src/ats/plugin_ats_ril.c

   1 /*
   2    This file is part of GNUnet.
   3    Copyright (C) 2011-2014 GNUnet e.V.
   4
   5    GNUnet is free software: you can redistribute it and/or modify it
   6    under the terms of the GNU Affero General Public License as published
   7    by the Free Software Foundation, either version 3 of the License,
   8    or (at your option) any later version.
   9
  10    GNUnet is distributed in the hope that it will be useful, but
  11    WITHOUT ANY WARRANTY; without even the implied warranty of
  12    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  13    Affero General Public License for more details.
  14
  15    You should have received a copy of the GNU Affero General Public License
  16    along with this program.  If not, see <http://www.gnu.org/licenses/>.
  17
  18      SPDX-License-Identifier: AGPL3.0-or-later
  19  */
  20
  21 /**
  22  * @file ats/plugin_ats_ril.c
  23  * @brief ATS reinforcement learning solver
  24  * @author Fabian Oehlmann
  25  * @author Matthias Wachs
  26  */
  27 #include "platform.h"
  28 #include <float.h>
  29 #include <math.h>
  30 #include "gnunet_ats_plugin.h"
  31 #include "gnunet-service-ats_addresses.h"
  32
  33
  34
  35 #define LOG(kind, ...) GNUNET_log_from(kind, "ats-ril", __VA_ARGS__)
  36
  37 #define RIL_MIN_BW                      (5 * ntohl(GNUNET_CONSTANTS_DEFAULT_BW_IN_OUT.value__))
  38 #define RIL_MAX_BW                      GNUNET_ATS_MaxBandwidth
  39
  40 #define RIL_ACTION_INVALID              -1
  41 #define RIL_INTERVAL_EXPONENT           10
  42 #define RIL_UTILITY_DELAY_MAX           1000
  43
  44 #define RIL_DEFAULT_STEP_TIME_MIN       GNUNET_TIME_relative_multiply(GNUNET_TIME_UNIT_MILLISECONDS, 200)
  45 #define RIL_DEFAULT_STEP_TIME_MAX       GNUNET_TIME_relative_multiply(GNUNET_TIME_UNIT_MILLISECONDS, 2000)
  46 #define RIL_DEFAULT_ALGORITHM           RIL_ALGO_SARSA
  47 #define RIL_DEFAULT_SELECT              RIL_SELECT_SOFTMAX
  48 #define RIL_DEFAULT_WELFARE             RIL_WELFARE_NASH
  49 #define RIL_DEFAULT_DISCOUNT_BETA       0.6
  50 #define RIL_DEFAULT_DISCOUNT_GAMMA      0.5
  51 #define RIL_DEFAULT_GRADIENT_STEP_SIZE  0.01
  52 #define RIL_DEFAULT_TRACE_DECAY         0.5
  53 #define RIL_DEFAULT_EXPLORE_RATIO       1
  54 #define RIL_DEFAULT_EXPLORE_DECAY       0.95
  55 #define RIL_DEFAULT_RBF_DIVISOR         50
  56 #define RIL_DEFAULT_TEMPERATURE         0.1
  57 #define RIL_DEFAULT_TEMPERATURE_DECAY   1
  58
  59 #define RIL_INC_DEC_STEP_SIZE           1
  60 #define RIL_NOP_DECAY                   0.5
  61
  62 /**
  63  * ATS reinforcement learning solver
  64  *
  65  * General description
  66  */
  67
  68 /**
  69  * The actions, how an agent can manipulate the current assignment. I.e. how the bandwidth can be
  70  * changed for the currently chosen address. Not depicted in the enum are the actions of switching
  71  * to a particular address. The action of switching to address with index i is depicted by the
  72  * number (RIL_ACTION_TYPE_NUM + i).
  73  */
  74 enum RIL_Action_Type {
  75   RIL_ACTION_NOTHING = 0,
  76   RIL_ACTION_BW_IN_DBL = -2, //TODO? Potentially add more actions
  77   RIL_ACTION_BW_IN_HLV = -3,
  78   RIL_ACTION_BW_IN_INC = 1,
  79   RIL_ACTION_BW_IN_DEC = 2,
  80   RIL_ACTION_BW_OUT_DBL = -4,
  81   RIL_ACTION_BW_OUT_HLV = -5,
  82   RIL_ACTION_BW_OUT_INC = 3,
  83   RIL_ACTION_BW_OUT_DEC = 4,
  84   RIL_ACTION_TYPE_NUM = 5
  85 };
  86
  87 enum RIL_Algorithm {
  88   RIL_ALGO_SARSA = 0,
  89   RIL_ALGO_Q = 1
  90 };
  91
  92 enum RIL_Select {
  93   RIL_SELECT_SOFTMAX = 0,
  94   RIL_SELECT_EGREEDY = 1
  95 };
  96
  97 enum RIL_Welfare {
  98   RIL_WELFARE_NASH,
  99   RIL_WELFARE_EGALITARIAN
 100 };
 101
 102 enum RIL_E_Modification {
 103   RIL_E_DECAY,
 104   RIL_E_ZERO,
 105   RIL_E_ACCUMULATE,
 106   RIL_E_REPLACE
 107 };
 108
 109 /**
 110  * Global learning parameters
 111  */
 112 struct RIL_Learning_Parameters {
 113   /**
 114    * The TD-algorithm to use
 115    */
 116   enum RIL_Algorithm algorithm;
 117
 118   /**
 119    * Gradient-descent step-size
 120    */
 121   double alpha;
 122
 123   /**
 124    * Learning discount variable in the TD-update for semi-MDPs
 125    */
 126   double beta;
 127
 128   /**
 129    * Learning discount factor in the TD-update for MDPs
 130    */
 131   double gamma;
 132
 133   /**
 134    * Trace-decay factor for eligibility traces
 135    */
 136   double lambda;
 137
 138   /**
 139    * Whether to accumulate or replace eligibility traces
 140    */
 141   enum RIL_E_Modification eligibility_trace_mode;
 142
 143   /**
 144    * Initial softmax action-selection temperature
 145    */
 146   double temperature_init;
 147
 148   /**
 149    * Softmax action-selection temperature
 150    */
 151   double temperature;
 152
 153   /**
 154    * Decay factor of the temperature value
 155    */
 156   double temperature_decay;
 157
 158   /**
 159    * Which measure of social welfare should be used
 160    */
 161   enum RIL_Welfare social_welfare;
 162
 163   /**
 164    * State space divisor
 165    */
 166   unsigned long long rbf_divisor;
 167
 168   /**
 169    * Action selection strategy;
 170    */
 171   enum RIL_Select select;
 172
 173   /**
 174    * Initial exploration ratio value
 175    */
 176   double epsilon_init;
 177
 178   /**
 179    * Ratio, with what probability an agent should explore in the e-greed policy
 180    */
 181   double epsilon;
 182
 183   /**
 184    * Decay factor of the explore ratio
 185    */
 186   double epsilon_decay;
 187
 188   /**
 189    * Minimal interval time between steps in milliseconds
 190    */
 191   struct GNUNET_TIME_Relative step_time_min;
 192
 193   /**
 194    * Maximum interval time between steps in milliseconds
 195    */
 196   struct GNUNET_TIME_Relative step_time_max;
 197 };
 198
 199 /**
 200  * Wrapper for addresses to store them in agent's linked list
 201  */
 202 struct RIL_Address_Wrapped {
 203   /**
 204    * Next in DLL
 205    */
 206   struct RIL_Address_Wrapped *next;
 207
 208   /**
 209    * Previous in DLL
 210    */
 211   struct RIL_Address_Wrapped *prev;
 212
 213   /**
 214    * The address
 215    */
 216   struct ATS_Address *address_naked;
 217 };
 218
 219
 220 struct RIL_Peer_Agent {
 221   /**
 222    * Next agent in solver's linked list
 223    */
 224   struct RIL_Peer_Agent *next;
 225
 226   /**
 227    * Previous agent in solver's linked list
 228    */
 229   struct RIL_Peer_Agent *prev;
 230
 231   /**
 232    * Environment handle
 233    */
 234   struct GAS_RIL_Handle *envi;
 235
 236   /**
 237    * Peer ID
 238    */
 239   struct GNUNET_PeerIdentity peer;
 240
 241   /**
 242    * Whether the agent is active or not
 243    */
 244   int is_active;
 245
 246   /**
 247    * Number of performed time-steps
 248    */
 249   unsigned long long step_count;
 250
 251   /**
 252    * Experience matrix W
 253    */
 254   double ** W;
 255
 256   /**
 257    * Number of rows of W / Number of state-vector features
 258    */
 259   unsigned int m;
 260
 261   /**
 262    * Number of columns of W / Number of actions
 263    */
 264   unsigned int n;
 265
 266   /**
 267    * Last perceived state feature vector
 268    */
 269   double *s_old;
 270
 271   /**
 272    * Last chosen action
 273    */
 274   int a_old;
 275
 276   /**
 277    * Eligibility traces
 278    */
 279   double ** E;
 280
 281   /**
 282    * Whether to reset the eligibility traces to 0 after a Q-exploration step
 283    */
 284   int eligibility_reset;
 285
 286   /**
 287    * Address in use
 288    */
 289   struct ATS_Address *address_inuse;
 290
 291   /**
 292    * Head of addresses DLL
 293    */
 294   struct RIL_Address_Wrapped *addresses_head;
 295
 296   /**
 297    * Tail of addresses DLL
 298    */
 299   struct RIL_Address_Wrapped *addresses_tail;
 300
 301   /**
 302    * Inbound bandwidth assigned by the agent
 303    */
 304   uint32_t bw_in;
 305
 306   /**
 307    * Outbound bandwidth assigned by the agent
 308    */
 309   uint32_t bw_out;
 310
 311   /**
 312    * Flag whether a suggestion has to be issued
 313    */
 314   int suggestion_issue;
 315
 316   /**
 317    * The address which has to be issued
 318    */
 319   struct ATS_Address *suggestion_address;
 320
 321   /**
 322    * The agent's last objective value
 323    */
 324   double objective_old;
 325
 326   /**
 327    * NOP bonus
 328    */
 329   double nop_bonus;
 330 };
 331
 332 struct RIL_Scope {
 333   /**
 334    * ATS network type
 335    */
 336   enum GNUNET_NetworkType type;
 337
 338   /**
 339    * Total available inbound bandwidth
 340    */
 341   uint32_t bw_in_available;
 342
 343   /**
 344    * Bandwidth inbound assigned in network after last step
 345    */
 346   uint32_t bw_in_assigned;
 347
 348   /**
 349    * Bandwidth inbound actually utilized in the network
 350    */
 351   uint32_t bw_in_utilized;
 352
 353   /**
 354    * Total available outbound bandwidth
 355    */
 356   uint32_t bw_out_available;
 357
 358   /**
 359    * Bandwidth outbound assigned in network after last step
 360    */
 361   unsigned long long bw_out_assigned;
 362
 363   /**
 364    * Bandwidth outbound actually utilized in the network
 365    */
 366   unsigned long long bw_out_utilized;
 367
 368   /**
 369    * Number of active agents in scope
 370    */
 371   unsigned int active_agent_count;
 372
 373   /**
 374    * The social welfare achieved in the scope
 375    */
 376   double social_welfare;
 377 };
 378
 379 /**
 380  * A handle for the reinforcement learning solver
 381  */
 382 struct GAS_RIL_Handle {
 383   /**
 384    * The solver-plugin environment of the solver-plugin API
 385    */
 386   struct GNUNET_ATS_PluginEnvironment *env;
 387
 388   /**
 389    * Number of performed steps
 390    */
 391   unsigned long long step_count;
 392
 393   /**
 394    * Timestamp for the last time-step
 395    */
 396   struct GNUNET_TIME_Absolute step_time_last;
 397
 398   /**
 399    * Task identifier of the next time-step to be executed
 400    */
 401   struct GNUNET_SCHEDULER_Task * step_next_task_id;
 402
 403   /**
 404    * Variable discount factor, dependent on time between steps
 405    */
 406   double global_discount_variable;
 407
 408   /**
 409    * Integrated variable discount factor, dependent on time between steps
 410    */
 411   double global_discount_integrated;
 412
 413   /**
 414    * Lock for bulk operations
 415    */
 416   int bulk_lock;
 417
 418   /**
 419    * Number of changes during a lock
 420    */
 421   int bulk_changes;
 422
 423   /**
 424    * Learning parameters
 425    */
 426   struct RIL_Learning_Parameters parameters;
 427
 428   /**
 429    * Array of networks with global assignment state
 430    */
 431   struct RIL_Scope * network_entries;
 432
 433   /**
 434    * Networks count
 435    */
 436   unsigned int networks_count;
 437
 438   /**
 439    * List of active peer-agents
 440    */
 441   struct RIL_Peer_Agent * agents_head;
 442   struct RIL_Peer_Agent * agents_tail;
 443
 444   /**
 445    * Shutdown
 446    */
 447   int done;
 448
 449   /**
 450    * Simulate steps, i.e. schedule steps immediately
 451    */
 452   unsigned long long simulate;
 453 };
 454
 455 /*
 456  *  "Private" functions
 457  *  ---------------------------
 458  */
 459
 460 /**
 461  * Estimate the current action-value for state s and action a
 462  *
 463  * @param agent agent performing the estimation
 464  * @param state s
 465  * @param action a
 466  * @return estimation value
 467  */
 468 static double
 469 agent_q(struct RIL_Peer_Agent *agent,
 470         const double *state,
 471         int action)
 472 {
 473   unsigned int i;
 474   double result = 0.0;
 475
 476   for (i = 0; i < agent->m; i++)
 477     result += state[i] * agent->W[action][i];
 478
 479   /* prevent crashes if learning diverges */
 480   if (isnan(result))
 481     return isnan(result) * UINT32_MAX;
 482   if (isinf(result))
 483     return isinf(result) * UINT32_MAX;
 484   return result;
 485 }
 486
 487
 488 /**
 489  * Get the index of the address in the agent's list.
