From e5733d8c89c3b57c8fcd40b8acf508388fabaa42 Mon Sep 17 00:00:00 2001 From: Reinette Chatre Date: Wed, 2 Feb 2022 11:41:12 -0800 Subject: x86/sgx: Fix missing poison handling in reclaimer The SGX reclaimer code lacks page poison handling in its main free path. This can lead to avoidable machine checks if a poisoned page is freed and reallocated instead of being isolated. A troublesome scenario is: 1. Machine check (#MC) occurs (asynchronous, !MF_ACTION_REQUIRED) 2. arch_memory_failure() is eventually called 3. (SGX) page->poison set to 1 4. Page is reclaimed 5. Page added to normal free lists by sgx_reclaim_pages() ^ This is the bug (poison pages should be isolated on the sgx_poison_page_list instead) 6. Page is reallocated by some innocent enclave, a second (synchronous) in-kernel #MC is induced, probably during EADD instruction. ^ This is the fallout from the bug (6) is unfortunate and can be avoided by replacing the open coded enclave page freeing code in the reclaimer with sgx_free_epc_page() to obtain support for poison page handling that includes placing the poisoned page on the correct list. Fixes: d6d261bded8a ("x86/sgx: Add new sgx_epc_page flag bit to mark free pages") Fixes: 992801ae9243 ("x86/sgx: Initial poison handling for dirty and free pages") Signed-off-by: Reinette Chatre Signed-off-by: Dave Hansen Reviewed-by: Jarkko Sakkinen Link: https://lkml.kernel.org/r/dcc95eb2aaefb042527ac50d0a50738c7c160dac.1643830353.git.reinette.chatre@intel.com --- arch/x86/kernel/cpu/sgx/main.c | 10 +--------- 1 file changed, 1 insertion(+), 9 deletions(-) (limited to 'arch/x86/kernel/cpu') diff --git a/arch/x86/kernel/cpu/sgx/main.c b/arch/x86/kernel/cpu/sgx/main.c index 4b41efc9e367..8e4bc6453d26 100644 --- a/arch/x86/kernel/cpu/sgx/main.c +++ b/arch/x86/kernel/cpu/sgx/main.c @@ -344,10 +344,8 @@ static void sgx_reclaim_pages(void) { struct sgx_epc_page *chunk[SGX_NR_TO_SCAN]; struct sgx_backing backing[SGX_NR_TO_SCAN]; - struct sgx_epc_section *section; struct sgx_encl_page *encl_page; struct sgx_epc_page *epc_page; - struct sgx_numa_node *node; pgoff_t page_index; int cnt = 0; int ret; @@ -418,13 +416,7 @@ skip: kref_put(&encl_page->encl->refcount, sgx_encl_release); epc_page->flags &= ~SGX_EPC_PAGE_RECLAIMER_TRACKED; - section = &sgx_epc_sections[epc_page->section]; - node = section->node; - - spin_lock(&node->lock); - list_add_tail(&epc_page->list, &node->free_page_list); - spin_unlock(&node->lock); - atomic_long_inc(&sgx_nr_free_pages); + sgx_free_epc_page(epc_page); } } -- cgit From d45476d9832409371537013ebdd8dc1a7781f97a Mon Sep 17 00:00:00 2001 From: "Peter Zijlstra (Intel)" Date: Wed, 16 Feb 2022 20:57:00 +0100 Subject: x86/speculation: Rename RETPOLINE_AMD to RETPOLINE_LFENCE The RETPOLINE_AMD name is unfortunate since it isn't necessarily AMD only, in fact Hygon also uses it. Furthermore it will likely be sufficient for some Intel processors. Therefore rename the thing to RETPOLINE_LFENCE to better describe what it is. Add the spectre_v2=retpoline,lfence option as an alias to spectre_v2=retpoline,amd to preserve existing setups. However, the output of /sys/devices/system/cpu/vulnerabilities/spectre_v2 will be changed. [ bp: Fix typos, massage. ] Co-developed-by: Josh Poimboeuf Signed-off-by: Josh