public inbox for gdb-cvs@sourceware.org
help / color / mirror / Atom feed
From: Andrew Burgess <aburgess@sourceware.org>
To: gdb-cvs@sourceware.org
Subject: [binutils-gdb] gdb/tui: fairer distribution of excess space during apply
Date: Sun,  3 Apr 2022 14:43:28 +0000 (GMT)	[thread overview]
Message-ID: <20220403144328.D5EDA385843E@sourceware.org> (raw)

https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;h=b0fcf3e344438a4d115afdfaa7468bf90cc8c880

commit b0fcf3e344438a4d115afdfaa7468bf90cc8c880
Author: Andrew Burgess <aburgess@redhat.com>
Date:   Tue Feb 1 10:28:18 2022 +0000

    gdb/tui: fairer distribution of excess space during apply
    
    When applying layouts gdb computes the size of each window (or rather,
    each sub-layout within a layout) using integer arithmetic.  As this
    rounds down the results, then, when all sub-layouts are sized, there
    is the possibility that we have some space left over.
    
    Currently, this space is just assigned to an arbitrary sub-layout.
    
    This can result in some unbalanced results.  Consider this set of
    steps with current master:
    
      (gdb) tui enable
      (gdb) layout regs
      (gdb) info win
      Name       Lines Columns Focus
      regs           7      80
      src            9      80 (has focus)
      status         1      80
      cmd            8      80
    
    Notice the weird split between the src and regs windows, the original
    layout specification has these windows given equal weight.  The
    problem is that, with rounding, both the regs and src windows are
    initially sized to 7, the extra 2 lines are then arbitrarily added to
    the src window.
    
    In this commit, rather than add all the extra space to one single
    window, I instead hand out the extra space 1 line at a time, looping
    over all the sub-layouts.  We take care to respect the min/max sizes,
    and so, we now get this result:
    
      (gdb) tui enable
      (gdb) layout regs
      (gdb) info win
      Name       Lines Columns Focus
      regs           8      80
      src            8      80 (has focus)
      status         1      80
      cmd            8      80
    
    This looks more natural to me.
    
    This is obviously a change in behaviour, and so, lots of the existing
    tests need to be updated to take this into account.  None of the
    changes are huge, it's just a line or two (or column for width) moved
    between windows.

Diff:
---
 gdb/testsuite/gdb.tui/basic.exp    |  4 ++--
 gdb/testsuite/gdb.tui/empty.exp    | 14 +++++++++-----
 gdb/testsuite/gdb.tui/regs.exp     |  4 ++--
 gdb/testsuite/gdb.tui/winwidth.exp | 17 +++++++++--------
 gdb/tui/tui-layout.c               | 38 +++++++++++++++++++++++++++++++-------
 5 files changed, 53 insertions(+), 24 deletions(-)

diff --git a/gdb/testsuite/gdb.tui/basic.exp b/gdb/testsuite/gdb.tui/basic.exp
index 0e9f2e3eab2..7afb28b0858 100644
--- a/gdb/testsuite/gdb.tui/basic.exp
+++ b/gdb/testsuite/gdb.tui/basic.exp
@@ -104,5 +104,5 @@ Term::command "layout split"
 Term::check_contents "split layout contents" \
     "$main_line *$main_re.*$hex <main>"
 
-Term::check_box "source box in split layout" 0 0 80 7
-Term::check_box "asm box in split layout" 0 6 80 9
+Term::check_box "source box in split layout" 0 0 80 8
+Term::check_box "asm box in split layout" 0 7 80 8
diff --git a/gdb/testsuite/gdb.tui/empty.exp b/gdb/testsuite/gdb.tui/empty.exp
index ef14fe4dac0..3efa55d812e 100644
--- a/gdb/testsuite/gdb.tui/empty.exp
+++ b/gdb/testsuite/gdb.tui/empty.exp
@@ -35,26 +35,26 @@ if {![Term::enter_tui]} {
 set layouts {
     {src src {{0 0 80 15}} {{0 0 90 26}}
 	{{"no source" "No Source Available"}}}
-    {regs src-regs {{0 0 80 7} {0 6 80 9}} {{0 0 90 13} {0 12 90 13}}
+    {regs src-regs {{0 0 80 8} {0 7 80 8}} {{0 0 90 13} {0 12 90 13}}
 	{
 	    {"no source" "No Source Available"}
 	    {"no regs" "Register Values Unavailable"}
 	}}
-    {asm asm {{0 0 80 13}} {{0 0 90 26}}
+    {asm asm {{0 0 80 15}} {{0 0 90 26}}
 	{
 	    {"no asm" "No Assembly Available"}
 	}}
-    {regs asm-regs {{0 0 80 7} {0 6 80 9}} {{0 0 90 13} {0 12 90 13}}
+    {regs asm-regs {{0 0 80 8} {0 7 80 8}} {{0 0 90 13} {0 12 90 13}}
 	{
 	    {"no asm" "No Assembly Available"}
 	    {"no regs" "Register Values Unavailable"}
 	}}
-    {split split {{0 0 80 6} {0 5 80 8}} {{0 0 90 13} {0 12 90 13}}
+    {split split {{0 0 80 8} {0 7 80 8}} {{0 0 90 13} {0 12 90 13}}
 	{
 	    {"no source" "No Source Available"}
 	    {"no asm" "No Assembly Available"}
 	}}
-    {regs split-regs {{0 0 80 6} {0 5 80 8}} {{0 0 90 13} {0 12 90 13}}
+    {regs split-regs {{0 0 80 8} {0 7 80 8}} {{0 0 90 13} {0 12 90 13}}
 	{
 	    {"no asm" "No Assembly Available"}
 	    {"no regs" "Register Values Unavailable"}
@@ -97,5 +97,9 @@ foreach layout $layouts {
 	    check_text $text_list
 	}
 	Term::resize 24 80
+	with_test_prefix "80x24 again" {
+	    check_boxes $small_boxes
+	    check_text $text_list
+	}
     }
 }
diff --git a/gdb/testsuite/gdb.tui/regs.exp b/gdb/testsuite/gdb.tui/regs.exp
index 178eba03f19..2f3482f5d38 100644
--- a/gdb/testsuite/gdb.tui/regs.exp
+++ b/gdb/testsuite/gdb.tui/regs.exp
@@ -38,8 +38,8 @@ if {![Term::enter_tui]} {
 Term::check_contents "source at startup" "\\|.*21 *return 0"
 
