Received: by 2002:ac0:8c9a:0:0:0:0:0 with SMTP id r26csp5057459ima; Tue, 5 Feb 2019 05:49:16 -0800 (PST) X-Google-Smtp-Source: AHgI3IZM577WqrazvmZ8UEYtJ9HnCJWeN/A7oRqRyPSDW3EcgnJkFL++ORNBWV1CrPbNqQAxylBa X-Received: by 2002:a63:e950:: with SMTP id q16mr4701190pgj.138.1549374556208; Tue, 05 Feb 2019 05:49:16 -0800 (PST) ARC-Seal: i=1; a=rsa-sha256; t=1549374556; cv=none; d=google.com; s=arc-20160816; b=rW9voxMI74HyLwKy1sXTzHjDC+sPyq1UAYTxa7xT8zq1L5js2j7AMjiMXcIW4y6BE3 LRrMZWHo8hmjt/P2P6F8hj6aQoY5iG4j6gsdwK1bsQYbx6JQbvdUHkvTAz2Lpb5TqNOV kkM8j4lvof7hRnp8O6e2KiBY4wLY7YlaZYmTEpUCDgDHGWa+wNXdQt4NEqp9xEAf0Xl0 +0+5TDMpv7GsHnwfLpc8dly2lRrsRIgN4v4oNZY9oVeVnRhglf3J6SAWoysVQwU2c3S3 guUx97jqHNidCXe06XLjbg3doEwLQXUZKkJ8Vb4M1VRDXcdlqHGQJEKbv6GdyYn1Lz+g mcCQ== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=list-id:precedence:sender:content-transfer-encoding :content-language:in-reply-to:mime-version:user-agent:date :message-id:from:references:cc:to:subject:dkim-signature; bh=RHYMo3OkeYdCIxueyQ5gRTTu8UX7Tnicvz9bUlAAwBM=; b=caLbiDz+t08iFc1MkDRfBcPpSzPO3NYoyFaQD5NtpKLXpBBnXEs43FWw+B/zgCMQoq VbJqJh7f0woqLr23VNgXT9pFoBJoY9iavmn8CgPbUGrjgamELtBLpbV8b0V/DKB3WW/X 4vLvA54dxMv6LD8CFNMmLWLNvhzmGyIhY+iwVCv15fwI71zVX0+LYzdimXBwMWTEpHyY 7CND9BjjA+5vBsMAGz/CqqirvTx0lGyCbIaZkBVPcicLaOIZq7VjRnPAhxrZnCjHbjNr R36MztBAZqH7YUgQbe9sBHFww0GhS4hQaSdAqJIul9xgIm68XSfVdwqAEFlErl/7mZRx 1oiw== ARC-Authentication-Results: i=1; mx.google.com; dkim=pass header.i=@ti.com header.s=ti-com-17Q1 header.b=Gam58mnf; spf=pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=pass (p=QUARANTINE sp=NONE dis=NONE) header.from=ti.com Return-Path: Received: from vger.kernel.org (vger.kernel.org. [209.132.180.67]) by mx.google.com with ESMTP id 23si3240998pfk.287.2019.02.05.05.49.00; Tue, 05 Feb 2019 05:49:16 -0800 (PST) Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67; Authentication-Results: mx.google.com; dkim=pass header.i=@ti.com header.s=ti-com-17Q1 header.b=Gam58mnf; spf=pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=pass (p=QUARANTINE sp=NONE dis=NONE) header.from=ti.com Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1729496AbfBENmx (ORCPT + 99 others); Tue, 5 Feb 2019 08:42:53 -0500 Received: from fllv0015.ext.ti.com ([198.47.19.141]:54342 "EHLO fllv0015.ext.ti.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1727523AbfBENmx (ORCPT ); Tue, 5 Feb 2019 08:42:53 -0500 Received: from lelv0265.itg.ti.com ([10.180.67.224]) by fllv0015.ext.ti.com (8.15.2/8.15.2) with ESMTP id x15DgQKq006849; Tue, 5 Feb 2019 07:42:26 -0600 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=ti.com; s=ti-com-17Q1; t=1549374146; bh=RHYMo3OkeYdCIxueyQ5gRTTu8UX7Tnicvz9bUlAAwBM=; h=Subject:To:CC:References:From:Date:In-Reply-To; b=Gam58mnfLMjn4WC5sxsosccVQ4ARtVnY1WPoB+HGe021XseYtDadWuKy4UcaGQRkC ItvLaD51bMZPdP0tjS7DZV4IbT7tAvgfI9Tk2hhAT0YFV7sTmfx/0dckN7rvUHNaaK T4FHYThrZlpBftHyXJ/iexR2xtw8tMLplMQEVVkU= Received: from DFLE103.ent.ti.com (dfle103.ent.ti.com [10.64.6.24]) by lelv0265.itg.ti.com (8.15.2/8.15.2) with ESMTPS id x15DgQRx112661 (version=TLSv1.2 cipher=AES256-GCM-SHA384 bits=256 verify=FAIL); Tue, 5 Feb 2019 07:42:26 -0600 Received: from DFLE101.ent.ti.com (10.64.6.22) by DFLE103.ent.ti.com (10.64.6.24) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_CBC_SHA256_P256) id 15.1.1591.10; Tue, 5 Feb 2019 07:42:26 -0600 Received: from dflp32.itg.ti.com (10.64.6.15) by DFLE101.ent.ti.com (10.64.6.22) with Microsoft SMTP Server (version=TLS1_0, cipher=TLS_RSA_WITH_AES_256_CBC_SHA) id 15.1.1591.10 via Frontend Transport; Tue, 5 Feb 2019 07:42:26 -0600 Received: from [172.24.190.117] (ileax41-snat.itg.ti.com [10.172.224.153]) by dflp32.itg.ti.com (8.14.3/8.13.8) with ESMTP id x15DgLjh026492; Tue, 5 Feb 2019 07:42:22 -0600 Subject: Re: [RFC PATCH v4 08/13] genirq/msi: Add support for allocating single MSI for a device To: Marc Zyngier , Nishanth Menon , Santosh Shilimkar , Rob Herring , "tglx@linutronix.de" , "jason@lakedaemon.net" CC: Linux ARM Mailing List , "linux-kernel@vger.kernel.org" , Tero Kristo , Sekhar Nori , Device Tree Mailing List , Peter Ujfalusi References: <20181227060829.5080-1-lokeshvutla@ti.com> <20181227061313.5451-1-lokeshvutla@ti.com> <20181227061313.5451-8-lokeshvutla@ti.com> <5be38277-3348-a6d9-4b67-3ead308c009a@arm.com> <523c4a23-b5b9-ff10-051d-c90e6e8341d4@arm.com> From: Lokesh Vutla Message-ID: <69699155-49f8-ce1e-3eb4-5b63ab032a94@ti.com> Date: Tue, 5 Feb 2019 19:12:02 +0530 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:60.0) Gecko/20100101 Thunderbird/60.2.1 MIME-Version: 1.0 In-Reply-To: <523c4a23-b5b9-ff10-051d-c90e6e8341d4@arm.com> Content-Type: text/plain; charset="utf-8" Content-Language: en-US Content-Transfer-Encoding: 7bit X-EXCLAIMER-MD-CONFIG: e1e8a2fd-e40a-4ac6-ac9b-f7e9cc9ee180 Sender: linux-kernel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi Marc, On 04/02/19 4:03 PM, Marc Zyngier wrote: > On 24/01/2019 10:19, Lokesh Vutla wrote: >> Hi Marc, >> Sorry for the delayed response. Just back from vacation. >> >> On 17/01/19 12:00 AM, Marc Zyngier wrote: >>> On 27/12/2018 06:13, Lokesh Vutla wrote: >>>> Previously all msi for a device are allocated in one go >>>> by calling msi_domain_alloc_irq() from a bus layer. This might >>>> not be the case when a device is trying to allocate interrupts >>>> dynamically based on a request to it. >>>> >>>> So introduce msi_domain_alloc/free_irq() apis to allocate a single >>>> msi. prepare and activate operations to be handled by bus layer >>>> calling msi_domain_alloc/free_irq() apis. >>>> >>>> Signed-off-by: Lokesh Vutla >>>> --- >>>> include/linux/msi.h | 3 +++ >>>> kernel/irq/msi.c | 62 +++++++++++++++++++++++++++++---------------- >>>> 2 files changed, 43 insertions(+), 22 deletions(-) >>>> >>>> diff --git a/include/linux/msi.h b/include/linux/msi.h >>>> index 784fb52b9900..474490826f8c 100644 >>>> --- a/include/linux/msi.h >>>> +++ b/include/linux/msi.h >>>> @@ -301,8 +301,11 @@ int msi_domain_set_affinity(struct irq_data *data, const struct cpumask *mask, >>>> struct irq_domain *msi_create_irq_domain(struct fwnode_handle *fwnode, >>>> struct msi_domain_info *info, >>>> struct irq_domain *parent); >>>> +int msi_domain_alloc_irq(struct irq_domain *domain, struct device *dev, >>>> + struct msi_desc *desc, msi_alloc_info_t *arg); >>>> int msi_domain_alloc_irqs(struct irq_domain *domain, struct device *dev, >>>> int nvec); >>>> +void msi_domain_free_irq(struct msi_desc *desc); >>>> void msi_domain_free_irqs(struct irq_domain *domain, struct device *dev); >>>> struct msi_domain_info *msi_get_domain_info(struct irq_domain *domain); >>>> >>>> diff --git a/kernel/irq/msi.c b/kernel/irq/msi.c >>>> index ad26fbcfbfc8..eb7459324113 100644 >>>> --- a/kernel/irq/msi.c >>>> +++ b/kernel/irq/msi.c >>>> @@ -387,6 +387,35 @@ static bool msi_check_reservation_mode(struct irq_domain *domain, >>>> return desc->msi_attrib.is_msix || desc->msi_attrib.maskbit; >>>> } >>>> >>>> +int msi_domain_alloc_irq(struct irq_domain *domain, struct device *dev, >>>> + struct msi_desc *desc, msi_alloc_info_t *arg) >>>> +{ >>>> +struct msi_domain_info *info = domain->host_data; >>>> +struct msi_domain_ops *ops = info->ops; >>>> +int i, ret, virq; >>>> + >>>> +ops->set_desc(arg, desc); >>>> + >>>> +virq = __irq_domain_alloc_irqs(domain, -1, desc->nvec_used, >>>> + dev_to_node(dev), arg, false, >>>> + desc->affinity); >>>> +if (virq < 0) { >>>> +ret = -ENOSPC; >>>> +if (ops->handle_error) >>>> +ret = ops->handle_error(domain, desc, ret); >>>> +if (ops->msi_finish) >>>> +ops->msi_finish(arg, ret); >>>> +return