Bartosz Taudul
e68214a88a
Change LockState to bitmask.
2023-04-15 22:56:44 +02:00
Bartosz Taudul
f7a8998663
Push locks draw lists to DrawThread().
2023-04-15 22:56:44 +02:00
Bartosz Taudul
7f301dfec0
Precalculate draw lists for locks.
2023-04-15 22:56:44 +02:00
Bartosz Taudul
9b270c8e11
Another fix for determining if there's CPU usage data to draw.
2023-04-15 22:56:44 +02:00
Bartosz Taudul
f0acb73d09
Add accessor for active lock info window lock id.
2023-04-15 22:56:43 +02:00
Bartosz Taudul
d737b911e9
Make View::Vis() public.
2023-04-15 22:56:43 +02:00
Bartosz Taudul
6e815d13a0
Move common lock helper functions to a separate header.
2023-04-15 22:56:43 +02:00
Bartosz Taudul
4c0e6fe3ca
Merge pull request #544 from simplyWiri/timeline-scrolling-tweak
...
Clamp scrolling to the difference between the deepest zone near the m…
2023-04-13 11:35:08 +02:00
Bartosz Taudul
f17b44c212
Take DPI scaling into account when drawing thread migration lines.
2023-04-07 22:37:23 +02:00
Bartosz Taudul
492e6611b9
No need to perform domain conversion.
...
Pixel values will be needed anyways later on. It doesn't make sense to
perform comparison in time domain here.
2023-04-07 22:37:23 +02:00
Bartosz Taudul
d0ffca56bf
Fix logic determining if there's CPU core usage data to draw.
2023-04-07 22:37:22 +02:00
Bartosz Taudul
1b824797a9
Do not project running regions end time to last time.
...
Just display known running regions, keeping the unended ones infinitely
collapsed. This makes the CPU core usage graph possibly display wrong
data at the end of capture. Note that this behavior was also present in
previous releases.
2023-04-07 22:37:22 +02:00
Bartosz Taudul
1ff9e0012b
Add a note why context switches only have coarse visibility check.
2023-04-05 19:39:39 +02:00
Bartosz Taudul
51f832c031
Don't recalculate sty.
2023-04-05 19:24:58 +02:00
Bartosz Taudul
f48415315a
Build samples draw list only if visible.
2023-04-05 19:03:40 +02:00
Bartosz Taudul
a9ee4c499e
Build messages draw list only if visible.
2023-04-05 18:55:23 +02:00
Bartosz Taudul
4b8ac41345
Calculate CPU data draw lists only if visible.
2023-04-05 18:45:00 +02:00
Bartosz Taudul
bb9219384f
Do not recalculate sty.
2023-04-05 18:44:36 +02:00
Bartosz Taudul
f2a2669ca8
Decouple CPU graph availability from having the draw data.
2023-04-05 18:21:12 +02:00
Bartosz Taudul
9500add83b
Push yPos to TimelineItem::Preprocess().
2023-04-05 18:07:09 +02:00
Bartosz Taudul
77402eb9c2
libbacktrace: minor fixes for zstd decompression (cdb64b68)
2023-04-05 17:24:19 +02:00
Bartosz Taudul
0f19727dc6
libbacktrace: change PC variables from uint64_t to uintptr_t (afe2967c)
2023-04-05 17:23:04 +02:00
Bartosz Taudul
eb8f485f99
Update nfd-extended to 1.0.2.
2023-04-05 17:19:48 +02:00
Bartosz Taudul
ab6c492117
Update zstd to 1.5.5.
2023-04-05 17:16:24 +02:00
Bartosz Taudul
856027a27d
Merge pull request #545 from topolarity/fix-no-sampling
...
linux: respect `TRACY_NO_SAMPLING` for sys-tracing
2023-04-04 23:29:08 +02:00
Cody Tapscott
6249999153
linux: respect TRACY_NO_SAMPLING for sys-tracing
...
This compile-time flag was being ignored on Linux. This change adds
gating for software-sampled stack trace sampling following the same
pattern as other `TRACY_NO_SAMPLE_*` options.