 Term::command "layout regs"
-Term::check_box "register box" 0 0 80 7
-Term::check_box "source box in regs layout" 0 6 80 9
+Term::check_box "register box" 0 0 80 8
+Term::check_box "source box in regs layout" 0 7 80 8
 
 set text [Term::get_line 1]
 # Just check for any register window content at all.
diff --git a/gdb/testsuite/gdb.tui/winwidth.exp b/gdb/testsuite/gdb.tui/winwidth.exp
index b0a838b578f..1767b25c8d3 100644
--- a/gdb/testsuite/gdb.tui/winwidth.exp
+++ b/gdb/testsuite/gdb.tui/winwidth.exp
@@ -39,24 +39,25 @@ with_test_prefix "original window sizes" {
 }
 
 with_test_prefix "after src +5" {
-    Term::check_box "source box" 0 0 44 15
-    Term::check_box "asm box" 43 0 37 15
+    Term::check_box "source box" 0 0 45 15
+    Term::check_box "asm box" 44 0 36 15
     Term::command "winwidth asm -5"
 }
 
 with_test_prefix "after asm -5" {
-    Term::check_box "source box" 0 0 48 15
-    Term::check_box "asm box" 47 0 33 15
+    Term::dump_screen
+    Term::check_box "source box" 0 0 50 15
+    Term::check_box "asm box" 49 0 31 15
     Term::command "winwidth asm +8"
 }
 
 with_test_prefix "after asm +8" {
-    Term::check_box "source box" 0 0 39 15
-    Term::check_box "asm box" 38 0 42 15
+    Term::check_box "source box" 0 0 42 15
+    Term::check_box "asm box" 41 0 39 15
     Term::command "winwidth src -2"
 }
 
 with_test_prefix "after src -2" {
-    Term::check_box "source box" 0 0 36 15
-    Term::check_box "asm box" 35 0 45 15
+    Term::check_box "source box" 0 0 40 15
+    Term::check_box "asm box" 39 0 41 15
 }
diff --git a/gdb/tui/tui-layout.c b/gdb/tui/tui-layout.c
index 52105cca612..30beefddfdb 100644
--- a/gdb/tui/tui-layout.c
+++ b/gdb/tui/tui-layout.c
@@ -810,11 +810,11 @@ tui_layout_split::apply (int x_, int y_, int width_, int height_)
 	    info[i].size = info[i].max_size;
 	  if (info[i].size < info[i].min_size)
 	    info[i].size = info[i].min_size;
-	  /* If there is any leftover size, just redistribute it to the
-	     last resizeable window, by dropping it from the allocated
-	     size.  We could try to be fancier here perhaps, by
-	     redistributing this size among all windows, not just the
-	     last window.  */
+	  /* Keep a total of all the size we've used so far (we gain some
+	     size back if this window can share a border with a preceding
+	     window).  Any unused space will be distributed between all of
+	     the other windows (while respecting min/max sizes) later in
+	     this function.  */
 	  used_size += info[i].size;
 	  if (info[i].share_box)
 	    --used_size;
@@ -834,9 +834,33 @@ tui_layout_split::apply (int x_, int y_, int width_, int height_)
     }
 
   /* Allocate any leftover size.  */
-  if (available_size >= used_size && last_index != -1)
+  if (available_size > used_size && last_index != -1)
     {
-      info[last_index].size += available_size - used_size;
+      /* Loop over all windows until all available space is used up.  */
+      bool found_window_that_can_grow_p = true;
+      for (int idx = last_index;
+	   available_size > used_size;
+	   idx = (idx + 1) % m_splits.size ())
+	{
+	  /* Once we have visited all of the windows, check that we did
+	     manage to allocate some more space.  This prevents us getting
+	     stuck in the loop forever if we can't allocate anything
+	     more.  */
+	  if (idx == last_index)
+	    {
+	      if (!found_window_that_can_grow_p)
+		break;
+	      found_window_that_can_grow_p = false;
+	    }
+
+	  if (available_size > used_size
+	      && info[idx].size < info[idx].max_size)
+	    {
+	      found_window_that_can_grow_p = true;
+	      info[idx].size += 1;
+	      used_size += 1;
+	    }
+	}
 
       if (debug_tui)
 	{


                 reply	other threads:[~2022-04-03 14:43 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20220403144328.D5EDA385843E@sourceware.org \
    --to=aburgess@sourceware.org \
    --cc=gdb-cvs@sourceware.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).