From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755065AbdABIFl (ORCPT ); Mon, 2 Jan 2017 03:05:41 -0500 Received: from mailout4.samsung.com ([203.254.224.34]:35518 "EHLO mailout4.samsung.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750893AbdABIFk (ORCPT ); Mon, 2 Jan 2017 03:05:40 -0500 MIME-version: 1.0 Content-type: text/plain; charset=utf-8 X-AuditID: b6c32a35-f799b6d0000016fb-f0-586a09d08eaf Content-transfer-encoding: 8BIT Subject: Re: [PATCH] mm: cma: print allocation failure reason and bitmap status To: Michal Nazarewicz , Michal Hocko Cc: gregkh@linuxfoundation.org, akpm@linux-foundation.org, labbott@redhat.com, m.szyprowski@samsung.com, linux-mm@kvack.org, linux-kernel@vger.kernel.org, jaewon31.kim@gmail.com From: Jaewon Kim Message-id: <586A0A05.4040708@samsung.com> Date: Mon, 02 Jan 2017 17:06:29 +0900 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101 Thunderbird/38.7.2 In-reply-to: X-Brightmail-Tracker: H4sIAAAAAAAAA01SfUgTcRjmt9tut3BxTVs/FpUd9IeWlzedd30YRhFH9ocRxSrKjvlrmtvN 7mZURFhZli5n9oGNJMOwHKa1SmaEiRnLIJOWQYImfZBZJH0L2ce2W9R/z/u+z/N+PLwEZmjC TUSR6EaSKDgofIq6/W4Kndav22FN93hN3Lm2Fpw71NiGc1XXzwKu+faEigvfOodzwy2/NdyV niEt9+5bE8Y1hMo1OTq+wzek5QP+Yzgf+FSr5e/X/VDzd+pbtHzr5JiKH+8cwPnqG37Afw7M ztNtQksLkVCApGQk2lwFRaI9m8pdl78i35KVzqQxiziWShYFJ8qmVq7JS1tV5IgsSSXvEhyl kVSeIMvUwmVLJVepGyUXumR3NrWZYcw0k87SZrOZzmS3LDZbIpRtqPBy50u8pGfR7urGEF4G wgsqgY6AZCb81nwXU7AR9g+34ZVgCmEggwDeavDHgwoVvDfUhf9VvOnsUCmFSwA+HA3HCnpy Gpw4OayuBASBkXNgz+PiaBojU+Dol1q1wn8OYNVFrypaSCTXwqvXJmM4icyFI+3NMRJG3oyQ OipiO+HkfDjeUKtRBqTCdy/7YgI1OQ9eGn+qjeLppBUG6z+CKNZFOOVvPSDaCJKPtdA7MIBF N4LkLBjoisOVMDzJKcckwrHQDa2CZ0L/r0GVIvUCGLhQhimBB8Cj5yviJmXA455+TDltKvzw 1aNRmurh0SMGhcLD188Oxd1aDi+MlMev92CwKlSD14A5vv8M8/0zzPefYQ0A8wMjKpGddiQz JQwtC065VLTTNpczAGI/mmoJglN9a7oBSQAqQU9uLrIaNMIueY+zG0ACo5L0uT8jKX2BsGcv klz5UqkDyd3AErHvBGaabnNFPl505zOZLMeyWZkZDMdaqBn6xgNLrAbSLrhRMUIlSPqrUxE6 Uxl40CQmdLxvO6xNywrW9HqPVaKUE8H1pjHZbnxUIW1l9B9W79xAp++osx7PebK9XkeH2g8k ZLxghcmZn34O2sbWzz29V5Sm0mf37R8xut6+KlhnTtL8KL6+lt3y3VHWenD7uK3LGOaq99WV A/zImURdr/XRtNbujTmDz6p9jKaFUsuFApOKSbLwB5A44ZS5AwAA X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFjrHIsWRmVeSWpSXmKPExsVy+t9jAd0LnFkRBhIWc9avYbNoXryezaJ7 80xGi5V7fjBZXN41h83i3pr/rBZrj9xlt3j9bRmzxYLjLawOnB47Z91l99i0qpPNY9OnSewe J2b8ZvHYP3cNu8e6P6+YPN7vu8rm0bdlFaPH501yAZxRbjYZqYkpqUUKqXnJ+SmZeem2SqEh broWSgp5ibmptkoRur4hQUoKZYk5pUCekQEacHAOcA9W0rdLcMtYse8xW8ERy4q+xcfZGhgv 63QxcnJICJhIPN+3kwnCFpO4cG89G4gtJLCUUeLlwUgQm1dAUOLH5HssXYwcHMwC8hJHLmVD mOoSU6bkdjFyAVU/ZJQ4cnQF2BhhAX+JBbc2gdkiAt4SD7atZIEo6mOWePDnGBOIwyywlVFi 29m97CBVbALaEu8XTGKFWKYl8frxObBuFgFVieXvr4HViApESKxed40ZxOYEqml52cM4gVFg FpL7ZiHcNwvhvgWMzKsYJVILkguKk9JzDfNSy/WKE3OLS/PS9ZLzczcxguPymdQOxoO73A8x CnAwKvHwZsRmRgixJpYVV+YeYpTgYFYS4fX+CxTiTUmsrEotyo8vKs1JLT7EaAp04kRmKdHk fGDKyCuJNzQxNzE3NrAwt7Q0MVIS522c/SxcSCA9sSQ1OzW1ILUIpo+Jg1OqgZFr7fEf1xfq JMibbvz/s/1C+YFd1VqPLf7nR0e/3frtq5jWg62yZ50/N2+dt00x5NlRF+1Vf76czucUUSgy nTkvRG75sovVO/7b9/Jz/bzVFGq4WfISe4yi4EKuYyHcNrcfbTsfeyRwZ5t37xr+pr7DGw/+ +W6rn7h869HG8uUzCgt7bXrdS5WUWIozEg21mIuKEwFB8HEJ4QIAAA== X-MTR: 20000000000000000@CPGS X-CMS-MailID: 20170102080536epcas1p4b4fb35fd5df8fcf82aa9d173e28f8007 X-Msg-Generator: CA X-Sender-IP: 203.254.230.26 X-Local-Sender: =?UTF-8?B?6rmA7J6s7JuQG1N5c3RlbSBTL1fqsJzrsJwy6re466O5KA==?= =?UTF-8?B?66y07ISgKRvsgrzshLHsoITsnpAbUzUo7LGF7J6EKS/ssYXsnoQ=?= X-Global-Sender: =?UTF-8?B?SmFld29uIEtpbRtTeXN0ZW0gUy9XIFImRCBHcm91cCAyG1Nh?= =?UTF-8?B?bXN1bmcgRWxlY3Ryb25pY3MbUzUvU2VuaW9yIEVuZ2luZWVy?