Poimboeuf Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Borislav Petkov Reviewed-by: Thomas Gleixner --- arch/x86/kernel/cpu/bugs.c | 29 ++++++++++++++++++----------- 1 file changed, 18 insertions(+), 11 deletions(-) (limited to 'arch/x86/kernel/cpu') diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index 1c1f218a701d..e3bb8afc6768 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -664,7 +664,7 @@ enum spectre_v2_mitigation_cmd { SPECTRE_V2_CMD_FORCE, SPECTRE_V2_CMD_RETPOLINE, SPECTRE_V2_CMD_RETPOLINE_GENERIC, - SPECTRE_V2_CMD_RETPOLINE_AMD, + SPECTRE_V2_CMD_RETPOLINE_LFENCE, }; enum spectre_v2_user_cmd { @@ -824,8 +824,8 @@ set_mode: static const char * const spectre_v2_strings[] = { [SPECTRE_V2_NONE] = "Vulnerable", - [SPECTRE_V2_RETPOLINE_GENERIC] = "Mitigation: Full generic retpoline", - [SPECTRE_V2_RETPOLINE_AMD] = "Mitigation: Full AMD retpoline", + [SPECTRE_V2_RETPOLINE] = "Mitigation: Retpolines", + [SPECTRE_V2_LFENCE] = "Mitigation: LFENCE", [SPECTRE_V2_IBRS_ENHANCED] = "Mitigation: Enhanced IBRS", }; @@ -837,7 +837,8 @@ static const struct { { "off", SPECTRE_V2_CMD_NONE, false }, { "on", SPECTRE_V2_CMD_FORCE, true }, { "retpoline", SPECTRE_V2_CMD_RETPOLINE, false }, - { "retpoline,amd", SPECTRE_V2_CMD_RETPOLINE_AMD, false }, + { "retpoline,amd", SPECTRE_V2_CMD_RETPOLINE_LFENCE, false }, + { "retpoline,lfence", SPECTRE_V2_CMD_RETPOLINE_LFENCE, false }, { "retpoline,generic", SPECTRE_V2_CMD_RETPOLINE_GENERIC, false }, { "auto", SPECTRE_V2_CMD_AUTO, false }, }; @@ -875,13 +876,19 @@ static enum spectre_v2_mitigation_cmd __init spectre_v2_parse_cmdline(void) } if ((cmd == SPECTRE_V2_CMD_RETPOLINE || - cmd == SPECTRE_V2_CMD_RETPOLINE_AMD || + cmd == SPECTRE_V2_CMD_RETPOLINE_LFENCE || cmd == SPECTRE_V2_CMD_RETPOLINE_GENERIC) && !IS_ENABLED(CONFIG_RETPOLINE)) { pr_err("%s selected but not compiled in. Switching to AUTO select\n", mitigation_options[i].option); return SPECTRE_V2_CMD_AUTO; } + if ((cmd == SPECTRE_V2_CMD_RETPOLINE_LFENCE) && + !boot_cpu_has(X86_FEATURE_LFENCE_RDTSC)) { + pr_err("%s selected, but CPU doesn't have a serializing LFENCE. Switching to AUTO select\n", mitigation_options[i].option); + return SPECTRE_V2_CMD_AUTO; + } + spec_v2_print_cond(mitigation_options[i].option, mitigation_options[i].secure); return cmd; @@ -916,9 +923,9 @@ static void __init spectre_v2_select_mitigation(void) if (IS_ENABLED(CONFIG_RETPOLINE)) goto retpoline_auto; break; - case SPECTRE_V2_CMD_RETPOLINE_AMD: + case SPECTRE_V2_CMD_RETPOLINE_LFENCE: if (IS_ENABLED(CONFIG_RETPOLINE)) - goto retpoline_amd; + goto retpoline_lfence; break; case SPECTRE_V2_CMD_RETPOLINE_GENERIC: if (IS_ENABLED(CONFIG_RETPOLINE)) @@ -935,17 +942,17 @@ static void __init spectre_v2_select_mitigation(void) retpoline_auto: if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD || boot_cpu_data.x86_vendor == X86_VENDOR_HYGON) { - retpoline_amd: + retpoline_lfence: if (!boot_cpu_has(X86_FEATURE_LFENCE_RDTSC)) { pr_err("Spectre mitigation: LFENCE not serializing, switching to generic retpoline\n"); goto retpoline_generic; } - mode = SPECTRE_V2_RETPOLINE_AMD; - setup_force_cpu_cap(X86_FEATURE_RETPOLINE_AMD); + mode = SPECTRE_V2_LFENCE; + setup_force_cpu_cap(X86_FEATURE_RETPOLINE_LFENCE); setup_force_cpu_cap(X86_FEATURE_RETPOLINE); } else { retpoline_generic: - mode = SPECTRE_V2_RETPOLINE_GENERIC; + mode = SPECTRE_V2_RETPOLINE; setup_force_cpu_cap(X86_FEATURE_RETPOLINE); } -- cgit From 1e19da8522c81bf46b335f84137165741e0d82b7 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 16 Feb 2022 20:57:01 +0100 Subject: x86/speculation: Add eIBRS + Retpoline options Thanks to the chaps at VUsec it is now clear that eIBRS is not sufficient, therefore allow enabling of retpolines along with eIBRS. Add spectre_v2=eibrs, spectre_v2=eibrs,lfence and spectre_v2=eibrs,retpoline options to explicitly pick your preferred means of mitigation. Since there's new mitigations there's also user visible changes in /sys/devices/system/cpu/vulnerabilities/spectre_v2 to reflect these new mitigations. [ bp: Massage commit message, trim error messages, do more precise eIBRS mode checking. ] Co-developed-by: Josh Poimboeuf Signed-off-by: Josh Poimboeuf Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Borislav Petkov Reviewed-by: Patrick Colp Reviewed-by: Thomas Gleixner --- arch/x86/kernel/cpu/bugs.c | 133 ++++++++++++++++++++++++++++++++------------- 1 file changed, 96 insertions(+), 37 deletions(-) (limited to 'arch/x86/kernel/cpu') diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index e3bb8afc6768..79c52dd6c597 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -665,6 +665,9 @@ enum spectre_v2_mitigation_cmd { SPECTRE_V2_CMD_RETPOLINE, SPECTRE_V2_CMD_RETPOLINE_GENERIC, SPECTRE_V2_CMD_RETPOLINE_LFENCE, + SPECTRE_V2_CMD_EIBRS, + SPECTRE_V2_CMD_EIBRS_RETPOLINE, + SPECTRE_V2_CMD_EIBRS_LFENCE, }; enum spectre_v2_user_cmd { @@ -737,6 +740,13 @@ spectre_v2_parse_user_cmdline(enum spectre_v2_mitigation_cmd v2_cmd) return SPECTRE_V2_USER_CMD_AUTO; } +static inline bool spectre_v2_in_eibrs_mode(enum spectre_v2_mitigation mode) +{ + return (mode == SPECTRE_V2_EIBRS || + mode == SPECTRE_V2_EIBRS_RETPOLINE || + mode == SPECTRE_V2_EIBRS_LFENCE); +} + static void __init spectre_v2_user_select_mitigation(enum spectre_v2_mitigation_cmd v2_cmd) { @@ -804,7 +814,7 @@ spectre_v2_user_select_mitigation(enum spectre_v2_mitigation_cmd v2_cmd) */ if (!boot_cpu_has(X86_FEATURE_STIBP) || !smt_possible || - spectre_v2_enabled == SPECTRE_V2_IBRS_ENHANCED) + spectre_v2_in_eibrs_mode(spectre_v2_enabled)) return; /* @@ -826,7 +836,9 @@ static const char * const spectre_v2_strings[] = { [SPECTRE_V2_NONE] = "Vulnerable", [SPECTRE_V2_RETPOLINE] = "Mitigation: Retpolines", [SPECTRE_V2_LFENCE] = "Mitigation: LFENCE", - [SPECTRE_V2_IBRS_ENHANCED] = "Mitigation: Enhanced IBRS", + [SPECTRE_V2_EIBRS] = "Mitigation: Enhanced IBRS", + [SPECTRE_V2_EIBRS_LFENCE] = "Mitigation: Enhanced IBRS + LFENCE", + [SPECTRE_V2_EIBRS_RETPOLINE] = "Mitigation: Enhanced IBRS + Retpolines", }; static const struct { @@ -840,6 +852,9 @@ static const struct { { "retpoline,amd", SPECTRE_V2_CMD_RETPOLINE_LFENCE, false }, { "retpoline,lfence", SPECTRE_V2_CMD_RETPOLINE_LFENCE, false }, { "retpoline,generic", SPECTRE_V2_CMD_RETPOLINE_GENERIC, false }, + { "eibrs", SPECTRE_V2_CMD_EIBRS, false }, + { "eibrs,lfence", SPECTRE_V2_CMD_EIBRS_LFENCE, false }, + { "eibrs,retpoline", SPECTRE_V2_CMD_EIBRS_RETPOLINE, false }, { "auto", SPECTRE_V2_CMD_AUTO, false }, }; @@ -877,15 +892,29 @@ static enum spectre_v2_mitigation_cmd __init spectre_v2_parse_cmdline(void) if ((cmd == SPECTRE_V2_CMD_RETPOLINE || cmd == SPECTRE_V2_CMD_RETPOLINE_LFENCE || - cmd == SPECTRE_V2_CMD_RETPOLINE_GENERIC) && + cmd == SPECTRE_V2_CMD_RETPOLINE_GENERIC || + cmd == SPECTRE_V2_CMD_EIBRS_LFENCE || + cmd == SPECTRE_V2_CMD_EIBRS_RETPOLINE) && !IS_ENABLED(CONFIG_RETPOLINE)) { - pr_err("%s selected but not compiled in. Switching to AUTO select\n", mitigation_options[i].option); + pr_err("%s selected but not compiled in. Switching to AUTO select\n", + mitigation_options[i].option); + return SPECTRE_V2_CMD_AUTO; + } + + if ((cmd == SPECTRE_V2_CMD_EIBRS || + cmd == SPECTRE_V2_CMD_EIBRS_LFENCE || + cmd == SPECTRE_V2_CMD_EIBRS_RETPOLINE) && + !boot_cpu_has(X86_FEATURE_IBRS_ENHANCED)) { + pr_err("%s selected but CPU doesn't have eIBRS. Switching to AUTO select\n", + mitigation_options[i].option); return SPECTRE_V2_CMD_AUTO; } - if ((cmd == SPECTRE_V2_CMD_RETPOLINE_LFENCE) && + if ((cmd == SPECTRE_V2_CMD_RETPOLINE_LFENCE || + cmd == SPECTRE_V2_CMD_EIBRS_LFENCE) && !boot_cpu_has(X86_FEATURE_LFENCE_RDTSC)) { - pr_err("%s selected, but CPU doesn't have a serializing LFENCE. Switching to AUTO select\n", mitigation_options[i].option); + pr_err("%s selected, but CPU doesn't have a serializing LFENCE. Switching to AUTO select\n", + mitigation_options[i].option); return SPECTRE_V2_CMD_AUTO; } @@ -894,6 +923,25 @@ static enum spectre_v2_mitigation_cmd __init spectre_v2_parse_cmdline(void) return cmd; } +static enum spectre_v2_mitigation __init spectre_v2_select_retpoline(void) +{ + if (!IS_ENABLED(CONFIG_RETPOLINE)) { + pr_err("Kernel not compiled with retpoline; no mitigation available!"); + return SPECTRE_V2_NONE; + } + + if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD || + boot_cpu_data.x86_vendor == X86_VENDOR_HYGON) { + if (!boot_cpu_has(X86_FEATURE_LFENCE_RDTSC)) { + pr_err("LFENCE not serializing, switching to generic retpoline\n"); + return SPECTRE_V2_RETPOLINE; + } + return SPECTRE_V2_LFENCE; + } + + return SPECTRE_V2_RETPOLINE; +} + static void __init spectre_v2_select_mitigation(void) { enum spectre_v2_mitigation_cmd cmd = spectre_v2_parse_cmdline(); @@ -914,49 +962,60 @@ static void __init spectre_v2_select_mitigation(void) case SPECTRE_V2_CMD_FORCE: case SPECTRE_V2_CMD_AUTO: if (boot_cpu_has(X86_FEATURE_IBRS_ENHANCED)) { - mode = SPECTRE_V2_IBRS_ENHANCED; - /* Force it so VMEXIT will restore correctly */ - x86_spec_ctrl_base |= SPEC_CTRL_IBRS; - wrmsrl(MSR_IA32_SPEC_CTRL, x86_spec_ctrl_base); - goto specv2_set_mode; + mode = SPECTRE_V2_EIBRS; + break; } - if (IS_ENABLED(CONFIG_RETPOLINE)) - goto retpoline_auto; + + mode = spectre_v2_select_retpoline(); break; + case SPECTRE_V2_CMD_RETPOLINE_LFENCE: - if (IS_ENABLED(CONFIG_RETPOLINE)) - goto retpoline_lfence; + mode = SPECTRE_V2_LFENCE; break; + case SPECTRE_V2_CMD_RETPOLINE_GENERIC: - if (IS_ENABLED(CONFIG_RETPOLINE)) - goto retpoline_generic; + mode = SPECTRE_V2_RETPOLINE; break; + case SPECTRE_V2_CMD_RETPOLINE: - if (IS_ENABLED(CONFIG_RETPOLINE)) - goto retpoline_auto; + mode = spectre_v2_select_retpoline(); + break; + + case SPECTRE_V2_CMD_EIBRS: + mode = SPECTRE_V2_EIBRS; + break; + + case SPECTRE_V2_CMD_EIBRS_LFENCE: + mode = SPECTRE_V2_EIBRS_LFENCE; + break; + + case SPECTRE_V2_CMD_EIBRS_RETPOLINE: + mode = SPECTRE_V2_EIBRS_RETPOLINE; break; } - pr_err("Spectre