From 2a689e4e83bdc90cd00ca21aa28d337d202f4950 Mon Sep 17 00:00:00 2001 From: SeongJae Park Date: Tue, 4 Mar 2025 13:19:08 -0800 Subject: [PATCH 01/16] mm/damon/core: put ops-handled filters to damos->ops_filters damos->ops_filters has introduced to be used for all operations layer handled filters. But DAMON kernel API callers can put any type of DAMOS filters to any of damos->filters and damos->ops_filters. DAMON user-space ABI users have no way to use ->ops_filters at all. Update damos_add_filter(), which should be used by API callers to install DAMOS filters, to add filters to ->filters and ->ops_filters depending on their handling layer. The change forces both API callers and ABI users to use proper lists since ABI users use the API internally. Link: https://lkml.kernel.org/r/20250304211913.53574-5-sj@kernel.org Signed-off-by: SeongJae Park Cc: Jonathan Corbet Signed-off-by: Andrew Morton --- mm/damon/core.c | 17 ++++++++++++++++- 1 file changed, 16 insertions(+), 1 deletion(-) diff --git a/mm/damon/core.c b/mm/damon/core.c index 1daccccb5d67..3fbc31d17239 100644 --- a/mm/damon/core.c +++ b/mm/damon/core.c @@ -281,9 +281,24 @@ struct damos_filter *damos_new_filter(enum damos_filter_type type, return filter; } +static bool damos_filter_for_ops(enum damos_filter_type type) +{ + switch (type) { + case DAMOS_FILTER_TYPE_ADDR: + case DAMOS_FILTER_TYPE_TARGET: + return false; + default: + break; + } + return true; +} + void damos_add_filter(struct damos *s, struct damos_filter *f) { - list_add_tail(&f->list, &s->filters); + if (damos_filter_for_ops(f->type)) + list_add_tail(&f->list, &s->ops_filters); + else + list_add_tail(&f->list, &s->filters); } static void damos_del_filter(struct damos_filter *f) -- 2.50.1 From 627983a55221d429db4fe9ecb75c4ef2f04acd15 Mon Sep 17 00:00:00 2001 From: SeongJae Park Date: Tue, 4 Mar 2025 13:19:09 -0800 Subject: [PATCH 02/16] mm/damon/paddr: support only damos->ops_filters DAMON physical address space operation set implementation (paddr) started handling both damos->filters and damos->ops_filters to avoid breakage during the change for the ->ops_filters setup. Now the change is done, so paddr's support of ->filters is only a waste that can safely be dropped. Remove it. Link: https://lkml.kernel.org/r/20250304211913.53574-6-sj@kernel.org Signed-off-by: SeongJae Park Cc: Jonathan Corbet Signed-off-by: Andrew Morton --- mm/damon/paddr.c | 12 ------------ 1 file changed, 12 deletions(-) diff --git a/mm/damon/paddr.c b/mm/damon/paddr.c index 2b1ea568a431..dded659bb110 100644 --- a/mm/damon/paddr.c +++ b/mm/damon/paddr.c @@ -256,10 +256,6 @@ static bool damos_pa_filter_out(struct damos *scheme, struct folio *folio) if (scheme->core_filters_allowed) return false; - damos_for_each_filter(filter, scheme) { - if (damos_pa_filter_match(filter, folio)) - return !filter->allow; - } damos_for_each_ops_filter(filter, scheme) { if (damos_pa_filter_match(filter, folio)) return !filter->allow; @@ -288,12 +284,6 @@ static unsigned long damon_pa_pageout(struct damon_region *r, struct damos *s, struct folio *folio; /* check access in page level again by default */ - damos_for_each_filter(filter, s) { - if (filter->type == DAMOS_FILTER_TYPE_YOUNG) { - install_young_filter = false; - break; - } - } damos_for_each_ops_filter(filter, s) { if (filter->type == DAMOS_FILTER_TYPE_YOUNG) { install_young_filter = false; @@ -546,8 +536,6 @@ static bool damon_pa_scheme_has_filter(struct damos *s) { struct damos_filter *f; - damos_for_each_filter(f, s) - return true; damos_for_each_ops_filter(f, s) return true; return false; -- 2.50.1 From dd038b728c8a2a0e1a632b767a50f09f076dab79 Mon Sep 17 00:00:00 2001 From: SeongJae Park Date: Tue, 4 Mar 2025 13:19:10 -0800 Subject: [PATCH 03/16] mm/damon: add default allow/reject behavior fields to struct damos Current default allow/reject behavior of filters handling stage has made before introduction of the allow behavior. For allow-filters usage, it is confusing and inefficient. It is more intuitive to decide the default filtering stage allow/reject behavior as opposite to the last filter's behavior. The decision should be made separately for core and operations layers' filtering stages, since last core layer-handled filter is not really a last filter if there are operations layer handling filters. Keeping separate decisions for the two categories can make the logic simpler. Add fields for storing the two decisions. Link: https://lkml.kernel.org/r/20250304211913.53574-7-sj@kernel.org Signed-off-by: SeongJae Park Cc: Jonathan Corbet Signed-off-by: Andrew Morton --- include/linux/damon.h | 3 +++ 1 file changed, 3 insertions(+) diff --git a/include/linux/damon.h b/include/linux/damon.h index 7f76e2e99f37..52559475dbe7 100644 --- a/include/linux/damon.h +++ b/include/linux/damon.h @@ -502,6 +502,9 @@ struct damos { * layer-handled filters. If true, operations layer allows it, too. */ bool core_filters_allowed; + /* whether to reject core/ops filters umatched regions */ + bool core_filters_default_reject; + bool ops_filters_default_reject; /* public: */ struct damos_quota quota; struct damos_watermarks wmarks; -- 2.50.1 From 961df88e4688bf94cfa49d644e49b74d34806d3d Mon Sep 17 00:00:00 2001 From: SeongJae Park Date: Tue, 4 Mar 2025 13:19:11 -0800 Subject: [PATCH 04/16] mm/damon/core: set damos_filter default allowance behavior based on installed filters Decide whether to allow or reject by default on core and opertions layer handled filters evaluation stages. It is decided as the opposite of the last installed filter's behavior. If there is no filter at all, allow by default. If there is any operations layer handled filters, core layer's filtering stage sets allowing as the default behavior regardless of the last filter of core layer-handling ones, since the last filter of core layer handled filters in the case is not really the last filter of the entire filtering stage. Also, make the core layer's DAMOS filters handling stage uses the newly set behavior field. [sj@kernel.org: setup damos->{core,ops}_filters_default_reject for initial start] Link: https://lkml.kernel.org/r/20250315222610.35245-1-sj@kernel.org Link: https://lkml.kernel.org/r/20250304211913.53574-8-sj@kernel.org