From bee90675a701d7267caba40921b4606db5072247 Mon Sep 17 00:00:00 2001 From: Boris Kolpackov Date: Thu, 7 Apr 2022 09:33:50 +0200 Subject: Rename {match,execute}() to *_sync(), add *_complete() In particular, the match() rename makes sure it doesn't clash with rule::match() which, after removal of the hint argument in simple_rule, has exactly the same signature, thus making it error-prone to calling recursively. --- libbuild2/algorithm.cxx | 94 ++++++++++++++++++------------------------------- 1 file changed, 35 insertions(+), 59 deletions(-) (limited to 'libbuild2/algorithm.cxx') diff --git a/libbuild2/algorithm.cxx b/libbuild2/algorithm.cxx index 61ab92c..9e4d77a 100644 --- a/libbuild2/algorithm.cxx +++ b/libbuild2/algorithm.cxx @@ -751,7 +751,7 @@ namespace build2 dr << info << "while matching group rule to " << diag_do (a, t); }); - pair r (match (a, g, 0, nullptr, try_match)); + pair r (match_impl (a, g, 0, nullptr, try_match)); if (r.first) { @@ -843,11 +843,11 @@ namespace build2 // the first half of the result. // pair - match (action a, - const target& ct, - size_t start_count, - atomic_count* task_count, - bool try_match) + match_impl (action a, + const target& ct, + size_t start_count, + atomic_count* task_count, + bool try_match) { // If we are blocking then work our own queue one task at a time. The // logic here is that we may have already queued other tasks before this @@ -997,15 +997,15 @@ namespace build2 { // Execute (unlocked). // - // Note that we use execute_direct() rather than execute() here to - // sidestep the dependents count logic. In this context, this is by - // definition the first attempt to execute this rule (otherwise we - // would have already known the members list) and we really do need + // Note that we use execute_direct_sync() rather than execute_sync() + // here to sidestep the dependents count logic. In this context, this + // is by definition the first attempt to execute this rule (otherwise + // we would have already known the members list) and we really do need // to execute it now. // { phase_switch ps (g.ctx, run_phase::execute); - execute_direct (a, g); + execute_direct_sync (a, g); } r = g.group_members (a); @@ -1099,7 +1099,7 @@ namespace build2 for (size_t n (pts.size ()); i != n; ++i) { const target& pt (*pts[i]); - match (a, pt); + match_complete (a, pt); } } @@ -1149,7 +1149,7 @@ namespace build2 if (m == nullptr || marked (m)) continue; - match (a, *m); + match_complete (a, *m); } } @@ -1217,7 +1217,7 @@ namespace build2 // Make it ad hoc so that it doesn't end up in prerequisite_targets // after execution. // - match (a, *r); + match_sync (a, *r); t.prerequisite_targets[a].emplace_back (r, include_type::adhoc); } @@ -1903,10 +1903,10 @@ namespace build2 } target_state - execute (action a, - const target& ct, - size_t start_count, - atomic_count* task_count) + execute_impl (action a, + const target& ct, + size_t start_count, + atomic_count* task_count) { target& t (const_cast (ct)); // MT-aware. target::opstate& s (t[a]); @@ -2002,17 +2002,17 @@ namespace build2 } target_state - execute_direct (action a, - const target& ct, - size_t start_count, - atomic_count* task_count) + execute_direct_impl (action a, + const target& ct, + size_t start_count, + atomic_count* task_count) { context& ctx (ct.ctx); target& t (const_cast (ct)); // MT-aware. target::opstate& s (t[a]); - // Similar logic to execute() above. + // Similar logic to execute_impl() above. // size_t tc (ctx.count_applied ()); @@ -2078,8 +2078,8 @@ namespace build2 // date (and which is where it originated). // // There would normally be a lot of headers for every source file (think - // all the system headers) and just calling execute_direct() on all of - // them can get expensive. At the same time, most of these headers are + // all the system headers) and just calling execute_direct_sync() on all + // of them can get expensive. At the same time, most of these headers are // existing files that we will never be updating (again, system headers, // for example) and the rule that will match them is the fallback // file_rule. That rule has an optimization: it returns noop_recipe (which @@ -2117,7 +2117,7 @@ namespace build2 // any generated header. // phase_switch ps (t.ctx, run_phase::execute); - target_state ns (execute_direct (a, t)); + target_state ns (execute_direct_sync (a, t)); if (ns != os && ns != target_state::unchanged) { @@ -2188,7 +2188,7 @@ namespace build2 const target& pt (*p.target); target_state os (static_cast (p.data)); - target_state ns (execute_direct (a, pt)); + target_state ns (execute_direct_sync (a, pt)); if (ns != os && ns != target_state::unchanged) { @@ -2213,7 +2213,6 @@ namespace build2 atomic_count& tc (t[a].task_count); size_t busy (ctx.count_busy ()); - size_t exec (ctx.count_executed ()); wait_guard wg (ctx, busy, tc); @@ -2234,13 +2233,8 @@ namespace build2 if ((p.include & mask) != 0 && p.data != 0) { const target& pt (*p.target); - - // If the target is still busy, wait for its completion. - // - ctx.sched.wait (exec, pt[a].task_count, scheduler::work_none); - + target_state ns (execute_complete (a, pt)); target_state os (static_cast (p.data)); - target_state ns (pt.executed_state (a)); if (ns != os && ns != target_state::unchanged) { @@ -2278,7 +2272,6 @@ namespace build2 target_state r (target_state::unchanged); size_t busy (ctx.count_busy ()); - size_t exec (ctx.count_executed ()); // Start asynchronous execution of prerequisites. // @@ -2313,12 +2306,7 @@ namespace build2 continue; const target& mt (*ts[i]); - - // If the target is still busy, wait for its completion. - // - ctx.sched.wait (exec, mt[a].task_count, scheduler::work_none); - - r |= mt.executed_state (a); + r |= execute_complete (a, mt); blank_adhoc_member (ts[i]); } @@ -2336,7 +2324,6 @@ namespace build2 target_state r (target_state::unchanged); size_t busy (ctx.count_busy ()); - size_t exec (ctx.count_executed ()); wait_guard wg (ctx, busy, tc); @@ -2365,10 +2352,7 @@ namespace build2 continue; const target& mt (*ts[i]); - - ctx.sched.wait (exec, mt[a].task_count, scheduler::work_none); - - r |= mt.executed_state (a); + r |= execute_complete (a, mt); blank_adhoc_member (ts[i]); } @@ -2405,7 +2389,6 @@ namespace build2 context& ctx (t.ctx); size_t busy (ctx.count_busy ()); - size_t exec (ctx.count_executed ()); auto& pts (t.prerequisite_targets[a]); @@ -2447,10 +2430,7 @@ namespace build2 continue; const target& pt (*p.target); - - ctx.sched.wait (exec, pt[a].task_count, scheduler::work_none); - - target_state s (pt.executed_state (a)); + target_state s (execute_complete (a, pt)); rs |= s; // Should we compare the timestamp to this target's? @@ -2499,7 +2479,6 @@ namespace build2 context& ctx (t.ctx); size_t busy (ctx.count_busy ()); - size_t exec (ctx.count_executed ()); auto& pts (t.prerequisite_targets[a]); @@ -2541,10 +2520,7 @@ namespace build2 continue; const target& pt (*p.target); - - ctx.sched.wait (exec, pt[a].task_count, scheduler::work_none); - - target_state s (pt.executed_state (a)); + target_state s (execute_complete (a, pt)); rs |= s; // Should we compare the timestamp to this target's? @@ -2601,10 +2577,10 @@ namespace build2 // const target& g (*t.group); - // This is execute(a, t, false) but that saves a call to executed_state() - // (which we don't need). + // This is execute_sync(a, t, false) but that saves a call to + // executed_state() (which we don't need). // - target_state gs (execute (a, g, 0, nullptr)); + target_state gs (execute_impl (a, g, 0, nullptr)); if (gs == target_state::busy) ctx.sched.wait (ctx.count_executed (), -- cgit v1.1