mitigation: kernel not compiled with retpoline; no mitigation available!"); - return; -retpoline_auto: - if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD || - boot_cpu_data.x86_vendor == X86_VENDOR_HYGON) { - retpoline_lfence: - if (!boot_cpu_has(X86_FEATURE_LFENCE_RDTSC)) { - pr_err("Spectre mitigation: LFENCE not serializing, switching to generic retpoline\n"); - goto retpoline_generic; - } - mode = SPECTRE_V2_LFENCE; + if (spectre_v2_in_eibrs_mode(mode)) { + /* Force it so VMEXIT will restore correctly */ + x86_spec_ctrl_base |= SPEC_CTRL_IBRS; + wrmsrl(MSR_IA32_SPEC_CTRL, x86_spec_ctrl_base); + } + + switch (mode) { + case SPECTRE_V2_NONE: + case SPECTRE_V2_EIBRS: + break; + + case SPECTRE_V2_LFENCE: + case SPECTRE_V2_EIBRS_LFENCE: setup_force_cpu_cap(X86_FEATURE_RETPOLINE_LFENCE); + fallthrough; + + case SPECTRE_V2_RETPOLINE: + case SPECTRE_V2_EIBRS_RETPOLINE: setup_force_cpu_cap(X86_FEATURE_RETPOLINE); - } else { - retpoline_generic: - mode = SPECTRE_V2_RETPOLINE; - setup_force_cpu_cap(X86_FEATURE_RETPOLINE); + break; } -specv2_set_mode: spectre_v2_enabled = mode; pr_info("%s\n", spectre_v2_strings[mode]); @@ -982,7 +1041,7 @@ specv2_set_mode: * the CPU supports Enhanced IBRS, kernel might un-intentionally not * enable IBRS around firmware calls. */ - if (boot_cpu_has(X86_FEATURE_IBRS) && mode != SPECTRE_V2_IBRS_ENHANCED) { + if (boot_cpu_has(X86_FEATURE_IBRS) && !spectre_v2_in_eibrs_mode(mode)) { setup_force_cpu_cap(X86_FEATURE_USE_IBRS_FW); pr_info("Enabling Restricted Speculation for firmware calls\n"); } @@ -1691,7 +1750,7 @@ static ssize_t tsx_async_abort_show_state(char *buf) static char *stibp_state(void) { - if (spectre_v2_enabled == SPECTRE_V2_IBRS_ENHANCED) + if (spectre_v2_in_eibrs_mode(spectre_v2_enabled)) return ""; switch (spectre_v2_user_stibp) { -- cgit From 44a3918c8245ab10c6c9719dd12e7a8d291980d8 Mon Sep 17 00:00:00 2001 From: Josh Poimboeuf Date: Fri, 18 Feb 2022 11:49:08 -0800 Subject: x86/speculation: Include unprivileged eBPF status in Spectre v2 mitigation reporting With unprivileged eBPF enabled, eIBRS (without retpoline) is vulnerable to Spectre v2 BHB-based attacks. When both are enabled, print a warning message and report it in the 'spectre_v2' sysfs vulnerabilities file. Signed-off-by: Josh Poimboeuf Signed-off-by: Borislav Petkov Reviewed-by: Thomas Gleixner --- arch/x86/kernel/cpu/bugs.c | 35 +++++++++++++++++++++++++++++------ 1 file changed, 29 insertions(+), 6 deletions(-) (limited to 'arch/x86/kernel/cpu') diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index 79c52dd6c597..0a4267c63d3b 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -16,6 +16,7 @@ #include #include #include +#include #include #include @@ -650,6 +651,16 @@ static inline const char *spectre_v2_module_string(void) static inline const char *spectre_v2_module_string(void) { return ""; } #endif +#define SPECTRE_V2_EIBRS_EBPF_MSG "WARNING: Unprivileged eBPF is enabled with eIBRS on, data leaks possible via Spectre v2 BHB attacks!