Signed-off-by: SeongJae Park Cc: Jonathan Corbet Signed-off-by: Andrew Morton --- mm/damon/core.c | 41 ++++++++++++++++++++++++++++++++++++----- 1 file changed, 36 insertions(+), 5 deletions(-) diff --git a/mm/damon/core.c b/mm/damon/core.c index 3fbc31d17239..511c464adcc5 100644 --- a/mm/damon/core.c +++ b/mm/damon/core.c @@ -518,7 +518,7 @@ struct damon_ctx *damon_new_ctx(void) ctx->attrs.ops_update_interval = 60 * 1000 * 1000; ctx->passed_sample_intervals = 0; - /* These will be set from kdamond_init_intervals_sis() */ + /* These will be set from kdamond_init_ctx() */ ctx->next_aggregation_sis = 0; ctx->next_ops_update_sis = 0; @@ -891,6 +891,32 @@ static int damos_commit_ops_filters(struct damos *dst, struct damos *src) return 0; } +/** + * damos_filters_default_reject() - decide whether to reject memory that didn't + * match with any given filter. + * @filters: Given DAMOS filters of a group. + */ +static bool damos_filters_default_reject(struct list_head *filters) +{ + struct damos_filter *last_filter; + + if (list_empty(filters)) + return false; + last_filter = list_last_entry(filters, struct damos_filter, list); + return last_filter->allow; +} + +static void damos_set_filters_default_reject(struct damos *s) +{ + if (!list_empty(&s->ops_filters)) + s->core_filters_default_reject = false; + else + s->core_filters_default_reject = + damos_filters_default_reject(&s->filters); + s->ops_filters_default_reject = + damos_filters_default_reject(&s->ops_filters); +} + static int damos_commit_filters(struct damos *dst, struct damos *src) { int err; @@ -898,7 +924,11 @@ static int damos_commit_filters(struct damos *dst, struct damos *src) err = damos_commit_core_filters(dst, src); if (err) return err; - return damos_commit_ops_filters(dst, src); + err = damos_commit_ops_filters(dst, src); + if (err) + return err; + damos_set_filters_default_reject(dst); + return 0; } static struct damos *damon_nth_scheme(int n, struct damon_ctx *ctx) @@ -1580,7 +1610,7 @@ static bool damos_filter_out(struct damon_ctx *ctx, struct damon_target *t, return !filter->allow; } } - return false; + return s->core_filters_default_reject; } /* @@ -2315,7 +2345,7 @@ static int kdamond_wait_activation(struct damon_ctx *ctx) return -EBUSY; } -static void kdamond_init_intervals_sis(struct damon_ctx *ctx) +static void kdamond_init_ctx(struct damon_ctx *ctx) { unsigned long sample_interval = ctx->attrs.sample_interval ? ctx->attrs.sample_interval : 1; @@ -2333,6 +2363,7 @@ static void kdamond_init_intervals_sis(struct damon_ctx *ctx) apply_interval = scheme->apply_interval_us ? scheme->apply_interval_us : ctx->attrs.aggr_interval; scheme->next_apply_sis = apply_interval / sample_interval; + damos_set_filters_default_reject(scheme); } } @@ -2350,7 +2381,7 @@ static int kdamond_fn(void *data) pr_debug("kdamond (%d) starts\n", current->pid); complete(&ctx->kdamond_started); - kdamond_init_intervals_sis(ctx); + kdamond_init_ctx(ctx); if (ctx->ops.init) ctx->ops.init(ctx); -- 2.50.1 From a54c42f6873d0fc9d7667433112e34a732c3b228 Mon Sep 17 00:00:00 2001 From: SeongJae Park Date: Tue, 4 Mar 2025 13:19:12 -0800 Subject: [PATCH 05/16] mm/damon/paddr: respect ops_filters_default_reject Use damos->ops_filters_default_reject, which is set based on the installed filters' behaviors, from physical address space DAMON operations set. Link: https://lkml.kernel.org/r/20250304211913.53574-9-sj@kernel.org Signed-off-by: SeongJae Park Cc: Jonathan Corbet Signed-off-by: Andrew Morton --- mm/damon/paddr.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mm/damon/paddr.c b/mm/damon/paddr.c index dded659bb110..fba8b3c8ba30 100644 --- a/mm/damon/paddr.c +++ b/mm/damon/paddr.c @@ -260,7 +260,7 @@ static bool damos_pa_filter_out(struct damos *scheme, struct folio *folio) if (damos_pa_filter_match(filter, folio)) return !filter->allow; } - return false; + return scheme->ops_filters_default_reject; } static bool damon_pa_invalid_damos_folio(struct folio *folio, struct damos *s) -- 2.50.1 From 9ea705a54badbc3f33daf60c2da989c24c467e77 Mon Sep 17 00:00:00 2001 From: SeongJae Park Date: Tue, 4 Mar 2025 13:19:13 -0800 Subject: [PATCH 06/16] Docs/mm/damon/design: update for changed filter-default behavior Update the design documentation for changed DAMOS filters default allowance behaviors. Link: https://lkml.kernel.org/r/20250304211913.53574-10-sj@kernel.org Signed-off-by: SeongJae Park Cc: Jonathan Corbet Signed-off-by: Andrew Morton --- Documentation/mm/damon/design.rst | 10 +++------- 1 file changed, 3 insertions(+), 7 deletions(-) diff --git a/Documentation/mm/damon/design.rst b/Documentation/mm/damon/design.rst index e6fd3b604e70..aae3a691ee69 100644 --- a/Documentation/mm/damon/design.rst +++ b/Documentation/mm/damon/design.rst @@ -631,9 +631,10 @@ When multiple filters are installed, the group of filters that handled by the core layer are evaluated first. After that, the group of filters that handled by the operations layer are evaluated. Filters in each of the groups are evaluated in the installed order. If a part of memory is matched to one of the -filter, next filters are ignored. If the memory passes through the filters +filter, next filters are ignored. If the part passes through the filters evaluation stage because it is not matched to any of the filters, applying the -scheme's action to it is allowed, same to the behavior when no filter exists. +scheme's action to it depends on the last filter's allowance type. If the last +filter was for allowing, the part of memory will be rejected, and vice versa. For example, let's assume 1) a filter for allowing anonymous pages and 2) another filter for rejecting young pages are installed in the order. If a page @@ -645,11 +646,6 @@ second reject-filter blocks it. If the page is neither anonymous nor young, the page will pass through the filters evaluation stage since there is no matching filter, and the action will be applied to the page. -Note that the action can equally be applied to memory that either explicitly -filter-allowed or filters evaluation stage passed. It means that installing -allow-filters at the end of the list makes no practical change but only -filters-checking overhead. - Below ``type`` of filters are currently supported. - Core layer handled -- 2.50.1 From ac55b38fe2f9b486031439c5c4ed7fce07d0d838 Mon Sep 17 00:00:00 2001 From: Liu Ye Date: Wed, 5 Mar 2025 15:17:59 +0800 Subject: [PATCH 07/16] mm/shrinker: fix name consistency issue in shrinker_debugfs_rename() MIME-Version: 1.0 Content-Type: text/plain; charset=utf8 Content-Transfer-Encoding: 8bit After calling debugfs_change_name function, the return value should be checked and the old name restored. If debugfs_change_name fails, the new name memory should be freed. The effect is that the shrinker->name is not consistent with the name displayed in debugfs. Link: https://lkml.kernel.org/r/20250305071759.661055-1-liuye@kylinos.cn Signed-off-by: Liu Ye Reviewed-by: Muchun Song Reviewed-by：Qi Zheng Cc: Dave Chinner Cc: Muchun Song Cc: Qi Zheng Signed-off-by: Andrew Morton --- mm/shrinker_debug.