 490  *
 491  * @param agent agent handle
 492  * @param address address handle
 493  * @return the index, starting with zero
 494  */
 495 static int
 496 agent_address_get_index(struct RIL_Peer_Agent *agent, struct ATS_Address *address)
 497 {
 498   int i;
 499   struct RIL_Address_Wrapped *cur;
 500
 501   i = -1;
 502   for (cur = agent->addresses_head; NULL != cur; cur = cur->next)
 503     {
 504       i++;
 505       if (cur->address_naked == address)
 506         return i;
 507     }
 508   return i;
 509 }
 510
 511
 512 /**
 513  * Gets the wrapped address from the agent's list
 514  *
 515  * @param agent agent handle
 516  * @param address address handle
 517  * @return wrapped address
 518  */
 519 static struct RIL_Address_Wrapped *
 520 agent_address_get_wrapped(struct RIL_Peer_Agent *agent, struct ATS_Address *address)
 521 {
 522   struct RIL_Address_Wrapped *cur;
 523
 524   for (cur = agent->addresses_head; NULL != cur; cur = cur->next)
 525     if (cur->address_naked == address)
 526       return cur;
 527   return NULL;
 528 }
 529
 530
 531 static int
 532 agent_action_is_possible(struct RIL_Peer_Agent *agent, int action)
 533 {
 534   int address_index;
 535
 536   switch (action)
 537     {
 538     case RIL_ACTION_NOTHING:
 539       return GNUNET_YES;
 540       break;
 541
 542     case RIL_ACTION_BW_IN_INC:
 543     case RIL_ACTION_BW_IN_DBL:
 544       if (agent->bw_in >= RIL_MAX_BW)
 545         return GNUNET_NO;
 546       else
 547         return GNUNET_YES;
 548       break;
 549
 550     case RIL_ACTION_BW_IN_DEC:
 551     case RIL_ACTION_BW_IN_HLV:
 552       if (agent->bw_in <= 0)
 553         return GNUNET_NO;
 554       else
 555         return GNUNET_YES;
 556       break;
 557
 558     case RIL_ACTION_BW_OUT_INC:
 559     case RIL_ACTION_BW_OUT_DBL:
 560       if (agent->bw_out >= RIL_MAX_BW)
 561         return GNUNET_NO;
 562       else
 563         return GNUNET_YES;
 564       break;
 565
 566     case RIL_ACTION_BW_OUT_DEC:
 567     case RIL_ACTION_BW_OUT_HLV:
 568       if (agent->bw_out <= 0)
 569         return GNUNET_NO;
 570       else
 571         return GNUNET_YES;
 572       break;
 573
 574     default:
 575       if ((action >= RIL_ACTION_TYPE_NUM) && (action < agent->n)) //switch address action
 576         {
 577           address_index = action - RIL_ACTION_TYPE_NUM;
 578
 579           GNUNET_assert(address_index >= 0);
 580           GNUNET_assert(
 581             address_index <= agent_address_get_index(agent, agent->addresses_tail->address_naked));
 582
 583           if ((agent_address_get_index(agent, agent->address_inuse) == address_index) ||
 584               agent->address_inuse->active)
 585             return GNUNET_NO;
 586           else
 587             return GNUNET_YES;
 588           break;
 589         }
 590       // error - action does not exist
 591       GNUNET_assert(GNUNET_NO);
 592     }
 593 }
 594
 595
 596 /**
 597  * Gets the action, with the maximal estimated Q-value (i.e. the one currently estimated to bring the
 598  * most reward in the future)
 599  *
 600  * @param agent agent performing the calculation
 601  * @param state the state from which to take the action
 602  * @return the action promising most future reward
 603  */
 604 static int
 605 agent_get_action_max(struct RIL_Peer_Agent *agent, double *state)
 606 {
 607   int i;
 608   int max_i = RIL_ACTION_INVALID;
 609   double cur_q;
 610   double max_q = -DBL_MAX;
 611
 612   for (i = 0; i < agent->n; i++)
 613     {
 614       if (agent_action_is_possible(agent, i))
 615         {
 616           cur_q = agent_q(agent, state, i);
 617           if (cur_q > max_q)
 618             {
 619               max_q = cur_q;
 620               max_i = i;
 621             }
 622         }
 623     }
 624
 625   GNUNET_assert(RIL_ACTION_INVALID != max_i);
 626
 627   return max_i;
 628 }
 629
 630 /**
 631  * Chooses a random action from the set of possible ones
 632  *
 633  * @param agent the agent performing the action
 634  * @return the action index
 635  */
 636 static int
 637 agent_get_action_random(struct RIL_Peer_Agent *agent)
 638 {
 639   int i;
 640   int is_possible[agent->n];
 641   int sum = 0;
 642   int r;
 643
 644   for (i = 0; i < agent->n; i++)
 645     {
 646       if (agent_action_is_possible(agent, i))
 647         {
 648           is_possible[i] = GNUNET_YES;
 649           sum++;
 650         }
 651       else
 652         {
 653           is_possible[i] = GNUNET_NO;
 654         }
 655     }
 656
 657   r = GNUNET_CRYPTO_random_u32(GNUNET_CRYPTO_QUALITY_WEAK, sum);
 658
 659   sum = -1;
 660   for (i = 0; i < agent->n; i++)
 661     {
 662       if (is_possible[i])
 663         {
 664           sum++;
 665           if (sum == r)
 666             return i;
 667         }
 668     }
 669
 670   GNUNET_assert(GNUNET_NO);
 671   return RIL_ACTION_INVALID;
 672 }
 673
 674
 675 /**
 676  * Updates the weights (i.e. coefficients) of the weight vector in matrix W for action a
 677  *
 678  * @param agent the agent performing the update
 679  * @param reward the reward received for the last action
 680  * @param s_next the new state, the last step got the agent into
 681  * @param a_prime the new
 682  */
 683 static void
 684 agent_update(struct RIL_Peer_Agent *agent, double reward, double *s_next, int a_prime)
 685 {
 686   int i;
 687   int k;
 688   double delta;
 689   double **theta = agent->W;
 690
 691   delta = agent->envi->global_discount_integrated * reward; //reward
 692   delta += agent->envi->global_discount_variable * agent_q(agent, s_next, a_prime);  //discounted future value
 693   delta -= agent_q(agent, agent->s_old, agent->a_old);  //one step
 694
 695 //  LOG(GNUNET_ERROR_TYPE_INFO, "update()   Step# %llu  Q(s,a): %f  a: %f  r: %f  y: %f  Q(s+1,a+1) = %f  delta: %f\n",
 696 //      agent->step_count,
 697 //      agent_q (agent, agent->s_old, agent->a_old),
 698 //      agent->envi->parameters.alpha,
 699 //      reward,
 700 //      agent->envi->global_discount_variable,
 701 //      agent_q (agent, s_next, a_prime),
 702 //      delta);
 703
 704   for (k = 0; k < agent->n; k++)
 705     {
 706       for (i = 0; i < agent->m; i++)
 707         {
 708           //    LOG(GNUNET_ERROR_TYPE_INFO, "alpha = %f   delta = %f   e[%d] = %f\n",
 709           //        agent->envi->parameters.alpha,
 710           //        delta,
 711           //        i,
 712           //        agent->e[i]);
 713           theta[k][i] += agent->envi->parameters.alpha * delta * agent->E[k][i];
 714         }
 715     }
 716 }
 717
 718
 719 /**
 720  * Changes the eligibility trace vector e in various manners:
 721  * #RIL_E_ACCUMULATE - adds @a feature to each component as in accumulating eligibility traces
 722  * #RIL_E_REPLACE - resets each component to @a feature  as in replacing traces
 723  * #RIL_E_DECAY - multiplies e with discount factor and lambda as in the update rule
 724  * #RIL_E_ZERO - sets e to 0 as in Watkin's Q-learning algorithm when exploring and when initializing
 725  *
 726  * @param agent the agent handle
 727  * @param mod the kind of modification
 728  * @param feature the feature vector
 729  * @param action the action to take
 730  */
 731 static void
 732 agent_modify_eligibility(struct RIL_Peer_Agent *agent,
 733                          enum RIL_E_Modification mod,
 734                          double *feature,
 735                          int action)
 736 {
 737   int i;
 738   int k;
 739
 740   for (i = 0; i < agent->m; i++)
 741     {
 742       switch (mod)
 743         {
 744         case RIL_E_ACCUMULATE:
 745           agent->E[action][i] += feature[i];
 746           break;
 747
 748         case RIL_E_REPLACE:
 749           agent->E[action][i] = agent->E[action][i] > feature[i] ? agent->E[action][i] : feature[i];
 750           break;
 751
 752         case RIL_E_DECAY:
 753           for (k = 0; k < agent->n; k++)
 754             {
 755               agent->E[k][i] *= agent->envi->global_discount_variable * agent->envi->parameters.lambda;
 756             }
 757           break;
 758
 759         case RIL_E_ZERO:
 760           for (k = 0; k < agent->n; k++)
 761             {
 762               agent->E[k][i] = 0;
 763             }
 764           break;
 765         }
 766     }
 767 }
 768
 769 /**
 770  * Informs the environment about the status of the solver
 771  *
 772  * @param solver
 773  * @param op
 774  * @param stat
 775  */
 776 static void
 777 ril_inform(struct GAS_RIL_Handle *solver,
 778            enum GAS_Solver_Operation op,
 779            enum GAS_Solver_Status stat)
 780 {
 781   solver->env->info_cb(solver->env->cls,
 782                        op,
 783                        stat,
 784                        GAS_INFO_NONE);
 785 }
 786
 787 /**
 788  * Calculates the maximum bandwidth an agent can assign in a network scope
 789  *
 790  * @param net
 791  */
 792 static unsigned long long
 793 ril_get_max_bw(struct RIL_Scope *net)
 794 {
 795   return GNUNET_MIN(2 * GNUNET_MAX(net->bw_in_available, net->bw_out_available), GNUNET_ATS_MaxBandwidth);
 796 }
 797
 798 /**
 799  * Changes the active assignment suggestion of the handler and invokes the bw_changed callback to
 800  * notify ATS of its new decision
 801  *
 802  * @param solver solver handle
 803  * @param agent agent handle
 804  * @param new_address the address which is to be used
 805  * @param new_bw_in the new amount of inbound bandwidth set for this address
 806  * @param new_bw_out the new amount of outbound bandwidth set for this address
 807  * @param silent disables invocation of the bw_changed callback, if #GNUNET_YES
 808  */
 809 static void
 810 envi_set_active_suggestion(struct GAS_RIL_Handle *solver,
 811                            struct RIL_Peer_Agent *agent,
 812                            struct ATS_Address *new_address,
 813                            unsigned long long new_bw_in,
 814                            unsigned long long new_bw_out,
 815                            int silent)
 816 {
 817   int notify = GNUNET_NO;
 818
 819   LOG(GNUNET_ERROR_TYPE_DEBUG,
 820       "    set_active_suggestion() for peer '%s'\n",
 821       GNUNET_i2s(&agent->peer));
 822
 823   //address change
 824   if (agent->address_inuse != new_address)
 825     {
 826       if (NULL != agent->address_inuse)
 827         {
 828           agent->address_inuse->active = GNUNET_NO;
 829           agent->address_inuse->assigned_bw_in = 0;
 830           agent->address_inuse->assigned_bw_out = 0;
 831         }
 832       if (NULL != new_address)
 833         {
 834           LOG(GNUNET_ERROR_TYPE_DEBUG, "    set address active: %s\n", agent->is_active ? "yes" : "no");
 835           new_address->active = agent->is_active;
 836           new_address->assigned_bw_in = agent->bw_in;
 837           new_address->assigned_bw_out = agent->bw_out;
 838         }
 839       notify |= GNUNET_YES;
 840     }
 841
 842   if (new_address)
 843     {
 844       //activity change
 845       if (new_address->active != agent->is_active)
 846         {
 847           new_address->active = agent->is_active;
 848           notify |= GNUNET_YES;
 849         }
 850
 851       //bw change
 852       if (agent->bw_in != new_bw_in)
 853         {
 854           agent->bw_in = new_bw_in;
 855           new_address->assigned_bw_in = new_bw_in;
