From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 0D26A46A23; Mon, 30 Jun 2025 10:50:32 +0200 (CEST) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id CB95F40264; Mon, 30 Jun 2025 10:50:31 +0200 (CEST) Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) by mails.dpdk.org (Postfix) with ESMTP id E77C64025D for ; Mon, 30 Jun 2025 10:50:29 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1751273429; h=from:from: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:autocrypt:autocrypt; bh=wLNT8C51kN6eu4Nw5xR4op4dSG3hb/caNnQth7GDoF4=; b=UbM1URmpmmmZzVFFwz4jDf8o8yPg5zXtwBqu4+fH/Jwbk7I2XecHGfjb0gp4TpdsAADWvR vtrWoxAs3AEPOJEOOp0JE2LggXfJsboyxqDcycmRkDBY4vOadxHxNkdYYGBq1HkAde4+pG fpU67PJhJ41VqYIz0Acz74Zjg7C8h5I= Received: from mx-prod-mc-08.mail-002.prod.us-west-2.aws.redhat.com (ec2-35-165-154-97.us-west-2.compute.amazonaws.com [35.165.154.97]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-540-pW7n4n3-P42FGCa__LP2pQ-1; Mon, 30 Jun 2025 04:50:26 -0400 X-MC-Unique: pW7n4n3-P42FGCa__LP2pQ-1 X-Mimecast-MFC-AGG-ID: pW7n4n3-P42FGCa__LP2pQ_1751273424 Received: from mx-prod-int-01.mail-002.prod.us-west-2.aws.redhat.com (mx-prod-int-01.mail-002.prod.us-west-2.aws.redhat.com [10.30.177.4]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) by mx-prod-mc-08.mail-002.prod.us-west-2.aws.redhat.com (Postfix) with ESMTPS id A12601809C8B; Mon, 30 Jun 2025 08:50:23 +0000 (UTC) Received: from [10.44.22.8] (unknown [10.44.22.8]) by mx-prod-int-01.mail-002.prod.us-west-2.aws.redhat.com (Postfix) with ESMTPS id C3E3E30001B1; Mon, 30 Jun 2025 08:50:15 +0000 (UTC) Message-ID: <108a1a35-b8ed-4b1f-b232-d0e01c9c9609@redhat.com> Date: Mon, 30 Jun 2025 10:50:11 +0200 MIME-Version: 1.0 User-Agent: Mozilla Thunderbird Subject: Re: [PATCH v4 1/1] vhost: handle virtqueue locking for memory hotplug To: Danylo Vodopianov , "thomas@monjalon.net" , "aman.deep.singh@intel.com" , "yuying.zhang@intel.com" , "orika@nvidia.com" , "mcoqueli@redhat.com" , Christian Koue Muf , "matan@mellanox.com" , "david.marchand@redhat.com" , Mykola Kostenok , Serhii Iliushyk Cc: "stephen@networkplumber.org" , "dev@dpdk.org" , Chenbo Xia References: <20250602084025.1881768-2-dvo-plv@napatech.com> <20250602085005.1882499-1-dvo-plv@napatech.com> <20250602085005.1882499-2-dvo-plv@napatech.com> <6ee82aef-dbb1-4e5f-8a53-6a956161c1db@redhat.com> From: Maxime Coquelin Autocrypt: addr=maxime.coquelin@redhat.com; keydata= xsFNBFOEQQIBEADjNLYZZqghYuWv1nlLisptPJp+TSxE/KuP7x47e1Gr5/oMDJ1OKNG8rlNg kLgBQUki3voWhUbMb69ybqdMUHOl21DGCj0BTU3lXwapYXOAnsh8q6RRM+deUpasyT+Jvf3a gU35dgZcomRh5HPmKMU4KfeA38cVUebsFec1HuJAWzOb/UdtQkYyZR4rbzw8SbsOemtMtwOx YdXodneQD7KuRU9IhJKiEfipwqk2pufm2VSGl570l5ANyWMA/XADNhcEXhpkZ1Iwj3TWO7XR uH4xfvPl8nBsLo/EbEI7fbuUULcAnHfowQslPUm6/yaGv6cT5160SPXT1t8U9QDO6aTSo59N jH519JS8oeKZB1n1eLDslCfBpIpWkW8ZElGkOGWAN0vmpLfdyiqBNNyS3eGAfMkJ6b1A24un /TKc6j2QxM0QK4yZGfAxDxtvDv9LFXec8ENJYsbiR6WHRHq7wXl/n8guyh5AuBNQ3LIK44x0 KjGXP1FJkUhUuruGyZsMrDLBRHYi+hhDAgRjqHgoXi5XGETA1PAiNBNnQwMf5aubt+mE2Q5r