)]}'
{"/COMMIT_MSG":[{"author":{"_account_id":15343,"name":"Tim Burke","email":"tburke@nvidia.com","username":"tburke"},"change_message_id":"9bfa91572229308113059878fcc530d98f4f5a53","unresolved":false,"context_lines":[{"line_number":17,"context_line":"obvious thing."},{"line_number":18,"context_line":""},{"line_number":19,"context_line":"Apropos the reviewer confusion, the code in Phazr.IO looks weird."},{"line_number":20,"context_line":"It uses (word_size - hamming_distance) to compute the necessary"},{"line_number":21,"context_line":"number of blocks... but then returns the amount of memory needed"},{"line_number":22,"context_line":"to store blocks of a different size (word_size). We left all of it"},{"line_number":23,"context_line":"alone and return exactly the same values that the old computation"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"ff0f0b1f_75aa2a19","line":20,"updated":"2017-05-24 00:30:23.000000000","message":"Phazr overloads the hd arg because I was nervous about making changes to the ec_args struct -- see https://review.openstack.org/#/c/424353/3/include/erasurecode/erasurecode.h@77\n\nMight also be worth checking out the thread at https://review.openstack.org/#/c/424353/5/src/erasurecode_preprocessing.c@72 if we want to go spelunking?","commit_id":"b9d62b6f08cab1e8c2032649843073fedfee9e00"},{"author":{"_account_id":15343,"name":"Tim Burke","email":"tburke@nvidia.com","username":"tburke"},"change_message_id":"9bfa91572229308113059878fcc530d98f4f5a53","unresolved":false,"context_lines":[{"line_number":24,"context_line":"returned."},{"line_number":25,"context_line":""},{"line_number":26,"context_line":"All these computations were the only thing in the code that used"},{"line_number":27,"context_line":"-lm, so drop that too."},{"line_number":28,"context_line":""},{"line_number":29,"context_line":"Coincidentially, this patch solves the crash of distro-built"},{"line_number":30,"context_line":"packages of liberasurecode (see Red Hat bug #1454543). But it\u0027s"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"ff0f0b1f_d510d602","line":27,"updated":"2017-05-24 00:30:23.000000000","message":"nice","commit_id":"b9d62b6f08cab1e8c2032649843073fedfee9e00"}],"src/erasurecode.c":[{"author":{"_account_id":15343,"name":"Tim Burke","email":"tburke@nvidia.com","username":"tburke"},"change_message_id":"9bfa91572229308113059878fcc530d98f4f5a53","unresolved":false,"context_lines":[{"line_number":1219,"context_line":"    alignment_multiple \u003d k * word_size;"},{"line_number":1220,"context_line":""},{"line_number":1221,"context_line":"    ret \u003d ((data_len + alignment_multiple - 1) / alignment_multiple)"},{"line_number":1222,"context_line":"\t    * alignment_multiple;"},{"line_number":1223,"context_line":""},{"line_number":1224,"context_line":"out:"},{"line_number":1225,"context_line":"    return ret;"}],"source_content_type":"text/x-csrc","patch_set":1,"id":"ff0f0b1f_f5b4faf3","line":1222,"updated":"2017-05-24 00:30:23.000000000","message":"tab?\n\nPart of me wonders whether it would make more sense to do something like\n\n ret \u003d data_len + (alignment_multiple - data_len % alignment_multiple) % alignment_multiple\n\nbut maybe the two modulos come out about the same as the div+multiply.","commit_id":"b9d62b6f08cab1e8c2032649843073fedfee9e00"},{"author":{"_account_id":597,"name":"Pete Zaitcev","email":"zaitcev@kotori.zaitcev.us","username":"zaitcev"},"change_message_id":"ba623fd5df9e13d0963ea775d11d19f72ca27e59","unresolved":false,"context_lines":[{"line_number":1219,"context_line":"    alignment_multiple \u003d k * word_size;"},{"line_number":1220,"context_line":""},{"line_number":1221,"context_line":"    ret \u003d ((data_len + alignment_multiple - 1) / alignment_multiple)"},{"line_number":1222,"context_line":"\t    * alignment_multiple;"},{"line_number":1223,"context_line":""},{"line_number":1224,"context_line":"out:"},{"line_number":1225,"context_line":"    return ret;"}],"source_content_type":"text/x-csrc","patch_set":1,"id":"df140735_7999c7ce","line":1222,"in_reply_to":"ff0f0b1f_f5b4faf3","updated":"2017-06-06 16:50:46.000000000","message":"Well it\u0027s 4-space indent, so... dunno.\n\nThe main impetus of the patch is to use an ideom. The old code was correct, there was nothing wrong with it, but it used an oddball method to accomplish what we need here. If you invent yet another way to do it, you defeat the purpose of the patch.","commit_id":"b9d62b6f08cab1e8c2032649843073fedfee9e00"}],"src/erasurecode_helpers.c":[{"author":{"_account_id":15343,"name":"Tim Burke","email":"tburke@nvidia.com","username":"tburke"},"change_message_id":"385f1d34664fa42079c79fda3ba184069ee6c621","unresolved":false,"context_lines":[{"line_number":200,"context_line":"    }"},{"line_number":201,"context_line":""},{"line_number":202,"context_line":"    aligned_size \u003d ((data_len + alignment_multiple - 1) / alignment_multiple)"},{"line_number":203,"context_line":"\t* alignment_multiple;"},{"line_number":204,"context_line":""},{"line_number":205,"context_line":"    return aligned_size;"},{"line_number":206,"context_line":"}"}],"source_content_type":"text/x-csrc","patch_set":3,"id":"df140735_4dcf5929","line":203,"updated":"2017-06-06 18:57:49.000000000","message":"There\u0027s another one here, but\n\n- we have whole files that use tabs instead of spaces (like src/utils/chksum/md5.c and src/builtin/xor_codes/xor_hd_code.c), and\n- we (obviously) don\u0027t have any jobs in place to prevent regressions\n\n...so... meh?","commit_id":"960cdd09dcabe5e125d8ee1c54fcc1f908e8ef11"}]}