From 0d6526f82c3cdefcca47f73f5fc08dc6f335eac6 Mon Sep 17 00:00:00 2001 From: Tim Chen Date: Mon, 21 Sep 2026 17:37:22 -0700 Subject: sched/cache: Keep nr_pref_llc_running in the runnable domain, to fix LLC mis-scheduling bug alb_break_llc() decides whether to break LLC preference during active load balance. It does so by testing that every runnable fair task on the source rq prefers its LLC: env->src_rq->nr_pref_llc_running == env->src_rq->cfs.h_nr_runnable But the two counters cover different sets. nr_pref_llc_running is updated in account_llc_enqueue()/account_llc_dequeue(), next to cfs_rq->nr_queued, so it follows queued tasks. h_nr_runnable is updated in set_delayed()/ clear_delayed() and drops delay-dequeued tasks. So under DELAY_DEQUEUE, a preferring task that goes to sleep stays counted in nr_pref_llc_running while h_nr_runnable falls. The equality then breaks, alb_break_llc() returns false, and active balance is free to pull a task off its preferred LLC. Active balance only moves runnable tasks, and this is the only LLC check it consults: once the stopper runs, LBF_ACTIVE_LB skips the per-task test in can_migrate_task(). The runnable set is the one we want. Fix it on the counter side. A task should be counted in nr_pref_llc_running exactly while it is both queued on its preferred LLC (pref_llc_queued) and runnable (!sched_delayed). Define that membership once in task_pref_llc_runnable(), and adjust the counter only through pref_llc_running_inc()/pref_llc_running_dec() from the four sites that change either input: account_llc_enqueue(), account_llc_dequeue(), set_delayed() and clear_delayed(). Gating every update on the same predicate keeps the delay, wake and dequeue paths from double-counting or underflowing; see the comments at those sites for the ordering. nr_llc_running and sd->llc_counts are not touched and stay on queued semantics. Fixes: 714059f79ff0 ("sched/cache: Handle moving single tasks to/from their preferred LLC") Closes: https://lore.kernel.org/lkml/20260827135000.735138-1-zhanxusheng@xiaomi.com/ Reported-by: Zhan Xusheng Suggested-by: Chen Yu Signed-off-by: Tim Chen Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Ingo Molnar Reviewed-by: Kayra Cizmeci Cc: # v7.2.x Link: https://patch.msgid.link/06af61afedac32e6477f57feb4d658f6c411c3af.1790035273.git.tim.c.chen@linux.intel.com --- fs/squashfs/zlib_wrapper.c | 137 +++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 137 insertions(+) create mode 100644 fs/squashfs/zlib_wrapper.c (limited to 'fs/squashfs/zlib_wrapper.c') diff --git a/fs/squashfs/zlib_wrapper.c b/fs/squashfs/zlib_wrapper.c new file mode 100644 index 000000000..a29001311 --- /dev/null +++ b/fs/squashfs/zlib_wrapper.c @@ -0,0 +1,137 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * Squashfs - a compressed read only filesystem for Linux + * + * Copyright (c) 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009 + * Phillip Lougher + * + * zlib_wrapper.c + */ + + +#include +#include +#include +#include +#include + +#include "squashfs_fs.h" +#include "squashfs_fs_sb.h" +#include "squashfs.h" +#include "decompressor.h" +#include "page_actor.h" + +static void *zlib_init(struct squashfs_sb_info *dummy, void *buff) +{ + z_stream *stream = kmalloc_obj(z_stream); + if (stream == NULL) + goto failed; + stream->workspace = vmalloc(zlib_inflate_workspacesize()); + if (stream->workspace == NULL) + goto failed; + + return stream; + +failed: + ERROR("Failed to allocate zlib workspace\n"); + kfree(stream); + return ERR_PTR(-ENOMEM); +} + + +static void zlib_free(void *strm) +{ + z_stream *stream = strm; + + if (stream) + vfree(stream->workspace); + kfree(stream); +} + + +static int zlib_uncompress(struct squashfs_sb_info *msblk, void *strm, + struct bio *bio, int offset, int length, + struct squashfs_page_actor *output) +{ + struct bvec_iter_all iter_all = {}; + struct bio_vec *bvec = bvec_init_iter_all(&iter_all); + int zlib_init = 0, error = 0; + z_stream *stream = strm; + + stream->avail_out = PAGE_SIZE; + stream->next_out = squashfs_first_page(output); + stream->avail_in = 0; + + if (IS_ERR(stream->next_out)) { + error = PTR_ERR(stream->next_out); + goto finish; + } + + for (;;) { + int zlib_err; + + if (stream->avail_in == 0) { + const void *data; + int avail; + + if (!bio_next_segment(bio, &iter_all)) { + /* Z_STREAM_END must be reached. */ + error = -EIO; + break; + } + + avail = min(length, ((int)bvec->bv_len) - offset); + data = bvec_virt(bvec); + length -= avail; + stream->next_in = data + offset; + stream->avail_in = avail; + offset = 0; + } + + if (stream->avail_out == 0) { + stream->next_out = squashfs_next_page(output); + if (IS_ERR(stream->next_out)) { + error = PTR_ERR(stream->next_out); + break; + } else if (stream->next_out != NULL) + stream->avail_out = PAGE_SIZE; + } + + if (!zlib_init) { + zlib_err = zlib_inflateInit(stream); + if (zlib_err != Z_OK) { + error = -EIO; + break; + } + zlib_init = 1; + } + + zlib_err = zlib_inflate(stream, Z_SYNC_FLUSH); + if (zlib_err == Z_STREAM_END) + break; + if (zlib_err != Z_OK) { + error = -EIO; + break; + } + } + +finish: + squashfs_finish_page(output); + + if (!error) + if (zlib_inflateEnd(stream) != Z_OK) + error = -EIO; + + return error ? error : stream->total_out; +} + +const struct squashfs_decompressor squashfs_zlib_comp_ops = { + .init = zlib_init, + .free = zlib_free, + .decompress = zlib_uncompress, + .id = ZLIB_COMPRESSION, + .name = "zlib", + .alloc_buffer = 1, + .supported = 1 +}; + -- cgit v1.3.1