_idatui_spans: one capturing re.split over the tag pairs instead of finditer+char-slicing, and collapse whitespace with ' '.join(txt.split()) instead of a regex sub. 13.15 -> 11.07 us/line (the previous finditer attempt was 14.4, i.e. SLOWER than the original char loop it replaced).

Result: {"status":"keep","total_ms":26050.2,"lg_boot_ms":727.4,"lg_decomp_ms":2618.1,"lg_graph_ms":935.1,"lg_hex_ms":954.2,"lg_index_ms":67.4,"lg_listing_cold_ms":459.3,"lg_listing_warm_ms":545.7,"lg_nav_ms":9257.8,"lg_palette_ms":4.7,"lg_render_ms":238.8,"lg_search_ms":5472.5,"pure_graph_ms":530.7,"sm_boot_ms":534.6,"sm_decomp_ms":635.8,"sm_graph_ms":748.1,"sm_hex_ms":858.7,"sm_index_ms":0,"sm_listing_cold_ms":268.7,"sm_listing_warm_ms":269.2,"sm_nav_ms":441.6,"sm_palette_ms":0.3,"sm_render_ms":281.7,"sm_search_ms":200,"fails":0}

2 files changed, 57 insertions, 46 deletions

diff --git a/.auto/log.jsonl b/.auto/log.jsonl
index ebef486..55322e0 100644
--- a/.auto/log.jsonl
+++ b/.auto/log.jsonl
@@ -1,3 +1,4 @@
 {"type":"config","name":"ida-tui performance: cut the latency of navigation, listing, decomp, graph and search","metricName":"total_ms","metricUnit":"ms","bestDirection":"lower"}
 {"run":1,"commit":"2910c93","metric":46572.1,"metrics":{"lg_boot_ms":863.5,"lg_decomp_ms":2881.3,"lg_graph_ms":951.7,"lg_hex_ms":900.9,"lg_index_ms":219.3,"lg_listing_cold_ms":434.4,"lg_listing_warm_ms":552.3,"lg_nav_ms":29106.9,"lg_palette_ms":4.8,"lg_render_ms":231.2,"lg_search_ms":5590.1,"pure_graph_ms":236.2,"sm_boot_ms":539.4,"sm_decomp_ms":621.4,"sm_graph_ms":702,"sm_hex_ms":850.5,"sm_index_ms":0,"sm_listing_cold_ms":270.8,"sm_listing_warm_ms":270.6,"sm_nav_ms":875.1,"sm_palette_ms":0.3,"sm_render_ms":272.3,"sm_search_ms":197.1,"fails":0},"status":"checks_failed","description":"Baseline run of the new bench harness. Benchmark clean (fails=0) but the pilot scenario suite reported 300 passed / 1 failed with ZERO code changes -> flaky, and checks.sh printed the tail instead of the FAIL line so the name is unknown.","timestamp":1786058161497,"segment":0,"confidence":null,"asi":{"hypothesis":"establish a baseline for total_ms","bottleneck":"lg_nav_ms=29107 is 62% of total_ms; lg_nav_worst_ms=28274 is ONE cold jump to a high address in bash. ListingModel.ensure_ea walks the segment forward in 500-head pages from seg_start, so landing near the end of a 224k-row listing costs ~450 sequential worker round trips.","second_bottleneck":"lg_search_ms=5590 (91783 hits over the whole segment)","cheap_phases":"palette/index/render/pure_graph are all <600ms; not where the time is","rollback_reason":"checks.sh flagged 1 scenario failure with no code change (flake)","next_action_hint":"make checks.sh print the FAIL line name on non-zero exit, re-run baseline, then attack ListingModel address->row lookup (needs a backend primitive in server/patch_server.py: heads walking anchored at an address, or a segment head-index built in one call)"}}
