In `27f27d1`, we changed the image loading logic to conform to the various options (`-a|--load-all` & `-c|--crashed-only`) and loaded them concurrently. However, instead of the subset of images that matched the user option, the thread pool would always run on all the crashlog images, causing them to be all loaded in the target everytime. This matches the `-a|--load-all` option behaviour but depending on the report, it can cause lldb to load thousands of images, which can take a very long time if the images are downloaded over the network. This patch fixes that issue by keeping a list of `images_to_load` based of the user-provided option. This list will be used with our executor thread pool to load the images according to the user selection, and reinstates the expected default behaviour, by only loading the crashed thread images and skipping all the others. This patch also unifies the way we load images into a single method that's shared by both the batch mode & the interactive scripted process. rdar://123694062 Signed-off-by: Med Ismail Bennani <ismail@bennani.ma>
7.3 KiB
7.3 KiB