 856           notify |= GNUNET_YES;
 857         }
 858       if (agent->bw_out != new_bw_out)
 859         {
 860           agent->bw_out = new_bw_out;
 861           new_address->assigned_bw_out = new_bw_out;
 862           notify |= GNUNET_YES;
 863         }
 864     }
 865
 866   if (notify && agent->is_active && (GNUNET_NO == silent))
 867     {
 868       if (new_address)
 869         {
 870           LOG(GNUNET_ERROR_TYPE_DEBUG, "    envi_set_active_suggestion() notify\n");
 871           agent->suggestion_issue = GNUNET_YES;
 872           agent->suggestion_address = new_address;
 873         }
 874       else if (agent->address_inuse)
 875         {
 876           /* disconnect case, no new address */
 877           GNUNET_assert(0 == agent->address_inuse->assigned_bw_in);
 878           GNUNET_assert(0 == agent->address_inuse->assigned_bw_out);
 879           agent->bw_in = 0;
 880           agent->bw_out = 0;
 881
 882           agent->suggestion_issue = GNUNET_YES;
 883           agent->suggestion_address = agent->address_inuse;
 884         }
 885     }
 886   agent->address_inuse = new_address;
 887 }
 888
 889
 890 /**
 891  * Allocates a state vector and fills it with the features present
 892  * @param solver the solver handle
 893  * @param agent the agent handle
 894  * @return pointer to the state vector
 895  */
 896 static double *
 897 envi_get_state(struct GAS_RIL_Handle *solver, struct RIL_Peer_Agent *agent)
 898 {
 899   double *state;
 900   double y[2];
 901   double x[2];
 902   double d[2];
 903   double sigma;
 904   double f;
 905   int m;
 906   int i;
 907   int k;
 908   unsigned long long max_bw;
 909
 910   state = GNUNET_malloc(sizeof(double) * agent->m);
 911
 912   max_bw = ril_get_max_bw((struct RIL_Scope *)agent->address_inuse->solver_information);
 913
 914   y[0] = (double)agent->bw_out;
 915   y[1] = (double)agent->bw_in;
 916
 917   m = agent_address_get_index(agent, agent->address_inuse) * (solver->parameters.rbf_divisor + 1) * (solver->parameters.rbf_divisor + 1);
 918   for (i = 0; i <= solver->parameters.rbf_divisor; i++)
 919     {
 920       for (k = 0; k <= solver->parameters.rbf_divisor; k++)
 921         {
 922           x[0] = (double)i * (double)max_bw / (double)solver->parameters.rbf_divisor;
 923           x[1] = (double)k * (double)max_bw / (double)solver->parameters.rbf_divisor;
 924           d[0] = x[0] - y[0];
 925           d[1] = x[1] - y[1];
 926           sigma = (((double)max_bw / ((double)solver->parameters.rbf_divisor + 1)) * 0.5);
 927           f = exp(-((d[0] * d[0] + d[1] * d[1]) / (2 * sigma * sigma)));
 928           state[m++] = f;
 929         }
 930     }
 931
 932   return state;
 933 }
 934
 935
 936 /**
 937  * Returns the utility value of the connection an agent manages
 938  *
 939  * @param agent the agent in question
 940  * @return the utility value
 941  */
 942 static double
 943 agent_get_utility(struct RIL_Peer_Agent *agent)
 944 {
 945   const double *preferences;
 946   double delay_atsi;
 947   double delay_norm;
 948   double pref_match;
 949
 950   preferences = agent->envi->env->get_preferences(agent->envi->env->cls,
 951                                                   &agent->peer);
 952
 953   delay_atsi = agent->address_inuse->norm_delay.norm;
 954   delay_norm = RIL_UTILITY_DELAY_MAX * exp(-delay_atsi * 0.00001);
 955
 956   pref_match = preferences[GNUNET_ATS_PREFERENCE_LATENCY] * delay_norm;
 957   pref_match += preferences[GNUNET_ATS_PREFERENCE_BANDWIDTH] *
 958                 sqrt((double)(agent->bw_in / RIL_MIN_BW) * (double)(agent->bw_out / RIL_MIN_BW));
 959   return pref_match;
 960 }
 961
 962 /**
 963  * Calculates the social welfare within a network scope according to what social
 964  * welfare measure is set in the configuration.
 965  *
 966  * @param solver the solver handle
 967  * @param scope the network scope in question
 968  * @return the social welfare value
 969  */
 970 static double
 971 ril_network_get_social_welfare(struct GAS_RIL_Handle *solver, struct RIL_Scope *scope)
 972 {
 973   struct RIL_Peer_Agent *cur;
 974   double result;
 975
 976   switch (solver->parameters.social_welfare)
 977     {
 978     case RIL_WELFARE_EGALITARIAN:
 979       result = DBL_MAX;
 980       for (cur = solver->agents_head; NULL != cur; cur = cur->next)
 981         {
 982           if (cur->is_active && cur->address_inuse && (cur->address_inuse->solver_information == scope))
 983             {
 984               result = GNUNET_MIN(result, agent_get_utility(cur));
 985             }
 986         }
 987       return result;
 988
 989     case RIL_WELFARE_NASH:
 990       result = 0;
 991       for (cur = solver->agents_head; NULL != cur; cur = cur->next)
 992         {
 993           if (cur->is_active && cur->address_inuse && (cur->address_inuse->solver_information == scope))
 994             {
 995               result *= pow(agent_get_utility(cur), 1.0 / (double)scope->active_agent_count);
 996             }
 997         }
 998       return result;
 999     }
1000   GNUNET_assert(GNUNET_NO);
1001   return 1;
1002 }
1003
1004 static double
1005 envi_get_penalty(struct GAS_RIL_Handle *solver, struct RIL_Peer_Agent *agent)
1006 {
1007   struct RIL_Scope *net;
1008   unsigned long long over_max;
1009   unsigned long long over_in = 0;
1010   unsigned long long over_out = 0;
1011
1012   net = agent->address_inuse->solver_information;
1013
1014   if (net->bw_in_utilized > net->bw_in_available)
1015     {
1016       over_in = net->bw_in_utilized - net->bw_in_available;
1017       if (RIL_ACTION_BW_IN_INC == agent->a_old)
1018         {
1019           /* increase quadratically */
1020           over_in *= over_in;
1021         }
1022     }
1023   if (net->bw_out_utilized > net->bw_out_available)
1024     {
1025       over_out = net->bw_out_utilized - net->bw_out_available;
1026       if (RIL_ACTION_BW_OUT_INC == agent->a_old)
1027         {
1028           /* increase quadratically */
1029           over_out *= over_out;
1030         }
1031     }
1032   over_max = (over_in + over_out) / (RIL_MIN_BW * RIL_MIN_BW);
1033
1034   return -1.0 * (double)over_max;
1035 }
1036
1037 /**
1038  * Gets the reward for the last performed step, which is calculated in equal
1039  * parts from the local (the peer specific) and the global (for all peers
1040  * identical) reward.
1041  *
1042  * @param solver the solver handle
1043  * @param agent the agent handle
1044  * @return the reward
1045  */
1046 static double
1047 envi_get_reward(struct GAS_RIL_Handle *solver, struct RIL_Peer_Agent *agent)
1048 {
1049   struct RIL_Scope *net;
1050   double objective;
1051   double delta;
1052   double steady;
1053   double penalty;
1054   double reward;
1055
1056   net = agent->address_inuse->solver_information;
1057
1058   penalty = envi_get_penalty(solver, agent);
1059   objective = (agent_get_utility(agent) + net->social_welfare) / 2;
1060   delta = objective - agent->objective_old;
1061   agent->objective_old = objective;
1062
1063   if (delta != 0 && penalty == 0)
1064     {
1065       agent->nop_bonus = delta * RIL_NOP_DECAY;
1066     }
1067   else
1068     {
1069       agent->nop_bonus *= RIL_NOP_DECAY;
1070     }
1071
1072   steady = (RIL_ACTION_NOTHING == agent->a_old) ? agent->nop_bonus : 0;
1073
1074   reward = delta + steady;
1075   return reward + penalty;
1076 }
1077
1078 /**
1079  * Doubles the bandwidth for the active address
1080  *
1081  * @param solver solver handle
1082  * @param agent agent handle
1083  * @param direction_in if GNUNET_YES, change inbound bandwidth, otherwise the outbound bandwidth
1084  */
1085 static void
1086 envi_action_bw_double(struct GAS_RIL_Handle *solver,
1087                       struct RIL_Peer_Agent *agent,
1088                       int direction_in)
1089 {
1090   unsigned long long new_bw;
1091   unsigned long long max_bw;
1092
1093   max_bw = ril_get_max_bw((struct RIL_Scope *)agent->address_inuse->solver_information);
1094
1095   if (direction_in)
1096     {
1097       new_bw = agent->bw_in * 2;
1098       if (new_bw < agent->bw_in || new_bw > max_bw)
1099         new_bw = max_bw;
1100       envi_set_active_suggestion(solver, agent, agent->address_inuse, new_bw,
1101                                  agent->bw_out, GNUNET_NO);
1102     }
1103   else
1104     {
1105       new_bw = agent->bw_out * 2;
1106       if (new_bw < agent->bw_out || new_bw > max_bw)
1107         new_bw = max_bw;
1108       envi_set_active_suggestion(solver, agent, agent->address_inuse, agent->bw_in,
1109                                  new_bw, GNUNET_NO);
1110     }
1111 }
1112
1113 /**
1114  * Cuts the bandwidth for the active address in half. The least amount of bandwidth suggested, is
1115  * the minimum bandwidth for a peer, in order to not invoke a disconnect.
1116  *
1117  * @param solver solver handle
1118  * @param agent agent handle
1119  * @param direction_in if GNUNET_YES, change inbound bandwidth, otherwise change the outbound
1120  * bandwidth
1121  */
1122 static void
1123 envi_action_bw_halven(struct GAS_RIL_Handle *solver,
1124                       struct RIL_Peer_Agent *agent,
1125                       int direction_in)
1126 {
1127   unsigned long long new_bw;
1128
1129   if (direction_in)
1130     {
1131       new_bw = agent->bw_in / 2;
1132       if (new_bw <= 0 || new_bw > agent->bw_in)
1133         new_bw = 0;
1134       envi_set_active_suggestion(solver, agent, agent->address_inuse, new_bw, agent->bw_out,
1135                                  GNUNET_NO);
1136     }
1137   else
1138     {
1139       new_bw = agent->bw_out / 2;
1140       if (new_bw <= 0 || new_bw > agent->bw_out)
1141         new_bw = 0;
1142       envi_set_active_suggestion(solver, agent, agent->address_inuse, agent->bw_in, new_bw,
1143                                  GNUNET_NO);
1144     }
1145 }
1146
1147 /**
1148  * Increases the bandwidth by 5 times the minimum bandwidth for the active address.
1149  *
1150  * @param solver solver handle
1151  * @param agent agent handle
1152  * @param direction_in if GNUNET_YES, change inbound bandwidth, otherwise change the outbound
1153  * bandwidth
1154  */
1155 static void
1156 envi_action_bw_inc(struct GAS_RIL_Handle *solver, struct RIL_Peer_Agent *agent, int direction_in)
1157 {
1158   unsigned long long new_bw;
1159   unsigned long long max_bw;
1160
1161   max_bw = ril_get_max_bw((struct RIL_Scope *)agent->address_inuse->solver_information);
1162
1163   if (direction_in)
1164     {
1165       new_bw = agent->bw_in + (RIL_INC_DEC_STEP_SIZE * RIL_MIN_BW);
1166       if (new_bw < agent->bw_in || new_bw > max_bw)
1167         new_bw = max_bw;
1168       envi_set_active_suggestion(solver, agent, agent->address_inuse, new_bw,
1169                                  agent->bw_out, GNUNET_NO);
1170     }
1171   else
1172     {
1173       new_bw = agent->bw_out + (RIL_INC_DEC_STEP_SIZE * RIL_MIN_BW);
1174       if (new_bw < agent->bw_out || new_bw > max_bw)
1175         new_bw = max_bw;
1176       envi_set_active_suggestion(solver, agent, agent->address_inuse, agent->bw_in,
1177                                  new_bw, GNUNET_NO);
1178     }
1179 }
1180
1181 /**
1182  * Decreases the bandwidth by 5 times the minimum bandwidth for the active address. The least amount
1183  * of bandwidth suggested, is the minimum bandwidth for a peer, in order to not invoke a disconnect.
