From nobody Sun Sep 14 13:00:54 2025 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 vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id B9A10C05027 for ; Fri, 20 Jan 2023 21:52:13 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S229750AbjATVwL (ORCPT ); Fri, 20 Jan 2023 16:52:11 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:50122 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229711AbjATVwJ (ORCPT ); Fri, 20 Jan 2023 16:52:09 -0500 Received: from mail-pf1-x44a.google.com (mail-pf1-x44a.google.com [IPv6:2607:f8b0:4864:20::44a]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 1119380BB1 for ; Fri, 20 Jan 2023 13:52:08 -0800 (PST) Received: by mail-pf1-x44a.google.com with SMTP id f22-20020a056a00239600b0058d956679f5so2979194pfc.5 for ; Fri, 20 Jan 2023 13:52:08 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20210112; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:from:to:cc:subject:date:message-id:reply-to; bh=kSxrMzG1Hv7pdQ1Ytw036hc1FJB6QyL44T06ZBbgDu4=; b=DGNmPG5xe76kN+gBDq5mU0YsgJej+WFDU/Tvr3GYRdJ5GgTvQ+csWUH6fEI6DSEp7T rT7NXuhVTAUCsGI5hm6oy8KY3yBz3Qyx0R3PtD08QCfq6iBmHRqhPPj8Ly4jfV8j7gZp 2hlnw1qCkBdZDlMYRHXcaDmlQy+PX9wGTSKrM7F+l+DP62zubLRIDXVMGlmGhXggtOPT YbXWRp+CKmbd0glr2I9mleU9HE0JuQPVts+BWfaLhQw3oxFpjM2XdiprknUIJK296KFt sytBP0DCk+F+yp98jBnfto7x1qF3yuxWQ1Fj2bvxLilOJ4FzHCS5C2Z0LreUpipW+o/b /IAA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=kSxrMzG1Hv7pdQ1Ytw036hc1FJB6QyL44T06ZBbgDu4=; b=R2NKs4O/Bqj9UfArbESpSEA2A27V93yZEMfofL/2JMPgRXyyGL4t/jyF0pYZOX3DmW 88PCIbU+MyJxw7iQZZUKh4eHfFhB2uz1A6zkGfvbiHQvXde3RZSyMHSjryOFiNzKjzq9 RNZsK4xrYtYoolSnIhWs/0HQsSQALmTZHvuxoHiU3CGFG6VG6Fx1Cz352SzZmzC2Q0NN rZZWAMIjzo3UalKxroB0mF61BCw7NeKMg9NKjZllCslMPkqwpRVXoZTWh3xRUhhUZ/Y4 ycjbK7I9ZSv/lvRC+41ryG5Wj91jpZdDo+kfUnujNCEa47e3rr/o0ZMLTY+uw2b8rYEq Yp9A== X-Gm-Message-State: AFqh2kp9Fes8IQnbclQpi6NLb8ZV83OrG+Fmp5OITEAoSizldpTY/YXl Khk9MV2YtUZ2tsKiDZb9mX5EPKVRzvz5TEEFJA== X-Google-Smtp-Source: AMrXdXvRQYSwAtryjO3/9NuhUBXCtmfhNrGdDeWz28VwYmz1ZWFq5gSbOxfizxdka/O+/85ZxAhGpIIv7A0KjtrMHw== X-Received: from dionnaglaze.c.googlers.com ([fda3:e722:ac3:cc00:7f:e700:c0a8:2ee6]) (user=dionnaglaze job=sendgmr) by 2002:a17:902:8694:b0:189:6889:c30a with SMTP id g20-20020a170902869400b001896889c30amr1672842plo.6.1674251527474; Fri, 20 Jan 2023 13:52:07 -0800 (PST) Date: Fri, 20 Jan 2023 21:48:54 +0000 In-Reply-To: <20230120214857.835931-1-dionnaglaze@google.com> Mime-Version: 1.0 References: <20230120214857.835931-1-dionnaglaze@google.com> X-Mailer: git-send-email 2.39.0.246.g2a6d74b583-goog Message-ID: <20230120214857.835931-2-dionnaglaze@google.com> Subject: [PATCH v12 1/3] crypto: ccp - Name -1 return value as SEV_RET_NO_FW_CALL From: Dionna Glaze To: x86@kernel.org, linux-kernel@vger.kernel.org Cc: Peter Gonda , Thomas Lendacky , Paolo Bonzini , Joerg Roedel , Ingo Molnar , Andy Lutomirsky , John Allen , Herbert Xu , "David S. Miller" , Borislav Petkov , Dionna Glaze Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" From: Peter Gonda The PSP can return a "firmware error" code of -1 in circumstances where the PSP is not actually called. To make this protocol unambiguous, the value is named SEV_RET_NO_FW_CALL. Cc: Thomas Lendacky Cc: Paolo Bonzini Cc: Joerg Roedel Cc: Ingo Molnar Cc: Andy Lutomirsky Cc: John Allen Cc: Herbert Xu Cc: "David S. Miller" Cc: Borislav Petkov Signed-off-by: Peter Gonda Signed-off-by: Dionna Glaze --- Documentation/virt/coco/sev-guest.rst | 2 +- drivers/crypto/ccp/sev-dev.c | 22 ++++++++++++++-------- include/uapi/linux/psp-sev.h | 7 +++++++ 3 files changed, 22 insertions(+), 9 deletions(-) diff --git a/Documentation/virt/coco/sev-guest.rst b/Documentation/virt/coc= o/sev-guest.rst