<!DOCTYPE html><html><head><title></title><style type="text/css">p.MsoNormal,p.MsoNoSpacing{margin:0}</style></head><body><div><br></div><div><br></div><div>On Fri, Mar 1, 2019, at 13:39, Rémi Denis-Courmont wrote:<br></div><blockquote type="cite" id="fastmail-quoted"><div>Hi,<br></div><div><br></div><div>I do not know why a separate block is needed as opposed to flagging a discontinuity on the following successfully output block.<br></div></blockquote><div><br></div><div>I will investigate that approach but it feels more complicated. I must make sure the flag is not lost in the filter chain. Should I check in every modules or handle it in the pipeline (aout/filters.c) ?<br></div><div><br></div><blockquote type="cite" id="fastmail-quoted"><div><br></div><div>But assuming that you have a good reason in mind, there are still some challenges with the error block approach:<br></div><div>- If it needs to be queued, it needs to be writable, and thus it needs to be allocated, and thus its allocation can fail.<br></div></blockquote><div><br></div><div>It won't, I put asserts in queue/release/realloc functions. </div><div><br></div><blockquote type="cite" id="fastmail-quoted"><div>- There are code paths that discards zero-size blocks specially, or fails to handle them at all.<br></div><div>- Some craptastic platforms might not support exported globals, only functions (not that I'd care personally).<br></div></blockquote><div>Indeed. </div><div><br></div><blockquote type="cite" id="fastmail-quoted"><div>-- <br></div><div>Envoyé de mon appareil Android avec Courriel K-9 Mail. Veuillez excuser ma brièveté. <br></div><div>_______________________________________________<br></div><div>vlc-devel mailing list<br></div><div>To unsubscribe or modify your subscription options:<br></div><div>https://mailman.videolan.org/listinfo/vlc-devel<br></div></blockquote><div><br></div></body></html>