If `TRACY_NO_SAMPLING=1` is provided as an environment variable,
software stack sampling is also disabled.
2023-04-04 17:22:31 -04:00
Bartosz Taudul
2b7b79352b
Parallelize calculation of per-CPU context switches.
2023-04-02 22:10:22 +02:00
simplyWiri
a8b42488ad
Clamp scrolling to the difference between the deepest zone near the mouse, and the height of the window.
...
This prevents from unncessary scrolling when the trace does not exceed the size of the screen
2023-04-02 19:01:58 +10:00
Bartosz Taudul
a4c200d242
Update manual.
2023-03-30 21:56:02 +02:00
Bartosz Taudul
df1b3e7feb
Update NEWS.
2023-03-30 21:52:29 +02:00
Bartosz Taudul
e2e55a77b5
Add TracySetProgramName() macro to set broadcast contents.
2023-03-30 21:51:00 +02:00
Bartosz Taudul
cade1d9705
Fix library/unix cleanup.
2023-03-30 21:30:25 +02:00
Bartosz Taudul
33a640f848
Name worker threads.
2023-03-25 22:14:34 +01:00
Bartosz Taudul
5eebc5d7cf
Preprocess CPU usage.
...
Things of note: Worker::GetCpuUsage() functionality was moved to
TimelineItemCpuData::PreprocessCpuUsage().
2023-03-25 17:55:15 +01:00
Bartosz Taudul
dcf27f4989
Provide small font height in TimelineContext.
2023-03-25 17:25:10 +01:00
Bartosz Taudul
c0136dbdcc
Don't do unnecessary work if items are not visible.
2023-03-25 16:54:38 +01:00
Bartosz Taudul
13b31db561
Cosmetics.
2023-03-25 16:54:38 +01:00
Bartosz Taudul
e3ec455aba
Decouple check for existence from building draw lists.
...
We need to know if samples, context switches and messages are present to be
able to correctly calculate thread height. However, if the thread is not
visible, it is not necessary to provide a list of items to draw.
2023-03-25 16:52:27 +01:00
Bartosz Taudul
96d60ce626
Change GetNextFrameHeight() to GetHeight().
...
The GetNextFrameHeight() name correctness was very situational. It was
often used in the meaning of "get current frame height".
2023-03-25 15:12:33 +01:00
Bartosz Taudul
7ec68f8b52
Push item visibility to Preprocess().
2023-03-25 15:10:58 +01:00
Bartosz Taudul
8c1b519fa1
Fix wasm cleanup.
2023-03-25 12:26:28 +01:00
Bartosz Taudul
bea3c0f4d1
Don't pass -march=native to emcc.
2023-03-25 12:25:40 +01:00
Bartosz Taudul
241fa68708
Make embed.tracy downloadable.
2023-03-25 12:20:36 +01:00
Bartosz Taudul
ecdecb7012
Ignore wasm build results.
2023-03-25 12:07:53 +01:00
Bartosz Taudul
8908b68557
Fix lockup in context switch processing.
2023-03-25 00:45:13 +01:00
Bartosz Taudul
1e63406ac7
Replace previous iterator manipulation with a boolean.
2023-03-25 00:30:44 +01:00
Bartosz Taudul
efa095e25f
Simplify context switch precalculation.
2023-03-25 00:30:44 +01:00
Bartosz Taudul
734753a941
Fix MinVisNs type in messages preprocessing.
2023-03-24 23:15:11 +01:00
Bartosz Taudul
b3e9ede557
Move context switch minpx calculation to rendering code.
2023-03-24 23:10:58 +01:00
Bartosz Taudul
f2700b2786
Remove rather useless check.
...
The prev == it condition could only fire on the first run of the loop,
and on all subsequent runs prev (=next-1) will never be "it" anymore.
Lack of this condition changes nothing, as the following lines checking
for time distance between next and prev satisfy the same exit condition
(i.e. next-1 will be "it" only if lower_bound does not find anything,
hence next is farther away than MinVisNs).
2023-03-23 22:17:36 +01:00