1184  *
1185  * @param solver solver handle
1186  * @param agent agent handle
1187  * @param direction_in if GNUNET_YES, change inbound bandwidth, otherwise change the outbound
1188  * bandwidth
1189  */
1190 static void
1191 envi_action_bw_dec(struct GAS_RIL_Handle *solver, struct RIL_Peer_Agent *agent, int direction_in)
1192 {
1193   unsigned long long new_bw;
1194
1195   if (direction_in)
1196     {
1197       new_bw = agent->bw_in - (RIL_INC_DEC_STEP_SIZE * RIL_MIN_BW);
1198       if (new_bw <= 0 || new_bw > agent->bw_in)
1199         new_bw = 0;
1200       envi_set_active_suggestion(solver, agent, agent->address_inuse, new_bw, agent->bw_out,
1201                                  GNUNET_NO);
1202     }
1203   else
1204     {
1205       new_bw = agent->bw_out - (RIL_INC_DEC_STEP_SIZE * RIL_MIN_BW);
1206       if (new_bw <= 0 || new_bw > agent->bw_out)
1207         new_bw = 0;
1208       envi_set_active_suggestion(solver, agent, agent->address_inuse, agent->bw_in, new_bw,
1209                                  GNUNET_NO);
1210     }
1211 }
1212
1213 /**
1214  * Switches to the address given by its index
1215  *
1216  * @param solver solver handle
1217  * @param agent agent handle
1218  * @param address_index index of the address as it is saved in the agent's list, starting with zero
1219  */
1220 static void
1221 envi_action_address_switch(struct GAS_RIL_Handle *solver,
1222                            struct RIL_Peer_Agent *agent,
1223                            unsigned int address_index)
1224 {
1225   struct RIL_Address_Wrapped *cur;
1226   int i = 0;
1227
1228   //cur = agent_address_get_wrapped(agent, agent->address_inuse);
1229
1230   for (cur = agent->addresses_head; NULL != cur; cur = cur->next)
1231     {
1232       if (i == address_index)
1233         {
1234           envi_set_active_suggestion(solver, agent, cur->address_naked, agent->bw_in, agent->bw_out,
1235                                      GNUNET_NO);
1236           return;
1237         }
1238
1239       i++;
1240     }
1241
1242   //no address with address_index exists, in this case this action should not be callable
1243   GNUNET_assert(GNUNET_NO);
1244 }
1245
1246 /**
1247  * Puts the action into effect by calling the according function
1248  *
1249  * @param solver the solver handle
1250  * @param agent the action handle
1251  * @param action the action to perform by the solver
1252  */
1253 static void
1254 envi_do_action(struct GAS_RIL_Handle *solver, struct RIL_Peer_Agent *agent, int action)
1255 {
1256   int address_index;
1257
1258   switch (action)
1259     {
1260     case RIL_ACTION_NOTHING:
1261       break;
1262
1263     case RIL_ACTION_BW_IN_DBL:
1264       envi_action_bw_double(solver, agent, GNUNET_YES);
1265       break;
1266
1267     case RIL_ACTION_BW_IN_HLV:
1268       envi_action_bw_halven(solver, agent, GNUNET_YES);
1269       break;
1270
1271     case RIL_ACTION_BW_IN_INC:
1272       envi_action_bw_inc(solver, agent, GNUNET_YES);
1273       break;
1274
1275     case RIL_ACTION_BW_IN_DEC:
1276       envi_action_bw_dec(solver, agent, GNUNET_YES);
1277       break;
1278
1279     case RIL_ACTION_BW_OUT_DBL:
1280       envi_action_bw_double(solver, agent, GNUNET_NO);
1281       break;
1282
1283     case RIL_ACTION_BW_OUT_HLV:
1284       envi_action_bw_halven(solver, agent, GNUNET_NO);
1285       break;
1286
1287     case RIL_ACTION_BW_OUT_INC:
1288       envi_action_bw_inc(solver, agent, GNUNET_NO);
1289       break;
1290
1291     case RIL_ACTION_BW_OUT_DEC:
1292       envi_action_bw_dec(solver, agent, GNUNET_NO);
1293       break;
1294
1295     default:
1296       if ((action >= RIL_ACTION_TYPE_NUM) && (action < agent->n)) //switch address action
1297         {
1298           address_index = action - RIL_ACTION_TYPE_NUM;
1299
1300           GNUNET_assert(address_index >= 0);
1301           GNUNET_assert(
1302             address_index <= agent_address_get_index(agent, agent->addresses_tail->address_naked));
1303
1304           envi_action_address_switch(solver, agent, address_index);
1305           break;
1306         }
1307       // error - action does not exist
1308       GNUNET_assert(GNUNET_NO);
1309     }
1310 }
1311
1312 /**
1313  * Selects the next action using the e-greedy strategy. I.e. with a probability
1314  * of (1-e) the action with the maximum expected return will be chosen
1315  * (=> exploitation) and with probability (e) a random action will be chosen.
1316  * In case the Q-learning rule is set, the function also resets the eligibility
1317  * traces in the exploration case (after Watkin's Q-learning).
1318  *
1319  * @param agent the agent selecting an action
1320  * @param state the current state-feature vector
1321  * @return the action index
1322  */
1323 static int
1324 agent_select_egreedy(struct RIL_Peer_Agent *agent, double *state)
1325 {
1326   int action;
1327   double r = (double)GNUNET_CRYPTO_random_u32(GNUNET_CRYPTO_QUALITY_WEAK,
1328                                               UINT32_MAX) / (double)UINT32_MAX;
1329
1330   if (r < agent->envi->parameters.epsilon) //explore
1331     {
1332       action = agent_get_action_random(agent);
1333       if (RIL_ALGO_Q == agent->envi->parameters.algorithm)
1334         {
1335           agent->eligibility_reset = GNUNET_YES;
1336         }
1337       agent->envi->parameters.epsilon *= agent->envi->parameters.epsilon_decay;
1338       return action;
1339     }
1340   else //exploit
1341     {
1342       action = agent_get_action_max(agent, state);
1343       return action;
1344     }
1345 }
1346
1347 /**
1348  * Selects the next action with a probability corresponding to its value. The
1349  * probability is calculated using a Boltzmann distribution with a temperature
1350  * value. The higher the temperature, the more are the action selection
1351  * probabilities the same. With a temperature of 0, the selection is greedy,
1352  * i.e. always the action with the highest value is chosen.
1353  * @param agent
1354  * @param state
1355  * @return
1356  */
1357 static int
1358 agent_select_softmax(struct RIL_Peer_Agent *agent, double *state)
1359 {
1360   int i;
1361   int a_max;
1362   double eqt[agent->n];
1363   double p[agent->n];
1364   double sum = 0;
1365   double r;
1366
1367   a_max = agent_get_action_max(agent, state);
1368
1369   for (i = 0; i < agent->n; i++)
1370     {
1371       if (agent_action_is_possible(agent, i))
1372         {
1373           eqt[i] = exp(agent_q(agent, state, i) / agent->envi->parameters.temperature);
1374           if (isinf(eqt[i]))
1375             eqt[i] = isinf(eqt[i]) * UINT32_MAX;
1376           sum += eqt[i];
1377         }
1378     }
1379   for (i = 0; i < agent->n; i++)
1380     {
1381       if (agent_action_is_possible(agent, i))
1382         {
1383           p[i] = eqt[i] / sum;
1384         }
1385       else
1386         {
1387           p[i] = 0;
1388         }
1389     }
1390   r = (double)GNUNET_CRYPTO_random_u32(GNUNET_CRYPTO_QUALITY_WEAK,
1391                                        UINT32_MAX) / (double)UINT32_MAX;
1392   sum = 0;
1393   for (i = 0; i < agent->n; i++)
1394     {
1395       if (sum + p[i] > r)
1396         {
1397           if (i != a_max)
1398             {
1399               if (RIL_ALGO_Q == agent->envi->parameters.algorithm)
1400                 agent->eligibility_reset = GNUNET_YES;
1401               agent->envi->parameters.temperature *= agent->envi->parameters.temperature_decay;
1402             }
1403           return i;
1404         }
1405       sum += p[i];
1406     }
1407   GNUNET_assert(GNUNET_NO);
1408   return RIL_ACTION_INVALID;
1409 }
1410
1411 /**
1412  * Select the next action of an agent either according to the e-greedy strategy
1413  * or the softmax strategy.
1414  *
1415  * @param agent the agent in question
1416  * @param state the current state-feature vector
1417  * @return the action index
1418  */
1419 static int
1420 agent_select_action(struct RIL_Peer_Agent *agent, double *state)
1421 {
1422   if (agent->envi->parameters.select == RIL_SELECT_EGREEDY)
1423     {
1424       return agent_select_egreedy(agent, state);
1425     }
1426   else
1427     {
1428       return agent_select_softmax(agent, state);
1429     }
1430 }
1431
1432 /**
1433  * Performs one step of the Markov Decision Process. Other than in the literature the step starts
1434  * after having done the last action a_old. It observes the new state s_next and the reward
1435  * received. Then the coefficient update is done according to the SARSA or Q-learning method. The
1436  * next action is put into effect.
1437  *
1438  * @param agent the agent performing the step
1439  */
1440 static void
1441 agent_step(struct RIL_Peer_Agent *agent)
1442 {
1443   int a_next = RIL_ACTION_INVALID;
1444   int a_max;
1445   double *s_next;
1446   double reward;
1447
1448   LOG(GNUNET_ERROR_TYPE_DEBUG, "    agent_step() Peer '%s', algorithm %s\n",
1449       GNUNET_i2s(&agent->peer),
1450       agent->envi->parameters.algorithm ? "Q" : "SARSA");
1451
1452   s_next = envi_get_state(agent->envi, agent);
1453   reward = envi_get_reward(agent->envi, agent);
1454
1455   if (agent->eligibility_reset)
1456     {
1457       agent_modify_eligibility(agent, RIL_E_ZERO, NULL, -1);
1458       agent->eligibility_reset = GNUNET_NO;
1459     }
1460   else
1461     {
1462       agent_modify_eligibility(agent, RIL_E_DECAY, NULL, -1);
1463     }
1464   if (RIL_ACTION_INVALID != agent->a_old)
1465     {
1466       agent_modify_eligibility(agent, agent->envi->parameters.eligibility_trace_mode, agent->s_old, agent->a_old);
1467     }
1468
1469   switch (agent->envi->parameters.algorithm)
1470     {
1471     case RIL_ALGO_SARSA:
1472       a_next = agent_select_action(agent, s_next);
1473       if (RIL_ACTION_INVALID != agent->a_old)
1474         {
1475           //updates weights with selected action (on-policy), if not first step
1476           agent_update(agent, reward, s_next, a_next);
1477         }
1478       break;
1479
1480     case RIL_ALGO_Q:
1481       a_max = agent_get_action_max(agent, s_next);
1482       if (RIL_ACTION_INVALID != agent->a_old)
1483         {
1484           //updates weights with best action, disregarding actually selected action (off-policy), if not first step
1485           agent_update(agent, reward, s_next, a_max);
1486         }
1487       a_next = agent_select_action(agent, s_next);
1488       break;
1489     }
1490
1491   GNUNET_assert(RIL_ACTION_INVALID != a_next);
1492
1493   LOG(GNUNET_ERROR_TYPE_DEBUG, "step()  Step# %llu  R: %f  IN %llu  OUT %llu  A: %d\n",
1494       agent->step_count,
1495       reward,
1496       agent->bw_in / 1024,
1497       agent->bw_out / 1024,
1498       a_next);
1499
1500   envi_do_action(agent->envi, agent, a_next);
1501
1502   GNUNET_free(agent->s_old);
1503   agent->s_old = s_next;
1504   agent->a_old = a_next;
1505
1506   agent->step_count += 1;
1507 }
1508
1509 /**
1510  * Prototype of the ril_step() procedure
1511  *
1512  * @param solver the solver handle
1513  */
1514 static void
1515 ril_step(struct GAS_RIL_Handle *solver);
1516
1517
1518 /**
1519  * Task for the scheduler, which performs one step and lets the solver know that
1520  * no further step is scheduled.
1521  *
1522  * @param cls the solver handle
1523  */
1524 static void
1525 ril_step_scheduler_task(void *cls)
1526 {
1527   struct GAS_RIL_Handle *solver = cls;
1528
1529   solver->step_next_task_id = NULL;
1530   ril_step(solver);
1531 }
1532
1533 /**
1534  * Determines how much of the available bandwidth is assigned. If more is
1535  * assigned than available it returns 1. The function is used to determine the
1536  * step size of the adaptive stepping.
