06:13 <_whitenotifier> [scopehal-apps] azonenberg opened issue #544: Auto-scale X axis to fit entire acquisition when first connecting to a scope - https://github.com/glscopeclient/scopehal-apps/issues/544

06:13 <_whitenotifier> [scopehal-apps] azonenberg labeled issue #544: Auto-scale X axis to fit entire acquisition when first connecting to a scope - https://github.com/glscopeclient/scopehal-apps/issues/544

07:57 <_whitenotifier> [scopehal] azonenberg pushed 2 commits to master [+0/-0/±3] https://github.com/glscopeclient/scopehal/compare/b562791e932e...62d6124f2b88

07:57 <_whitenotifier> [scopehal] azonenberg 299f4d3 - Filter::FillDurations: fixed crash if length is zero

07:57 <_whitenotifier> [scopehal] azonenberg 62d6124 - Fixed a bunch of bugs in eSPI decode, allowed explicit user selection of x1 and x4 mode vs existing autodetection. Fixes #730.

07:57 <_whitenotifier> [scopehal] azonenberg closed issue #730: eSPI: support x1 mode - https://github.com/glscopeclient/scopehal/issues/730

07:57 <_whitenotifier> [scopehal-apps] azonenberg pushed 2 commits to master [+0/-0/±2] https://github.com/glscopeclient/scopehal-apps/compare/0ab974056dbb...df3d128cee08

07:57 <_whitenotifier> [scopehal-apps] azonenberg 9bd90b5 - OscilloscopeWindow: fixed bug where timestamps were not marked modified from CPU

07:58 <_whitenotifier> [scopehal-apps] azonenberg df3d128 - Updated submodules

17:48 <azonenberg> yaaaay everybody's favorite

17:48 <azonenberg> protocols that dynamically negotiate how many lanes they run on

17:48 <azonenberg> such that you can't just statically configure the decode to use X signals

17:48 <azonenberg> you have to actually parse SFR writes

17:48 <azonenberg> and change the decode config on the fly from packet to packet

18:10 <_whitenotifier> [scopehal] azonenberg pushed 2 commits to master [+0/-0/±3] https://github.com/glscopeclient/scopehal/compare/62d6124f2b88...5e8c4c2d56ad

18:10 <_whitenotifier> [scopehal] azonenberg 46da6d1 - eSPI: initial decoding support for Set Configuration packets to the General Capabilities register

18:10 <_whitenotifier> [scopehal] azonenberg 5e8c4c2 - eSPI: implemented dynamic bus width switching

23:20 <_whitenotifier> [scopehal-apps] lainy opened pull request #545: Fix glscopeclient protocol shader copy step. - https://github.com/glscopeclient/scopehal-apps/pull/545

23:20 <_whitenotifier> [scopehal-apps] lainy synchronize pull request #545: Fix glscopeclient protocol shader copy step. - https://github.com/glscopeclient/scopehal-apps/pull/545

23:27 <_whitenotifier> [scopehal-apps] azonenberg closed pull request #545: Fix glscopeclient protocol shader copy step. - https://github.com/glscopeclient/scopehal-apps/pull/545

23:27 <_whitenotifier> [scopehal-apps] azonenberg pushed 3 commits to master [+4/-0/±22] https://github.com/glscopeclient/scopehal-apps/compare/df3d128cee08...f662e9b24138

23:27 <_whitenotifier> [scopehal-apps] lain a96c8c2 - Fix glscopeclient protocol shader copy step.

23:27 <_whitenotifier> [scopehal-apps] lainy 9923660 - Merge branch 'glscopeclient:master' into master

23:27 <_whitenotifier> [scopehal-apps] azonenberg f662e9b - Merge pull request #545 from lainy/master Fix glscopeclient protocol shader copy step.

23:46 <azonenberg> just found a pathological case in which ImportFilter::TryNormalizeTimebase() gave bogus results

23:46 <azonenberg> so that's fun

23:46 <azonenberg> investigating now

23:47 <azonenberg> tl;dr if most of your samples are at uniform spacing but a few are very long (think bursts of squarewave separated by long flatlines - like a clock)

23:48 <azonenberg> it finds the standard deviation of sample intervals to be very small

23:48 <azonenberg> since *almost* all are at Fclk

23:48 <azonenberg> then takes the average - including the super long ones - and declares that your uniform sample rate :p