 {"run":2,"commit":"a3f3400","metric":46685.5,"metrics":{"lg_boot_ms":885.7,"lg_decomp_ms":2871.8,"lg_graph_ms":930.9,"lg_hex_ms":947.3,"lg_index_ms":207.8,"lg_listing_cold_ms":564.3,"lg_listing_warm_ms":454.7,"lg_nav_ms":29018.5,"lg_palette_ms":4.6,"lg_render_ms":223.6,"lg_search_ms":5448.7,"pure_graph_ms":535.4,"sm_boot_ms":537.9,"sm_decomp_ms":643.8,"sm_graph_ms":680.4,"sm_hex_ms":854.9,"sm_index_ms":0,"sm_listing_cold_ms":265.5,"sm_listing_warm_ms":263.7,"sm_nav_ms":887.2,"sm_palette_ms":0.3,"sm_render_ms":264.1,"sm_search_ms":194.5,"fails":0},"status":"discard","description":"Baseline re-run with the fixed checks gate. Checks pass; total_ms reproduces to within 0.24% of run #1 (46572 -> 46686), so the noise floor is ~115ms on a 46.6s metric.","timestamp":1786058370297,"segment":0,"confidence":null,"asi":{"hypothesis":"confirm the baseline is reproducible and the checks gate is green","noise_floor_ms":115,"reproducibility":"run1 46572 / run2 46686 -> 0.24% spread; pure_graph_ms is the jumpiest single phase (236 -> 535, it is CPU-only and gets descheduled)","checks":"flaky-scenario retry logic works; suite green on a clean tree","next_action_hint":"attack ListingModel.ensure_ea / the heads tool: 29s of 46.6s is one cold address->row walk over bash"}}
+{"run":3,"commit":"93240e2","metric":26923.9,"metrics":{"lg_boot_ms":762.2,"lg_decomp_ms":2631.7,"lg_graph_ms":941.8,"lg_hex_ms":1052,"lg_index_ms":67.2,"lg_listing_cold_ms":440.6,"lg_listing_warm_ms":530.5,"lg_nav_ms":10598.3,"lg_palette_ms":4.6,"lg_render_ms":227.8,"lg_search_ms":5265.5,"pure_graph_ms":237.9,"sm_boot_ms":535.3,"sm_decomp_ms":631.8,"sm_graph_ms":702.1,"sm_hex_ms":841.1,"sm_index_ms":0,"sm_listing_cold_ms":268.2,"sm_listing_warm_ms":269.4,"sm_nav_ms":443.6,"sm_palette_ms":0.3,"sm_render_ms":277.8,"sm_search_ms":194.4,"fails":0},"status":"keep","description":"Stop ida-pro-mcp installing a sys.setprofile hook around every tool call. Its deadline mechanism profiles every python call/return so a pure-python tool loop can be interrupted; our tools are call-heavy, so it taxed the whole backend 3.3x. Worker now sets IDA_MCP_TOOL_TIMEOUT_SEC=0 and arms the deadline itself with one polling watchdog thread + ida_kernwin.set_cancelled() (the half that actually frees the IDA main thread). Also rewrote _idatui_spans to jump between colour tags instead of walking characters (byte-identical over 258k real lines).","timestamp":1786059117542,"segment":0,"confidence":173.264550264548,"asi":{"hypothesis":"the heads tool is not IDA-bound; the ida-pro-mcp sync wrapper's sys.setprofile deadline is the tax","evidence":"in-process A/B on targets/bash: heads(count=500,annotate) 92.2us/row with IDA_MCP_TOOL_TIMEOUT_SEC unset vs 27.5us/row with it 0; domain ListingModel._load_next_page 116 -> 37 us/row","where_it_lives":"site-packages/ida_pro_mcp/ida_mcp/sync.py sync_wrapper() installs profilefunc via sys.setprofile whenever timeout>0; default _DEFAULT_TOOL_TIMEOUT_SEC=60","what_we_kept":"the native half: threading watchdog + ida_kernwin.set_cancelled(), which is what decompile/auto_wait/find_bytes actually poll. Only the interruption of a pure-python loop inside a tool body is gone, and every idatui tool loop is bounded by its count argument (<=2000).","watchdog_design":"polling at 0.25s instead of signalling, because arm/disarm is on the path of EVERY worker call (a scroll is dozens); arming is two attribute writes + clr_cancelled()","spans_rewrite":"_idatui_spans now