{"id":2233165,"url":"http://patchwork.ozlabs.org/api/1.2/patches/2233165/?format=json","web_url":"http://patchwork.ozlabs.org/project/linux-pci/patch/20260505200510.2954-2-alifm@linux.ibm.com/","project":{"id":28,"url":"http://patchwork.ozlabs.org/api/1.2/projects/28/?format=json","name":"Linux PCI development","link_name":"linux-pci","list_id":"linux-pci.vger.kernel.org","list_email":"linux-pci@vger.kernel.org","web_url":null,"scm_url":null,"webscm_url":null,"list_archive_url":"","list_archive_url_format":"","commit_url_format":""},"msgid":"<20260505200510.2954-2-alifm@linux.ibm.com>","list_archive_url":null,"date":"2026-05-05T20:05:04","name":"[v15,1/7] PCI: Allow per function PCI slots to fix slot reset on s390","commit_ref":null,"pull_url":null,"state":"new","archived":false,"hash":"8e7f2b0457c7e98f0c0436d5e4c94443ac3c933b","submitter":{"id":73785,"url":"http://patchwork.ozlabs.org/api/1.2/people/73785/?format=json","name":"Farhan Ali","email":"alifm@linux.ibm.com"},"delegate":null,"mbox":"http://patchwork.ozlabs.org/project/linux-pci/patch/20260505200510.2954-2-alifm@linux.ibm.com/mbox/","series":[{"id":502892,"url":"http://patchwork.ozlabs.org/api/1.2/series/502892/?format=json","web_url":"http://patchwork.ozlabs.org/project/linux-pci/list/?series=502892","date":"2026-05-05T20:05:06","name":"Error recovery for vfio-pci devices on s390x","version":15,"mbox":"http://patchwork.ozlabs.org/series/502892/mbox/"}],"comments":"http://patchwork.ozlabs.org/api/patches/2233165/comments/","check":"pending","checks":"http://patchwork.ozlabs.org/api/patches/2233165/checks/","tags":{},"related":[],"headers":{"Return-Path":"\n ","X-Original-To":["incoming@patchwork.ozlabs.org","linux-pci@vger.kernel.org"],"Delivered-To":"patchwork-incoming@legolas.ozlabs.org","Authentication-Results":["legolas.ozlabs.org;\n\tdkim=pass (2048-bit key;\n unprotected) header.d=ibm.com header.i=@ibm.com header.a=rsa-sha256\n header.s=pp1 header.b=bCm5IyuY;\n\tdkim-atps=neutral","legolas.ozlabs.org;\n spf=pass (sender SPF authorized) smtp.mailfrom=vger.kernel.org\n (client-ip=2600:3c09:e001:a7::12fc:5321; helo=sto.lore.kernel.org;\n envelope-from=linux-pci+bounces-53784-incoming=patchwork.ozlabs.org@vger.kernel.org;\n receiver=patchwork.ozlabs.org)","smtp.subspace.kernel.org;\n\tdkim=pass (2048-bit key) header.d=ibm.com header.i=@ibm.com\n header.b=\"bCm5IyuY\"","smtp.subspace.kernel.org;\n arc=none smtp.client-ip=148.163.156.1","smtp.subspace.kernel.org;\n dmarc=pass (p=none dis=none) header.from=linux.ibm.com","smtp.subspace.kernel.org;\n spf=pass smtp.mailfrom=linux.ibm.com"],"Received":["from sto.lore.kernel.org (sto.lore.kernel.org\n [IPv6:2600:3c09:e001:a7::12fc:5321])\n\t(using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits)\n\t key-exchange x25519)\n\t(No client certificate requested)\n\tby legolas.ozlabs.org (Postfix) with ESMTPS id 4g98fc4d4Cz1yJx\n\tfor ; Wed, 06 May 2026 06:06:32 +1000 (AEST)","from smtp.subspace.kernel.org (conduit.subspace.kernel.org\n [100.90.174.1])\n\tby sto.lore.kernel.org (Postfix) with ESMTP id E3EEB302B921\n\tfor ; Tue, 5 May 2026 20:05:54 +0000 (UTC)","from localhost.localdomain (localhost.localdomain [127.0.0.1])\n\tby smtp.subspace.kernel.org (Postfix) with ESMTP id 2AE143DC4D3;\n\tTue, 5 May 2026 20:05:22 +0000 (UTC)","from mx0a-001b2d01.pphosted.com (mx0a-001b2d01.pphosted.com\n [148.163.156.1])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits))\n\t(No client certificate requested)\n\tby smtp.subspace.kernel.org (Postfix) with ESMTPS id 969853D6CB2;\n\tTue, 5 May 2026 20:05:19 +0000 (UTC)","from pps.filterd (m0353729.ppops.net [127.0.0.1])\n\tby mx0a-001b2d01.pphosted.com (8.18.1.11/8.18.1.11) with ESMTP id\n 645IxNTX2692843;\n\tTue, 5 May 2026 20:05:15 GMT","from ppma13.dal12v.mail.ibm.com\n (dd.9e.1632.ip4.static.sl-reverse.com [50.22.158.221])\n\tby mx0a-001b2d01.pphosted.com (PPS) with ESMTPS id 4dw9x4ngy8-1\n\t(version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT);\n\tTue, 05 May 2026 20:05:14 +0000 (GMT)","from pps.filterd (ppma13.dal12v.mail.ibm.com [127.0.0.1])\n\tby ppma13.dal12v.mail.ibm.com (8.18.1.7/8.18.1.7) with ESMTP id\n 645JsUQc014212;\n\tTue, 5 May 2026 20:05:14 GMT","from smtprelay05.wdc07v.mail.ibm.com ([172.16.1.72])\n\tby ppma13.dal12v.mail.ibm.com (PPS) with ESMTPS id 4dwwtgb3a5-1\n\t(version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT);\n\tTue, 05 May 2026 20:05:14 +0000 (GMT)","from smtpav06.dal12v.mail.ibm.com (smtpav06.dal12v.mail.ibm.com\n [10.241.53.105])\n\tby smtprelay05.wdc07v.mail.ibm.com (8.14.9/8.14.9/NCO v10.0) with ESMTP id\n 645K5C1P27198046\n\t(version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256 verify=OK);\n\tTue, 5 May 2026 20:05:12 GMT","from smtpav06.dal12v.mail.ibm.com (unknown [127.0.0.1])\n\tby IMSVA (Postfix) with ESMTP id 2365D58055;\n\tTue, 5 May 2026 20:05:12 +0000 (GMT)","from smtpav06.dal12v.mail.ibm.com (unknown [127.0.0.1])\n\tby IMSVA (Postfix) with ESMTP id 67E6358043;\n\tTue, 5 May 2026 20:05:11 +0000 (GMT)","from IBM-D32RQW3.ibm.com (unknown [9.61.242.219])\n\tby smtpav06.dal12v.mail.ibm.com (Postfix) with ESMTP;\n\tTue, 5 May 2026 20:05:11 +0000 (GMT)"],"ARC-Seal":"i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116;\n\tt=1778011521; cv=none;\n