1537  *
1538  * @param solver the solver handle
1539  * @return the ratio
1540  */
1541 static double
1542 ril_get_used_resource_ratio(struct GAS_RIL_Handle *solver)
1543 {
1544   int i;
1545   struct RIL_Scope net;
1546   unsigned long long sum_assigned = 0;
1547   unsigned long long sum_available = 0;
1548   double ratio;
1549
1550   for (i = 0; i < solver->networks_count; i++)
1551     {
1552       net = solver->network_entries[i];
1553       if (net.bw_in_assigned > 0) //only consider scopes where an address is actually active
1554         {
1555           sum_assigned += net.bw_in_utilized;
1556           sum_assigned += net.bw_out_utilized;
1557           sum_available += net.bw_in_available;
1558           sum_available += net.bw_out_available;
1559         }
1560     }
1561   if (sum_available > 0)
1562     {
1563       ratio = ((double)sum_assigned) / ((double)sum_available);
1564     }
1565   else
1566     {
1567       ratio = 0;
1568     }
1569
1570   return ratio > 1 ? 1 : ratio; //overutilization is possible, cap at 1
1571 }
1572
1573 /**
1574  * Lookup network struct by type
1575  *
1576  * @param s the solver handle
1577  * @param type the network type
1578  * @return the network struct
1579  */
1580 static struct RIL_Scope *
1581 ril_get_network(struct GAS_RIL_Handle *s, uint32_t type)
1582 {
1583   int i;
1584
1585   for (i = 0; i < s->networks_count; i++)
1586     {
1587       if (s->network_entries[i].type == type)
1588         {
1589           return &s->network_entries[i];
1590         }
1591     }
1592   return NULL;
1593 }
1594
1595 /**
1596  * Determines whether more connections are allocated in a network scope, than
1597  * they would theoretically fit. This is used as a heuristic to determine,
1598  * whether a new connection can be allocated or not.
1599  *
1600  * @param solver the solver handle
1601  * @param network the network scope in question
1602  * @return GNUNET_YES if there are theoretically enough resources left
1603  */
1604 static int
1605 ril_network_is_not_full(struct GAS_RIL_Handle *solver, enum GNUNET_NetworkType network)
1606 {
1607   struct RIL_Scope *net;
1608   struct RIL_Peer_Agent *agent;
1609   unsigned long long address_count = 0;
1610
1611   for (agent = solver->agents_head; NULL != agent; agent = agent->next)
1612     {
1613       if (agent->address_inuse && agent->is_active)
1614         {
1615           net = agent->address_inuse->solver_information;
1616           if (net->type == network)
1617             {
1618               address_count++;
1619             }
1620         }
1621     }
1622
1623   net = ril_get_network(solver, network);
1624   return (net->bw_in_available > RIL_MIN_BW * address_count) && (net->bw_out_available > RIL_MIN_BW * address_count);
1625 }
1626
1627 /**
1628  * Unblocks an agent for which a connection request is there, that could not
1629  * be satisfied. Iterates over the addresses of the agent, if one of its
1630  * addresses can now be allocated in its scope the agent is unblocked,
1631  * otherwise it remains unchanged.
1632  *
1633  * @param solver the solver handle
1634  * @param agent the agent in question
1635  * @param silent
1636  */
1637 static void
1638 ril_try_unblock_agent(struct GAS_RIL_Handle *solver, struct RIL_Peer_Agent *agent, int silent)
1639 {
1640   struct RIL_Address_Wrapped *addr_wrap;
1641   struct RIL_Scope *net;
1642   unsigned long long start_in;
1643   unsigned long long start_out;
1644
1645   for (addr_wrap = agent->addresses_head; NULL != addr_wrap; addr_wrap = addr_wrap->next)
1646     {
1647       net = addr_wrap->address_naked->solver_information;
1648       if (ril_network_is_not_full(solver, net->type))
1649         {
1650           if (NULL == agent->address_inuse)
1651             {
1652               start_in = net->bw_in_available < net->bw_in_utilized ? (net->bw_in_available - net->bw_in_utilized) / 2 : RIL_MIN_BW;
1653               start_out = net->bw_out_available < net->bw_out_utilized ? (net->bw_out_available - net->bw_out_utilized) / 2 : RIL_MIN_BW;
1654               envi_set_active_suggestion(solver, agent, addr_wrap->address_naked, start_in, start_out, silent);
1655             }
1656           return;
1657         }
1658     }
1659   agent->address_inuse = NULL;
1660 }
1661
1662 /**
1663  * Determines how much the reward needs to be discounted depending on the amount
1664  * of time, which has passed since the last time-step.
1665  *
1666  * @param solver the solver handle
1667  */
1668 static void
1669 ril_calculate_discount(struct GAS_RIL_Handle *solver)
1670 {
1671   struct GNUNET_TIME_Absolute time_now;
1672   struct GNUNET_TIME_Relative time_delta;
1673   double tau;
1674
1675   // MDP case only for debugging purposes
1676   if (solver->simulate)
1677     {
1678       solver->global_discount_variable = solver->parameters.gamma;
1679       solver->global_discount_integrated = 1;
1680       return;
1681     }
1682
1683   // semi-MDP case
1684
1685   //calculate tau, i.e. how many real valued time units have passed, one time unit is one minimum time step
1686   time_now = GNUNET_TIME_absolute_get();
1687   time_delta = GNUNET_TIME_absolute_get_difference(solver->step_time_last, time_now);
1688   solver->step_time_last = time_now;
1689   tau = (double)time_delta.rel_value_us
1690         / (double)solver->parameters.step_time_min.rel_value_us;
1691
1692   //calculate reward discounts (once per step for all agents)
1693   solver->global_discount_variable = pow(M_E, ((-1.0) * ((double)solver->parameters.beta) * tau));
1694   solver->global_discount_integrated = (1.0 - solver->global_discount_variable)
1695                                        / (double)solver->parameters.beta;
1696 }
1697
1698 /**
1699  * Count the number of active agents/connections in a network scope
1700  *
1701  * @param solver the solver handle
1702  * @param scope the network scope in question
1703  * @return the number of allocated connections
1704  */
1705 static int
1706 ril_network_count_active_agents(struct GAS_RIL_Handle *solver, struct RIL_Scope *scope)
1707 {
1708   int c = 0;
1709   struct RIL_Peer_Agent *cur_agent;
1710
1711   for (cur_agent = solver->agents_head; NULL != cur_agent; cur_agent = cur_agent->next)
1712     {
1713       if (cur_agent->is_active && cur_agent->address_inuse && (cur_agent->address_inuse->solver_information == scope))
1714         {
1715           c++;
1716         }
1717     }
1718   return c;
1719 }
1720
1721 /**
1722  * Calculates how much bandwidth is assigned in sum in a network scope, either
1723  * in the inbound or in the outbound direction.
1724  *
1725  * @param solver the solver handle
1726  * @param type the type of the network scope in question
1727  * @param direction_in GNUNET_YES if the inbound direction should be summed up,
1728  *   otherwise the outbound direction will be summed up
1729  * @return the sum of the assigned bandwidths
1730  */
1731 static unsigned long long
1732 ril_network_get_assigned(struct GAS_RIL_Handle *solver, enum GNUNET_NetworkType type, int direction_in)
1733 {
1734   struct RIL_Peer_Agent *cur;
1735   struct RIL_Scope *net;
1736   unsigned long long sum = 0;
1737
1738   for (cur = solver->agents_head; NULL != cur; cur = cur->next)
1739     {
1740       if (cur->is_active && cur->address_inuse)
1741         {
1742           net = cur->address_inuse->solver_information;
1743           if (net->type == type)
1744             {
1745               if (direction_in)
1746                 sum += cur->bw_in;
1747               else
1748                 sum += cur->bw_out;
1749             }
1750         }
1751     }
1752
1753   return sum;
1754 }
1755
1756 /**
1757  * Calculates how much bandwidth is actually utilized in sum in a network scope,
1758  * either in the inbound or in the outbound direction.
1759  *
1760  * @param solver the solver handle
1761  * @param type the type of the network scope in question
1762  * @param direction_in GNUNET_YES if the inbound direction should be summed up,
1763  *   otherwise the outbound direction will be summed up
1764  * @return the sum of the utilized bandwidths (in bytes/second)
1765  */
1766 static unsigned long long
1767 ril_network_get_utilized(struct GAS_RIL_Handle *solver, enum GNUNET_NetworkType type, int direction_in)
1768 {
1769   struct RIL_Peer_Agent *cur;
1770   struct RIL_Scope *net;
1771   unsigned long long sum = 0;
1772
1773   for (cur = solver->agents_head; NULL != cur; cur = cur->next)
1774     {
1775       if (cur->is_active && cur->address_inuse)
1776         {
1777           net = cur->address_inuse->solver_information;
1778           if (net->type == type)
1779             {
1780               if (direction_in)
1781                 sum += cur->address_inuse->norm_utilization_in.norm;
1782               else
1783                 sum += cur->address_inuse->norm_utilization_out.norm;
1784             }
1785         }
1786     }
1787
1788   return sum;
1789 }
1790
1791 /**
1792  * Retrieves the state of the network scope, so that its attributes are up-to-
1793  * date.
1794  *
1795  * @param solver the solver handle
1796  */
1797 static void
1798 ril_networks_update_state(struct GAS_RIL_Handle *solver)
1799 {
1800   int c;
1801   struct RIL_Scope *net;
1802
1803   for (c = 0; c < solver->networks_count; c++)
1804     {
1805       net = &solver->network_entries[c];
1806       net->bw_in_assigned = ril_network_get_assigned(solver, net->type, GNUNET_YES);
1807       net->bw_in_utilized = ril_network_get_utilized(solver, net->type, GNUNET_YES);
1808       net->bw_out_assigned = ril_network_get_assigned(solver, net->type, GNUNET_NO);
1809       net->bw_out_utilized = ril_network_get_utilized(solver, net->type, GNUNET_NO);
1810       net->active_agent_count = ril_network_count_active_agents(solver, net);
1811       net->social_welfare = ril_network_get_social_welfare(solver, net);
1812     }
1813 }
1814
1815 /**
1816  * Schedules the next global step in an adaptive way. The more resources are
1817  * left, the earlier the next step is scheduled. This serves the reactivity of
1818  * the solver to changed inputs.
1819  *
1820  * @param solver the solver handle
1821  */
1822 static void
1823 ril_step_schedule_next(struct GAS_RIL_Handle *solver)
1824 {
1825   double used_ratio;
1826   double factor;
1827   double y;
1828   double offset;
1829   struct GNUNET_TIME_Relative time_next;
1830
1831   used_ratio = ril_get_used_resource_ratio(solver);
1832
1833   GNUNET_assert(
1834     solver->parameters.step_time_min.rel_value_us
1835     <= solver->parameters.step_time_max.rel_value_us);
1836
1837   factor = (double)GNUNET_TIME_relative_subtract(solver->parameters.step_time_max,
1838                                                  solver->parameters.step_time_min).rel_value_us;
1839   offset = (double)solver->parameters.step_time_min.rel_value_us;
1840   y = factor * pow(used_ratio, RIL_INTERVAL_EXPONENT) + offset;
1841
1842   GNUNET_assert(y <= (double)solver->parameters.step_time_max.rel_value_us);
1843   GNUNET_assert(y >= (double)solver->parameters.step_time_min.rel_value_us);
1844
1845   time_next = GNUNET_TIME_relative_saturating_multiply(GNUNET_TIME_UNIT_MICROSECONDS, (unsigned long long)y);
1846
1847 //  LOG (GNUNET_ERROR_TYPE_INFO, "ratio: %f, factor: %f, offset: %f, y: %f\n",
1848 //      used_ratio,
1849 //      factor,
1850 //      offset,
1851 //      y);
1852
1853   if (solver->simulate)
1854     {
1855       time_next = GNUNET_TIME_UNIT_ZERO;
1856     }
1857
1858   if ((NULL == solver->step_next_task_id) && (GNUNET_NO == solver->done))
1859     {
1860       solver->step_next_task_id = GNUNET_SCHEDULER_add_delayed(time_next, &ril_step_scheduler_task,
1861                                                                solver);
1862     }
1863 }
1864
1865 /**
1866  * Triggers one step per agent
1867  *
1868  * @param solver
1869  */
1870 static void
1871 ril_step(struct GAS_RIL_Handle *solver)
1872 {
1873   struct RIL_Peer_Agent *cur;
1874
1875   if (GNUNET_YES == solver->bulk_lock)
1876     {
1877       solver->bulk_changes++;
1878       return;
1879     }
1880
1881   ril_inform(solver, GAS_OP_SOLVE_START, GAS_STAT_SUCCESS);
1882
1883   LOG(GNUNET_ERROR_TYPE_DEBUG, "    RIL step number %d\n", solver->step_count);
1884
1885   if (0 == solver->step_count)
1886     {
1887       solver->step_time_last = GNUNET_TIME_absolute_get();
1888     }
1889
1890   ril_calculate_discount(solver);
1891   ril_networks_update_state(solver);
1892
1893   //trigger one step per active, unblocked agent
1894   for (cur = solver->agents_head; NULL != cur; cur = cur->next)
1895     {
1896       if (cur->is_active)
1897         {
1898           if (NULL == cur->address_inuse)
1899             {
1900               ril_try_unblock_agent(solver, cur, GNUNET_NO);
1901             }
1902           if (cur->address_inuse)
1903             {
1904               agent_step(cur);
1905             }
1906         }
1907     }
1908
1909   ril_networks_update_state(solver);
1910
1911   solver->step_count++;
1912   ril_step_schedule_next(solver);
1913
1914   ril_inform(solver, GAS_OP_SOLVE_STOP, GAS_STAT_SUCCESS);
1915
1916   ril_inform(solver, GAS_OP_SOLVE_UPDATE_NOTIFICATION_START, GAS_STAT_SUCCESS);
1917   for (cur = solver->agents_head; NULL != cur; cur = cur->next)
1918     {
1919       if (cur->suggestion_issue)
1920         {
1921           solver->env->bandwidth_changed_cb(solver->env->cls,
1922                                             cur->suggestion_address);
1923           cur->suggestion_issue = GNUNET_NO;
1924         }
1925     }
1926   ril_inform(solver, GAS_OP_SOLVE_UPDATE_NOTIFICATION_STOP, GAS_STAT_SUCCESS);
1927 }
1928
1929 /**
1930  * Initializes the matrix W of parameter vectors theta with small random numbers.