index bf593e88cfd9..e76393e389eb 100644 --- a/Documentation/virt/coco/sev-guest.rst +++ b/Documentation/virt/coco/sev-guest.rst @@ -41,7 +41,7 @@ The guest ioctl should be issued on a file descriptor of = the /dev/sev-guest devi The ioctl accepts struct snp_user_guest_request. The input and output stru= cture is specified through the req_data and resp_data field respectively. If the io= ctl fails to execute due to a firmware error, then fw_err code will be set otherwise= the -fw_err will be set to 0x00000000000000ff. +fw_err will be set to 0x00000000ffffffff, i.e., the lower 32-bits are -1. =20 The firmware checks that the message sequence counter is one greater than the guests message sequence counter. If guest driver fails to increment me= ssage diff --git a/drivers/crypto/ccp/sev-dev.c b/drivers/crypto/ccp/sev-dev.c index 06fc7156c04f..ac205f78a595 100644 --- a/drivers/crypto/ccp/sev-dev.c +++ b/drivers/crypto/ccp/sev-dev.c @@ -440,12 +440,19 @@ static int __sev_init_ex_locked(int *error) return __sev_do_cmd_locked(SEV_CMD_INIT_EX, &data, error); } =20 +static inline int __sev_do_init_locked(int *psp_ret) +{ + if (sev_init_ex_buffer) + return __sev_init_ex_locked(psp_ret); + else + return __sev_init_locked(psp_ret); +} + static int __sev_platform_init_locked(int *error) { struct psp_device *psp =3D psp_master; struct sev_device *sev; - int rc =3D 0, psp_ret =3D -1; - int (*init_function)(int *error); + int rc =3D 0, psp_ret =3D SEV_RET_NO_FW_CALL; =20 if (!psp || !psp->sev_data) return -ENODEV; @@ -456,15 +463,12 @@ static int __sev_platform_init_locked(int *error) return 0; =20 if (sev_init_ex_buffer) { - init_function =3D __sev_init_ex_locked; rc =3D sev_read_init_ex_file(); if (rc) return rc; - } else { - init_function =3D __sev_init_locked; } =20 - rc =3D init_function(&psp_ret); + rc =3D __sev_do_init_locked(&psp_ret); if (rc && psp_ret =3D=3D SEV_RET_SECURE_DATA_INVALID) { /* * Initialization command returned an integrity check failure @@ -473,9 +477,11 @@ static int __sev_platform_init_locked(int *error) * initialization function should succeed by replacing the state * with a reset state. */ - dev_err(sev->dev, "SEV: retrying INIT command because of SECURE_DATA_INV= ALID error. Retrying once to reset PSP SEV state."); - rc =3D init_function(&psp_ret); + dev_err(sev->dev, +"SEV: retrying INIT command because of SECURE_DATA_INVALID error. Retrying= once to reset PSP SEV state."); + rc =3D __sev_do_init_locked(&psp_ret); } + if (error) *error =3D psp_ret; =20 diff --git a/include/uapi/linux/psp-sev.h b/include/uapi/linux/psp-sev.h index 91b4c63d5cbf..e8cfb8bde0d7 100644 --- a/include/uapi/linux/psp-sev.h +++ b/include/uapi/linux/psp-sev.h @@ -36,6 +36,13 @@ enum { * SEV Firmware status code */ typedef enum { + /* + * This error code is not in the SEV spec but is added to convey that + * there was an error that prevented the SEV Firmware from being called. + * The SEV API error codes are 16 bits, so the -1 value will not overlap + * with possible values from the specification. + */ + SEV_RET_NO_FW_CALL =3D -1, SEV_RET_SUCCESS =3D 0, SEV_RET_INVALID_PLATFORM_STATE, SEV_RET_INVALID_GUEST_STATE, --=20 2.39.0.246.g2a6d74b583-goog From nobody Sun Sep 14 13:00:54 2025 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 vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id B47A8C05027 for ; Fri, 20 Jan 2023 21:52:16 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S229780AbjATVwO (ORCPT ); Fri, 20 Jan 2023 16:52:14 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:50144 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229744AbjATVwL (ORCPT ); Fri, 20 Jan 2023 16:52:11 -0500 Received: from mail-pg1-x54a.google.com (mail-pg1-x54a.google.com [IPv6:2607:f8b0:4864:20::54a]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 7CD6280BB0 for ; Fri, 20 Jan 2023 13:52:09 -0800 (PST) Received: by mail-pg1-x54a.google.com with SMTP id q130-20020a632a88000000b004a03cfb3ac6so3134095pgq.6 for ; Fri, 20 Jan 2023 13:52:09 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20210112; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:from:to:cc:subject:date:message-id:reply-to; bh=vZLGH6GKpprsZryglGGsmlbnC99kz7Hmw/lP8KSRAIA=; b=H4S6MZ3Pu39FsRTnCqJAOdecUr98F5KodOtdYhzSnskQsZxaVRchBP/S8QXaAyCyeF pzl8QHq3Cognwhb+VTsm0Pw8X7dRVoRHokVU6LlRnsUPaOvRwyr597df0RFYLSkM2ly+ MedNBuHyilOTTmDHZhqHeeH8hlSY9NvEH/zY2aPHkup/8sYOY53czXUT34wRZ/HRcCJC oYV4ecWyZejeGKtPvnJOmbgeWf/JyvZu+5qRDPkb4ayJyYiHwgdLKuCAPoSv7Uxit5sJ NoUYagoBzw2wc4y3EeEPrJlYVV13qRKbX/SeUDZwqdbDnzS5HetwWWSJznTvh3/XokSY fKgA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=vZLGH6GKpprsZryglGGsmlbnC99kz7Hmw/lP8KSRAIA=; b=o+B4/KlkzcrqUjrad3hBRhYckySDmstTwbuHvyNMImU0+jnUCqGbxUkHbXUAeCUoeN 494lGnCsF7SlJqnr/3XUMazfQIhDcGSJ8tR2qpQdWJCokfiyHIjPtRFr99jGpx/vuCDs vWh136En+Cm2jUGhEXsYquExE1UqSvMQNPC9c1aIjYqSEvWgMzxBmIhd3eLOKWTkSNRS T7Zv4HNlw45ukWj3BJu1jWjoC7tP3/adNDi1HBRFMRIA1BXL2Z6dmkA1WhnN0R47JLh9 kRWOwNGMmlgCOuYvvoJBwkMT2nVUI0J23IerBU7QTGY88Q0pdXJaYjB9DQtl7COk8/qQ RkFQ== X-Gm-Message-State: AFqh2krKaPoOXhhrui8PSzEBqW7mIc6zrMUNdaV/p5b1BtRf1ZpLv+FY sQiqSvXFfLIv8YeYe2yZIGBAq6xvt0bh/kZF8A== X-Google-Smtp-Source: AMrXdXvgnhUOxHnbYYU2gTk3XLNC4wEojLvl5cRHswvnKmIzlXmAibhm7cHfTYwECk4VrKSdyiTFkWrg5f4UKbDgpw== X-Received: from dionnaglaze.c.googlers.com ([fda3:e722:ac3:cc00:7f:e700:c0a8:2ee6]) (user=dionnaglaze job=sendgmr) by 2002:a63:455a:0:b0:49f:5dc6:caf8 with SMTP id u26-20020a63455a000000b0049f5dc6caf8mr1692899pgk.41.1674251528854; Fri, 20 Jan 2023 13:52:08 -0800 (PST) Date: Fri, 20 Jan 2023 21:48:55 +0000 In-Reply-To: <20230120214857.835931-1-dionnaglaze@google.com> Mime-Version: 1.0 References: <20230120214857.835931-1-dionnaglaze@google.com> X-Mailer: git-send-email 2.39.0.246.g2a6d74b583-goog Message-ID: <20230120214857.835931-3-dionnaglaze@google.com> Subject: [PATCH v12 2/3] x86/sev: Change snp_guest_issue_request's fw_err From: Dionna Glaze To: x86@kernel.org, linux-kernel@vger.kernel.org Cc: Dionna Glaze , Tom Lendacky , Paolo Bonzini , Joerg Roedel , Peter Gonda , Thomas Gleixner , Dave Hansen , Ingo Molnar , Borislav Petkov , "H. Peter Anvin" , Venu Busireddy , Michael Roth , "Kirill A. Shutemov" , Michael Sterritt Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" The GHCB specification declares that the firmware error value for a guest request will be stored in the lower 32 bits of EXIT_INFO_2. The upper 32 bits are for the VMM's own error code. The fw_err argument is thus a misnomer, and callers will need access to all 64 bits. The type of unsigned long also causes problems, since sw_exit_info2 is u64 (unsigned long long) vs the argument's unsigned long*. This type is changed for issuing the guest request. The ioctl command struct's error field is passed directly instead of a local variable, since an incomplete guest request may not set the error code, and uninitialized stack memory would be written back to user space. The firmware might not even be called, so the call is bookended with the no firmware call error and clearing the error. Since the "fw_err" field is really exitinfo2 split into the upper bits' vmm error code and lower bits' firmware error code, sev-guest.h is updated to represent the 64 bit value as a union. Cc: Tom Lendacky Cc: Paolo Bonzini Cc: Joerg Roedel Cc: Peter Gonda Cc: Thomas Gleixner Cc: Dave Hansen Cc: Ingo Molnar Cc: Borislav Petkov Cc: "H. Peter Anvin" Cc: Venu Busireddy Cc: Michael Roth Cc: "Kirill A. Shutemov" Cc: Michael Sterritt Fixes: d5af44dde546 ("x86/sev: Provide support for SNP guest request NAEs") Reviewed-by: Tom Lendacky Reviewed-by: Borislav Petkov Reviewed-by: Peter Gonda Signed-off-by: Dionna Glaze --- arch/x86/include/asm/sev-common.h | 3 -- arch/x86/include/asm/sev.h | 4 +-- arch/x86/kernel/sev.c | 13 ++++--- drivers/virt/coco/sev-guest/sev-guest.c | 46 ++++++++++++------------- include/uapi/linux/sev-guest.h | 18 ++++++++-- 5 files changed, 48 insertions(+), 36 deletions(-) diff --git a/arch/x86/include/asm/sev-common.h b/arch/x86/include/asm/sev-c= ommon.h index b8357d6ecd47..0759af9b1acf 100644 --- a/arch/x86/include/asm/sev-common.h +++ b/arch/x86/include/asm/sev-common.h @@ -128,9 +128,6 @@ struct snp_psc_desc { struct psc_entry entries[VMGEXIT_PSC_MAX_ENTRY]; } __packed; =20 -/* Guest message request error code */ -#define SNP_GUEST_REQ_INVALID_LEN BIT_ULL(32) - #define GHCB_MSR_TERM_REQ 0x100 #define GHCB_MSR_TERM_REASON_SET_POS 12 #define GHCB_MSR_TERM_REASON_SET_MASK 0xf diff --git a/arch/x86/include/asm/sev.h b/arch/x86/include/asm/sev.h index ebc271bb6d8e..05de34d10d89 100644 --- a/arch/x86/include/asm/sev.h +++ b/arch/x86/include/asm/sev.h @@ -196,7 +196,7 @@ void snp_set_memory_private(unsigned long vaddr, unsign= ed int npages); void snp_set_wakeup_secondary_cpu(void); bool snp_init(struct boot_params *bp); void __init __noreturn snp_abort(void); -int snp_issue_guest_request(u64 exit_code, struct snp_req_data *input, uns= igned long *fw_err); +int snp_issue_guest_request(u64 exit_code, struct snp_req_data *input, u64= *exitinfo2); #else static inline void sev_es_ist_enter(struct pt_regs *regs) { } static inline void sev_es_ist_exit(void) { } @@ -217,7 +217,7 @@ static inline void snp_set_wakeup_secondary_cpu(void) {= } static inline bool snp_init(struct boot_params *bp) { return false; } static inline void snp_abort(void) { } static inline int snp_issue_guest_request(u64 exit_code, struct snp_req_da= ta *input, - unsigned long *fw_err) + u64 *exitinfo2) { return -ENOTTY; } diff --git a/arch/x86/kernel/sev.c b/arch/x86/kernel/sev.c index 679026a640ef..82aa81fafb71 100644 --- a/arch/x86/kernel/sev.c +++ b/arch/x86/kernel/sev.c @@ -22,6 +22,8 @@ #include #include #include +#include +#include =20 #include #include @@ -2175,7 +2177,7 @@ static int __init init_sev_config(char *str) } __setup("sev=3D", init_sev_config); =20 -int snp_issue_guest_request(u64 exit_code, struct snp_req_data *input, uns= igned long *fw_err) +int snp_issue_guest_request(u64 exit_code, struct snp_req_data *input, u64= *exitinfo2) { struct ghcb_state state; struct es_em_ctxt ctxt; @@ -2186,9 +2188,11 @@ int snp_issue_guest_request(u64 exit_code, struct sn= p_req_data *input, unsigned if (!cc_platform_has(CC_ATTR_GUEST_SEV_SNP)) return -ENODEV; =20 - if (!fw_err) + if (!exitinfo2) return -EINVAL; =20 + *exitinfo2 =3D SEV_RET_NO_FW_CALL; + /* * __sev_get_ghcb() needs to run with IRQs disabled because it is using * a per-CPU GHCB. @@ -2212,14 +2216,13 @@ int snp_issue_guest_request(u64 exit_code, struct s= np_req_data *input, unsigned if (ret) goto e_put; =20 + *exitinfo2 =3D ghcb->save.sw_exit_info_2; if (ghcb->save.sw_exit_info_2) { /* Number of expected pages are returned in RBX */ if (exit_code =3D=3D SVM_VMGEXIT_EXT_GUEST_REQUEST && - ghcb->save.sw_exit_info_2 =3D=3D SNP_GUEST_REQ_INVALID_LEN) + ghcb->save.sw_exit_info_2 =3D=3D SNP_GUEST_VMM_ERR(SNP_GUEST_VMM_ERR_IN= VALID_LEN)) input->data_npages =3D ghcb_get_rbx(ghcb); =20 - *fw_err =3D ghcb->save.sw_exit_info_2; - ret =3D -EIO; } =20 diff --git a/drivers/virt/coco/sev-guest/sev-guest.c b/drivers/virt/coco/se= v-guest/sev-guest.c index 4ec4174e05a3..ed6fc14f6b00 