b=mZYEGd58sS8yNrUUVydMic1bIaU0bCb5JZEMHltgkDnZmxSoOwTjRQmXZp606H0a0Ouqf6YgGfaDR+jeXxgBTkYcgT2XNCwa3lmPZ9oSKlPo9yX1GsfKUj0gEWNkV6adGemUHkta3i07WrPyC80X4K4XEe+1wd8SFaWG5Ua5wkk=","ARC-Message-Signature":"i=1; a=rsa-sha256; d=subspace.kernel.org;\n\ts=arc-20240116; t=1778011521; c=relaxed/simple;\n\tbh=0O2tkwQnRY2NgTfcng5FhRNS7XbWmh46GDptlDcRPm4=;\n\th=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References:\n\t MIME-Version;\n b=ZTFCIg9AumMWnoS2oRB9kx+wP1zrIpw74U7aTAJvUATsWC9S+PYaBP4KySvjJrU1cDlQ5h1k0avNGf9+eNv7vnPDEMwmbZpFxqnTHpIRo2d2k8xUu88zsUCkxn+ou7wYkLPOX+bHyJGH/fqGwxm6dPGWd617GNJJNYZCPgYqR84=","ARC-Authentication-Results":"i=1; smtp.subspace.kernel.org;\n dmarc=pass (p=none dis=none) header.from=linux.ibm.com;\n spf=pass smtp.mailfrom=linux.ibm.com;\n dkim=pass (2048-bit key) header.d=ibm.com header.i=@ibm.com\n header.b=bCm5IyuY; arc=none smtp.client-ip=148.163.156.1","DKIM-Signature":"v=1; a=rsa-sha256; c=relaxed/relaxed; d=ibm.com; h=cc\n\t:content-transfer-encoding:date:from:in-reply-to:message-id\n\t:mime-version:references:subject:to; s=pp1; bh=2g2+8cwzfC472wSkG\n\tK/bPeX8sWH0JpNCZ8A2YKaMGcQ=; b=bCm5IyuYgjZTvFdIMiKcupetvhG9nbQrk\n\tkopiLD29ZsIqx3780jvUbQNjRLq62hxL7107xw4SLd2g2LCvs/OAjoQeP6Pj6sLl\n\tHKRYZnywH2NCntbzz9JmYd3OwlmTnxOcgl8So1NnBLYkcpLHGgq7oygWrB/mX1xR\n\tC/x9GKGjDN00hxz/ITnxqVtT/csID1W08L0NzfVGt9T9oHq5rgKpfYcOXLhMoHz2\n\tRW7tRPmTDjKE+UXYtP18s5L7bSN5W9XmHO6aHiwklMOPXLrc5TLO+wCblOxhVDWe\n\tGSANYJ4yux28Q4Ktf0H7hdatwIiW9D56unbKk+2QZsdihVBRmJJNg==","From":"Farhan Ali ","To":"linux-s390@vger.kernel.org, linux-kernel@vger.kernel.org,\n linux-pci@vger.kernel.org","Cc":"helgaas@kernel.org, alex@shazbot.org, alifm@linux.ibm.com,\n schnelle@linux.ibm.com, mjrosato@linux.ibm.com, stable@vger.kernel.org","Subject":"[PATCH v15 1/7] PCI: Allow per function PCI slots to fix slot reset\n on s390","Date":"Tue, 5 May 2026 13:05:04 -0700","Message-ID":"<20260505200510.2954-2-alifm@linux.ibm.com>","X-Mailer":"git-send-email 2.43.0","In-Reply-To":"<20260505200510.2954-1-alifm@linux.ibm.com>","References":"<20260505200510.2954-1-alifm@linux.ibm.com>","Precedence":"bulk","X-Mailing-List":"linux-pci@vger.kernel.org","List-Id":"","List-Subscribe":"","List-Unsubscribe":"","MIME-Version":"1.0","Content-Transfer-Encoding":"8bit","X-TM-AS-GCONF":"00","X-Proofpoint-Spam-Details-Enc":"AW1haW4tMjYwNTA1MDE5MSBTYWx0ZWRfX9vcBQyBKbgaD\n 9mJxWkO7AyG8tlp39XHpD+l4oJhXS6n/UaCUO91KK5fOF8vsPjINTqaPy1mcAeSF+moWkfWXZlF\n 5X4FW3arV80Oo0gMVlXhHvzhJdSJAwtiSm30yitLOFE5P234hvWFZhtrkwCOyjLGla0RDSFHZIt\n 6ZQdICk3LOh9jPEc8c2vs1SytXdwnIvxlTSnm2dNA8oq02HTTYuRmFj4uCpICT1yJCugImpdFw3\n TUtBTUZA41iwFYbZr/qCb3nmapTQq97cSdXRwUV3QzGJkxmTgm3fdjq4UFGHxKlMC7SlX/zvCqt\n ebHb7gWbeUJoIDLP5mD8cYJdHc2cN1PfvwcUWJaF7agR3VqDI+WiZO43+iEKdwX7FnUg4SGIhkF\n