re.finditer's over the 3 control chars instead of walking characters. Verified byte-identical vs HEAD over 258k real disasm lines across echo/ls_ttl/bash/libcrypto (.auto/diff_spans.py). Standalone it measured 12.5 -> 13.6 us/line i.e. slightly SLOWER, so it is probably a small net loss that the setprofile win is hiding -- A/B it on its own next.","gains":"total 46572 -> 26924 (-42%); lg_nav 29107 -> 10598; lg_nav_worst 28274 -> 10170; sm_nav_worst 656 -> 224; lg_index 219 -> 67","work_unchanged":"nav_rows/decomp_ok/graph_ok/search_hits/render_cells all identical -> the speedup is not less work","next_action_hint":"1) A/B the _idatui_spans rewrite alone (suspect regression). 2) lg_nav is still 10.6s: ListingModel.ensure_ea is O(distance) at ~34us/row; a skeleton (ea,size,kind only) walk or a bigger PAGE would cut it further. 3) lg_search_ms 5265 is now the #2 term."}}
diff --git a/server/patch_server.py b/server/patch_server.py
index b806be2..b75b120 100644
--- a/server/patch_server.py
+++ b/server/patch_server.py
@@ -364,8 +364,7 @@ def _idatui_tag_map():
 
 _IDATUI_TAGS = None
 _IDATUI_OPND_TAGS = None
-_IDATUI_CTL = None   # re: the three control characters a tagged line can hold
-_IDATUI_WS = None    # re: a run of whitespace, exactly what str.isspace() calls one
+_IDATUI_CTL = None   # re: a tag = one of three control chars plus its argument
 
 
 def _idatui_opnd_tag_map():
@@ -395,7 +394,7 @@ def _idatui_spans(line):
     Unknown tags become 'text' rather than being dropped: a processor module can
     emit a colour we don't classify, and losing the characters would corrupt the
     line."""
-    global _IDATUI_TAGS, _IDATUI_OPND_TAGS, _IDATUI_CTL, _IDATUI_WS
+    global _IDATUI_TAGS, _IDATUI_OPND_TAGS, _IDATUI_CTL
     import ida_lines
     if _IDATUI_TAGS is None:
         _IDATUI_TAGS = _idatui_tag_map()
@@ -403,48 +402,54 @@ def _idatui_spans(line):
         _IDATUI_OPND_TAGS = _idatui_opnd_tag_map()
     if _IDATUI_CTL is None:
         import re as _re
-        _IDATUI_CTL = _re.compile("[\\x01\\x02\\x03]")
-        # str.isspace() is true for \\x1c-\\x1f and \\x85 as well as the \\s
-        # class, so spell those out: this substitution has to agree with the
-        # plain-text collapse character for character (checked over every
-        # codepoint) or the row silently loses its highlighting.
-        _IDATUI_WS = _re.compile("[\\\\s\\x1c\\x1d\\x1e\\x1f\\x85]+")
-    tags, opnds, ctl = _IDATUI_TAGS, _IDATUI_OPND_TAGS, _IDATUI_CTL
+        # One capturing split gives [text, tag, text, tag, ..., text] in a
+        # single C pass. A per-character python loop over the line used to be
+        # the most expensive thing the `heads` tool did, and a line is ~54
+        # characters but only ~13 tags -- everything between two tags is already
+        # exactly one span's worth of text.
+        _IDATUI_CTL = _re.compile("([\\x01\\x02\\x03][\\s\\S])")
+    tags, opnds = _IDATUI_TAGS, _IDATUI_OPND_TAGS
     on, off, esc = "\x01", "\x02", "\x03"
     addr_tag = chr(getattr(ida_lines, "COLOR_ADDR", 0x28))
     addr_len = int(getattr(ida_lines, "COLOR_ADDR_SIZE", 16))
-    # Jump between control characters and take the text in between as one slice.
