)]}'
{"/COMMIT_MSG":[{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"cfa96175586f698c5b5dada0a0b745b30d94648b","unresolved":true,"context_lines":[{"line_number":17,"context_line":"So, get ahead of things and move to the much faster"},{"line_number":18,"context_line":"recommended by the kernel developers alternative."},{"line_number":19,"context_line":""},{"line_number":20,"context_line":"[1]: https://lore.kernel.org/all/20240904233343.933462-1-yosryahmed@google.com/T/#u"},{"line_number":21,"context_line":""},{"line_number":22,"context_line":"Change-Id: I7c137114dd7585d3179a8d5dee818bb379bbcb1f"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"e180b1a5_efdd429c","line":20,"updated":"2025-05-08 18:35:08.000000000","message":"ya, I was vaguely aware of that zbud is the default which is the other alternitive we could have used.  when i first looked into it z3fold was better but its been a while so i agree better to change now and let zsmalloc bake in our ci before we are forced to move.","commit_id":"1d100182a8dd302ec257e4b6da7819c249915fcc"}],"/PATCHSET_LEVEL":[{"author":{"_account_id":15334,"name":"Stephen Finucane","display_name":"stephenfin","email":"stephenfin@redhat.com","username":"sfinucan"},"change_message_id":"3daee9343cf46ff8e531d0edb18219a36d0c2524","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"34818da0_dc99fe8c","updated":"2025-05-08 18:18:50.000000000","message":"@mnaser@vexxhost.com Any chance you could build on this and skip zswap configuration if it\u0027s not enabled in the kernel? Chatting to @smooney@redhat.com, it seems this is only a performance optimisation so that would be a reasonable thing to do. fwict, you can check like so:\n\n```\ncat /boot/config-$(uname -r) | grep CONFIG_ZSWAP\u003dy\n```\n\nI had started doing so, but seeing as you\u0027re already here 😄","commit_id":"1d100182a8dd302ec257e4b6da7819c249915fcc"},{"author":{"_account_id":1004,"name":"Mohammed Naser","email":"mnaser@vexxhost.com","username":"mnaser"},"change_message_id":"857f90ca3f967609d94373a2fe793c7666c0754e","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"76f332ae_6cf5a9d2","updated":"2025-02-27 21:18:30.000000000","message":"recheck unrelated issue","commit_id":"1d100182a8dd302ec257e4b6da7819c249915fcc"},{"author":{"_account_id":1004,"name":"Mohammed Naser","email":"mnaser@vexxhost.com","username":"mnaser"},"change_message_id":"cec9f432e4d3984860b14ef664b4c41fe381459a","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"027efadb_0daaa661","updated":"2025-05-08 21:42:28.000000000","message":"Thanks for the rebase.  This should totally be fixed sooner than later 😊","commit_id":"5cb2abf79ef103838ab0f922643f4a62ddf16cfb"},{"author":{"_account_id":8556,"name":"Ghanshyam Maan","display_name":"Ghanshyam Maan","email":"gmaan.os14@gmail.com","username":"ghanshyam"},"change_message_id":"2ea5ea9b2c0a880ea322d63f7fd1086fc96aeb8d","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"c54f6f17_701f2b04","updated":"2025-05-09 17:31:48.000000000","message":"lgtm,","commit_id":"5cb2abf79ef103838ab0f922643f4a62ddf16cfb"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"25f1d556ad8fc3a40b7cd9363bcfdbaf1997a3b1","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"f3b9828d_b8e28c2c","updated":"2025-05-08 19:05:58.000000000","message":"th ci logs have rotated so jsut rebased this via the ui to get new results instead of doign a recheck","commit_id":"5cb2abf79ef103838ab0f922643f4a62ddf16cfb"}],"lib/host":[{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"cfa96175586f698c5b5dada0a0b745b30d94648b","unresolved":true,"context_lines":[{"line_number":35,"context_line":"# lz4 is very fast although it does not have the best compression"},{"line_number":36,"context_line":"# zstd has much better compression but more latency"},{"line_number":37,"context_line":"ZSWAP_COMPRESSOR\u003d${ZSWAP_COMPRESSOR:\u003d\"lz4\"}"},{"line_number":38,"context_line":"ZSWAP_ZPOOL\u003d${ZSWAP_ZPOOL:\u003d\"zsmalloc\"}"},{"line_number":39,"context_line":"function configure_zswap {"},{"line_number":40,"context_line":"    if [[ $ENABLE_ZSWAP \u003d\u003d \"True\" ]] ; then"},{"line_number":41,"context_line":"        # Centos 9 stream seems to only support enabling but not run time"}],"source_content_type":"application/x-shellscript","patch_set":1,"id":"e3b49058_278b9d57","line":38,"updated":"2025-05-08 18:35:08.000000000","message":"yep the other option that i dicussed with stephenfin is we could default these to empty string and only set them if you defeint them in your local.conf and or the job\n\nwithout seting these we will jsut use the kernel default which shoudl be ok in most case but i choose thsee default based on some local testing and https://waldorf.waveform.org.uk/2021/6-months-with-the-pi-desktop.html\n\n\nBut I\u0027m totally ok with moving to zsmalloc i dotn think that was aviabel at the time.","commit_id":"1d100182a8dd302ec257e4b6da7819c249915fcc"},{"author":{"_account_id":4146,"name":"Clark Boylan","email":"cboylan@sapwetik.org","username":"cboylan"},"change_message_id":"9f397d5e7b4b3cb74c564ef815dd93bc7e23f646","unresolved":true,"context_lines":[{"line_number":35,"context_line":"# lz4 is very fast although it does not have the best compression"},{"line_number":36,"context_line":"# zstd has much better compression but more latency"},{"line_number":37,"context_line":"ZSWAP_COMPRESSOR\u003d${ZSWAP_COMPRESSOR:\u003d\"lz4\"}"},{"line_number":38,"context_line":"ZSWAP_ZPOOL\u003d${ZSWAP_ZPOOL:\u003d\"zsmalloc\"}"},{"line_number":39,"context_line":"function configure_zswap {"},{"line_number":40,"context_line":"    if [[ $ENABLE_ZSWAP \u003d\u003d \"True\" ]] ; then"},{"line_number":41,"context_line":"        # Centos 9 stream seems to only support enabling but not run time"}],"source_content_type":"application/x-shellscript","patch_set":1,"id":"fb71ed67_a8de88c8","line":38,"in_reply_to":"e3b49058_278b9d57","updated":"2025-05-12 22:22:52.000000000","message":"One limitation of zsmalloc is that it cannot evict pages from swap it can only reject new pages. I think that is probably ok for devstack\u0027s use case primarily as a CI and development tool. But I wanted to call that out here in case there are problems in the future maybe this comment will be seen.\n\nWorst case we revert and/or switch to zbud.\n\nhttps://www.kernel.org/doc/html/v4.18/vm/zswap.html#design","commit_id":"1d100182a8dd302ec257e4b6da7819c249915fcc"}]}