ret; >>>> +} >>>> + >>>> +for (i = 0; i < desc->nvec_used; i++) { >>>> +irq_set_msi_desc_off(virq, i, desc); >>>> +irq_debugfs_copy_devname(virq + i, dev); >>>> +} >>>> + >>>> +return 0; >>>> +} >>>> + >>>> /** >>>> * msi_domain_alloc_irqs - Allocate interrupts from a MSI interrupt domain >>>> * @domain:The domain to allocate from >>>> @@ -404,7 +433,7 @@ int msi_domain_alloc_irqs(struct irq_domain *domain, struct device *dev, >>>> struct irq_data *irq_data; >>>> struct msi_desc *desc; >>>> msi_alloc_info_t arg; >>>> -int i, ret, virq; >>>> +int ret, virq; >>>> bool can_reserve; >>>> >>>> ret = msi_domain_prepare_irqs(domain, dev, nvec, &arg); >>>> @@ -412,24 +441,9 @@ int msi_domain_alloc_irqs(struct irq_domain *domain, struct device *dev, >>>> return ret; >>>> >>>> for_each_msi_entry(desc, dev) { >>>> -ops->set_desc(&arg, desc); >>>> - >>>> -virq = __irq_domain_alloc_irqs(domain, -1, desc->nvec_used, >>>> - dev_to_node(dev), &arg, false, >>>> - desc->affinity); >>>> -if (virq < 0) { >>>> -ret = -ENOSPC; >>>> -if (ops->handle_error) >>>> -ret = ops->handle_error(domain, desc, ret); >>>> -if (ops->msi_finish) >>>> -ops->msi_finish(&arg, ret); >>>> +ret = msi_domain_alloc_irq(domain, dev, desc, &arg); >>>> +if (ret) >>>> return ret; >>>> -} >>>> - >>>> -for (i = 0; i < desc->nvec_used; i++) { >>>> -irq_set_msi_desc_off(virq, i, desc); >>>> -irq_debugfs_copy_devname(virq + i, dev); >>>> -} >>>> } >>>> >>>> if (ops->msi_finish) >>>> @@ -487,6 +501,12 @@ int msi_domain_alloc_irqs(struct irq_domain *domain, struct device *dev, >>>> return ret; >>>> } >>>> >>>> +void msi_domain_free_irq(struct msi_desc *desc) >>>> +{ >>>> +irq_domain_free_irqs(desc->irq, desc->nvec_used); >>>> +desc->irq = 0; >>>> +} >>>> + >>>> /** >>>> * msi_domain_free_irqs - Free interrupts from a MSI interrupt @domain associated tp @dev >>>> * @domain:The domain to managing the interrupts >>>> @@ -503,10 +523,8 @@ void msi_domain_free_irqs(struct irq_domain *domain, struct device *dev) >>>> * enough that there is no IRQ associated to this >>>> * entry. If that's the case, don't do anything. >>>> */ >>>> -if (desc->irq) { >>>> -irq_domain_free_irqs(desc->irq, desc->nvec_used); >>>> -desc->irq = 0; >>>> -} >>>> +if (desc->irq) >>>> +msi_domain_free_irq(desc); >>>> } >>>> } >>>> >>>> >>> >>> I can see some interesting issues with this API. >>> >>> At the moment, MSIs are allocated upfront, and that's usually done >>> before the driver can do anything else. With what you're suggesting >>> here, MSIs can now be allocated at any time, which sounds great. But how >>> does it work when MSIs get added/freed in parallel? I can't see any >>> locking here... >>> >>> It is also pretty nasty that the user of this API has to know about the >>> MSI descriptor. Really, nobody should have to deal with this outside of >>> the MSI layer. >>> >>> The real question is why you need to need to allocate MSIs on demand for >>> a given device. Usually, you allocate them because this is a per-CPU >>> resource, or something similar. What makes it so variable that you need >>> to resort to fine grained MSI allocation? >> >> I added this after the discussion we had in the previous version[1] of this >> series. Let me provide the details again: >> >> As you must be aware INTR is interrupt re-director and INTA is the interrupt >> multiplexer is the SoC. Here we are trying to address the interrupt connection >> route as below: >> Device(Global event) --> INTA --> INTR --> GIC >> >> For the above case you suggested to have the following sw IRQ domain hierarchy: >> INTA multi MSI --> INTA --> INTR --> GIC >> >> The problem here with the INTA MSI is that all the interrupts for a device >> should be pre-allocated during the device probe time. But this is not what we >> wanted especially