100644 --- a/drivers/virt/coco/sev-guest/sev-guest.c +++ b/drivers/virt/coco/sev-guest/sev-guest.c @@ -318,11 +318,11 @@ static int enc_payload(struct snp_guest_dev *snp_dev,= u64 seqno, int version, u8 return __enc_payload(snp_dev, req, payload, sz); } =20 -static int handle_guest_request(struct snp_guest_dev *snp_dev, u64 exit_co= de, int msg_ver, - u8 type, void *req_buf, size_t req_sz, void *resp_buf, - u32 resp_sz, __u64 *fw_err) +static int handle_guest_request(struct snp_guest_dev *snp_dev, u64 exit_co= de, + struct snp_guest_request_ioctl *arg, + u8 type, void *req_buf, size_t req_sz, + void *resp_buf, u32 resp_sz) { - unsigned long err; u64 seqno; int rc; =20 @@ -334,7 +334,7 @@ static int handle_guest_request(struct snp_guest_dev *s= np_dev, u64 exit_code, in memset(snp_dev->response, 0, sizeof(struct snp_guest_msg)); =20 /* Encrypt the userspace provided payload */ - rc =3D enc_payload(snp_dev, seqno, msg_ver, type, req_buf, req_sz); + rc =3D enc_payload(snp_dev, seqno, arg->msg_version, type, req_buf, req_s= z); if (rc) return rc; =20 @@ -344,7 +344,8 @@ static int handle_guest_request(struct snp_guest_dev *s= np_dev, u64 exit_code, in * sequence number must be incremented or the VMPCK must be deleted to * prevent reuse of the IV. */ - rc =3D snp_issue_guest_request(exit_code, &snp_dev->input, &err); + rc =3D snp_issue_guest_request(exit_code, &snp_dev->input, + &arg->exitinfo2); =20 /* * If the extended guest request fails due to having too small of a @@ -353,7 +354,7 @@ static int handle_guest_request(struct snp_guest_dev *s= np_dev, u64 exit_code, in * and thus avoid IV reuse. */ if (exit_code =3D=3D SVM_VMGEXIT_EXT_GUEST_REQUEST && - err =3D=3D SNP_GUEST_REQ_INVALID_LEN) { + arg->vmm_error =3D=3D SNP_GUEST_VMM_ERR_INVALID_LEN) { const unsigned int certs_npages =3D snp_dev->input.data_npages; =20 exit_code =3D SVM_VMGEXIT_GUEST_REQUEST; @@ -366,24 +367,22 @@ static int handle_guest_request(struct snp_guest_dev = *snp_dev, u64 exit_code, in * of the VMPCK and the error code being propagated back to the * user as an ioctl() return code. */ - rc =3D snp_issue_guest_request(exit_code, &snp_dev->input, &err); + rc =3D snp_issue_guest_request(exit_code, &snp_dev->input, + &arg->exitinfo2); =20 /* * Override the error to inform callers the given extended * request buffer size was too small and give the caller the * required buffer size. */ - err =3D SNP_GUEST_REQ_INVALID_LEN; + arg->vmm_error =3D SNP_GUEST_VMM_ERR_INVALID_LEN; snp_dev->input.data_npages =3D certs_npages; } =20 - if (fw_err) - *fw_err =3D err; - if (rc) { dev_alert(snp_dev->dev, - "Detected error from ASP request. rc: %d, fw_err: %llu\n", - rc, *fw_err); + "Detected error from ASP request. rc: %d, exitinfo2: %llu\n", + rc, arg->exitinfo2); goto disable_vmpck; } =20 @@ -430,9 +429,9 @@ static int get_report(struct snp_guest_dev *snp_dev, st= ruct snp_guest_request_io if (!resp) return -ENOMEM; =20 - rc =3D handle_guest_request(snp_dev, SVM_VMGEXIT_GUEST_REQUEST, arg->msg_= version, + rc =3D handle_guest_request(snp_dev, SVM_VMGEXIT_GUEST_REQUEST, arg, SNP_MSG_REPORT_REQ, &req, sizeof(req), resp->data, - resp_len, &arg->fw_err); + resp_len); if (rc) goto e_free; =20 @@ -470,9 +469,8 @@ static int get_derived_key(struct snp_guest_dev *snp_de= v, struct snp_guest_reque if (copy_from_user(&req, (void __user *)arg->req_data, sizeof(req))) return -EFAULT; =20 - rc =3D handle_guest_request(snp_dev, SVM_VMGEXIT_GUEST_REQUEST, arg->msg_= version, - SNP_MSG_KEY_REQ, &req, sizeof(req), buf, resp_len, - &arg->fw_err); + rc =3D handle_guest_request(snp_dev, SVM_VMGEXIT_GUEST_REQUEST, arg, + SNP_MSG_KEY_REQ, &req, sizeof(req), buf, resp_len); if (rc) return