\n" + +#ifdef CONFIG_BPF_SYSCALL +void unpriv_ebpf_notify(int new_state) +{ + if (spectre_v2_enabled == SPECTRE_V2_EIBRS && !new_state) + pr_err(SPECTRE_V2_EIBRS_EBPF_MSG); +} +#endif + static inline bool match_option(const char *arg, int arglen, const char *opt) { int len = strlen(opt); @@ -994,6 +1005,9 @@ static void __init spectre_v2_select_mitigation(void) break; } + if (mode == SPECTRE_V2_EIBRS && unprivileged_ebpf_enabled()) + pr_err(SPECTRE_V2_EIBRS_EBPF_MSG); + if (spectre_v2_in_eibrs_mode(mode)) { /* Force it so VMEXIT will restore correctly */ x86_spec_ctrl_base |= SPEC_CTRL_IBRS; @@ -1780,6 +1794,20 @@ static char *ibpb_state(void) return ""; } +static ssize_t spectre_v2_show_state(char *buf) +{ + if (spectre_v2_enabled == SPECTRE_V2_EIBRS && unprivileged_ebpf_enabled()) + return sprintf(buf, "Vulnerable: Unprivileged eBPF enabled\n"); + + return sprintf(buf, "%s%s%s%s%s%s\n", + spectre_v2_strings[spectre_v2_enabled], + ibpb_state(), + boot_cpu_has(X86_FEATURE_USE_IBRS_FW) ? ", IBRS_FW" : "", + stibp_state(), + boot_cpu_has(X86_FEATURE_RSB_CTXSW) ? ", RSB filling" : "", + spectre_v2_module_string()); +} + static ssize_t srbds_show_state(char *buf) { return sprintf(buf, "%s\n", srbds_strings[srbds_mitigation]); @@ -1805,12 +1833,7 @@ static ssize_t cpu_show_common(struct device *dev, struct device_attribute *attr return sprintf(buf, "%s\n", spectre_v1_strings[spectre_v1_mitigation]); case X86_BUG_SPECTRE_V2: - return sprintf(buf, "%s%s%s%s%s%s\n", spectre_v2_strings[spectre_v2_enabled], - ibpb_state(), - boot_cpu_has(X86_FEATURE_USE_IBRS_FW) ? ", IBRS_FW" : "", - stibp_state(), - boot_cpu_has(X86_FEATURE_RSB_CTXSW) ? ", RSB filling" : "", - spectre_v2_module_string()); + return spectre_v2_show_state(buf); case X86_BUG_SPEC_STORE_BYPASS: return sprintf(buf, "%s\n", ssb_strings[ssb_mode]); -- cgit From 244d00b5dd4755f8df892c86cab35fb2cfd4f14b Mon Sep 17 00:00:00 2001 From: Kim Phillips Date: Mon, 28 Feb 2022 11:23:15 -0600 Subject: x86/speculation: Use generic retpoline by default on AMD AMD retpoline may be susceptible to speculation. The speculation execution window for an incorrect indirect branch prediction using LFENCE/JMP sequence may potentially be large enough to allow exploitation using Spectre V2. By default, don't use retpoline,lfence on AMD. Instead, use the generic retpoline. Signed-off-by: Kim Phillips Signed-off-by: Borislav Petkov --- arch/x86/kernel/cpu/bugs.c | 9 --------- 1 file changed, 9 deletions(-) (limited to 'arch/x86/kernel/cpu') diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index 0a4267c63d3b..a36bfe2c2480 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -941,15 +941,6 @@ static enum spectre_v2_mitigation __init spectre_v2_select_retpoline(void) return SPECTRE_V2_NONE; } - if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD || - boot_cpu_data.x86_vendor == X86_VENDOR_HYGON) { - if (!boot_cpu_has(X86_FEATURE_LFENCE_RDTSC)) { - pr_err("LFENCE not serializing, switching to generic retpoline\n"); - return SPECTRE_V2_RETPOLINE; - } - return SPECTRE_V2_LFENCE; - } - return SPECTRE_V2_RETPOLINE; } -- cgit From eafd987d4a82c7bb5aa12f0e3b4f8f3dea93e678 Mon Sep 17 00:00:00 2001 From: Josh Poimboeuf Date: Fri, 25 Feb 2022 14:31:49 -0800 Subject: x86/speculation: Warn about Spectre v2 LFENCE mitigation With: f8a66d608a3e ("x86,bugs: Unconditionally allow spectre_v2=retpoline,amd") it became possible to enable the LFENCE "retpoline" on Intel. However, Intel doesn't recommend it, as it has some weaknesses compared to retpoline. Now AMD doesn't recommend it either. It can still be left available as a cmdline option. It's faster than retpoline but is weaker in certain scenarios -- particularly SMT, but even non-SMT may be vulnerable in some cases. So just unconditionally warn if the user requests it on the cmdline. [ bp: Massage commit message. ] Signed-off-by: Josh Poimboeuf Signed-off-by: Borislav Petkov --- arch/x86/kernel/cpu/bugs.