-    # A per-character loop here was 68% of the whole `heads` tool: a disasm line
-    # is ~50 characters but only ~15 tags, and everything between two tags is
-    # already exactly one span's worth of text.
-    spans, stack, buf = [], [], []   # stack entries: (kind, operand index|None)
+    parts = _IDATUI_CTL.split(line)
+    spans, stack = [], []            # stack entries: (kind, operand index|None)
     kind, opnd = "text", None        # state the current run of text belongs to
-    i, n = 0, len(line)
-    for m in ctl.finditer(line):
-        j = m.start()
-        if j < i:                    # inside an address payload / after an esc
-            continue
-        if j + 1 >= n:               # a trailing control char is literal text
+    pend = ""
+    skip = 0                         # characters of an address payload still due
+    i, n = 0, len(parts)
+    while i < n:
+        txt = parts[i]
+        i += 1
+        if skip:
+            if len(txt) <= skip:
+                skip -= len(txt)
+                txt = ""
+            else:
+                txt = txt[skip:]
+                skip = 0
+        if txt:
+            pend += txt
+        if i >= n:
             break
-        ch = line[j]
+        pair = parts[i]
+        i += 1
+        if skip:                     # a tag INSIDE an address payload: 2 chars
+            skip = skip - 2 if skip > 2 else 0
+            continue
+        ch = pair[0]
         if ch == esc:                # escaped literal: keep the char it guards
-            buf.append(line[i:j])
-            buf.append(line[j + 1])
-            i = j + 2
+            pend += pair[1]
             continue
-        tag = line[j + 1]
+        tag = pair[1]
         if ch == on and tag == addr_tag:
             # An embedded target address, not display text: 16 hex digits that
             # must not reach the screen. Deliberately NOT a span boundary.
-            buf.append(line[i:j])
-            i = j + 2 + addr_len
+            skip = addr_len
             continue
-        buf.append(line[i:j])
-        i = j + 2
-        txt = "".join(buf)
-        if txt:
-            spans.append([kind, txt, opnd])
-        del buf[:]
+        if pend:
+            spans.append([kind, pend, opnd])
+            pend = ""
         if ch == on:
             stack.append((kind, opnd))
             kind = tags.get(tag, "text")
@@ -455,24 +460,29 @@ def _idatui_spans(line):
             kind, opnd = stack.pop()
         else:
             kind, opnd = "text", None
-    if i < n:
-        buf.append(line[i:])
-    txt = "".join(buf)
-    if txt:
-        spans.append([kind, txt, opnd])
+    if pend:
+        spans.append([kind, pend, opnd])
     # Collapse IDA's column padding EXACTLY as the plain text does. A run of
     # spaces can straddle two spans, so the leading space of a span is dropped
     # when the previous one ended in space — otherwise the spans and `text`
     # disagree about the line and the row silently loses its highlighting.
+    # ``" ".join(txt.split())`` splits on exactly what str.isspace() calls
+    # whitespace, which is what the character walk this replaces tested.
     out, prev_space = [], False
-    ws = _IDATUI_WS
     for kind, txt, opnd in spans:
-        acc = ws.sub(" ", txt)
-        if prev_space and acc[:1] == " ":
-            acc = acc[1:]
-        if acc:
-            prev_space = acc[-1] == " "
-            out.append([kind, acc, opnd])
+        core = " ".join(txt.split())
+        if not core:                 # the span is nothing but padding
+            if not prev_space:
+                prev_space = True
+                out.append([kind, " ", opnd])
+            continue
+        acc = core
+        if txt[0].isspace() and not prev_space:
+            acc = " " + acc
+        if txt[-1].isspace():
+            acc += " "
+        prev_space = acc[-1] == " "
+        out.append([kind, acc, opnd])
     while out and out[0][1] == " ":
         out.pop(0)
     while out and out[-1][1] == " ":