aboutsummaryrefslogtreecommitdiffstats
path: root/doc/lispref/debugging.texi
diff options
context:
space:
mode:
authorEli Zaretskii2021-06-19 15:39:11 +0300
committerEli Zaretskii2021-06-19 15:39:11 +0300
commit8d5c70d73acf2fe05494dde5fdffa13e257bf819 (patch)
tree798203b5c8648dbf7bfbf7c0a9c77ab1de6179b8 /doc/lispref/debugging.texi
parent0ffef0b46b4e68e3f4113042f036f3a295498855 (diff)
downloademacs-8d5c70d73acf2fe05494dde5fdffa13e257bf819.tar.gz
emacs-8d5c70d73acf2fe05494dde5fdffa13e257bf819.zip
Improve documentation of profiler
* doc/lispref/debugging.texi (Profiling): Stop misleading users about what "memory" profiling really is.
Diffstat (limited to 'doc/lispref/debugging.texi')
-rw-r--r--doc/lispref/debugging.texi13
1 files changed, 7 insertions, 6 deletions
diff --git a/doc/lispref/debugging.texi b/doc/lispref/debugging.texi
index 1b28bf7aa5d..ed3160f4548 100644
--- a/doc/lispref/debugging.texi
+++ b/doc/lispref/debugging.texi
@@ -997,11 +997,12 @@ start looking for ways to optimize that piece.
997@findex profiler-report 997@findex profiler-report
998@findex profiler-stop 998@findex profiler-stop
999Emacs has built-in support for this. To begin profiling, type 999Emacs has built-in support for this. To begin profiling, type
1000@kbd{M-x profiler-start}. You can choose to profile by processor 1000@w{@kbd{M-x profiler-start}}. You can choose to sample CPU usage
1001usage, memory usage, or both. Then run the code you'd like to speed 1001periodically (@code{cpu}), when memory is allocated (@code{memory}),
1002up. After that, type @kbd{M-x profiler-report} to display a summary 1002or both. Then run the code you'd like to speed up. After that, type
1003buffer for each resource (cpu and memory) that you chose to profile. 1003@kbd{M-x profiler-report} to display a summary buffer for CPU usage
1004The names of the report buffers include the times at which the reports 1004sampled by each type (cpu and memory) that you chose to profile. The
1005names of the report buffers include the times at which the reports
1005were generated, so you can generate another report later on without 1006were generated, so you can generate another report later on without
1006erasing previous results. When you have finished profiling, type 1007erasing previous results. When you have finished profiling, type
1007@kbd{M-x profiler-stop} (there is a small overhead associated with 1008@kbd{M-x profiler-stop} (there is a small overhead associated with
@@ -1009,7 +1010,7 @@ profiling, so we don't recommend leaving it active except when you are
1009actually running the code you want to examine). 1010actually running the code you want to examine).
1010 1011
1011The profiler report buffer shows, on each line, a function that was 1012The profiler report buffer shows, on each line, a function that was
1012called, followed by how much resources (cpu or memory) it used in 1013called, followed by how much CPU resources it used in
1013absolute and percentage terms since profiling started. If a given 1014absolute and percentage terms since profiling started. If a given
1014line has a @samp{+} symbol at the left-hand side, you can expand that 1015line has a @samp{+} symbol at the left-hand side, you can expand that
1015line by typing @kbd{@key{RET}}, in order to see the function(s) called 1016line by typing @kbd{@key{RET}}, in order to see the function(s) called