LEuZayQvvugyZZNn3cgE98S+3zCHra1CT1gHqvd81BXscAyBAjR8N2Tq+WYgJIbS69eY4oQjdB2\n ri/kbbgxtFwmyIbV3EQ==","X-Proofpoint-ORIG-GUID":"WvUca-KbaYvziepamqTsU66LvJ82-Sgb","X-Proofpoint-GUID":"WvUca-KbaYvziepamqTsU66LvJ82-Sgb","X-Authority-Analysis":"v=2.4 cv=W7UIkxWk c=1 sm=1 tr=0 ts=69fa4d7a cx=c_pps\n a=AfN7/Ok6k8XGzOShvHwTGQ==:117 a=AfN7/Ok6k8XGzOShvHwTGQ==:17\n a=NGcC8JguVDcA:10 a=VkNPw1HP01LnGYTKEx00:22 a=RnoormkPH1_aCDwRdu11:22\n a=uAbxVGIbfxUO_5tXvNgY:22 a=VwQbUJbxAAAA:8 a=VnNF1IyMAAAA:8\n a=9_SSeoq7eJPDN4OswFYA:9 a=O8hF6Hzn-FEA:10","X-Proofpoint-Virus-Version":"vendor=baseguard\n engine=ICAP:2.0.293,Aquarius:18.0.1143,Hydra:6.1.51,FMLib:17.12.100.49\n definitions=2026-05-05_02,2026-04-30_02,2025-10-01_01","X-Proofpoint-Spam-Details":"rule=outbound_notspam policy=outbound score=0\n priorityscore=1501 adultscore=0 lowpriorityscore=0 malwarescore=0\n suspectscore=0 spamscore=0 clxscore=1015 phishscore=0 bulkscore=0\n impostorscore=0 classifier=typeunknown authscore=0 authtc= authcc=\n route=outbound adjust=0 reason=mlx scancount=1 engine=8.22.0-2604200000\n definitions=main-2605050191"},"content":"On s390 systems, which use a machine level hypervisor, PCI devices are\nalways accessed through a form of PCI pass-through which fundamentally\noperates on a per PCI function granularity. This is also reflected in the\ns390 PCI hotplug driver which creates hotplug slots for individual PCI\nfunctions. Its reset_slot() function, which is a wrapper for\nzpci_hot_reset_device(), thus also resets individual functions.\n\nCurrently, the kernel's PCI_SLOT() macro assigns the same pci_slot object\nto multifunction devices. This approach worked fine on s390 systems that\nonly exposed virtual functions as individual PCI domains to the operating\nsystem. Since commit 44510d6fa0c0 (\"s390/pci: Handling multifunctions\")\ns390 supports exposing the topology of multifunction PCI devices by\ngrouping them in a shared PCI domain. This creates a problem when resetting\na function through the hotplug driver's slot_reset() interface.\n\nWhen attempting to reset a function through the hotplug driver, the shared\nslot assignment causes the wrong function to be reset instead of the\nintended one. It also leaks memory as we do create a pci_slot object for\nthe function, but don't correctly free it in pci_slot_release().\n\nAdd a flag for struct pci_slot to allow per function PCI slots for\nfunctions managed through a hypervisor, which exposes individual PCI\nfunctions while retaining the topology. Since we can use all 8 bits\nfor slot 'number' (for ARI devices), change slot 'number' u16 to\naccount for special values -1 and PCI_SLOT_ALL_DEVICES.