rc; =20 @@ -532,12 +530,12 @@ static int get_ext_report(struct snp_guest_dev *snp_d= ev, struct snp_guest_reques return -ENOMEM; =20 snp_dev->input.data_npages =3D npages; - ret =3D handle_guest_request(snp_dev, SVM_VMGEXIT_EXT_GUEST_REQUEST, arg-= >msg_version, + ret =3D handle_guest_request(snp_dev, SVM_VMGEXIT_EXT_GUEST_REQUEST, arg, SNP_MSG_REPORT_REQ, &req.data, - sizeof(req.data), resp->data, resp_len, &arg->fw_err); + sizeof(req.data), resp->data, resp_len); =20 /* If certs length is invalid then copy the returned length */ - if (arg->fw_err =3D=3D SNP_GUEST_REQ_INVALID_LEN) { + if (arg->vmm_error =3D=3D SNP_GUEST_VMM_ERR_INVALID_LEN) { req.certs_len =3D snp_dev->input.data_npages << PAGE_SHIFT; =20 if (copy_to_user((void __user *)arg->req_data, &req, sizeof(req))) @@ -572,7 +570,7 @@ static long snp_guest_ioctl(struct file *file, unsigned= int ioctl, unsigned long if (copy_from_user(&input, argp, sizeof(input))) return -EFAULT; =20 - input.fw_err =3D 0xff; + input.exitinfo2 =3D 0xff; =20 /* Message version must be non-zero */ if (!input.msg_version) @@ -603,7 +601,7 @@ static long snp_guest_ioctl(struct file *file, unsigned= int ioctl, unsigned long =20 mutex_unlock(&snp_cmd_mutex); =20 - if (input.fw_err && copy_to_user(argp, &input, sizeof(input))) + if (input.exitinfo2 && copy_to_user(argp, &input, sizeof(input))) return -EFAULT; =20 return ret; diff --git a/include/uapi/linux/sev-guest.h b/include/uapi/linux/sev-guest.h index 256aaeff7e65..2aa39112cf8d 100644 --- a/include/uapi/linux/sev-guest.h +++ b/include/uapi/linux/sev-guest.h @@ -52,8 +52,14 @@ struct snp_guest_request_ioctl { __u64 req_data; __u64 resp_data; =20 - /* firmware error code on failure (see psp-sev.h) */ - __u64 fw_err; + /* bits[63:32]: VMM error code, bits[31:0] firmware error code (see psp-s= ev.h) */ + union { + __u64 exitinfo2; + struct { + __u32 fw_error; + __u32 vmm_error; + }; + }; }; =20 struct snp_ext_report_req { @@ -77,4 +83,12 @@ struct snp_ext_report_req { /* Get SNP extended report as defined in the GHCB specification version 2.= */ #define SNP_GET_EXT_REPORT _IOWR(SNP_GUEST_REQ_IOC_TYPE, 0x2, struct snp_g= uest_request_ioctl) =20 +/* Guest message request EXIT_INFO_2 constants */ +#define SNP_GUEST_FW_ERR_MASK GENMASK_ULL(31, 0) +#define SNP_GUEST_VMM_ERR_SHIFT 32 +#define SNP_GUEST_VMM_ERR(x) (((u64)x) << SNP_GUEST_VMM_ERR_SHIFT) + +#define SNP_GUEST_VMM_ERR_INVALID_LEN 1 +#define SNP_GUEST_VMM_ERR_BUSY 2 + #endif /* __UAPI_LINUX_SEV_GUEST_H_ */ --=20 2.39.0.246.g2a6d74b583-goog From nobody Sun Sep 14 13:00:54 2025 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 vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 1582FC05027 for ; Fri, 20 Jan 2023 21:52:33 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S229811AbjATVwR (ORCPT ); Fri, 20 Jan 2023 16:52:17 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:50176 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229711AbjATVwM (ORCPT ); Fri, 20 Jan 2023 16:52:12 -0500 Received: from mail-pf1-x449.google.com (mail-pf1-x449.google.com [IPv6:2607:f8b0:4864:20::449]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 2C2B981021 for ; Fri, 20 Jan 2023 13:52:11 -0800 (PST) Received: by mail-pf1-x449.google.com with SMTP id a18-20020a62bd12000000b0056e7b61ec78so2980510pff.17 for ; Fri, 20 Jan 2023 13:52:11 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20210112; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:from:to:cc:subject:date:message-id:reply-to; bh=SUCHfM6ksm2j9ziQyxgA2TAJuZtKMQ5HI7hbyLcswG0=; b=kN5SBQN5rPmpYwqfPCYN+Nzpuhw94Z19rdN+KbYq0Q4f+yrl7w0UyU6ENTxKLl/EzR mS5pE2Hl9VvCMVThDGV0TqKROVmYCQcYcPfzmW4vs3iCvZ5K81LG3k1FD/Pu6+/W447M 