From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by smtp.lore.kernel.org (Postfix) with ESMTP id 310D4CD13CF for ; Mon, 2 Sep 2024 09:58:30 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id B659B8D00B6; Mon, 2 Sep 2024 05:58:29 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id B151A8D0098; Mon, 2 Sep 2024 05:58:29 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 9B53E8D00B6; Mon, 2 Sep 2024 05:58:29 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from relay.hostedemail.com (smtprelay0011.hostedemail.com [216.40.44.11]) by kanga.kvack.org (Postfix) with ESMTP id 7E0538D0098 for ; Mon, 2 Sep 2024 05:58:29 -0400 (EDT) Received: from smtpin22.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay06.hostedemail.com (Postfix) with ESMTP id E5F4FA7F25 for ; Mon, 2 Sep 2024 09:58:28 +0000 (UTC) X-FDA: 82519348296.22.AB7FDD0 Received: from foss.arm.com (foss.arm.com [217.140.110.172]) by imf09.hostedemail.com (Postfix) with ESMTP id E12C714000D for ; Mon, 2 Sep 2024 09:58:26 +0000 (UTC) Authentication-Results: imf09.hostedemail.com; dkim=none; spf=pass (imf09.hostedemail.com: domain of ryan.roberts@arm.com designates 217.140.110.172 as permitted sender) smtp.mailfrom=ryan.roberts@arm.com; dmarc=pass (policy=none) header.from=arm.com ARC-Seal: i=1; s=arc-20220608; d=hostedemail.com; t=1725271060; a=rsa-sha256; cv=none; b=MfkX5Czdb5MXJiDuBRF/lZ1lvS81Frl1X3BIbkNrAWSXuYmUNLHy1KN8BqyGAI2GbeqjEe KWmrvTHj0bR7iyoERsjBAvV3uRhxH4SK+I7GvkDANxvoDrxSl+X2oZHdIarUuHbJmYP/rq QxGVpd17eBql9qmhoNjUid+ItsBX51k= ARC-Authentication-Results: i=1; imf09.hostedemail.com; dkim=none; spf=pass (imf09.hostedemail.com: domain of ryan.roberts@arm.com designates 217.140.110.172 as permitted sender) smtp.mailfrom=ryan.roberts@arm.com; dmarc=pass (policy=none) header.from=arm.com ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=hostedemail.com; s=arc-20220608; t=1725271060; h=from:from:sender:reply-to:subject:subject:date:date: message-id:message-id:to:to:cc:cc:mime-version:mime-version: content-type:content-type: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=zjQtD4SSlncLVcYA3SAU4Pla3wdheLg2KWe2+NgBD6k=; b=BBr0RQdcAIjwTnWQBuil09jLgC/ctFO6fGoaYzh0B5XieOv2HUAbuu4EqQ7e8jkmLh0t1a pQ/L1hTGoo0FTTz4RGS2T1ouEwMu1MOFLOlWCUjePqOHb/RkKHEdbUF2iNSAMp/68dqzYn B2l6VwAOFGcyfaY3iDOCDSJa/MI5ZGs= Received: from usa-sjc-imap-foss1.foss.arm.com (unknown [10.121.207.14]) by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id 48CC8FEC; Mon, 2 Sep 2024 02:58:52 -0700 (PDT) Received: from [10.57.81.153] (unknown [10.57.81.153]) by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPSA id 3467D3F66E; Mon, 2 Sep 2024 02:58:24 -0700 (PDT) Message-ID: <14823123-79e3-4c7d-8501-8c46c6ec13c7@arm.com> Date: Mon, 2 Sep 2024 10:58:22 +0100 MIME-Version: 1.0 User-Agent: Mozilla Thunderbird Subject: Re: [PATCH v3 2/2] mm: Tidy up shmem mTHP controls and stats Content-Language: en-GB To: Baolin Wang , Andrew Morton , Hugh Dickins , "Matthew Wilcox (Oracle)" , David Hildenbrand , Barry Song , Lance Yang , Gavin Shan Cc: linux-kernel@vger.kernel.org, linux-mm@kvack.org References: <20240808111849.651867-1-ryan.roberts@arm.com> <20240808111849.651867-3-ryan.roberts@arm.com> <747d1319-f746-4379-bf88-a0f6c3f558b4@linux.alibaba.com> From: Ryan Roberts In-Reply-To: <747d1319-f746-4379-bf88-a0f6c3f558b4@linux.alibaba.com> Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-Stat-Signature: ig55txcmpguox5jaza4m5wcyfcwm1n6i X-Rspamd-Queue-Id: E12C714000D X-Rspam-User: X-Rspamd-Server: rspam10 X-HE-Tag: 1725271106-878700 X-HE-Meta: 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 tnkvcmYv