap"))) \
+struct _name##_table_t _name = { .max_entries = (_max_entries) }
+
 #define BPF_ARRAY_OF_MAPS(_name, _inner_map_name, _max_entries) \
  BPF_TABLE("array_of_maps$" _inner_map_name, int, int, _name, _max_entries)
 

src/cc/frontends/clang/b_frontend_action.cc

@@ -1259,6 +1259,8 @@ bool BTypeVisitor::VisitVarDecl(VarDecl *Decl) {
  map_type = BPF_MAP_TYPE_DEVMAP;
  } else if (section_attr == "maps/cpumap") {
  map_type = BPF_MAP_TYPE_CPUMAP;
+ } else if (section_attr == "maps/xskmap") {
+ map_type = BPF_MAP_TYPE_XSKMAP;
  } else if (section_attr == "maps/hash_of_maps") {
  map_type = BPF_MAP_TYPE_HASH_OF_MAPS;
  } else if (section_attr == "maps/array_of_maps") {

src/python/bcc/table.py

@@ -182,6 +182,8 @@ def Table(bpf, map_id, map_fd, keytype, leaftype, name, **kwargs):
  t = DevMap(bpf, map_id, map_fd, keytype, leaftype)
  elif ttype == BPF_MAP_TYPE_CPUMAP:
  t = CpuMap(bpf, map_id, map_fd, keytype, leaftype)
+ elif ttype == BPF_MAP_TYPE_XSKMAP:
+ t = XskMap(bpf, map_id, map_fd, keytype, leaftype)
  elif ttype == BPF_MAP_TYPE_ARRAY_OF_MAPS:
  t = MapInMapArray(bpf, map_id, map_fd, keytype, leaftype)
  elif ttype == BPF_MAP_TYPE_HASH_OF_MAPS:
@@ -905,6 +907,10 @@ class CpuMap(ArrayBase):
  def __init__(self, *args, **kwargs):
  super(CpuMap, self).__init__(*args, **kwargs)
 
+class XskMap(ArrayBase):
+ def __init__(self, *args, **kwargs):
+ super(XskMap, self).__init__(*args, **kwargs)
+
 class MapInMapArray(ArrayBase):
  def __init__(self, *args, **kwargs):
  super(MapInMapArray, self).__init__(*args, **kwargs)