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/mm/shrinker_debug.c b/mm/shrinker_debug.c index 794bd433cce0..20eaee3e97f7 100644 --- a/mm/shrinker_debug.c +++ b/mm/shrinker_debug.c @@ -214,10 +214,14 @@ int shrinker_debugfs_rename(struct shrinker *shrinker, const char *fmt, ...) ret = debugfs_change_name(shrinker->debugfs_entry, "%s-%d", shrinker->name, shrinker->debugfs_id); + if (ret) { + shrinker->name = old; + kfree_const(new); + } else { + kfree_const(old); + } mutex_unlock(&shrinker_mutex); - kfree_const(old); - return ret; } EXPORT_SYMBOL(shrinker_debugfs_rename); -- 2.50.1 From 9bbe033c75a56d72fc35e7c8ca6f3258d9782fa5 Mon Sep 17 00:00:00 2001 From: Yosry Ahmed Date: Wed, 5 Mar 2025 06:11:29 +0000 Subject: [PATCH 08/16] mm: zpool: add interfaces for object read/write APIs Patch series "Switch zswap to object read/write APIs". This patch series updates zswap to use the new object read/write APIs defined by zsmalloc in [1], and remove the old object mapping APIs and the related code from zpool and zsmalloc. This patch (of 5): Zsmalloc introduced new APIs to read/write objects besides mapping them. Add the necessary zpool interfaces. Link: https://lkml.kernel.org/r/20250305061134.4105762-1-yosry.ahmed@linux.dev Link: https://lkml.kernel.org/r/20250305061134.4105762-2-yosry.ahmed@linux.dev Signed-off-by: Yosry Ahmed Reviewed-by: Sergey Senozhatsky Acked-by: Johannes Weiner Acked-by: Nhat Pham Cc: Chengming Zhou Cc: Herbert Xu Cc: Minchan Kim Cc: Peter Zijlstra Cc: Thomas Gleixner Signed-off-by: Andrew Morton --- include/linux/zpool.h | 17 +++++++++++++++ mm/zpool.c | 48 +++++++++++++++++++++++++++++++++++++++++++ mm/zsmalloc.c | 21 +++++++++++++++++++ 3 files changed, 86 insertions(+) diff --git a/include/linux/zpool.h b/include/linux/zpool.h index 5e6dc46b8cc4..1784e735ee04 100644 --- a/include/linux/zpool.h +++ b/include/linux/zpool.h @@ -52,6 +52,16 @@ void *zpool_map_handle(struct zpool *pool, unsigned long handle, void zpool_unmap_handle(struct zpool *pool, unsigned long handle); + +void *zpool_obj_read_begin(struct zpool *zpool, unsigned long handle, + void *local_copy); + +void zpool_obj_read_end(struct zpool *zpool, unsigned long handle, + void *handle_mem); + +void zpool_obj_write(struct zpool *zpool, unsigned long handle, + void *handle_mem, size_t mem_len); + u64 zpool_get_total_pages(struct zpool *pool); @@ -90,6 +100,13 @@ struct zpool_driver { enum zpool_mapmode mm); void (*unmap)(void *pool, unsigned long handle); + void *(*obj_read_begin)(void *pool, unsigned long handle, + void *local_copy); + void (*obj_read_end)(void *pool, unsigned long handle, + void *handle_mem); + void (*obj_write)(void *pool, unsigned long handle, + void *handle_mem, size_t mem_len); + u64 (*total_pages)(void *pool); }; diff --git a/mm/zpool.c b/mm/zpool.c index 4bbd12d4b659..378c2d1e5638 100644 --- a/mm/zpool.c +++ b/mm/zpool.c @@ -320,6 +320,54 @@ void zpool_unmap_handle(struct zpool *zpool, unsigned long handle) zpool->driver->unmap(zpool->pool, handle); } +/** + * zpool_obj_read_begin() - Start reading from a previously allocated handle. + * @zpool: The zpool that the handle was allocated from + * @handle: The handle to read from + * @local_copy: A local buffer to use if needed. + * + * This starts a read operation of a previously allocated handle. The passed + * @local_copy buffer may be used if needed by copying the memory into. + * zpool_obj_read_end() MUST be called after the read is completed to undo any + * actions taken (e.g. release locks). + * + * Returns: A pointer to the handle memory to be read, if @local_copy is used, + * the returned pointer is @local_copy. + */ +void *zpool_obj_read_begin(struct zpool *zpool, unsigned long handle, + void *local_copy) +{ + return zpool->driver->obj_read_begin(zpool->pool, handle, local_copy); +} + +/** + * zpool_obj_read_end() - Finish reading from a previously allocated handle. + * @zpool: The zpool that the handle was allocated from + * @handle: The handle to read from + * @handle_mem: The pointer returned by zpool_obj_read_begin() + * + * Finishes a read operation previously started by zpool_obj_read_begin(). + */ +void zpool_obj_read_end(struct zpool *zpool, unsigned long handle, + void *handle_mem) +{ + zpool->driver->obj_read_end(zpool->pool, handle, handle_mem); +} + +/** + * zpool_obj_write() - Write to a previously allocated handle. + * @zpool: The zpool that the handle was allocated from + * @handle: The handle to read from + * @handle_mem: The memory to copy from into the handle. + * @mem_len: The length of memory to be written. + * + */ +void zpool_obj_write(struct zpool *zpool, unsigned long handle, + void *handle_mem, size_t mem_len) +{ + zpool->driver->obj_write(zpool->pool, handle, handle_mem, mem_len); +} + /** * zpool_get_total_pages() - The total size of the pool * @zpool: The zpool to check diff --git a/mm/zsmalloc.c b/mm/zsmalloc.c index 63c99db71dc1..d84b300db64e 100644 --- a/mm/zsmalloc.c +++ b/mm/zsmalloc.c @@ -507,6 +507,24 @@ static void zs_zpool_unmap(void *pool, unsigned long handle) zs_unmap_object(pool, handle); } +static void *zs_zpool_obj_read_begin(void *pool, unsigned long handle, + void *local_copy) +{ + return zs_obj_read_begin(pool, handle, local_copy); +} + +static void zs_zpool_obj_read_end(void *pool, unsigned long handle, + void *handle_mem) +{ + zs_obj_read_end(pool, handle, handle_mem); +} + +static void zs_zpool_obj_write(void *pool, unsigned long handle, + void *handle_mem, size_t mem_len) +{ + zs_obj_write(pool, handle, handle_mem, mem_len); +} + static u64 zs_zpool_total_pages(void *pool) { return zs_get_total_pages(pool); @@ -522,6 +540,9 @@ static struct zpool_driver zs_zpool_driver = { .free = zs_zpool_free, .map = zs_zpool_map, .unmap = zs_zpool_unmap, + .obj_read_begin = zs_zpool_obj_read_begin, + .obj_read_end = zs_zpool_obj_read_end, + .obj_write = zs_zpool_obj_write, .total_pages = zs_zpool_total_pages, }; -- 2.50.1 From 7d4c9629b74ff7ad3b58e57324e235d710e55c21 Mon Sep 17 00:00:00 2001 From: Yosry Ahmed Date: Wed, 5 Mar 2025 06:11:30 +0000 Subject: [PATCH 09/16] mm: zswap: use object read/write APIs instead of object mapping APIs Use the new object read/write APIs instead of mapping APIs. On compress side, zpool_obj_write() is more concise and provides exactly what zswap needs to write the compressed object to the zpool, instead of map->copy->unmap. On the decompress side, zpool_obj_read_begin() is sleepable, which allows avoiding the memcpy() for zsmalloc and slightly simplifying the code by: - Avoiding checking if the zpool driver is sleepable, reducing special cases and shrinking the huge comment. - Having a single zpool_obj_read_end() call rather than multiple conditional zpool_unmap_handle() calls. The !virt_addr_valid() case can be removed in the future if the crypto API supports kmap addresses or by using kmap_to_page(), completely eliminating the memcpy() path in zswap_decompress(). This a step toward that. In that spirit, opportunistically make the comment more specific about the kmap case instead of generic non-linear addresses. This is the only case that needs to be handled in practice, and the generic comment makes it seem like a bigger problem that it actually is. Link: https://lkml.kernel.org/r/20250305061134.4105762-3-yosry.ahmed@linux.dev Signed-off-by: Yosry Ahmed Acked-by: Johannes Weiner Acked-by: Nhat Pham Cc: Chengming Zhou Cc: Herbert Xu Cc: Minchan Kim Cc: Peter Zijlstra Cc: Sergey Senozhatsky Cc: Thomas Gleixner Signed-off-by: Andrew Morton --- mm/zswap.c | 33 +++++++++++++-------------------- 1 file changed, 13 insertions(+), 20 deletions(-) diff --git a/mm/zswap.c b/mm/zswap.c index 8a1ded8fa973..7de54f105d04 100644 --- a/mm/zswap.c +++ b/mm/zswap.c @@ -930,7 +930,6 @@ static bool zswap_compress(struct page *page, struct zswap_entry *entry, unsigned int dlen = PAGE_SIZE; unsigned long handle; struct zpool *zpool; - char *buf; gfp_t gfp; u8 *dst; @@ -972,10 +971,7 @@ static bool zswap_compress(struct page *page, struct zswap_entry *entry, if (alloc_ret) goto unlock; - buf = zpool_map_handle(zpool, handle, ZPOOL_MM_WO); - memcpy(buf, dst, dlen); - zpool_unmap_handle(zpool, handle); - + zpool_obj_write(zpool, handle, dst, dlen); entry->handle = handle; entry->length = dlen; @@ -996,24 +992,22 @@ static void zswap_decompress(struct zswap_entry *entry, struct folio *folio) struct zpool *zpool = entry->pool->zpool; struct scatterlist input, output; struct crypto_acomp_ctx *acomp_ctx; - u8 *src; + u8 *src, *obj; acomp_ctx = acomp_ctx_get_cpu_lock(entry->pool); - src = zpool_map_handle(zpool, entry->handle, ZPOOL_MM_RO); + obj = zpool_obj_read_begin(zpool, entry->handle, acomp_ctx->buffer); + /* - * If zpool_map_handle is atomic, we cannot reliably utilize its mapped buffer - * to do crypto_acomp_decompress() which might sleep. In such cases, we must - * resort to copying the buffer to a temporary one. - * Meanwhile, zpool_map_handle() might return a non-linearly mapped buffer, - * such as a kmap address of high memory or even ever a vmap address. - * However, sg_init_one is only equipped to handle linearly mapped low memory. - * In such cases, we also must copy the buffer to a temporary and lowmem one. + * zpool_obj_read_begin() might return a kmap address of highmem when + * acomp_ctx->buffer is not used. However, sg_init_one() does not + * handle highmem addresses, so copy the object to acomp_ctx->buffer. */ - if ((acomp_ctx->is_sleepable && !zpool_can_sleep_mapped(zpool)) || - !virt_addr_valid(src)) { - memcpy(acomp_ctx->buffer, src, entry->length); + if (virt_addr_valid(obj)) { + src = obj; + } else { + WARN_ON_ONCE(obj == acomp_ctx->buffer); + memcpy(acomp_ctx->buffer, obj, entry->length); src = acomp_ctx->buffer; - zpool_unmap_handle(zpool, entry->handle); } sg_init_one(&input, src, entry->length); @@ -1023,8 +1017,7 @@ static void zswap_decompress(struct zswap_entry *entry, struct folio *folio) BUG_ON(crypto_wait_req(crypto_acomp_decompress(acomp_ctx->req), &acomp_ctx->wait)); BUG_ON(acomp_ctx->req->dlen != PAGE_SIZE); - if (src != acomp_ctx->buffer) - zpool_unmap_handle(zpool, entry->handle); + zpool_obj_read_end(zpool, entry->handle, obj); acomp_ctx_put_unlock(acomp_ctx); } -- 2.50.1 From fcbea574754c63f7035d0c4ef7dfb161b60b5bde Mon Sep 17 00:00:00 2001 From: Yosry Ahmed Date: Wed, 5 Mar 2025 06:11:31 +0000 Subject: [PATCH 10/16] mm: zpool: remove object mapping APIs zpool_map_handle(), zpool_unmap_handle(), and zpool_can_sleep_mapped() are no longer used. Remove them with the underlying driver callbacks. Link: https://lkml.kernel.org/r/20250305061134.4105762-4-yosry.ahmed@linux.dev Signed-off-by: Yosry Ahmed Reviewed-by: Sergey Senozhatsky Acked-by: Johannes Weiner Acked-by: Nhat Pham Cc: Chengming Zhou Cc: Herbert Xu Cc: Minchan Kim Cc: Peter Zijlstra Cc: Thomas Gleixner Signed-off-by: Andrew Morton --- include/linux/zpool.h | 30 --------------------- mm/zpool.c | 61 ------------------------------------------- mm/zsmalloc.c | 27 ------------------- 3 files changed, 118 deletions(-) diff --git a/include/linux/zpool.h b/include/linux/zpool.h index 1784e735ee04..2c8a9d2654f6 100644 --- a/include/linux/zpool.h +++ b/include/linux/zpool.h @@ -13,25 +13,6 @@ struct zpool; -/* - * Control how a handle is mapped. It will be ignored if the - * implementation does not support it. Its use is optional. - * Note that this does not refer to memory protection, it - * refers to how the memory will be copied in/out if copying - * is necessary during mapping; read-write is the safest