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'arch/x86/kernel/cpu') diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index a36bfe2c2480..cfd116423908 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -651,6 +651,7 @@ static inline const char *spectre_v2_module_string(void) static inline const char *spectre_v2_module_string(void) { return ""; } #endif +#define SPECTRE_V2_LFENCE_MSG "WARNING: LFENCE mitigation is not recommended for this CPU, data leaks possible!\n" #define SPECTRE_V2_EIBRS_EBPF_MSG "WARNING: Unprivileged eBPF is enabled with eIBRS on, data leaks possible via Spectre v2 BHB attacks!\n" #ifdef CONFIG_BPF_SYSCALL @@ -972,6 +973,7 @@ static void __init spectre_v2_select_mitigation(void) break; case SPECTRE_V2_CMD_RETPOLINE_LFENCE: + pr_err(SPECTRE_V2_LFENCE_MSG); mode = SPECTRE_V2_LFENCE; break; @@ -1787,6 +1789,9 @@ static char *ibpb_state(void) static ssize_t spectre_v2_show_state(char *buf) { + if (spectre_v2_enabled == SPECTRE_V2_LFENCE) + return sprintf(buf, "Vulnerable: LFENCE\n"); + if (spectre_v2_enabled == SPECTRE_V2_EIBRS && unprivileged_ebpf_enabled()) return sprintf(buf, "Vulnerable: Unprivileged eBPF enabled\n"); -- cgit From 0de05d056afdb00eca8c7bbb0c79a3438daf700c Mon Sep 17 00:00:00 2001 From: Josh Poimboeuf Date: Fri, 25 Feb 2022 14:32:28 -0800 Subject: x86/speculation: Warn about eIBRS + LFENCE + Unprivileged eBPF + SMT The commit 44a3918c8245 ("x86/speculation: Include unprivileged eBPF status in Spectre v2 mitigation reporting") added a warning for the "eIBRS + unprivileged eBPF" combination, which has been shown to be vulnerable against Spectre v2 BHB-based attacks. However, there's no warning about the "eIBRS + LFENCE retpoline + unprivileged eBPF" combo. The LFENCE adds more protection by shortening the speculation window after a mispredicted branch. That makes an attack significantly more difficult, even with unprivileged eBPF. So at least for now the logic doesn't warn about that combination. But if you then add SMT into the mix, the SMT attack angle weakens the effectiveness of the LFENCE considerably. So extend the "eIBRS + unprivileged eBPF" warning to also include the "eIBRS + LFENCE + unprivileged eBPF + SMT" case. [ bp: Massage commit message. ] Suggested-by: Alyssa Milburn Signed-off-by: Josh Poimboeuf Signed-off-by: Borislav Petkov --- arch/x86/kernel/cpu/bugs.c | 27 +++++++++++++++++++++++++-- 1 file changed, 25 insertions(+), 2 deletions(-) (limited to 'arch/x86/kernel/cpu') diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index cfd116423908..6296e1ebed1d 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -653,12 +653,27 @@ static inline const char *spectre_v2_module_string(void) { return ""; } #define SPECTRE_V2_LFENCE_MSG "WARNING: LFENCE mitigation is not recommended for this CPU, data leaks possible!