qLNTgwSo2dpTU3+mJ3y3KlsIfoaxYI7XNsPRXGnZi4hbxmeb2NSXgdCXhX3nELUNYm4ArKBP LugOIT/zRwk0H0+RVwL2zHdMO1Tht1UOFGfOZpvuBF60jhMzbQARAQABzSxNYXhpbWUgQ29x dWVsaW4gPG1heGltZS5jb3F1ZWxpbkByZWRoYXQuY29tPsLBeAQTAQIAIgUCV3u/5QIbAwYL CQgHAwIGFQgCCQoLBBYCAwECHgECF4AACgkQyjiNKEaHD4ma2g/+P+Hg9WkONPaY1J4AR7Uf kBneosS4NO3CRy0x4WYmUSLYMLx1I3VH6SVjqZ6uBoYy6Fs6TbF6SHNc7QbB6Qjo3neqnQR1 71Ua1MFvIob8vUEl3jAR/+oaE1UJKrxjWztpppQTukIk4oJOmXbL0nj3d8dA2QgHdTyttZ1H xzZJWWz6vqxCrUqHU7RSH9iWg9R2iuTzii4/vk1oi4Qz7y/q8ONOq6ffOy/t5xSZOMtZCspu Mll2Szzpc/trFO0pLH4LZZfz/nXh2uuUbk8qRIJBIjZH3ZQfACffgfNefLe2PxMqJZ8mFJXc RQO0ONZvwoOoHL6CcnFZp2i0P5ddduzwPdGsPq1bnIXnZqJSl3dUfh3xG5ArkliZ/++zGF1O wvpGvpIuOgLqjyCNNRoR7cP7y8F24gWE/HqJBXs1qzdj/5Hr68NVPV1Tu/l2D1KMOcL5sOrz 2jLXauqDWn1Okk9hkXAP7+0Cmi6QwAPuBT3i6t2e8UdtMtCE4sLesWS/XohnSFFscZR6Vaf3 gKdWiJ/fW64L6b9gjkWtHd4jAJBAIAx1JM6xcA1xMbAFsD8gA2oDBWogHGYcScY/4riDNKXi lw92d6IEHnSf6y7KJCKq8F+Jrj2BwRJiFKTJ6ChbOpyyR6nGTckzsLgday2KxBIyuh4w+hMq TGDSp2rmWGJjASrOwU0EVPSbkwEQAMkaNc084Qvql+XW+wcUIY+Dn9A2D1gMr2BVwdSfVDN7 0ZYxo9PvSkzh6eQmnZNQtl8WSHl3VG3IEDQzsMQ2ftZn2sxjcCadexrQQv3Lu60Tgj7YVYRM H+fLYt9W5YuWduJ+FPLbjIKynBf6JCRMWr75QAOhhhaI0tsie3eDsKQBA0w7WCuPiZiheJaL 4MDe9hcH4rM3ybnRW7K2dLszWNhHVoYSFlZGYh+MGpuODeQKDS035+4H2rEWgg+iaOwqD7bg CQXwTZ1kSrm8NxIRVD3MBtzp9SZdUHLfmBl/tLVwDSZvHZhhvJHC6Lj6VL4jPXF5K2+Nn/Su CQmEBisOmwnXZhhu8ulAZ7S2tcl94DCo60ReheDoPBU8PR2TLg8rS5f9w6mLYarvQWL7cDtT d2eX3Z6TggfNINr/RTFrrAd7NHl5h3OnlXj7PQ1f0kfufduOeCQddJN4gsQfxo/qvWVB7PaE 1WTIggPmWS+Xxijk7xG6x9McTdmGhYaPZBpAxewK8ypl5+yubVsE9yOOhKMVo9DoVCjh5To5 aph7CQWfQsV7cd9PfSJjI2lXI0dhEXhQ7lRCFpf3V3mD6CyrhpcJpV6XVGjxJvGUale7+IOp sQIbPKUHpB2F+ZUPWds9yyVxGwDxD8WLqKKy0WLIjkkSsOb9UBNzgRyzrEC9lgQ/ABEBAAHC wV8EGAECAAkFAlT0m5MCGwwACgkQyjiNKEaHD4nU8hAAtt0xFJAy0sOWqSmyxTc7FUcX+pbD KVyPlpl6urKKMk1XtVMUPuae/+UwvIt0urk1mXi6DnrAN50TmQqvdjcPTQ6uoZ8zjgGeASZg jj0/bJGhgUr9U7oG7Hh2F8vzpOqZrdd65MRkxmc7bWj1k81tOU2woR/Gy8xLzi0k0KUa8ueB iYOcZcIGTcs9CssVwQjYaXRoeT65LJnTxYZif2pfNxfINFzCGw42s3EtZFteczClKcVSJ1+L +QUY/J24x0/ocQX/M1PwtZbB4c/2Pg/t5FS+s6UB1Ce08xsJDcwyOPIH6O3tccZuriHgvqKP yKz/Ble76+NFlTK1mpUlfM7PVhD5XzrDUEHWRTeTJSvJ8TIPL4uyfzhjHhlkCU0mw7Pscyxn DE8G0UYMEaNgaZap8dcGMYH/96EfE5s/nTX0M6MXV0yots7U2BDb4soLCxLOJz4tAFDtNFtA wLBhXRSvWhdBJZiig/9CG3dXmKfi2H+wdUCSvEFHRpgo7GK8/Kh3vGhgKmnnxhl8ACBaGy9n fxjSxjSO6rj4/MeenmlJw1yebzkX8ZmaSi8BHe+n6jTGEFNrbiOdWpJgc5yHIZZnwXaW54QT UhhSjDL1rV2B4F28w30jYmlRmm2RdN7iCZfbyP3dvFQTzQ4ySquuPkIGcOOHrvZzxbRjzMx1 Mwqu3GQ= In-Reply-To: X-Scanned-By: MIMEDefang 3.4.1 on 10.30.177.4 X-Mimecast-Spam-Score: 0 X-Mimecast-MFC-PROC-ID: Zo8tsLOFKvUkehK0w5MFa7ArdKgUNnyDSPbGISxfg0I_1751273424 X-Mimecast-Originator: redhat.com Content-Language: en-US Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org Hi, On 6/19/25 3:01 PM, Danylo Vodopianov wrote: > Hi, Maxime > > > I understand your point. However we coould have a situation like this, > could resize twice: > > VHOST_CONFIG: (/usr/local/var/run/stdvio4) read message > VHOST_USER_SET_MEM_TABLE > VHOST_CONFIG: (/usr/local/var/run/stdvio4) guest memory region size: > 0x40000000 > VHOST_CONFIG: (/usr/local/var/run/stdvio4)  guest physical addr: > 0x140000000 > VHOST_CONFIG: (/usr/local/var/run/stdvio4)  guest virtual  addr: > 0x140000000 > VHOST_CONFIG: (/usr/local/var/run/stdvio4)  host  virtual  addr: > 0x7fde00000000 > VHOST_CONFIG: (/usr/local/var/run/stdvio4)  mmap addr : 0x7fde00000000 > VHOST_CONFIG: (/usr/local/var/run/stdvio4)  mmap size : 0x40000000 > VHOST_CONFIG: (/usr/local/var/run/stdvio4)  mmap align: 0x40000000 > VHOST_CONFIG: (/usr/local/var/run/stdvio4)  mmap off  : 0x0 > VHOST_CONFIG: (/usr/local/var/run/stdvio4) guest memory region size: > 0x40000000 > VHOST_CONFIG: (/usr/local/var/run/stdvio4)  guest physical addr: > 0x11c0000000 > VHOST_CONFIG: (/usr/local/var/run/stdvio4)  guest virtual  addr: > 0x11c0000000 > VHOST_CONFIG: (/usr/local/var/run/stdvio4)  host  virtual  addr: > 0x7fddc0000000 > VHOST_CONFIG: (/usr/local/var/run/stdvio4)  mmap addr : 0x7fddc0000000 > VHOST_CONFIG: (/usr/local/var/run/stdvio4)  mmap size : 0x40000000 > VHOST_CONFIG: (/usr/local/var/run/stdvio4)  mmap align: 0x40000000 > VHOST_CONFIG: (/usr/local/var/run/stdvio4)  mmap off  : 0x0 > > > When we set memory region twice. After first iteration > VIRTIO_DEV_VDPA_CONFIGURED flag will be unset here: https://github.com/ > DPDK/dpdk/blob/main/lib/vhost/vhost_user.c#L1425 DPDK/dpdk/blob/main/lib/vhost/vhost_user.c#L1425> On the second > iteration, this leads to an rte_panic, as queues are accessed without a > lock. > > So we should check vhost message id ( VHOST_USER_SET_MEM_TABLE ). > > However, extend VHOST_USER_ASSERT_LOCK macros with additional check if > we work with this message VHOST_USER_SET_MEM_TABLE not handle this case, > therefore translate_ring_addresses function calls > q_assert_lock directly, without macros wrapper. In this function is > check access_lock vhost_virtqueue and this case should be handle. > > To address the described issue, we need to make the following changes: > > 1. *Extend VHOST_USER_ASSERT_LOCK macro*: > * Add a check for the VHOST_USER_SET_MEM_TABLE message ID. > * Skip rte_panic if the message ID matches VHOST_USER_SET_MEM_TABLE. > 2. *Modify translate_ring_addresses function*: > * Extend its signature to include the id parameter (message ID). > * Add logic to skip vq_assert_lock if the message ID matches > VHOST_USER_SET_MEM_TABLE. > > > 1. > Extend *VHOST_USER_ASSERT_LOCK* Macro: > > > #define VHOST_USER_ASSERT_LOCK(dev, vq, id) \ >     do { \ >         if ((id) == VHOST_USER_SET_MEM_TABLE) \ >             break; \ >         if (!