as - * it copies the existing memory in on map, and copies the - * changed memory back out on unmap. Write-only does not copy - * in the memory and should only be used for initialization. - * If in doubt, use ZPOOL_MM_DEFAULT which is read-write. - */ -enum zpool_mapmode { - ZPOOL_MM_RW, /* normal read-write mapping */ - ZPOOL_MM_RO, /* read-only (no copy-out at unmap time) */ - ZPOOL_MM_WO, /* write-only (no copy-in at map time) */ - - ZPOOL_MM_DEFAULT = ZPOOL_MM_RW -}; - bool zpool_has_pool(char *type); struct zpool *zpool_create_pool(const char *type, const char *name, gfp_t gfp); @@ -47,12 +28,6 @@ int zpool_malloc(struct zpool *pool, size_t size, gfp_t gfp, void zpool_free(struct zpool *pool, unsigned long handle); -void *zpool_map_handle(struct zpool *pool, unsigned long handle, - enum zpool_mapmode mm); - -void zpool_unmap_handle(struct zpool *pool, unsigned long handle); - - void *zpool_obj_read_begin(struct zpool *zpool, unsigned long handle, void *local_copy); @@ -95,11 +70,6 @@ struct zpool_driver { unsigned long *handle); void (*free)(void *pool, unsigned long handle); - bool sleep_mapped; - void *(*map)(void *pool, unsigned long handle, - enum zpool_mapmode mm); - void (*unmap)(void *pool, unsigned long handle); - void *(*obj_read_begin)(void *pool, unsigned long handle, void *local_copy); void (*obj_read_end)(void *pool, unsigned long handle, diff --git a/mm/zpool.c b/mm/zpool.c index 378c2d1e5638..4fc665b42f5e 100644 --- a/mm/zpool.c +++ b/mm/zpool.c @@ -277,49 +277,6 @@ void zpool_free(struct zpool *zpool, unsigned long handle) zpool->driver->free(zpool->pool, handle); } -/** - * zpool_map_handle() - Map a previously allocated handle into memory - * @zpool: The zpool that the handle was allocated from - * @handle: The handle to map - * @mapmode: How the memory should be mapped - * - * This maps a previously allocated handle into memory. The @mapmode - * param indicates to the implementation how the memory will be - * used, i.e. read-only, write-only, read-write. If the - * implementation does not support it, the memory will be treated - * as read-write. - * - * This may hold locks, disable interrupts, and/or preemption, - * and the zpool_unmap_handle() must be called to undo those - * actions. The code that uses the mapped handle should complete - * its operations on the mapped handle memory quickly and unmap - * as soon as possible. As the implementation may use per-cpu - * data, multiple handles should not be mapped concurrently on - * any cpu. - * - * Returns: A pointer to the handle's mapped memory area. - */ -void *zpool_map_handle(struct zpool *zpool, unsigned long handle, - enum zpool_mapmode mapmode) -{ - return zpool->driver->map(zpool->pool, handle, mapmode); -} - -/** - * zpool_unmap_handle() - Unmap a previously mapped handle - * @zpool: The zpool that the handle was allocated from - * @handle: The handle to unmap - * - * This unmaps a previously mapped handle. Any locks or other - * actions that the implementation took in zpool_map_handle() - * will be undone here. The memory area returned from - * zpool_map_handle() should no longer be used after this. - */ -void zpool_unmap_handle(struct zpool *zpool, unsigned long handle) -{ - zpool->driver->unmap(zpool->pool, handle); -} - /** * zpool_obj_read_begin() - Start reading from a previously allocated handle. * @zpool: The zpool that the handle was allocated from @@ -381,23 +338,5 @@ u64 zpool_get_total_pages(struct zpool *zpool) return zpool->driver->total_pages(zpool->pool); } -/** - * zpool_can_sleep_mapped - Test if zpool can sleep when do mapped. - * @zpool: The zpool to test - * - * Some allocators enter non-preemptible context in ->map() callback (e.g. - * disable pagefaults) and exit that context in ->unmap(), which limits what - * we can do with the mapped object. For instance, we cannot wait for - * asynchronous crypto API to decompress such an object or take mutexes - * since those will call into the scheduler. This function tells us whether - * we use such an allocator. - * - * Returns: true if zpool can sleep; false otherwise. - */ -bool zpool_can_sleep_mapped(struct zpool *zpool) -{ - return zpool->driver->sleep_mapped; -} - MODULE_AUTHOR("Dan Streetman "); MODULE_DESCRIPTION("Common API for compressed memory storage"); diff --git a/mm/zsmalloc.c b/mm/zsmalloc.c index d84b300db64e..56d6ed5c675b 100644 --- a/mm/zsmalloc.c +++ b/mm/zsmalloc.c @@ -482,31 +482,6 @@ static void zs_zpool_free(void *pool, unsigned long handle) zs_free(pool, handle); } -static void *zs_zpool_map(void *pool, unsigned long handle, - enum zpool_mapmode mm) -{ - enum zs_mapmode zs_mm; - - switch (mm) { - case ZPOOL_MM_RO: - zs_mm = ZS_MM_RO; - break; - case ZPOOL_MM_WO: - zs_mm = ZS_MM_WO; - break; - case ZPOOL_MM_RW: - default: - zs_mm = ZS_MM_RW; - break; - } - - return zs_map_object(pool, handle, zs_mm); -} -static void zs_zpool_unmap(void *pool, unsigned long handle) -{ - zs_unmap_object(pool, handle); -} - static void *zs_zpool_obj_read_begin(void *pool, unsigned long handle, void *local_copy) { @@ -538,8 +513,6 @@ static struct zpool_driver zs_zpool_driver = { .malloc_support_movable = true, .malloc = zs_zpool_malloc, .free = zs_zpool_free, - .map = zs_zpool_map, - .unmap = zs_zpool_unmap, .obj_read_begin = zs_zpool_obj_read_begin, .obj_read_end = zs_zpool_obj_read_end, .obj_write = zs_zpool_obj_write, -- 2.50.1 From 07864f1a57fb1f798a7d21f13e4929c9cb52daf7 Mon Sep 17 00:00:00 2001 From: Yosry Ahmed Date: Wed, 5 Mar 2025 06:11:32 +0000 Subject: [PATCH 11/16] mm: zsmalloc: remove object mapping APIs and per-CPU map areas zs_map_object() and zs_unmap_object() are no longer used, remove them. Since these are the only users of per-CPU mapping_areas, remove them and the associated CPU hotplug callbacks too. [yosry.ahmed@linux.dev: update the docs] Link: https://lkml.kernel.org/r/Z8ier-ZZp8T6MOTH@google.com Link: https://lkml.kernel.org/r/20250305061134.4105762-5-yosry.ahmed@linux.dev Signed-off-by: Yosry Ahmed Acked-by: Sergey Senozhatsky Acked-by: Johannes Weiner Acked-by: Nhat Pham Cc: Chengming Zhou Cc: Herbert Xu Cc: Minchan Kim Cc: Peter Zijlstra Cc: Thomas Gleixner Signed-off-by: Andrew Morton --- Documentation/mm/zsmalloc.rst | 5 +- include/linux/cpuhotplug.h | 1 - include/linux/zsmalloc.h | 21 ---- mm/zsmalloc.c | 226 +--------------------------------- 4 files changed, 3 insertions(+), 250 deletions(-) diff --git a/Documentation/mm/zsmalloc.rst b/Documentation/mm/zsmalloc.rst index 76902835e68e..d2bbecd78e14 100644 --- a/Documentation/mm/zsmalloc.rst +++ b/Documentation/mm/zsmalloc.rst @@ -27,9 +27,8 @@ Instead, it returns an opaque handle (unsigned long) which encodes actual location of the allocated object. The reason for this indirection is that zsmalloc does not keep zspages permanently mapped since that would cause issues on 32-bit systems where the VA region for kernel space mappings -is very small. So, before using the allocating memory, the object has to -be mapped using zs_map_object() to get a usable pointer and subsequently -unmapped using zs_unmap_object(). +is very small. So, using the allocated memory should be done through the +proper handle-based APIs. stat ==== diff --git a/include/linux/cpuhotplug.h b/include/linux/cpuhotplug.h index 6cc5e484547c..1987400000b4 100644 --- a/include/linux/cpuhotplug.h +++ b/include/linux/cpuhotplug.h @@ -116,7 +116,6 @@ enum cpuhp_state { CPUHP_NET_IUCV_PREPARE, CPUHP_ARM_BL_PREPARE, CPUHP_TRACE_RB_PREPARE, - CPUHP_MM_ZS_PREPARE, CPUHP_MM_ZSWP_POOL_PREPARE, CPUHP_KVM_PPC_BOOK3S_PREPARE, CPUHP_ZCOMP_PREPARE, diff --git a/include/linux/zsmalloc.h b/include/linux/zsmalloc.h index 7d70983cf398..c26baf9fb331 100644 --- a/include/linux/zsmalloc.h +++ b/include/linux/zsmalloc.h @@ -16,23 +16,6 @@ #include -/* - * zsmalloc mapping modes - * - * NOTE: These only make a difference when a mapped object spans pages. - */ -enum zs_mapmode { - ZS_MM_RW, /* normal read-write mapping */ - ZS_MM_RO, /* read-only (no copy-out at unmap time) */ - ZS_MM_WO /* write-only (no copy-in at map time) */ - /* - * NOTE: ZS_MM_WO should only be used for initializing new - * (uninitialized) allocations. Partial writes to already - * initialized allocations should use ZS_MM_RW to preserve the - * existing data. - */ -}; - struct zs_pool_stats { /* How many pages were migrated (freed) */ atomic_long_t pages_compacted; @@ -48,10 +31,6 @@ void zs_free(struct zs_pool *pool, unsigned long obj); size_t zs_huge_class_size(struct zs_pool *pool); -void *zs_map_object(struct zs_pool *pool, unsigned long handle, - enum zs_mapmode mm); -void zs_unmap_object(struct zs_pool *pool, unsigned long handle); - unsigned long zs_get_total_pages(struct zs_pool *pool); unsigned long zs_compact(struct zs_pool *pool); diff --git a/mm/zsmalloc.c b/mm/zsmalloc.c index 56d6ed5c675b..cd1c2a8ffef0 100644 --- a/mm/zsmalloc.c +++ b/mm/zsmalloc.c @@ -281,13 +281,6 @@ struct zspage { struct zspage_lock zsl; }; -struct mapping_area { - local_lock_t lock; - char *vm_buf; /* copy buffer for objects that span pages */ - char *vm_addr; /* address of kmap_local_page()'ed pages */ - enum zs_mapmode vm_mm; /* mapping mode */ -}; - static void zspage_lock_init(struct zspage *zspage) { static struct lock_class_key __key; @@ -522,11 +515,6 @@ static struct zpool_driver zs_zpool_driver = { MODULE_ALIAS("zpool-zsmalloc"); #endif /* CONFIG_ZPOOL */ -/* per-cpu VM mapping areas for zspage accesses that cross page boundaries */ -static DEFINE_PER_CPU(struct mapping_area, zs_map_area) = { - .lock = INIT_LOCAL_LOCK(lock), -}; - static inline bool __maybe_unused is_first_zpdesc(struct zpdesc *zpdesc) { return PagePrivate(zpdesc_page(zpdesc)); @@ -1111,93 +1099,6 @@ static struct zspage *find_get_zspage(struct size_class *class) return zspage; } -static inline int __zs_cpu_up(struct mapping_area *area) -{ - /* - * Make sure we don't leak memory if a cpu UP notification - * and zs_init() race and both call zs_cpu_up() on the same cpu - */ - if (area->vm_buf) - return 0; - area->vm_buf = kmalloc(ZS_MAX_ALLOC_SIZE, GFP_KERNEL); - if (!area->vm_buf) - return -ENOMEM; - return 0; -} - -static inline void __zs_cpu_down(struct mapping_area *area) -{ - kfree(area->vm_buf); - area->vm_buf = NULL; -} - -static void *__zs_map_object(struct mapping_area *area, - struct zpdesc *zpdescs[2], int off, int size) -{ - size_t sizes[2]; - char *buf = area->vm_buf; - - /* disable page faults to match kmap_local_page() return conditions */ - pagefault_disable(); - - /* no read fastpath */ - if (area->vm_mm == ZS_MM_WO) - goto out; - - sizes[0] = PAGE_SIZE - off; - sizes[1] = size - sizes[0]; - - /* copy object to per-cpu buffer */ - memcpy_from_page(buf, zpdesc_page(zpdescs[0]), off, sizes[0]); - memcpy_from_page(buf + sizes[0], zpdesc_page(zpdescs[1]), 0, sizes[1]); -out: - return area->vm_buf; -} - -static void __zs_unmap_object(struct mapping_area *area, - struct zpdesc *zpdescs[2], int off, int size) -{ - size_t sizes[2]; - char *buf; - - /* no write fastpath */ - if (area->vm_mm == ZS_MM_RO) - goto out; - - buf = area->vm_buf; - buf = buf + ZS_HANDLE_SIZE; - size -= ZS_HANDLE_SIZE; - off += ZS_HANDLE_SIZE; - - sizes[0] = PAGE_SIZE - off; - sizes[1] = size - sizes[0]; - - /* copy per-cpu buffer to object */ - memcpy_to_page(zpdesc_page(zpdescs[0]), off, buf, sizes[0]); - memcpy_to_page(zpdesc_page(zpdescs[1]), 0, buf + sizes[0], sizes[1]); - -out: - /* enable page faults to match kunmap_local() return conditions */ - pagefault_enable(); -} - -static int zs_cpu_prepare(unsigned int cpu) -{ - struct mapping_area *area; - - area = &per_cpu(zs_map_area, cpu); - return __zs_cpu_up(area); -} - -static int zs_cpu_dead(unsigned int cpu) -{ - struct mapping_area *area; - - area = &per_cpu(zs_map_area, cpu); - __zs_cpu_down(area); - return 0; -} - static bool can_merge(struct size_class *prev, int pages_per_zspage, int objs_per_zspage) { @@ -1245,117 +1146,6 @@ unsigned long zs_get_total_pages(struct zs_pool *pool) } EXPORT_SYMBOL_GPL(zs_get_total_pages); -/** - * zs_map_object - get address of allocated object from handle. - * @pool: pool from which the object was allocated - * @handle: handle returned from zs_malloc - * @mm: mapping mode to use - * - * Before using an object allocated from zs_malloc, it must be mapped using - * this function. When done with the object, it must be unmapped using - * zs_unmap_object. - * - * Only one object can be mapped per cpu at a