diff options
Diffstat (limited to 'chromium/docs/website/site/developers/design-documents/printing/index.md')
| -rw-r--r-- | chromium/docs/website/site/developers/design-documents/printing/index.md | 30 |
1 files changed, 0 insertions, 30 deletions
diff --git a/chromium/docs/website/site/developers/design-documents/printing/index.md b/chromium/docs/website/site/developers/design-documents/printing/index.md deleted file mode 100644 index 66bf83b91b4..00000000000 --- a/chromium/docs/website/site/developers/design-documents/printing/index.md +++ /dev/null @@ -1,30 +0,0 @@ ---- -breadcrumbs: -- - /developers - - For Developers -- - /developers/design-documents - - Design Documents -page_name: printing -title: Printing ---- - -* The printing is initiated by the yellow boxes; either by user - intents or by script. -* The actual printer spooling is done in the print worker thread. -* The renderer thread is blocking the whole time, which could lead to - deadlock where: - * The renderer sends a sync message to the browser UI thread - (which implicitly requires a round trip to the I/O thread) - * The browser thread displays a PrintDlgEx() dialog box for - instance - * The tab client area (e.g. the HTML display area) receives a - invalidation; requiring it to regenerate the display. - * (or) A windowed plugin receives a WM_PAINT window message. - * The windowed plugin calls NPAPI_ExecuteJavascript() - * This calls synchronously the renderer. - * But the renderer is not processing messages anymore -> - **Deadlock**. - -Because of this, there is work going on to not block the Renderer Thread by -duplicating the WebViewImpl/WebFrameImpl and use the "inert" duplicate frame for -the printing. This removes all the sync messages requirements.
\ No newline at end of file |