(vq)->access_ok) \ >             rte_panic("Virtqueue access lock not held\n"); \ >     } while (0 > > > 2. > Modify *translate_ring_addresses* Function: > > > static int > translate_ring_addresses(struct virtio_net *dev, > struct vhost_virtqueue *vq, uint32_t id) > { >     if (id != VHOST_USER_SET_MEM_TABLE) >         vq_assert_lock(dev, vq); > >     // Existing logic for translating ring addresses... >     // ...existing code... >     return 0; > } > > > > This approach requires extending more functions with conditional checks > to handle cases where queue locking should be ignored when the memory > table is impacted. > > Do you have any thoughts about this? Or I should rework my patchset > according to this described solution above ? And if instead of relying on the VIRTIO_DEV_VDPA_CONFIGURED flag as I suggested, we rely on the vdpa_dev being set? > > > ------------------------------------------------------------------------ > *От:* Maxime Coquelin > *Отправлено:* 12 июня 2025 г. 14:38 > *Кому:* Danylo Vodopianov ; thomas@monjalon.net > ; aman.deep.singh@intel.com > ; yuying.zhang@intel.com > ; orika@nvidia.com ; > mcoqueli@redhat.com ; Christian Koue Muf > ; matan@mellanox.com ; > david.marchand@redhat.com ; Mykola Kostenok > ; Serhii Iliushyk > *Копия:* stephen@networkplumber.org ; > dev@dpdk.org ; Chenbo Xia > *Тема:* Re: [PATCH v4 1/1] vhost: handle virtqueue locking for memory > hotplug > Hi Danylo, > > On 6/4/25 10:32 AM, Danylo Vodopianov wrote: >> Hello, Maxime >> >> Thank you for your review. >> If I understand correctly, you propose modifying the | >> VHOST_USER_ASSERT_LOCK()| macro so that a |VHOST_USER_SET_MEM_TABLE| >>   request does not trigger an assertion. >> However, I believe such modification would not be appropriate, as it >> would revert the logic introduced in commit |5e8fcc60b59d| ("vhost: >> enhance virtqueue access lock asserts"). With this approach, we would be >> performing memory hotplug without queue locking, which could lead to >> unintended consequences. >> Regarding VDPA device regression. We faced with this issue when we >> request the number of lcores that the default amount of memory on the >> socket cannot handle it. >> So, regression occurred during the startup part, during device >> configuration when it creates pkt mbuf pool. >> >> Let me know your thoughts regarding this. > > No, my point was to modify VHOST_USER_ASSERT_LOCK() to no trigger an > assertion in case vDPA is configured, as we don't want to lock in this > case. > > Regards, > Maxime > >> ------------------------------------------------------------------------ >> *От:* Maxime Coquelin >> *Отправлено:* 3 июня 2025 г. 15:30 >> *Кому:* Danylo Vodopianov ; thomas@monjalon.net >> ; aman.deep.singh@intel.com >> ; yuying.zhang@intel.com >> ; orika@nvidia.com ; >> mcoqueli@redhat.com ; Christian Koue Muf >> ; matan@mellanox.com ; >> david.marchand@redhat.com ; Mykola Kostenok >> ; Serhii Iliushyk >> *Копия:* stephen@networkplumber.org ; >> dev@dpdk.org ; Chenbo Xia >> *Тема:* Re: [PATCH v4 1/1] vhost: handle virtqueue locking for memory >> hotplug >> Hello Danylo, >> >> On 6/2/25 10:50 AM, Danylo Vodopianov wrote: >>> For vDPA devices, virtqueues are not locked once the device has been >>> configured. In the >>> commit 5e8fcc60b59d ("vhost: enhance virtqueue access lock asserts"), >>> the asserts