time. There is no protection - * against nested mappings. - * - * This function returns with preemption and page faults disabled. - */ -void *zs_map_object(struct zs_pool *pool, unsigned long handle, - enum zs_mapmode mm) -{ - struct zspage *zspage; - struct zpdesc *zpdesc; - unsigned long obj, off; - unsigned int obj_idx; - - struct size_class *class; - struct mapping_area *area; - struct zpdesc *zpdescs[2]; - void *ret; - - /* - * Because we use per-cpu mapping areas shared among the - * pools/users, we can't allow mapping in interrupt context - * because it can corrupt another users mappings. - */ - BUG_ON(in_interrupt()); - - /* It guarantees it can get zspage from handle safely */ - read_lock(&pool->lock); - obj = handle_to_obj(handle); - obj_to_location(obj, &zpdesc, &obj_idx); - zspage = get_zspage(zpdesc); - - /* - * migration cannot move any zpages in this zspage. Here, class->lock - * is too heavy since callers would take some time until they calls - * zs_unmap_object API so delegate the locking from class to zspage - * which is smaller granularity. - */ - zspage_read_lock(zspage); - read_unlock(&pool->lock); - - class = zspage_class(pool, zspage); - off = offset_in_page(class->size * obj_idx); - - local_lock(&zs_map_area.lock); - area = this_cpu_ptr(&zs_map_area); - area->vm_mm = mm; - if (off + class->size <= PAGE_SIZE) { - /* this object is contained entirely within a page */ - area->vm_addr = kmap_local_zpdesc(zpdesc); - ret = area->vm_addr + off; - goto out; - } - - /* this object spans two pages */ - zpdescs[0] = zpdesc; - zpdescs[1] = get_next_zpdesc(zpdesc); - BUG_ON(!zpdescs[1]); - - ret = __zs_map_object(area, zpdescs, off, class->size); -out: - if (likely(!ZsHugePage(zspage))) - ret += ZS_HANDLE_SIZE; - - return ret; -} -EXPORT_SYMBOL_GPL(zs_map_object); - -void zs_unmap_object(struct zs_pool *pool, unsigned long handle) -{ - struct zspage *zspage; - struct zpdesc *zpdesc; - unsigned long obj, off; - unsigned int obj_idx; - - struct size_class *class; - struct mapping_area *area; - - obj = handle_to_obj(handle); - obj_to_location(obj, &zpdesc, &obj_idx); - zspage = get_zspage(zpdesc); - class = zspage_class(pool, zspage); - off = offset_in_page(class->size * obj_idx); - - area = this_cpu_ptr(&zs_map_area); - if (off + class->size <= PAGE_SIZE) - kunmap_local(area->vm_addr); - else { - struct zpdesc *zpdescs[2]; - - zpdescs[0] = zpdesc; - zpdescs[1] = get_next_zpdesc(zpdesc); - BUG_ON(!zpdescs[1]); - - __zs_unmap_object(area, zpdescs, off, class->size); - } - local_unlock(&zs_map_area.lock); - - zspage_read_unlock(zspage); -} -EXPORT_SYMBOL_GPL(zs_unmap_object); - void *zs_obj_read_begin(struct zs_pool *pool, unsigned long handle, void *local_copy) { @@ -1975,7 +1765,7 @@ static int zs_page_migrate(struct page *newpage, struct page *page, * the class lock protects zpage alloc/free in the zspage. */ spin_lock(&class->lock); - /* the zspage write_lock protects zpage access via zs_map_object */ + /* the zspage write_lock protects zpage access via zs_obj_read/write() */ if (!zspage_write_trylock(zspage)) { spin_unlock(&class->lock); write_unlock(&pool->lock); @@ -2459,23 +2249,11 @@ EXPORT_SYMBOL_GPL(zs_destroy_pool); static int __init zs_init(void) { - int ret; - - ret = cpuhp_setup_state(CPUHP_MM_ZS_PREPARE, "mm/zsmalloc:prepare", - zs_cpu_prepare, zs_cpu_dead); - if (ret) - goto out; - #ifdef CONFIG_ZPOOL zpool_register_driver(&zs_zpool_driver); #endif - zs_stat_init(); - return 0; - -out: - return ret; } static void __exit zs_exit(void) @@ -2483,8 +2261,6 @@ static void __exit zs_exit(void) #ifdef CONFIG_ZPOOL zpool_unregister_driver(&zs_zpool_driver); #endif - cpuhp_remove_state(CPUHP_MM_ZS_PREPARE); - zs_stat_exit(); } -- 2.50.1 From 7b60041156079f256a7df3f7de469de7db618580 Mon Sep 17 00:00:00 2001 From: Yosry Ahmed Date: Wed, 5 Mar 2025 06:11:33 +0000 Subject: [PATCH 12/16] mm: zpool: remove zpool_malloc_support_movable() zpool_malloc_support_movable() always returns true for zsmalloc, the only remaining zpool driver. Remove it and set the gfp flags in zswap_compress() accordingly. Opportunistically use GFP_NOWAIT instead of __GFP_NOWARN | __GFP_KSWAPD_RECLAIM for conciseness as they are equivalent. Link: https://lkml.kernel.org/r/20250305061134.4105762-6-yosry.ahmed@linux.dev Signed-off-by: Yosry Ahmed Reviewed-by: Sergey Senozhatsky Acked-by: Johannes Weiner Acked-by: Nhat Pham Cc: Thomas Gleixner Cc: Chengming Zhou Cc: Herbert Xu Cc: Minchan Kim Cc: Peter Zijlstra Signed-off-by: Andrew Morton --- include/linux/zpool.h | 3 --- mm/zpool.c | 16 ---------------- mm/zsmalloc.c | 1 - mm/zswap.c | 4 +--- 4 files changed, 1 insertion(+), 23 deletions(-) diff --git a/include/linux/zpool.h b/include/linux/zpool.h index 2c8a9d2654f6..52f30e526607 100644 --- a/include/linux/zpool.h +++ b/include/linux/zpool.h @@ -21,8 +21,6 @@ const char *zpool_get_type(struct zpool *pool); void zpool_destroy_pool(struct zpool *pool); -bool zpool_malloc_support_movable(struct zpool *pool); - int zpool_malloc(struct zpool *pool, size_t size, gfp_t gfp, unsigned long *handle); @@ -65,7 +63,6 @@ struct zpool_driver { void *(*create)(const char *name, gfp_t gfp); void (*destroy)(void *pool); - bool malloc_support_movable; int (*malloc)(void *pool, size_t size, gfp_t gfp, unsigned long *handle); void (*free)(void *pool, unsigned long handle); diff --git a/mm/zpool.c b/mm/zpool.c index 4fc665b42f5e..6d6d88930932 100644 --- a/mm/zpool.c +++ b/mm/zpool.c @@ -220,22 +220,6 @@ const char *zpool_get_type(struct zpool *zpool) return zpool->driver->type; } -/** - * zpool_malloc_support_movable() - Check if the zpool supports - * allocating movable memory - * @zpool: The zpool to check - * - * This returns if the zpool supports allocating movable memory. - * - * Implementations must guarantee this to be thread-safe. - * - * Returns: true if the zpool supports allocating movable memory, false if not - */ -bool zpool_malloc_support_movable(struct zpool *zpool) -{ - return zpool->driver->malloc_support_movable; -} - /** * zpool_malloc() - Allocate memory * @zpool: The zpool to allocate from. diff --git a/mm/zsmalloc.c b/mm/zsmalloc.c index cd1c2a8ffef0..961b270f023c 100644 --- a/mm/zsmalloc.c +++ b/mm/zsmalloc.c @@ -503,7 +503,6 @@ static struct zpool_driver zs_zpool_driver = { .owner = THIS_MODULE, .create = zs_zpool_create, .destroy = zs_zpool_destroy, - .malloc_support_movable = true, .malloc = zs_zpool_malloc, .free = zs_zpool_free, .obj_read_begin = zs_zpool_obj_read_begin, diff --git a/mm/zswap.c b/mm/zswap.c index 7de54f105d04..5f0e62289444 100644 --- a/mm/zswap.c +++ b/mm/zswap.c @@ -964,9 +964,7 @@ static bool zswap_compress(struct page *page, struct zswap_entry *entry, goto unlock; zpool = pool->zpool; - gfp = __GFP_NORETRY | __GFP_NOWARN | __GFP_KSWAPD_RECLAIM; - if (zpool_malloc_support_movable(zpool)) - gfp |= __GFP_HIGHMEM | __GFP_MOVABLE; + gfp = GFP_NOWAIT | __GFP_NORETRY | __GFP_HIGHMEM | __GFP_MOVABLE; alloc_ret = zpool_malloc(zpool, dlen, gfp, &handle); if (alloc_ret) goto unlock; -- 2.50.1 From c0d017896b72d7dd251dabf64765196ff9a46a0f Mon Sep 17 00:00:00 2001 From: Baolin Wang Date: Fri, 7 Feb 2025 17:44:17 +0800 Subject: [PATCH 13/16] mm: shmem: drop the unused macro Patch series "Some trivial cleanups for shmem". Patch 1 - Patch 5 do some trivial cleanups and refactoring for shmem. Patch 6 adds myself as shmem reviewer. This patch (of 6): Drop the unused 'BLOCKS_PER_PAGE' macro. Link: https://lkml.kernel.org/r/cover.1738918357.git.baolin.wang@linux.alibaba.com Link: https://lkml.kernel.org/r/69264cee1d938442477e657004e4924f8a5c4dd4.1738918357.git.baolin.wang@linux.alibaba.com Signed-off-by: Baolin Wang Cc: Baolin Wang Cc: David Hildenbrand Cc: Hugh Dickins Signed-off-by: Andrew Morton --- mm/shmem.c | 1 - 1 file changed, 1 deletion(-) diff --git a/mm/shmem.c b/mm/shmem.c index b276ae233dfa..6d6d5fce2120 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -86,7 +86,6 @@ static struct vfsmount *shm_mnt __ro_after_init; #include "internal.h" -#define BLOCKS_PER_PAGE (PAGE_SIZE/512) #define VM_ACCT(size) (PAGE_ALIGN(size) >> PAGE_SHIFT) /* Pretend that each entry is of this size in directory's i_size */ -- 2.50.1 From 6d26a149f5483acdc8a9a7a8fcf5e737a324a2b6 Mon Sep 17 00:00:00 2001 From: Baolin Wang Date: Fri, 7 Feb 2025 17:44:18 +0800 Subject: [PATCH 14/16] mm: shmem: remove 'fadvise()' comments Similar to commit 255ff62d1586 ("docs: tmpfs: drop 'fadvise()' from the documentation"), fadvise() has no HUGEPAGE advise currently. Remove the confusing fadvise() comments. Link: https://lkml.kernel.org/r/fae702b9775f58b55b45be5eaad22d8586d0290a.1738918357.git.baolin.wang@linux.alibaba.com Signed-off-by: Baolin Wang Cc: David Hildenbrand Cc: Hugh Dickins Signed-off-by: Andrew Morton --- mm/shmem.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/mm/shmem.c b/mm/shmem.c index 6d6d5fce2120..c63fd18cea50 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -525,9 +525,9 @@ static bool shmem_confirm_swap(struct address_space *mapping, * enables huge pages for the mount; * SHMEM_HUGE_WITHIN_SIZE: * only allocate huge pages if the page will be fully within i_size, - * also respect fadvise()/madvise() hints; + * also respect madvise() hints; * SHMEM_HUGE_ADVISE: - * only allocate huge pages if requested with fadvise()/madvise(); + * only allocate huge pages if requested with madvise(); */ #define SHMEM_HUGE_NEVER 0 -- 2.50.1 From d5e4e147c0f59259cd527d58295ba1bfefbad481 Mon Sep 17 00:00:00 2001 From: Baolin Wang Date: Fri, 7 Feb 2025 17:44:19 +0800 Subject: [PATCH 15/16] mm: shmem: remove duplicate error validation Remove duplicate error code checks for 'start' and 'end', as the get_order_from_str() will only return -EINVAL if the cmdline string is configured incorrectly. Link: https://lkml.kernel.org/r/dfadaba4c8b24c5ae1467fe8b6744b654c65ec91.1738918357.git.baolin.wang@linux.alibaba.com Signed-off-by: Baolin Wang Cc: David Hildenbrand Cc: Hugh Dickins Signed-off-by: Andrew Morton --- mm/shmem.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/mm/shmem.c b/mm/shmem.c index c63fd18cea50..51bdeea828a0 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -5664,19 +5664,19 @@ static int __init setup_thp_shmem(char *str) THP_ORDERS_ALL_FILE_DEFAULT); } - if (start == -EINVAL) { + if (start < 0) { pr_err("invalid size %s in thp_shmem boot parameter\n", start_size); goto err; } - if (end == -EINVAL) { + if (end < 0) { pr_err("invalid size %s in thp_shmem boot parameter\n", end_size); goto err; } - if (start < 0 || end < 0 || start > end) + if (start > end) goto err; nr = end - start + 1; -- 2.50.1 From cd81c424b53fa174df1e18b021806bc978c8fb7f Mon Sep 17 00:00:00 2001 From: Baolin Wang Date: Fri, 7 Feb 2025 17:44:20 +0800 Subject: [PATCH 16/16] mm: shmem: change the return value of shmem_find_swap_entries() The shmem_find_swap_entries() originally returned the index corresponding to the swap entry, but no callers used this return value. It should return the number of entries that were found like other functions, which can be used by the callers. No functional changes. Link: https://lkml.kernel.org/r/070489b5946b8379b2a2d25f78115cef167cd145.1738918357.git.baolin.wang@linux.alibaba.com Signed-off-by: Baolin Wang Cc: David Hildenbrand Cc: Hugh Dickins Signed-off-by: Andrew Morton --- mm/shmem.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/mm/shmem.c b/mm/shmem.c index 51bdeea828a0..f5a081563022 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -1379,9 +1379,9 @@ static void shmem_evict_inode(struct inode *inode) #endif } -static int shmem_find_swap_entries(struct address_space *mapping, - pgoff_t start, struct folio_batch *fbatch, - pgoff_t *indices, unsigned int type) +static unsigned int shmem_find_swap_entries(struct address_space *mapping, + pgoff_t start, struct folio_batch *fbatch, + pgoff_t *indices, unsigned int type) { XA_STATE(xas, &mapping->i_pages, start); struct folio *folio; @@ -1414,7 +1414,7 @@ static int shmem_find_swap_entries(struct address_space *mapping, } rcu_read_unlock(); - return xas.xa_index; + return folio_batch_count(fbatch); } /* @@ -1461,8 +1461,8 @@ static int shmem_unuse_inode(struct inode *inode, unsigned int type) do { folio_batch_init(&fbatch); - shmem_find_swap_entries(mapping, start, &fbatch, indices, type); - if (folio_batch_count(&fbatch) == 0) { + if (!shmem_find_swap_entries(mapping, start, &fbatch, + indices, type)) { ret = 0; break; } -- 2.50.1