VY0wojakqs3lBk26yksC39oydnWKhGUYJaoxoTUPqQ+Cp1/u0bGpIPIp1gGg41polJPlEjn9C/FTQQ8jXjP0G3V5IAxHFwR6wIEYIXlfLswoP5ko/Byv/7aLDn4bq2UlirlYrWYZGO/is7T2LljnTEft4U7HsBiL/vHug3eBwaxCGVilXQUsbSe6dXoaMgwANHIVxw+DIYC2Tv7dY5vEgz5olhLk5nu0oVT+rmsNhgt5Y5sA= X-Bogosity: Ham, tests=bogofilter, spamicity=0.000000, version=1.2.4 Sender: owner-linux-mm@kvack.org Precedence: bulk X-Loop: owner-majordomo@kvack.org List-ID: List-Subscribe: List-Unsubscribe: Hi Baolin, Thanks for the review - I've been out on Paternity leave so only getting around to replying now... On 09/08/2024 09:31, Baolin Wang wrote: > > > On 2024/8/8 19:18, Ryan Roberts wrote: >> Previously we had a situation where shmem mTHP controls and stats were >> not exposed for some supported sizes and were exposed for some >> unsupported sizes. So let's clean that up. >> >> Anon mTHP can support all large orders [2, PMD_ORDER]. But shmem can >> support all large orders [1, MAX_PAGECACHE_ORDER]. However, per-size >> shmem controls and stats were previously being exposed for all the anon >> mTHP orders, meaning order-1 was not present, and for arm64 64K base >> pages, orders 12 and 13 were exposed but were not supported internally. >> >> Tidy this all up by defining ctrl and stats attribute groups for anon >> and file separately. Anon ctrl and stats groups are populated for all >> orders in THP_ORDERS_ALL_ANON and file ctrl and stats groups are >> populated for all orders in THP_ORDERS_ALL_FILE_DEFAULT. >> >> Additionally, create "any" ctrl and stats attribute groups which are >> populated for all orders in (THP_ORDERS_ALL_ANON | >> THP_ORDERS_ALL_FILE_DEFAULT). swpout stats use this since they apply to >> anon and shmem. >> >> The side-effect of all this is that different hugepage-*kB directories >> contain different sets of controls and stats, depending on which memory >> types support that size. This approach is preferred over the >> alternative, which is to populate dummy controls and stats for memory >> types that do not support a given size. >> >> Signed-off-by: Ryan Roberts >> --- >>   mm/huge_memory.c | 144 +++++++++++++++++++++++++++++++++++++---------- >>   1 file changed, 114 insertions(+), 30 deletions(-) >> >> diff --git a/mm/huge_memory.c b/mm/huge_memory.c >> index 0c3075ee00012..082d86b7c6c2f 100644 >> --- a/mm/huge_memory.c >> +++ b/mm/huge_memory.c >> @@ -482,8 +482,8 @@ static void thpsize_release(struct kobject *kobj); >>   static DEFINE_SPINLOCK(huge_anon_orders_lock); >>   static LIST_HEAD(thpsize_list); >>   -static ssize_t thpsize_enabled_show(struct kobject *kobj, >> -                    struct kobj_attribute *attr, char *buf) >> +static ssize_t anon_enabled_show(struct kobject *kobj, >> +                 struct kobj_attribute *attr, char *buf) >>   { >>       int order = to_thpsize(kobj)->order; >>       const char *output; >> @@ -500,9 +500,9 @@ static ssize_t thpsize_enabled_show(struct kobject *kobj, >>       return sysfs_emit(buf, "%s\n", output); >>   } >>   -static ssize_t thpsize_enabled_store(struct kobject *kobj, >> -                     struct kobj_attribute *attr, >> -                     const char *buf, size_t count) >> +static ssize_t anon_enabled_store(struct kobject *kobj, >> +                  struct kobj_attribute *attr, >> +                  const char *buf, size_t count) >>   { >>       int order = to_thpsize(kobj)->order; >>       ssize_t ret = count; >> @@ -544,19 +544,35 @@ static ssize_t thpsize_enabled_store(struct kobject *kobj, >>       return ret; >>   } >>   -static struct kobj_attribute thpsize_enabled_attr = >> -    __ATTR(enabled, 0644, thpsize_enabled_show, thpsize_enabled_store); >> +static struct kobj_attribute anon_enabled_attr = >> +    __ATTR(enabled, 