= X-Sender-Code: =?UTF-8?B?QzEwG1RFTEUbQzEwRDkxMjI=?= CMS-TYPE: 101P DLP-Filter: Pass X-CFilter-Loop: Reflected X-HopCount: 7 X-CMS-RootMailID: 20161229022722epcas5p4be0e1924f3c8d906cbfb461cab8f0374 X-RootMTR: 20161229022722epcas5p4be0e1924f3c8d906cbfb461cab8f0374 References: <1482978482-14007-1-git-send-email-jaewon31.kim@samsung.com> <20161229091449.GG29208@dhcp22.suse.cz> <58660BBE.1040807@samsung.com> <20161230094411.GD13301@dhcp22.suse.cz> <5869E849.1040605@samsung.com> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 2017년 01월 02일 15:46, Michal Nazarewicz wrote: > On Mon, Jan 02 2017, Jaewon Kim wrote: >> There are many reasons of CMA allocation failure such as EBUSY, ENOMEM, EINTR. >> But we did not know error reason so far. This patch prints the error value. >> >> Additionally if CONFIG_CMA_DEBUG is enabled, this patch shows bitmap status to >> know available pages. Actually CMA internally tries on all available regions >> because some regions can be failed because of EBUSY. Bitmap status is useful to >> know in detail on both ENONEM and EBUSY; >> ENOMEM: not tried at all because of no available region >> it could be too small total region or could be fragmentation issue >> EBUSY: tried some region but all failed >> >> This is an ENOMEM example with this patch. >> [ 12.415458] [2: Binder:714_1: 744] cma: cma_alloc: alloc failed, req-size: 256 pages, ret: -12 >> If CONFIG_CMA_DEBUG is enabled, avabile pages also will be shown as concatenated >> size@position format. So 4@572 means that there are 4 available pages at 572 >> position starting from 0 position. >> [ 12.415503] [2: Binder:714_1: 744] cma: number of available pages: 4@572+7@585+7@601+8@632+38@730+166@1114+127@1921=> 357 free of 2048 total pages >> >> Signed-off-by: Jaewon Kim >> Acked-by: Michal Nazarewicz >> --- >> mm/cma.c | 34 +++++++++++++++++++++++++++++++++- >> 1 file changed, 33 insertions(+), 1 deletion(-) >> >> diff --git a/mm/cma.c b/mm/cma.c >> index c960459..9e037541 100644 >> --- a/mm/cma.c >> +++ b/mm/cma.c >> @@ -353,6 +353,32 @@ int __init cma_declare_contiguous(phys_addr_t base, >> return ret; >> } >> >> +#ifdef CONFIG_CMA_DEBUG >> +static void debug_show_cma_areas(struct cma *cma) > Make it ‘cma_debug_show_areas’. All other functions have ‘cma’ as > prefix so that’s more consistent. OK no problem. > >> +{ >> + unsigned long next_zero_bit, next_set_bit; >> + unsigned long start = 0; >> + unsigned int nr_zero, nr_total = 0; >> + >> + mutex_lock(&cma->lock); >> + pr_info("number of available pages: "); >> + for (;;) { >> + next_zero_bit = find_next_zero_bit(cma->bitmap, cma->count, start); >> + if (next_zero_bit >= cma->count) >> + break; >> + next_set_bit = find_next_bit(cma->bitmap, cma->count, next_zero_bit); >> + nr_zero = next_set_bit - next_zero_bit; >> + pr_cont("%s%u@%lu", nr_total ? "+" : "", nr_zero, next_zero_bit); >> + nr_total += nr_zero; >> + start = next_zero_bit + nr_zero; >> + } >> + pr_cont("=> %u free of %lu total pages\n", nr_total, cma->count); >> + mutex_unlock(&cma->lock); >> +} >> +#else >> +static inline void debug_show_cma_areas(struct cma *cma) { } >> +#endif >> + >> /** >> * cma_alloc() - allocate pages from contiguous area >> * @cma: Contiguous memory region for which the allocation is performed. >> @@ -369,7 +395,7 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align) >> unsigned long start = 0; >> unsigned long bitmap_maxno, bitmap_no, bitmap_count; >> struct page *page = NULL; >> - int ret; >> + int ret = -ENOMEM; >> >> if (!cma || !cma->count) >> return