4Cky3a2MU+pFZg/plq4cCPI6wcK4QP5SPbxQhrsSG80NqTc/TXVocW9fHD8alPUSZ/5v iZeXt+zJ3/3s18HmJffjjOJ/bXI82/qy/pSlQxd/luv8Yx6PBEQ+tygwJayxzPpwkyP2 itWMd+m03O+XVXOLiNBmMdpUNXLUm8gA8dgxWv9MG7ZEodEjRjtHuMyuNBdAGBRP0k0k dHng== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=SUCHfM6ksm2j9ziQyxgA2TAJuZtKMQ5HI7hbyLcswG0=; b=dkhyNpeiJQJbMNbGOO9PgeA0onu/XuxRinVF/zvWpiryRYaAxhm6jxAODoIvMdK6sR lgofT/dZr5LsYbpWGqb9UxLV3TGWTxtawbWIQ50Aal/VEGTOFkIufrP02SdN9sj7sB2O /vLFaIom6raQUmgm9M3MxsU8y/Qhdhxk3cNzJ63v5wYP/RiypR6IJwEXWCCz1KZCRQ49 f0p9nQRZ0CBVQZueN6z6fXLLLXzCUJU3syFOedQ1EUWhu+AsGjnNxDLfwsp3BcE8BjsQ H/ntlYp912PkuLaL3/RAdJilzfenDwflLArK8THXyHGrNt1BS7uP6vlBYlVqx5XUrhfk QWvA== X-Gm-Message-State: AFqh2krC6a9SME82SZwzpSfV1cC7LpoaBYz/Rb1B3fX4et3/tPKkxzGj 1HJzeyERIAyPKphmEewGDFn59GWAD/yC5A6HpA== X-Google-Smtp-Source: AMrXdXuHtzIs/NyctfTZ+MNIanuRRuoorwxpbCfb60PtSNh3IU4zGA0bfk7B09c5iLR+3FTnoRPuJTEmVja//toMiQ== X-Received: from dionnaglaze.c.googlers.com ([fda3:e722:ac3:cc00:7f:e700:c0a8:2ee6]) (user=dionnaglaze job=sendgmr) by 2002:a63:fc41:0:b0:49b:2004:b42 with SMTP id r1-20020a63fc41000000b0049b20040b42mr1281142pgk.170.1674251530579; Fri, 20 Jan 2023 13:52:10 -0800 (PST) Date: Fri, 20 Jan 2023 21:48:56 +0000 In-Reply-To: <20230120214857.835931-1-dionnaglaze@google.com> Mime-Version: 1.0 References: <20230120214857.835931-1-dionnaglaze@google.com> X-Mailer: git-send-email 2.39.0.246.g2a6d74b583-goog Message-ID: <20230120214857.835931-4-dionnaglaze@google.com> Subject: [PATCH v12 3/3] virt: sev-guest: interpret VMM errors from guest request From: Dionna Glaze To: x86@kernel.org, linux-kernel@vger.kernel.org Cc: Dionna Glaze , Tom Lendacky , Peter Gonda , Borislav Petkov , Tom Lendacky , Liam Merwick , Yang Yingliang , Haowen Bai Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" The GHCB specification states that the upper 32 bits of exitinfo2 are for the VMM's error codes. The sev-guest ABI has already locked in that the fw_err status of the input will be 64 bits, and that BIT_ULL(32) means that the extended guest request's data buffer was too small. Add BIT_ULL(33) for "host refused the request for throttling reasons". The driver interprets the upper 32 bits of exitinfo2 for the user anyway in case the request gets throttled. For safety, since the encryption algorithm in GHCBv2 is AES_GCM, control must remain in the kernel to complete the request with the current sequence number. Returning without finishing the request allows the the guest to make another request but with different message contents. This is IV reuse, and breaks cryptographic protections. When throttled, the driver will reschedule itself and then try again after sleeping half its ratelimit time to avoid a big wait queue. The ioctl may block indefinitely, but that has always been the case when deferring these requests to the host. Cc: Tom Lendacky Cc: Peter Gonda Cc: Borislav Petkov Cc: Tom Lendacky Cc: Liam Merwick Cc: Yang Yingliang Cc: Haowen Bai Reviewed-by: Tom Lendacky Reviewed-by: Peter Gonda Signed-off-by: Dionna Glaze --- Documentation/virt/coco/sev-guest.rst | 21 +++++++++++------ drivers/virt/coco/sev-guest/sev-guest.c | 31 +++++++++++++++++++++++++ 2 files changed, 45 insertions(+), 7 deletions(-) diff --git a/Documentation/virt/coco/sev-guest.rst b/Documentation/virt/coc= o/sev-guest.rst index e76393e389eb..fd8bf833f1cd 100644 --- a/Documentation/virt/coco/sev-guest.rst +++ b/Documentation/virt/coco/sev-guest.rst @@ -37,11 +37,12 @@ along with a description: the return value. General error numbers (-ENOMEM, -EINVAL) are not detailed, but errors with specific meanings are. =20 -The guest ioctl should be issued on a file descriptor of the /dev/sev-gues= t device. -The ioctl accepts struct