1931  *
1932  * @param agent The respective agent
1933  */
1934 static void
1935 agent_w_init(struct RIL_Peer_Agent *agent)
1936 {
1937   int i;
1938   int k;
1939
1940   for (i = 0; i < agent->n; i++)
1941     {
1942       for (k = 0; k < agent->m; k++)
1943         {
1944           agent->W[i][k] = agent->envi->parameters.alpha * (1.0 - 2.0 * ((double)GNUNET_CRYPTO_random_u32(GNUNET_CRYPTO_QUALITY_WEAK, UINT32_MAX) / (double)UINT32_MAX));
1945         }
1946     }
1947 }
1948
1949 /**
1950  * Initialize an agent without addresses and its knowledge base
1951  *
1952  * @param s ril solver
1953  * @param peer the one in question
1954  * @return handle to the new agent
1955  */
1956 static struct RIL_Peer_Agent *
1957 agent_init(void *s, const struct GNUNET_PeerIdentity *peer)
1958 {
1959   int i;
1960   struct GAS_RIL_Handle * solver = s;
1961   struct RIL_Peer_Agent * agent = GNUNET_new(struct RIL_Peer_Agent);
1962
1963   agent->envi = solver;
1964   agent->peer = *peer;
1965   agent->step_count = 0;
1966   agent->is_active = GNUNET_NO;
1967   agent->bw_in = RIL_MIN_BW;
1968   agent->bw_out = RIL_MIN_BW;
1969   agent->suggestion_issue = GNUNET_NO;
1970   agent->n = RIL_ACTION_TYPE_NUM;
1971   agent->m = 0;
1972   agent->W = (double **)GNUNET_malloc(sizeof(double *) * agent->n);
1973   agent->E = (double **)GNUNET_malloc(sizeof(double *) * agent->n);
1974   for (i = 0; i < agent->n; i++)
1975     {
1976       agent->W[i] = (double *)GNUNET_malloc(sizeof(double) * agent->m);
1977       agent->E[i] = (double *)GNUNET_malloc(sizeof(double) * agent->m);
1978     }
1979   agent_w_init(agent);
1980   agent->eligibility_reset = GNUNET_NO;
1981   agent->a_old = RIL_ACTION_INVALID;
1982   agent->s_old = GNUNET_malloc(sizeof(double) * agent->m);
1983   agent->address_inuse = NULL;
1984   agent->objective_old = 0;
1985   agent->nop_bonus = 0;
1986
1987   return agent;
1988 }
1989
1990 /**
1991  * Deallocate agent
1992  *
1993  * @param solver the solver handle
1994  * @param agent the agent to retire
1995  */
1996 static void
1997 agent_die(struct GAS_RIL_Handle *solver, struct RIL_Peer_Agent *agent)
1998 {
1999   int i;
2000
2001   for (i = 0; i < agent->n; i++)
2002     {
2003       GNUNET_free_non_null(agent->W[i]);
2004       GNUNET_free_non_null(agent->E[i]);
2005     }
2006   GNUNET_free_non_null(agent->W);
2007   GNUNET_free_non_null(agent->E);
2008   GNUNET_free_non_null(agent->s_old);
2009   GNUNET_free(agent);
2010 }
2011
2012 /**
2013  * Returns the agent for a peer
2014  *
2015  * @param solver the solver handle
2016  * @param peer the identity of the peer
2017  * @param create whether or not to create an agent, if none is allocated yet
2018  * @return the agent
2019  */
2020 static struct RIL_Peer_Agent *
2021 ril_get_agent(struct GAS_RIL_Handle *solver, const struct GNUNET_PeerIdentity *peer, int create)
2022 {
2023   struct RIL_Peer_Agent *cur;
2024
2025   for (cur = solver->agents_head; NULL != cur; cur = cur->next)
2026     {
2027       if (0 == GNUNET_memcmp(peer, &cur->peer))
2028         {
2029           return cur;
2030         }
2031     }
2032
2033   if (create)
2034     {
2035       cur = agent_init(solver, peer);
2036       GNUNET_CONTAINER_DLL_insert_tail(solver->agents_head, solver->agents_tail, cur);
2037       return cur;
2038     }
2039   return NULL;
2040 }
2041
2042 /**
2043  * Determine whether at least the minimum bandwidth is set for the network. Otherwise the network is
2044  * considered inactive and not used. Addresses in an inactive network are ignored.
2045  *
2046  * @param solver solver handle
2047  * @param network the network type
2048  * @return whether or not the network is considered active
2049  */
2050 static int
2051 ril_network_is_active(struct GAS_RIL_Handle *solver, enum GNUNET_NetworkType network)
2052 {
2053   struct RIL_Scope *net;
2054
2055   net = ril_get_network(solver, network);
2056   return net->bw_out_available >= RIL_MIN_BW;
2057 }
2058
2059 /**
2060  * Cuts a slice out of a vector of elements. This is used to decrease the size of the matrix storing
2061  * the reward function approximation. It copies the memory, which is not cut, to the new vector,
2062  * frees the memory of the old vector, and redirects the pointer to the new one.
2063  *
2064  * @param old pointer to the pointer to the first element of the vector
2065  * @param element_size byte size of the vector elements
2066  * @param hole_start the first element to cut out
2067  * @param hole_length the number of elements to cut out
2068  * @param old_length the length of the old vector
2069  */
2070 static void
2071 ril_cut_from_vector(void **old,
2072                     size_t element_size,
2073                     unsigned int hole_start,
2074                     unsigned int hole_length,
2075                     unsigned int old_length)
2076 {
2077   char *tmpptr;
2078   char *oldptr = (char *)*old;
2079   size_t size;
2080   unsigned int bytes_before;
2081   unsigned int bytes_hole;
2082   unsigned int bytes_after;
2083
2084   GNUNET_assert(old_length >= hole_length);
2085   GNUNET_assert(old_length >= (hole_start + hole_length));
2086
2087   size = element_size * (old_length - hole_length);
2088
2089   bytes_before = element_size * hole_start;
2090   bytes_hole = element_size * hole_length;
2091   bytes_after = element_size * (old_length - hole_start - hole_length);
2092
2093   if (0 == size)
2094     {
2095       tmpptr = NULL;
2096     }
2097   else
2098     {
2099       tmpptr = GNUNET_malloc(size);
2100       GNUNET_memcpy(tmpptr, oldptr, bytes_before);
2101       GNUNET_memcpy(tmpptr + bytes_before, oldptr + (bytes_before + bytes_hole), bytes_after);
2102     }
2103   if (NULL != *old)
2104     {
2105       GNUNET_free(*old);
2106     }
2107   *old = (void *)tmpptr;
2108 }
2109
2110 /*
2111  *  Solver API functions
2112  *  ---------------------------
2113  */
2114
2115 /**
2116  * Change relative preference for quality in solver
2117  *
2118  * @param solver the solver handle
2119  * @param peer the peer to change the preference for
2120  * @param kind the kind to change the preference
2121  * @param pref_rel the normalized preference value for this kind over all clients
2122  */
2123 static void
2124 GAS_ril_address_change_preference(void *solver,
2125                                   const struct GNUNET_PeerIdentity *peer,
2126                                   enum GNUNET_ATS_PreferenceKind kind,
2127                                   double pref_rel)
2128 {
2129   LOG(GNUNET_ERROR_TYPE_DEBUG,
2130       "API_address_change_preference() Preference '%s' for peer '%s' changed to %.2f \n",
2131       GNUNET_ATS_print_preference_type(kind), GNUNET_i2s(peer), pref_rel);
2132
2133   struct GAS_RIL_Handle *s = solver;
2134
2135   s->parameters.temperature = s->parameters.temperature_init;
2136   s->parameters.epsilon = s->parameters.epsilon_init;
2137   ril_step(s);
2138 }
2139
2140
2141 /**
2142  * Add a new address for a peer to the solver
2143  *
2144  * The address is already contained in the addresses hashmap!
2145  *
2146  * @param solver the solver Handle
2147  * @param address the address to add
2148  * @param network network type of this address
2149  */
2150 static void
2151 GAS_ril_address_add(void *solver,
2152                     struct ATS_Address *address,
2153                     uint32_t network)
2154 {
2155   struct GAS_RIL_Handle *s = solver;
2156   struct RIL_Peer_Agent *agent;
2157   struct RIL_Address_Wrapped *address_wrapped;
2158   struct RIL_Scope *net;
2159   unsigned int m_new;
2160   unsigned int m_old;
2161   unsigned int n_new;
2162   unsigned int n_old;
2163   int i;
2164   unsigned int zero;
2165
2166   LOG(GNUNET_ERROR_TYPE_DEBUG,
2167       "API_address_add()\n");
2168
2169   net = ril_get_network(s, network);
2170   address->solver_information = net;
2171
2172   if (!ril_network_is_active(s, network))
2173     {
2174       LOG(GNUNET_ERROR_TYPE_DEBUG,
2175           "API_address_add() Did not add %s address %s for peer '%s', network does not have enough bandwidth\n",
2176           address->plugin, address->addr, GNUNET_i2s(&address->peer));
2177       return;
2178     }
2179
2180   s->parameters.temperature = s->parameters.temperature_init;
2181   s->parameters.epsilon = s->parameters.epsilon_init;
2182
2183   agent = ril_get_agent(s, &address->peer, GNUNET_YES);
2184
2185   //add address
2186   address_wrapped = GNUNET_new(struct RIL_Address_Wrapped);
2187   address_wrapped->address_naked = address;
2188   GNUNET_CONTAINER_DLL_insert_tail(agent->addresses_head, agent->addresses_tail, address_wrapped);
2189
2190   //increase size of W
2191   m_new = agent->m + ((s->parameters.rbf_divisor + 1) * (s->parameters.rbf_divisor + 1));
2192   m_old = agent->m;
2193   n_new = agent->n + 1;
2194   n_old = agent->n;
2195
2196   GNUNET_array_grow(agent->W, agent->n, n_new);
2197   agent->n = n_old;
2198   GNUNET_array_grow(agent->E, agent->n, n_new);
2199   for (i = 0; i < n_new; i++)
2200     {
2201       if (i < n_old)
2202         {
2203           agent->m = m_old;
2204           GNUNET_array_grow(agent->W[i], agent->m, m_new);
2205           agent->m = m_old;
2206           GNUNET_array_grow(agent->E[i], agent->m, m_new);
2207         }
2208       else
2209         {
2210           zero = 0;
2211           GNUNET_array_grow(agent->W[i], zero, m_new);
2212           zero = 0;
2213           GNUNET_array_grow(agent->E[i], zero, m_new);
2214         }
2215     }
2216
2217   //increase size of old state vector
2218   agent->m = m_old;
2219   GNUNET_array_grow(agent->s_old, agent->m, m_new);
2220
2221   ril_try_unblock_agent(s, agent, GNUNET_NO);
2222
2223   ril_step(s);
2224
2225   LOG(GNUNET_ERROR_TYPE_DEBUG, "API_address_add() Added %s %s address %s for peer '%s'\n",
2226       address->active ? "active" : "inactive", address->plugin, address->addr,
2227       GNUNET_i2s(&address->peer));
2228 }
2229
2230 /**
2231  * Delete an address in the solver
2232  *
2233  * The address is not contained in the address hashmap anymore!