for the DMA case. >> >> An example DMA ring connection would look like below[2]: >> >> +---------------------+ >> | IA | >> +--------+ | +------+ | +--------+ +------+ >> | ring 1 +----->evtA+->VintX+-------->+ IR +-- ---> GIC +--> >> +--------+ | +------+ | +--------+ +------+ >> Linux IRQ Y >> evtA | | >> | | >> +----------------------+ >> >> So when a DMA client driver requests a dma channel during probe, the DMA driver >> gets a free ring in its allocated range. Then DMA driver requests MSI layer for >> an IRQ. This is why I had to introduce on demand allocation of MSIs for a device. >> >> The reason why we avoided DMA driver to allocate interrupts during its probe as >> it is not aware of the exact no of channels that are going to be used. Also max >> allocation of interrupts will overrun the gic IRQs available to this INTA and >> the IPs that are connected to INTR directly will not get any interrupts. > > But surely there is an upper bound that does exist for a given system, > right? DMA rings are not allocated out of nowhere. Or are you saying Right, there is an upper limit on the DMA rings and DMA driver knows it. But when an MSI is requested for each ring then the global events allocated for the IA gets exhausted and MSI allocation gets failed. Either I did not explain the the complete picture properly or I must be missing something very badly here. Sorry for being dumb here. Let me give a pure software perspective: In the IRQ route we have the following variables: 1) source id 2) source index. 3) global events 4) vint 5) vint status bit(global event mapping to vint) 6) gic_irqs - Source id and source index are managed by MSI client driver. - Global event, vint and vint_status_bit allocation is managed by INTA driver - Grouping of global event and mapping to vint is handled by INTA driver(chained IRQ) - gic_irqs allocation is managed by INTR driver. When MSI client tries to allocate MSI for each source_index(rings for eg.) available in a source, msi domain tries to allocate parent IRQs. Then INTA driver will try to allocate a global event to each source index and fails when the global events are exhausted. What am I missing here? keeping in mind the above scenario, I started with single MSI allocations. I agree there are issues with $patch. But before doing that I was hoping if the above problem can be solved easily. Also the same problem occurs for INTA parent IRQ allocation. INTA probe cannot do a irq_create_fwspec_mapping() for all the available vints as gic_irqs will be exhausted. Where do you think is the best place to create parent IRQs for INTA? In the current implementation I am allocating it during msi_prepare(). So I had to introduce msi_unprepare for freeing parent IRQs. Thanks and regards, Lokesh > that when a DMA client requests DMA rings, it doesn't know how many it > wants? Or does it? I don't think this is new in any of TI's design (the > crossbar already had such limitation). > > That being said, I'm open to revisiting this, but you must then > introduce the right level of mutual exclusion in core code, as the > msi_desc list now becomes much more dynamic and things would otherwise > break badly. This has implications all over the place, and probably > requires quite a bit of work (things like for_each_msi_desc_entry will > break). > > Between the two solutions outlined above, the first one is much easier > to achieve than the second one. > > Another avenue to explore would be to let the clients allocate as many > MSIs as they want upfront, and use the "activate" callback to actually > perform the assignment when such interrupt gets requested. In a way, > this is not that different from what x86 does with the "early > reservation, late assignment" type mechanism. > > Thanks, > > M. >