\n\nFixes: 44510d6fa0c0 (\"s390/pci: Handling multifunctions\")\nCc: stable@vger.kernel.org\nSuggested-by: Niklas Schnelle \nReviewed-by: Niklas Schnelle \nSigned-off-by: Farhan Ali \n---\n drivers/pci/hotplug/rpaphp_slot.c | 2 +-\n drivers/pci/pci.c | 5 +++--\n drivers/pci/slot.c | 33 +++++++++++++++++++++++--------\n include/linux/pci.h | 8 ++++++--\n 4 files changed, 35 insertions(+), 13 deletions(-)","diff":"diff --git a/drivers/pci/hotplug/rpaphp_slot.c b/drivers/pci/hotplug/rpaphp_slot.c\nindex 67362e5b9971..92eabf5f61b9 100644\n--- a/drivers/pci/hotplug/rpaphp_slot.c\n+++ b/drivers/pci/hotplug/rpaphp_slot.c\n@@ -84,7 +84,7 @@ int rpaphp_register_slot(struct slot *slot)\n \tstruct hotplug_slot *php_slot = &slot->hotplug_slot;\n \tu32 my_index;\n \tint retval;\n-\tint slotno = -1;\n+\tint slotno = PCI_SLOT_PLACEHOLDER;\n \n \tdbg(\"%s registering slot:path[%pOF] index[%x], name[%s] pdomain[%x] type[%d]\\n\",\n \t\t__func__, slot->dn, slot->index, slot->name,\ndiff --git a/drivers/pci/pci.c b/drivers/pci/pci.c\nindex 8f7cfcc00090..d0c9f0166af5 100644\n--- a/drivers/pci/pci.c\n+++ b/drivers/pci/pci.c\n@@ -4865,8 +4865,9 @@ static int pci_reset_hotplug_slot(struct hotplug_slot *hotplug, bool probe)\n \n static int pci_dev_reset_slot_function(struct pci_dev *dev, bool probe)\n {\n-\tif (dev->multifunction || dev->subordinate || !dev->slot ||\n-\t dev->dev_flags & PCI_DEV_FLAGS_NO_BUS_RESET)\n+\tif (dev->subordinate || !dev->slot ||\n+\t dev->dev_flags & PCI_DEV_FLAGS_NO_BUS_RESET ||\n+\t (dev->multifunction && !dev->slot->per_func_slot))\n \t\treturn -ENOTTY;\n \n \treturn pci_reset_hotplug_slot(dev->slot->hotplug, probe);\ndiff --git a/drivers/pci/slot.c b/drivers/pci/slot.c\nindex 6d5cd37bfb1e..894d6213ed30 100644\n--- a/drivers/pci/slot.c\n+++ b/drivers/pci/slot.c\n@@ -37,7 +37,7 @@ static const struct sysfs_ops pci_slot_sysfs_ops = {\n \n static ssize_t address_read_file(struct pci_slot *slot, char *buf)\n {\n-\tif (slot->number == 0xff)\n+\tif (slot->number == (u16)PCI_SLOT_PLACEHOLDER)\n \t\treturn sysfs_emit(buf, \"%04x:%02x\\n\",\n \t\t\t\t pci_domain_nr(slot->bus),\n \t\t\t\t slot->bus->number);\n@@ -72,6 +72,23 @@ static ssize_t cur_speed_read_file(struct pci_slot *slot, char *buf)\n \treturn bus_speed_read(slot->bus->cur_bus_speed, buf);\n }\n \n+static bool pci_dev_matches_slot(struct pci_dev *dev, struct pci_slot *slot)\n+{\n+\tif (slot->per_func_slot)\n+\t\treturn dev->devfn == slot->number;\n+\n+\treturn slot->number == PCI_SLOT_ALL_DEVICES ||\n+\t\tPCI_SLOT(dev->devfn) == slot->number;\n+}\n+\n+static bool pci_slot_enabled_per_func(void)\n+{\n+\tif (IS_ENABLED(CONFIG_S390))\n+\t\treturn true;\n+\n+\treturn