NULL; >> @@ -426,6 +452,12 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align) >> >> trace_cma_alloc(pfn, page, count, align); >> >> + if (ret) { >> + pr_info("%s: alloc failed, req-size: %zu pages, ret: %d\n", >> + __func__, count, ret); >> + debug_show_cma_areas(cma); >> + } >> + >> pr_debug("%s(): returned %p\n", __func__, page); >> return page; >> } >> -- >> Added the latest. >>From ec724faf7ace5e690715e23e82f17f27c697b6b1 Mon Sep 17 00:00:00 2001 From: Jaewon Kim Date: Thu, 29 Dec 2016 11:00:16 +0900 Subject: [PATCH] mm: cma: print allocation failure reason and bitmap status There are many reasons of CMA allocation failure such as EBUSY, ENOMEM, EINTR. But we did not know error reason so far. This patch prints the error value. Additionally if CONFIG_CMA_DEBUG is enabled, this patch shows bitmap status to know available pages. Actually CMA internally tries on all available regions because some regions can be failed because of EBUSY. Bitmap status is useful to know in detail on both ENONEM and EBUSY; ENOMEM: not tried at all because of no available region it could be too small total region or could be fragmentation issue EBUSY: tried some region but all failed This is an ENOMEM example with this patch. [ 12.415458] [2: Binder:714_1: 744] cma: cma_alloc: alloc failed, req-size: 256 pages, ret: -12 If CONFIG_CMA_DEBUG is enabled, avabile pages also will be shown as concatenated size@position format. So 4@572 means that there are 4 available pages at 572 position starting from 0 position. [ 12.415503] [2: Binder:714_1: 744] cma: number of available pages: 4@572+7@585+7@601+8@632+38@730+166@1114+127@1921=> 357 free of 2048 total pages Signed-off-by: Jaewon Kim Acked-by: Michal Nazarewicz --- mm/cma.c | 34 +++++++++++++++++++++++++++++++++- 1 file changed, 33 insertions(+), 1 deletion(-) diff --git a/mm/cma.c b/mm/cma.c index c960459..c393229 100644 --- a/mm/cma.c +++ b/mm/cma.c @@ -353,6 +353,32 @@ int __init cma_declare_contiguous(phys_addr_t base, return ret; } +#ifdef CONFIG_CMA_DEBUG +static void cma_debug_show_areas(struct cma *cma) +{ + unsigned long next_zero_bit, next_set_bit; + unsigned long start = 0; + unsigned int nr_zero, nr_total = 0; + + mutex_lock(&cma->lock); + pr_info("number of available pages: "); + for (;;) { + next_zero_bit = find_next_zero_bit(cma->bitmap, cma->count, start); + if (next_zero_bit >= cma->count) + break; + next_set_bit = find_next_bit(cma->bitmap, cma->count, next_zero_bit); + nr_zero = next_set_bit - next_zero_bit; + pr_cont("%s%u@%lu", nr_total ? "+" : "", nr_zero, next_zero_bit); + nr_total += nr_zero; + start = next_zero_bit + nr_zero; + } + pr_cont("=> %u free of %lu total pages\n", nr_total, cma->count); + mutex_unlock(&cma->lock); +} +#else +static inline void cma_debug_show_areas(struct cma *cma) { } +#endif + /** * cma_alloc() - allocate pages from contiguous area * @cma: Contiguous memory region for which the allocation is performed. @@ -369,7 +395,7 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align) unsigned long start = 0; unsigned long bitmap_maxno, bitmap_no, bitmap_count; struct page *page = NULL; - int ret; + int ret = -ENOMEM; if (!cma || !cma->count) return NULL; @@ -426,6 +452,12 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align) trace_cma_alloc(pfn, page, count, align); + if (ret) { + pr_info("%s: alloc failed, req-size: %zu pages, ret: %d\n", + __func__, count, ret); + cma_debug_show_areas(cma); + } + pr_debug("%s(): returned %p\n", __func__, page); return page; } -- 1.9.1