0644, anon_enabled_show, anon_enabled_store); >>   -static struct attribute *thpsize_attrs[] = { >> -    &thpsize_enabled_attr.attr, >> +static struct attribute *anon_ctrl_attrs[] = { >> +    &anon_enabled_attr.attr, >> +    NULL, >> +}; >> + >> +static const struct attribute_group anon_ctrl_attr_grp = { >> +    .attrs = anon_ctrl_attrs, >> +}; >> + >> +static struct attribute *file_ctrl_attrs[] = { >>   #ifdef CONFIG_SHMEM >>       &thpsize_shmem_enabled_attr.attr, >>   #endif >>       NULL, >>   }; >>   -static const struct attribute_group thpsize_attr_group = { >> -    .attrs = thpsize_attrs, >> +static const struct attribute_group file_ctrl_attr_grp = { >> +    .attrs = file_ctrl_attrs, >> +}; >> + >> +static struct attribute *any_ctrl_attrs[] = { >> +    NULL, >> +}; >> + >> +static const struct attribute_group any_ctrl_attr_grp = { >> +    .attrs = any_ctrl_attrs, >>   }; > > I wonder why adding a NULL group? It made everything a bit more generic and therefore extensible. Its my preference to leave it as is, but will remove it if you insist. > >>     static const struct kobj_type thpsize_ktype = { >> @@ -595,64 +611,132 @@ DEFINE_MTHP_STAT_ATTR(anon_fault_fallback, >> MTHP_STAT_ANON_FAULT_FALLBACK); >>   DEFINE_MTHP_STAT_ATTR(anon_fault_fallback_charge, >> MTHP_STAT_ANON_FAULT_FALLBACK_CHARGE); >>   DEFINE_MTHP_STAT_ATTR(swpout, MTHP_STAT_SWPOUT); >>   DEFINE_MTHP_STAT_ATTR(swpout_fallback, MTHP_STAT_SWPOUT_FALLBACK); >> +#ifdef CONFIG_SHMEM >>   DEFINE_MTHP_STAT_ATTR(shmem_alloc, MTHP_STAT_SHMEM_ALLOC); >>   DEFINE_MTHP_STAT_ATTR(shmem_fallback, MTHP_STAT_SHMEM_FALLBACK); >>   DEFINE_MTHP_STAT_ATTR(shmem_fallback_charge, MTHP_STAT_SHMEM_FALLBACK_CHARGE); >> +#endif >>   DEFINE_MTHP_STAT_ATTR(split, MTHP_STAT_SPLIT); >>   DEFINE_MTHP_STAT_ATTR(split_failed, MTHP_STAT_SPLIT_FAILED); >>   DEFINE_MTHP_STAT_ATTR(split_deferred, MTHP_STAT_SPLIT_DEFERRED); >>   -static struct attribute *stats_attrs[] = { >> +static struct attribute *anon_stats_attrs[] = { >>       &anon_fault_alloc_attr.attr, >>       &anon_fault_fallback_attr.attr, >>       &anon_fault_fallback_charge_attr.attr, >> +#ifndef CONFIG_SHMEM >>       &swpout_attr.attr, >>       &swpout_fallback_attr.attr, >> -    &shmem_alloc_attr.attr, >> -    &shmem_fallback_attr.attr, >> -    &shmem_fallback_charge_attr.attr, >> +#endif >>       &split_attr.attr, >>       &split_failed_attr.attr, >>       &split_deferred_attr.attr, >>       NULL, >>   }; >>   -static struct attribute_group stats_attr_group = { >> +static struct attribute_group anon_stats_attr_grp = { >> +    .name = "stats", >> +    .attrs = anon_stats_attrs, >> +}; >> + >> +static struct attribute *file_stats_attrs[] = { >> +#ifdef CONFIG_SHMEM >> +    &shmem_alloc_attr.attr, >> +    &shmem_fallback_attr.attr, >> +    &shmem_fallback_charge_attr.attr, >> +#endif >> +    NULL, >> +}; >> + >> +static struct attribute_group file_stats_attr_grp = { >> +    .name = "stats", >> +    .attrs = file_stats_attrs, >> +}; >> + >> +static struct attribute *any_stats_attrs[] = { >> +#ifdef CONFIG_SHMEM >> +    &swpout_attr.attr, >> +    &swpout_fallback_attr.attr, >> +#endif > > Sorry I did not point it out in early version. I think file pages and shmem can > also be split, while 'split_deferred' is only for anonymous page. So I think the > any_stats_attrs should be: > static struct attribute *any_stats_attrs[] = { > #ifdef CONFIG_SHMEM >     &swpout_attr.attr, >     &swpout_fallback_attr.attr, > #endif >     &split_attr.attr, >     &split_failed_attr.attr, >     NULL, > }; Ahh yes, good point. I'll fix this in the next version (I'm back in office on Wednesday so will send it then). Thanks, Ryan