false;\n+}\n+\n static void pci_slot_release(struct kobject *kobj)\n {\n \tstruct pci_dev *dev;\n@@ -82,8 +99,7 @@ static void pci_slot_release(struct kobject *kobj)\n \n \tdown_read(&pci_bus_sem);\n \tlist_for_each_entry(dev, &slot->bus->devices, bus_list)\n-\t\tif (slot->number == PCI_SLOT_ALL_DEVICES ||\n-\t\t PCI_SLOT(dev->devfn) == slot->number)\n+\t\tif (pci_dev_matches_slot(dev, slot))\n \t\t\tdev->slot = NULL;\n \tup_read(&pci_bus_sem);\n \n@@ -187,8 +203,7 @@ void pci_dev_assign_slot(struct pci_dev *dev)\n \n \tmutex_lock(&pci_slot_mutex);\n \tlist_for_each_entry(slot, &dev->bus->slots, list)\n-\t\tif (slot->number == PCI_SLOT_ALL_DEVICES ||\n-\t\t PCI_SLOT(dev->devfn) == slot->number)\n+\t\tif (pci_dev_matches_slot(dev, slot))\n \t\t\tdev->slot = slot;\n \tmutex_unlock(&pci_slot_mutex);\n }\n@@ -267,7 +282,7 @@ struct pci_slot *pci_create_slot(struct pci_bus *parent, int slot_nr,\n \n \tmutex_lock(&pci_slot_mutex);\n \n-\tif (slot_nr == -1)\n+\tif (slot_nr == PCI_SLOT_PLACEHOLDER)\n \t\tgoto placeholder;\n \n \t/*\n@@ -298,6 +313,9 @@ struct pci_slot *pci_create_slot(struct pci_bus *parent, int slot_nr,\n \tslot->bus = pci_bus_get(parent);\n \tslot->number = slot_nr;\n \n+\tif (pci_slot_enabled_per_func())\n+\t\tslot->per_func_slot = 1;\n+\n \tslot->kobj.kset = pci_slots_kset;\n \n \tslot_name = make_slot_name(name);\n@@ -318,8 +336,7 @@ struct pci_slot *pci_create_slot(struct pci_bus *parent, int slot_nr,\n \n \tdown_read(&pci_bus_sem);\n \tlist_for_each_entry(dev, &parent->devices, bus_list)\n-\t\tif (slot_nr == PCI_SLOT_ALL_DEVICES ||\n-\t\t PCI_SLOT(dev->devfn) == slot_nr)\n+\t\tif (pci_dev_matches_slot(dev, slot))\n \t\t\tdev->slot = slot;\n \tup_read(&pci_bus_sem);\n \ndiff --git a/include/linux/pci.h b/include/linux/pci.h\nindex 2c4454583c11..d58982aa8730 100644\n--- a/include/linux/pci.h\n+++ b/include/linux/pci.h\n@@ -78,14 +78,18 @@\n * and, if ARI Forwarding is enabled, functions may appear to be on multiple\n * devices.\n */\n-#define PCI_SLOT_ALL_DEVICES\t0xfe\n+#define PCI_SLOT_ALL_DEVICES\t0xfeff\n+\n+/* Used to identify a slot as a placeholder */\n+#define PCI_SLOT_PLACEHOLDER\t-1\n \n /* pci_slot represents a physical slot */\n struct pci_slot {\n \tstruct pci_bus\t\t*bus;\t\t/* Bus this slot is on */\n \tstruct list_head\tlist;\t\t/* Node in list of slots */\n \tstruct hotplug_slot\t*hotplug;\t/* Hotplug info (move here) */\n-\tunsigned char\t\tnumber;\t\t/* Device nr, or PCI_SLOT_ALL_DEVICES */\n+\tu16\t\t\tnumber;\t\t/* Device nr, or PCI_SLOT_ALL_DEVICES */\n+\tunsigned int\t\tper_func_slot:1; /* Allow per function slot */\n \tstruct kobject\t\tkobj;\n };\n \n","prefixes":["v15","1/7"]}