2018-01-25 0:44 GMT+01:00 Ben Chang <b...@nvidia.com>: > Please help review this patch to reduce stack frame size per GPU > thread. The default allocation size per thread (1024 bytes) is > excessive and can be reduced to 128 bytes based on nvidia cuda > kernel compilation statistics. This should help with reducing video > memory usage per cuda context.
> Subject: [PATCH] Reduce cuda context's stack frame size limit through > cuCtxSetLimit. The default stack limit is 1024 byte per GPU thread. This > reduces limit to 128 byte as verified against current cuda kernel compilation > statistic. This will reduce local memory allocated per cuda context. This looks as if your commit message spans several lines, should be one line followed by an empty line and as many more lines as you need. Is there a reason why the error messages are different? > This email message is for the sole use of the intended recipient(s) > and may contain confidential information. Please remove this or use another email address. Carl Eugen _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