snp_user_guest_request. The input and output stru= cture is -specified through the req_data and resp_data field respectively. If the io= ctl fails -to execute due to a firmware error, then fw_err code will be set otherwise= the -fw_err will be set to 0x00000000ffffffff, i.e., the lower 32-bits are -1. +The guest ioctl should be issued on a file descriptor of the +/dev/sev-guest device. The ioctl accepts struct +snp_user_guest_request. The input and output structure is specified +through the req_data and resp_data field respectively. If the ioctl +fails to execute due to a firmware error, then the fw_error code will +be set, otherwise fw_error will be set to -1. =20 The firmware checks that the message sequence counter is one greater than the guests message sequence counter. If guest driver fails to increment me= ssage @@ -57,8 +58,14 @@ counter (e.g. counter overflow), then -EIO will be retur= ned. __u64 req_data; __u64 resp_data; =20 - /* firmware error code on failure (see psp-sev.h) */ - __u64 fw_err; + /* bits[63:32]: VMM error code, bits[31:0] firmware error code (see psp-= sev.h) */ + union { + __u64 exitinfo2; + struct { + __u32 fw_error; + __u32 vmm_error; + }; + }; }; =20 2.1 SNP_GET_REPORT diff --git a/drivers/virt/coco/sev-guest/sev-guest.c b/drivers/virt/coco/se= v-guest/sev-guest.c index ed6fc14f6b00..1bffd9b1b7dc 100644 --- a/drivers/virt/coco/sev-guest/sev-guest.c +++ b/drivers/virt/coco/sev-guest/sev-guest.c @@ -14,6 +14,7 @@ #include #include #include +#include #include #include #include @@ -48,12 +49,22 @@ struct snp_guest_dev { struct snp_req_data input; u32 *os_area_msg_seqno; u8 *vmpck; + + struct ratelimit_state rs; }; =20 static u32 vmpck_id; module_param(vmpck_id, uint, 0444); MODULE_PARM_DESC(vmpck_id, "The VMPCK ID to use when communicating with th= e PSP."); =20 +static int rate_s =3D 1; +module_param(rate_s, int, 0444); +MODULE_PARM_DESC(rate_s, "The rate limit interval in seconds to limit requ= ests to."); + +static int rate_burst =3D 2; +module_param(rate_burst, int, 0444); +MODULE_PARM_DESC(rate_burst, "The rate limit burst amount to limit request= s to."); + /* Mutex to serialize the shared buffer access and command handling. */ static DEFINE_MUTEX(snp_cmd_mutex); =20 @@ -338,6 +349,16 @@ static int handle_guest_request(struct snp_guest_dev *= snp_dev, u64 exit_code, if (rc) return rc; =20 +retry: + /* + * Rate limit commands internally since the host can also throttle, and + * the guest shouldn't create a tight request spin that could end up + * getting this VM throttled more heavily. + */ + if (!__ratelimit(&snp_dev->rs)) { + schedule_timeout_interruptible((rate_s * HZ) / 2); + goto retry; + } /* * Call firmware to process the request. In this function the encrypted * message enters shared memory with the host. So after this call the @@ -347,6 +368,14 @@ static int handle_guest_request(struct snp_guest_dev *= snp_dev, u64 exit_code, rc =3D snp_issue_guest_request(exit_code, &snp_dev->input, &arg->exitinfo2); =20 + /* + * The host may return EBUSY if the request has been throttled. + * We retry in the driver to avoid returning and reusing the message + * sequence number on a different message. + */ + if (arg->vmm_error =3D=3D SNP_GUEST_VMM_ERR_BUSY) + goto retry; + /* * If the extended guest request fails due to having too small of a * certificate data buffer, retry the same guest request without the @@ -751,6 +780,8 @@ static int __init sev_guest_probe(struct platform_devic= e *pdev) if (ret) goto e_free_cert_data; =20 + ratelimit_state_init(&snp_dev->rs, rate_s * HZ, rate_burst); + dev_info(dev, "Initialized SEV guest driver (using vmpck_id %d)\n", vmpck= _id); return 0; =20 --=20 2.39.0.246.g2a6d74b583-goog