2234  *
2235  * @param solver the solver handle
2236  * @param address the address to remove
2237  */
2238 static void
2239 GAS_ril_address_delete(void *solver,
2240                        struct ATS_Address *address)
2241 {
2242   struct GAS_RIL_Handle *s = solver;
2243   struct RIL_Peer_Agent *agent;
2244   struct RIL_Address_Wrapped *address_wrapped;
2245   int address_index;
2246   unsigned int m_new;
2247   unsigned int n_new;
2248   int i;
2249   struct RIL_Scope *net;
2250
2251   LOG(GNUNET_ERROR_TYPE_DEBUG,
2252       "API_address_delete() Delete %s %s address %s for peer '%s'\n",
2253       address->active ? "active" : "inactive",
2254       address->plugin,
2255       address->addr,
2256       GNUNET_i2s(&address->peer));
2257
2258   agent = ril_get_agent(s, &address->peer, GNUNET_NO);
2259   if (NULL == agent)
2260     {
2261       net = address->solver_information;
2262       GNUNET_assert(!ril_network_is_active(s, net->type));
2263       LOG(GNUNET_ERROR_TYPE_DEBUG,
2264           "No agent allocated for peer yet, since address was in inactive network\n");
2265       return;
2266     }
2267
2268   s->parameters.temperature = s->parameters.temperature_init;
2269   s->parameters.epsilon = s->parameters.epsilon_init;
2270
2271   address_index = agent_address_get_index(agent, address);
2272   address_wrapped = agent_address_get_wrapped(agent, address);
2273
2274   if (NULL == address_wrapped)
2275     {
2276       net = address->solver_information;
2277       LOG(GNUNET_ERROR_TYPE_DEBUG,
2278           "Address not considered by agent, address was in inactive network\n");
2279       return;
2280     }
2281   GNUNET_CONTAINER_DLL_remove(agent->addresses_head,
2282                               agent->addresses_tail,
2283                               address_wrapped);
2284   GNUNET_free(address_wrapped);
2285
2286   //decrease W
2287   m_new = agent->m - ((s->parameters.rbf_divisor + 1) * (s->parameters.rbf_divisor + 1));
2288   n_new = agent->n - 1;
2289
2290   for (i = 0; i < agent->n; i++)
2291     {
2292       ril_cut_from_vector((void **)&agent->W[i], sizeof(double),
2293                           address_index * ((s->parameters.rbf_divisor + 1) * (s->parameters.rbf_divisor + 1)),
2294                           ((s->parameters.rbf_divisor + 1) * (s->parameters.rbf_divisor + 1)), agent->m);
2295       ril_cut_from_vector((void **)&agent->E[i], sizeof(double),
2296                           address_index * ((s->parameters.rbf_divisor + 1) * (s->parameters.rbf_divisor + 1)),
2297                           ((s->parameters.rbf_divisor + 1) * (s->parameters.rbf_divisor + 1)), agent->m);
2298     }
2299   GNUNET_free_non_null(agent->W[RIL_ACTION_TYPE_NUM + address_index]);
2300   GNUNET_free_non_null(agent->E[RIL_ACTION_TYPE_NUM + address_index]);
2301   ril_cut_from_vector((void **)&agent->W, sizeof(double *), RIL_ACTION_TYPE_NUM + address_index,
2302                       1, agent->n);
2303   ril_cut_from_vector((void **)&agent->E, sizeof(double *), RIL_ACTION_TYPE_NUM + address_index,
2304                       1, agent->n);
2305   //correct last action
2306   if (agent->a_old > (RIL_ACTION_TYPE_NUM + address_index))
2307     {
2308       agent->a_old -= 1;
2309     }
2310   else if (agent->a_old == (RIL_ACTION_TYPE_NUM + address_index))
2311     {
2312       agent->a_old = RIL_ACTION_INVALID;
2313     }
2314   //decrease old state vector
2315   ril_cut_from_vector((void **)&agent->s_old, sizeof(double),
2316                       address_index * ((s->parameters.rbf_divisor + 1) * (s->parameters.rbf_divisor + 1)),
2317                       ((s->parameters.rbf_divisor + 1) * (s->parameters.rbf_divisor + 1)), agent->m);
2318   agent->m = m_new;
2319   agent->n = n_new;
2320
2321   if (agent->address_inuse == address)
2322     {
2323       if (NULL != agent->addresses_head) //if peer has an address left, use it
2324         {
2325           GNUNET_log(GNUNET_ERROR_TYPE_DEBUG,
2326                      "Active address died, suggesting alternative!\n");
2327           envi_set_active_suggestion(s,
2328                                      agent,
2329                                      agent->addresses_head->address_naked,
2330                                      agent->bw_in,
2331                                      agent->bw_out,
2332                                      GNUNET_YES);
2333         }
2334       else
2335         {
2336           GNUNET_log(GNUNET_ERROR_TYPE_DEBUG,
2337                      "Active address died, suggesting disconnect!\n");
2338           envi_set_active_suggestion(s, agent, NULL, 0, 0, GNUNET_NO);
2339         }
2340     }
2341   ril_step(solver);
2342   if (agent->suggestion_address == address)
2343     {
2344       agent->suggestion_issue = GNUNET_NO;
2345       agent->suggestion_address = NULL;
2346     }
2347   GNUNET_assert(agent->address_inuse != address);
2348 }
2349
2350
2351 /**
2352  * Update the properties of an address in the solver
2353  *
2354  * @param solver solver handle
2355  * @param address the address
2356  */
2357 static void
2358 GAS_ril_address_property_changed(void *solver,
2359                                  struct ATS_Address *address)
2360 {
2361   struct GAS_RIL_Handle *s = solver;
2362
2363   LOG(GNUNET_ERROR_TYPE_DEBUG,
2364       "Properties for peer '%s' address changed\n",
2365       GNUNET_i2s(&address->peer));
2366   s->parameters.temperature = s->parameters.temperature_init;
2367   s->parameters.epsilon = s->parameters.epsilon_init;
2368   ril_step(s);
2369 }
2370
2371
2372 /**
2373  * Give feedback about the current assignment
2374  *
2375  * @param solver the solver handle
2376  * @param application the application
2377  * @param peer the peer to change the preference for
2378  * @param scope the time interval for this feedback: [now - scope .. now]
2379  * @param kind the kind to change the preference
2380  * @param score the score
2381  */
2382 static void
2383 GAS_ril_address_preference_feedback(void *solver,
2384                                     struct GNUNET_SERVICE_Client *application,
2385                                     const struct GNUNET_PeerIdentity *peer,
2386                                     const struct GNUNET_TIME_Relative scope,
2387                                     enum GNUNET_ATS_PreferenceKind kind,
2388                                     double score)
2389 {
2390   LOG(GNUNET_ERROR_TYPE_DEBUG,
2391       "API_address_preference_feedback() Peer '%s' got a feedback of %+.3f from application %s for "
2392       "preference %s for %d seconds\n",
2393       GNUNET_i2s(peer),
2394       "UNKNOWN",
2395       GNUNET_ATS_print_preference_type(kind),
2396       scope.rel_value_us / 1000000);
2397 }
2398
2399
2400 /**
2401  * Start a bulk operation
2402  *
2403  * @param solver the solver
2404  */
2405 static void
2406 GAS_ril_bulk_start(void *solver)
2407 {
2408   struct GAS_RIL_Handle *s = solver;
2409
2410   LOG(GNUNET_ERROR_TYPE_DEBUG,
2411       "API_bulk_start() lock: %d\n", s->bulk_lock + 1);
2412
2413   s->bulk_lock++;
2414 }
2415
2416
2417 /**
2418  * Bulk operation done
2419  *
2420  * @param solver the solver handle
2421  */
2422 static void
2423 GAS_ril_bulk_stop(void *solver)
2424 {
2425   struct GAS_RIL_Handle *s = solver;
2426
2427   LOG(GNUNET_ERROR_TYPE_DEBUG,
2428       "API_bulk_stop() lock: %d\n",
2429       s->bulk_lock - 1);
2430
2431   if (s->bulk_lock < 1)
2432     {
2433       GNUNET_break(0);
2434       return;
2435     }
2436   s->bulk_lock--;
2437
2438   if (0 < s->bulk_changes)
2439     {
2440       ril_step(solver);
2441       s->bulk_changes = 0;
2442     }
2443 }
2444
2445
2446 /**
2447  * Tell solver to notify ATS if the address to use changes for a specific
2448  * peer using the bandwidth changed callback
2449  *
2450  * The solver must only notify about changes for peers with pending address
2451  * requests!
2452  *
2453  * @param solver the solver handle
2454  * @param peer the identity of the peer
2455  */
2456 static void
2457 GAS_ril_get_preferred_address(void *solver,
2458                               const struct GNUNET_PeerIdentity *peer)
2459 {
2460   struct GAS_RIL_Handle *s = solver;
2461   struct RIL_Peer_Agent *agent;
2462
2463   LOG(GNUNET_ERROR_TYPE_DEBUG, "API_get_preferred_address()\n");
2464
2465   agent = ril_get_agent(s, peer, GNUNET_YES);
2466
2467   agent->is_active = GNUNET_YES;
2468   envi_set_active_suggestion(solver, agent, agent->address_inuse, agent->bw_in, agent->bw_out, GNUNET_YES);
2469
2470   ril_try_unblock_agent(solver, agent, GNUNET_YES);
2471
2472   if (agent->address_inuse)
2473     {
2474       LOG(GNUNET_ERROR_TYPE_DEBUG,
2475           "API_get_preferred_address() Activated agent for peer '%s' with %s address %s\n",
2476           GNUNET_i2s(peer), agent->address_inuse->plugin, agent->address_inuse->addr);
2477     }
2478   else
2479     {
2480       LOG(GNUNET_ERROR_TYPE_DEBUG,
2481           "API_get_preferred_address() Activated agent for peer '%s', but no address available\n",
2482           GNUNET_i2s(peer));
2483       s->parameters.temperature = s->parameters.temperature_init;
2484       s->parameters.epsilon = s->parameters.epsilon_init;
2485     }
2486   if (NULL != agent->address_inuse)
2487     s->env->bandwidth_changed_cb(s->env->cls,
2488                                  agent->address_inuse);
2489 }
2490
2491
2492 /**
2493  * Tell solver stop notifying ATS about changes for this peers
2494  *
2495  * The solver must only notify about changes for peers with pending address
2496  * requests!
