Commit d1e385a
Eliminate looping transform in mcbp_parser::next (#347)
* Eliminate looping transform in mcbp_parser::next
I found that in mcbp_session_impl::do_read, 39% of the time was in
mcbp::parser::next, with the vast majority of that in the std::transform
call after uncompressing the data. By using vector::insert instead,
that dropped to only 11% of the time in do:read.
Probably the upshot here is to look for everywhere we are filling
containers with a back_inserter and considering a way to modify that to
not loop and instead copy in blocks.
* Same, but for the mcbp_parser::feed call.
* similarly, but for generate_binary
---------
Co-authored-by: Sergey Avseyev <sergey.avseyev@gmail.com>1 parent 7b5ae9a commit d1e385a
3 files changed
+8
-8
lines changed| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
51 | 51 | | |
52 | 52 | | |
53 | 53 | | |
54 | | - | |
| 54 | + | |
55 | 55 | | |
56 | 56 | | |
57 | 57 | | |
58 | 58 | | |
59 | 59 | | |
60 | 60 | | |
61 | 61 | | |
62 | | - | |
63 | | - | |
64 | | - | |
| 62 | + | |
| 63 | + | |
| 64 | + | |
65 | 65 | | |
66 | 66 | | |
67 | 67 | | |
68 | 68 | | |
69 | 69 | | |
70 | 70 | | |
71 | | - | |
| 71 | + | |
72 | 72 | | |
73 | 73 | | |
74 | 74 | | |
| |||
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
30 | 30 | | |
31 | 31 | | |
32 | 32 | | |
33 | | - | |
| 33 | + | |
34 | 34 | | |
35 | 35 | | |
36 | 36 | | |
| |||
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
96 | 96 | | |
97 | 97 | | |
98 | 98 | | |
99 | | - | |
| 99 | + | |
100 | 100 | | |
101 | 101 | | |
102 | 102 | | |
103 | 103 | | |
104 | 104 | | |
105 | | - | |
| 105 | + | |
106 | 106 | | |
107 | 107 | | |
108 | 108 | | |
| |||
0 commit comments