were enhanced to trigger rte_panic functionality, preventing >>> access to virtqueues without locking. However, this change introduced >>> an issue where the memory hotplug mechanism, added in the >>> commit 127f9c6f7b78 ("vhost: handle memory hotplug with vDPA devices"), >>> no longer works. Since it expects for all queues are locked. >>> >>> During the initialization of a vDPA device, the driver sets the >>> VIRTIO_DEV_VDPA_CONFIGURED flag, which prevents the >>> vhost_user_lock_all_queue_pairs function from locking the >>> virtqueues. This leads to the error: the VIRTIO_DEV_VDPA_CONFIGURED >>> flag allows the use of the hotplug mechanism, but it fails >>> because the virtqueues are not locked, while it expects to be locked >>> for VHOST_USER_SET_MEM_TABLE in the table VHOST_MESSAGE_HANDLERS. >>> >>> This patch addresses the issue by enhancing the conditional statement >>> to include a new condition. Specifically, when the device receives the >>> VHOST_USER_SET_MEM_TABLE request, the virtqueues are locked to update >>> the basic configurations and hotplug the guest memory. >>> >>> This fix does not impact access lock when vDPA driver is configured >>> for other unnecessary message handlers. >>> >>> Manual memory configuring with "--socket-mem" option allows to avoid >>> hotplug mechanism using. >> >> s/using/use/ >> >> It needs a fixes tag, and stable@dpdk.org should be cc'ed, so that it >> gets backported to LTS branches. >> >>> >>> Signed-off-by: Danylo Vodopianov >>> --- >>>   lib/vhost/vhost_user.c | 8 +++++++- >>>   1 file changed, 7 insertions(+), 1 deletion(-) >>> >>> diff --git a/lib/vhost/vhost_user.c b/lib/vhost/vhost_user.c >>> index ec950acf97..16d03e1033 100644 >>> --- a/lib/vhost/vhost_user.c >>> +++ b/lib/vhost/vhost_user.c >>> @@ -3178,7 +3178,13 @@ vhost_user_msg_handler(int vid, int fd) >>>         * would cause a dead lock. >>>         */ >>>        if (msg_handler != NULL && msg_handler->lock_all_qps) { >>> -             if (!(dev->flags & VIRTIO_DEV_VDPA_CONFIGURED)) { >>> +             /* Lock all queue pairs if the device is not configured for vDPA, >>> +              * or if it is configured for vDPA but the request is VHOST_USER_SET_MEM_TABLE. >>> +              * This ensures proper queue locking for memory table updates and guest >>> +              * memory hotplug. >>> +              */ >>> +             if (!(dev->flags & VIRTIO_DEV_VDPA_CONFIGURED) || >>> +                     request == VHOST_USER_SET_MEM_TABLE) { >> >> It looks like a workaround, and I'm afraid it could cause regression >> with some vDPA devices, or that it would not be enough and we would have >> to add other requests as exception. >> >> >> Wouldn't it better to modify VHOST_USER_ASSERT_LOCK() so that it takes >> into account the VIRTIO_DEV_VDPA_CONFIGURED flag? >> >> Thanks, >> Maxime >> >>>                        vhost_user_lock_all_queue_pairs(dev); >>>                        unlock_required = 1; >>>                } >> >