2497  *
2498  * @param solver the solver handle
2499  * @param peer the peer
2500  */
2501 static void
2502 GAS_ril_stop_get_preferred_address(void *solver,
2503                                    const struct GNUNET_PeerIdentity *peer)
2504 {
2505   struct GAS_RIL_Handle *s = solver;
2506   struct RIL_Peer_Agent *agent;
2507
2508   LOG(GNUNET_ERROR_TYPE_DEBUG,
2509       "API_stop_get_preferred_address()");
2510
2511   agent = ril_get_agent(s, peer, GNUNET_NO);
2512
2513   if (NULL == agent)
2514     {
2515       GNUNET_break(0);
2516       return;
2517     }
2518   if (GNUNET_NO == agent->is_active)
2519     {
2520       GNUNET_break(0);
2521       return;
2522     }
2523
2524   s->parameters.temperature = s->parameters.temperature_init;
2525   s->parameters.epsilon = s->parameters.epsilon_init;
2526
2527   agent->is_active = GNUNET_NO;
2528
2529   envi_set_active_suggestion(s, agent, agent->address_inuse, agent->bw_in, agent->bw_out,
2530                              GNUNET_YES);
2531
2532   ril_step(s);
2533
2534   LOG(GNUNET_ERROR_TYPE_DEBUG,
2535       "API_stop_get_preferred_address() Paused agent for peer '%s'\n",
2536       GNUNET_i2s(peer));
2537 }
2538
2539
2540 /**
2541  * Entry point for the plugin
2542  *
2543  * @param cls pointer to the 'struct GNUNET_ATS_PluginEnvironment'
2544  */
2545 void *
2546 libgnunet_plugin_ats_ril_init(void *cls)
2547 {
2548   static struct GNUNET_ATS_SolverFunctions sf;
2549   struct GNUNET_ATS_PluginEnvironment *env = cls;
2550   struct GAS_RIL_Handle *solver = GNUNET_new(struct GAS_RIL_Handle);
2551   struct RIL_Scope * cur;
2552   int c;
2553   char *string;
2554   float f_tmp;
2555
2556   LOG(GNUNET_ERROR_TYPE_DEBUG,
2557       "API_init() Initializing RIL solver\n");
2558
2559   GNUNET_assert(NULL != env);
2560   GNUNET_assert(NULL != env->cfg);
2561   GNUNET_assert(NULL != env->stats);
2562   GNUNET_assert(NULL != env->bandwidth_changed_cb);
2563   GNUNET_assert(NULL != env->get_preferences);
2564
2565   if (GNUNET_OK != GNUNET_CONFIGURATION_get_value_number(env->cfg, "ats", "RIL_RBF_DIVISOR", &solver->parameters.rbf_divisor))
2566     {
2567       solver->parameters.rbf_divisor = RIL_DEFAULT_RBF_DIVISOR;
2568     }
2569
2570   if (GNUNET_OK
2571       != GNUNET_CONFIGURATION_get_value_time(env->cfg, "ats", "RIL_STEP_TIME_MIN",
2572                                              &solver->parameters.step_time_min))
2573     {
2574       solver->parameters.step_time_min = RIL_DEFAULT_STEP_TIME_MIN;
2575     }
2576
2577   if (GNUNET_OK
2578       != GNUNET_CONFIGURATION_get_value_time(env->cfg, "ats", "RIL_STEP_TIME_MAX",
2579                                              &solver->parameters.step_time_max))
2580     {
2581       solver->parameters.step_time_max = RIL_DEFAULT_STEP_TIME_MAX;
2582     }
2583
2584   if (GNUNET_OK == GNUNET_CONFIGURATION_get_value_string(env->cfg, "ats", "RIL_ALGORITHM", &string))
2585     {
2586       GNUNET_STRINGS_utf8_toupper(string, string);
2587       if (0 == strcmp(string, "SARSA"))
2588         {
2589           solver->parameters.algorithm = RIL_ALGO_SARSA;
2590         }
2591       if (0 == strcmp(string, "Q-LEARNING"))
2592         {
2593           solver->parameters.algorithm = RIL_ALGO_Q;
2594         }
2595
2596       GNUNET_free(string);
2597     }
2598   else
2599     {
2600       solver->parameters.algorithm = RIL_DEFAULT_ALGORITHM;
2601     }
2602
2603   if (GNUNET_OK == GNUNET_CONFIGURATION_get_value_string(env->cfg, "ats", "RIL_SELECT", &string))
2604     {
2605       solver->parameters.select = !strcmp(string, "EGREEDY") ? RIL_SELECT_EGREEDY : RIL_SELECT_SOFTMAX;
2606       GNUNET_free(string);
2607     }
2608   else
2609     {
2610       solver->parameters.select = RIL_DEFAULT_SELECT;
2611     }
2612
2613
2614   solver->parameters.beta = RIL_DEFAULT_DISCOUNT_BETA;
2615   if (GNUNET_SYSERR != GNUNET_CONFIGURATION_get_value_float(env->cfg, "ats",
2616                                                             "RIL_DISCOUNT_BETA", &f_tmp))
2617     {
2618       if (f_tmp < 0.0)
2619         {
2620           LOG(GNUNET_ERROR_TYPE_ERROR, _("Invalid %s configuration %f \n"),
2621               "RIL_DISCOUNT_BETA", f_tmp);
2622         }
2623       else
2624         {
2625           solver->parameters.beta = f_tmp;
2626           LOG(GNUNET_ERROR_TYPE_INFO, "Using %s of %.3f\n",
2627               "RIL_DISCOUNT_BETA", f_tmp);
2628         }
2629     }
2630
2631   solver->parameters.gamma = RIL_DEFAULT_DISCOUNT_GAMMA;
2632   if (GNUNET_SYSERR != GNUNET_CONFIGURATION_get_value_float(env->cfg, "ats",
2633                                                             "RIL_DISCOUNT_GAMMA", &f_tmp))
2634     {
2635       if ((f_tmp < 0.0) || (f_tmp > 1.0))
2636         {
2637           LOG(GNUNET_ERROR_TYPE_ERROR, _("Invalid %s configuration %f \n"),
2638               "RIL_DISCOUNT_GAMMA", f_tmp);
2639         }
2640       else
2641         {
2642           solver->parameters.gamma = f_tmp;
2643           LOG(GNUNET_ERROR_TYPE_INFO, "Using %s of %.3f\n",
2644               "RIL_DISCOUNT_GAMMA", f_tmp);
2645         }
2646     }
2647
2648   solver->parameters.alpha = RIL_DEFAULT_GRADIENT_STEP_SIZE;
2649   if (GNUNET_SYSERR != GNUNET_CONFIGURATION_get_value_float(env->cfg, "ats",
2650                                                             "RIL_GRADIENT_STEP_SIZE", &f_tmp))
2651     {
2652       if ((f_tmp < 0.0) || (f_tmp > 1.0))
2653         {
2654           LOG(GNUNET_ERROR_TYPE_ERROR, _("Invalid %s configuration %f \n"),
2655               "RIL_GRADIENT_STEP_SIZE", f_tmp);
2656         }
2657       else
2658         {
2659           solver->parameters.alpha = f_tmp;
2660           LOG(GNUNET_ERROR_TYPE_INFO, "Using %s of %.3f\n",
2661               "RIL_GRADIENT_STEP_SIZE", f_tmp);
2662         }
2663     }
2664
2665   solver->parameters.lambda = RIL_DEFAULT_TRACE_DECAY;
2666   if (GNUNET_SYSERR != GNUNET_CONFIGURATION_get_value_float(env->cfg, "ats",
2667                                                             "RIL_TRACE_DECAY", &f_tmp))
2668     {
2669       if ((f_tmp < 0.0) || (f_tmp > 1.0))
2670         {
2671           LOG(GNUNET_ERROR_TYPE_ERROR, _("Invalid %s configuration %f \n"),
2672               "RIL_TRACE_DECAY", f_tmp);
2673         }
2674       else
2675         {
2676           solver->parameters.lambda = f_tmp;
2677           LOG(GNUNET_ERROR_TYPE_INFO, "Using %s of %.3f\n",
2678               "RIL_TRACE_DECAY", f_tmp);
2679         }
2680     }
2681
2682   solver->parameters.epsilon_init = RIL_DEFAULT_EXPLORE_RATIO;
2683   if (GNUNET_SYSERR != GNUNET_CONFIGURATION_get_value_float(env->cfg, "ats",
2684                                                             "RIL_EXPLORE_RATIO", &f_tmp))
2685     {
2686       if ((f_tmp < 0.0) || (f_tmp > 1.0))
2687         {
2688           LOG(GNUNET_ERROR_TYPE_ERROR, _("Invalid %s configuration %f \n"),
2689               "RIL_EXPLORE_RATIO", f_tmp);
2690         }
2691       else
2692         {
2693           solver->parameters.epsilon_init = f_tmp;
2694           LOG(GNUNET_ERROR_TYPE_INFO, "Using %s of %.3f\n",
2695               "RIL_EXPLORE_RATIO", f_tmp);
2696         }
2697     }
2698
2699   solver->parameters.epsilon_decay = RIL_DEFAULT_EXPLORE_DECAY;
2700   if (GNUNET_SYSERR != GNUNET_CONFIGURATION_get_value_float(env->cfg, "ats",
2701                                                             "RIL_EXPLORE_DECAY", &f_tmp))
2702     {
2703       if ((f_tmp < 0.0) || (f_tmp > 1.0))
2704         {
2705           LOG(GNUNET_ERROR_TYPE_ERROR, _("Invalid %s configuration %f \n"),
2706               "RIL_EXPLORE_DECAY", f_tmp);
2707         }
2708       else
2709         {
2710           solver->parameters.epsilon_decay = f_tmp;
2711           LOG(GNUNET_ERROR_TYPE_INFO, "Using %s of %.3f\n",
2712               "RIL_EXPLORE_DECAY", f_tmp);
2713         }
2714     }
2715
2716   solver->parameters.temperature_init = RIL_DEFAULT_TEMPERATURE;
2717   if (GNUNET_SYSERR != GNUNET_CONFIGURATION_get_value_float(env->cfg, "ats",
2718                                                             "RIL_TEMPERATURE", &f_tmp))
2719     {
2720       if (f_tmp <= 0.0)
2721         {
2722           LOG(GNUNET_ERROR_TYPE_ERROR, _("Invalid %s configuration %f \n"),
2723               "RIL_TEMPERATURE", f_tmp);
2724         }
2725       else
2726         {
2727           solver->parameters.temperature_init = f_tmp;
2728           LOG(GNUNET_ERROR_TYPE_INFO, "Using %s of %.3f\n",
2729               "RIL_TEMPERATURE", f_tmp);
2730         }
2731     }
2732
2733   solver->parameters.temperature_decay = RIL_DEFAULT_TEMPERATURE_DECAY;
2734   if (GNUNET_SYSERR != GNUNET_CONFIGURATION_get_value_float(env->cfg, "ats",
2735                                                             "RIL_TEMPERATURE_DECAY", &f_tmp))
2736     {
2737       if ((f_tmp <= 0.0) || solver->parameters.temperature_decay > 1)
2738         {
2739           LOG(GNUNET_ERROR_TYPE_ERROR, _("Invalid %s configuration %f \n"),
2740               "RIL_TEMPERATURE_DECAY", f_tmp);
2741         }
2742       else
2743         {
2744           solver->parameters.temperature_decay = f_tmp;
2745           LOG(GNUNET_ERROR_TYPE_INFO, "Using %s of %.3f\n",
2746               "RIL_TEMPERATURE_DECAY", f_tmp);
2747         }
2748     }
2749
2750   if (GNUNET_OK != GNUNET_CONFIGURATION_get_value_number(env->cfg, "ats", "RIL_SIMULATE", &solver->simulate))
2751     {
2752       solver->simulate = GNUNET_NO;
2753     }
2754
2755   if (GNUNET_YES == GNUNET_CONFIGURATION_get_value_yesno(env->cfg, "ats", "RIL_REPLACE_TRACES"))
2756     {
2757       solver->parameters.eligibility_trace_mode = RIL_E_REPLACE;
2758     }
2759   else
2760     {
2761       solver->parameters.eligibility_trace_mode = RIL_E_ACCUMULATE;
2762     }
2763
2764   if (GNUNET_OK == GNUNET_CONFIGURATION_get_value_string(env->cfg, "ats", "RIL_SOCIAL_WELFARE", &string))
2765     {
2766       solver->parameters.social_welfare = !strcmp(string, "NASH") ? RIL_WELFARE_NASH : RIL_WELFARE_EGALITARIAN;
2767       GNUNET_free(string);
2768     }
2769   else
2770     {
2771       solver->parameters.social_welfare = RIL_DEFAULT_WELFARE;
2772     }
2773
2774   solver->env = env;
2775   sf.cls = solver;
2776   sf.s_add = &GAS_ril_address_add;
2777   sf.s_address_update_property = &GAS_ril_address_property_changed;
2778   sf.s_get = &GAS_ril_get_preferred_address;
2779   sf.s_get_stop = &GAS_ril_stop_get_preferred_address;
2780   sf.s_pref = &GAS_ril_address_change_preference;
2781   sf.s_feedback = &GAS_ril_address_preference_feedback;
2782   sf.s_del = &GAS_ril_address_delete;
2783   sf.s_bulk_start = &GAS_ril_bulk_start;
2784   sf.s_bulk_stop = &GAS_ril_bulk_stop;
2785
2786   solver->networks_count = env->network_count;
2787   solver->network_entries = GNUNET_malloc(env->network_count * sizeof(struct RIL_Scope));
2788   solver->step_count = 0;
2789   solver->done = GNUNET_NO;
2790
2791   for (c = 0; c < env->network_count; c++)
2792     {
2793       cur = &solver->network_entries[c];
2794       cur->type = c;
2795       cur->bw_in_available = env->in_quota[c];
2796       cur->bw_out_available = env->out_quota[c];
2797       LOG(GNUNET_ERROR_TYPE_DEBUG,
2798           "init()  Quotas for %s network:  IN %llu - OUT %llu\n",
2799           GNUNET_NT_to_string(cur->type),
2800           cur->bw_in_available / 1024,
2801           cur->bw_out_available / 1024);
2802     }
2803
2804   LOG(GNUNET_ERROR_TYPE_DEBUG, "init()  Parameters:\n");
2805   LOG(GNUNET_ERROR_TYPE_DEBUG, "init()  Algorithm = %s, alpha = %f, beta = %f, lambda = %f\n",
2806       solver->parameters.algorithm ? "Q" : "SARSA",
2807       solver->parameters.alpha,
2808       solver->parameters.beta,
2809       solver->parameters.lambda);
2810   LOG(GNUNET_ERROR_TYPE_DEBUG, "init()  exploration_ratio = %f, temperature = %f, ActionSelection = %s\n",
2811       solver->parameters.epsilon,
2812       solver->parameters.temperature,
2813       solver->parameters.select ? "EGREEDY" : "SOFTMAX");
2814   LOG(GNUNET_ERROR_TYPE_DEBUG, "init()  RBF_DIVISOR = %llu\n",
2815       solver->parameters.rbf_divisor);
2816
2817   return &sf;
2818 }
2819
2820
2821 /**
2822  * Exit point for the plugin
2823  *
2824  * @param cls the solver handle
2825  */
2826 void *
2827 libgnunet_plugin_ats_ril_done(void *cls)
2828 {
2829   struct GNUNET_ATS_SolverFunctions *sf = cls;
2830   struct GAS_RIL_Handle *s = sf->cls;
2831   struct RIL_Peer_Agent *cur_agent;
2832   struct RIL_Peer_Agent *next_agent;
2833
2834   LOG(GNUNET_ERROR_TYPE_DEBUG, "API_done() Shutting down RIL solver\n");
2835
2836   s->done = GNUNET_YES;
2837
2838   cur_agent = s->agents_head;
2839   while (NULL != cur_agent)
2840     {
2841       next_agent = cur_agent->next;
2842       GNUNET_CONTAINER_DLL_remove(s->agents_head, s->agents_tail, cur_agent);
2843       agent_die(s, cur_agent);
2844       cur_agent = next_agent;
2845     }
2846
2847   if (NULL != s->step_next_task_id)
2848     {
2849       GNUNET_SCHEDULER_cancel(s->step_next_task_id);
2850     }
2851   GNUNET_free(s->network_entries);
2852   GNUNET_free(s);
2853
2854   return NULL;
2855 }
2856
2857
2858 /* end of plugin_ats_ril.c */