\n" #define SPECTRE_V2_EIBRS_EBPF_MSG "WARNING: Unprivileged eBPF is enabled with eIBRS on, data leaks possible via Spectre v2 BHB attacks!\n" +#define SPECTRE_V2_EIBRS_LFENCE_EBPF_SMT_MSG "WARNING: Unprivileged eBPF is enabled with eIBRS+LFENCE mitigation and SMT, data leaks possible via Spectre v2 BHB attacks!\n" #ifdef CONFIG_BPF_SYSCALL void unpriv_ebpf_notify(int new_state) { - if (spectre_v2_enabled == SPECTRE_V2_EIBRS && !new_state) + if (new_state) + return; + + /* Unprivileged eBPF is enabled */ + + switch (spectre_v2_enabled) { + case SPECTRE_V2_EIBRS: pr_err(SPECTRE_V2_EIBRS_EBPF_MSG); + break; + case SPECTRE_V2_EIBRS_LFENCE: + if (sched_smt_active()) + pr_err(SPECTRE_V2_EIBRS_LFENCE_EBPF_SMT_MSG); + break; + default: + break; + } } #endif @@ -1118,6 +1133,10 @@ void cpu_bugs_smt_update(void) { mutex_lock(&spec_ctrl_mutex); + if (sched_smt_active() && unprivileged_ebpf_enabled() && + spectre_v2_enabled == SPECTRE_V2_EIBRS_LFENCE) + pr_warn_once(SPECTRE_V2_EIBRS_LFENCE_EBPF_SMT_MSG); + switch (spectre_v2_user_stibp) { case SPECTRE_V2_USER_NONE: break; @@ -1793,7 +1812,11 @@ static ssize_t spectre_v2_show_state(char *buf) return sprintf(buf, "Vulnerable: LFENCE\n"); if (spectre_v2_enabled == SPECTRE_V2_EIBRS && unprivileged_ebpf_enabled()) - return sprintf(buf, "Vulnerable: Unprivileged eBPF enabled\n"); + return sprintf(buf, "Vulnerable: eIBRS with unprivileged eBPF\n"); + + if (sched_smt_active() && unprivileged_ebpf_enabled() && + spectre_v2_enabled == SPECTRE_V2_EIBRS_LFENCE) + return sprintf(buf, "Vulnerable: eIBRS+LFENCE with unprivileged eBPF and SMT\n"); return sprintf(buf, "%s%s%s%s%s%s\n", spectre_v2_strings[spectre_v2_enabled], -- cgit From 08999b2489b4c9b939d7483dbd03702ee4576d96 Mon Sep 17 00:00:00 2001 From: Jarkko Sakkinen Date: Fri, 4 Mar 2022 00:38:58 +0200 Subject: x86/sgx: Free backing memory after faulting the enclave page There is a limited amount of SGX memory (EPC) on each system. When that memory is used up, SGX has its own swapping mechanism which is similar in concept but totally separate from the core mm/* code. Instead of swapping to disk, SGX swaps from EPC to normal RAM. That normal RAM comes from a shared memory pseudo-file and can itself be swapped by the core mm code. There is a hierarchy like this: EPC <-> shmem <-> disk After data is swapped back in from shmem to EPC, the shmem backing storage needs to be freed. Currently, the backing shmem is not freed. This effectively wastes the shmem while the enclave is running. The memory is recovered when the enclave is destroyed and the backing storage freed. Sort this out by freeing memory with shmem_truncate_range(), as soon as a page is faulted back to the EPC. In addition, free the memory for PCMD pages as soon as all PCMD's in a page have been marked as unused by zeroing its contents. Cc: stable@vger.kernel.org Fixes: 1728ab54b4be ("x86/sgx: Add a page reclaimer") Reported-by: Dave Hansen Signed-off-by: Jarkko Sakkinen Signed-off-by: Dave Hansen Link: https://lkml.kernel.org/r/20220303223859.273187-1-jarkko@kernel.org --- arch/x86/kernel/cpu/sgx/encl.c | 57 +++++++++++++++++++++++++++++++++++------- 1 file changed, 48 insertions(+), 9 deletions(-) (limited to 'arch/x86/kernel/cpu') diff --git a/arch/x86/kernel/cpu/sgx/encl.c b/arch/x86/kernel/cpu/sgx/encl.c index 48afe96ae0f0..7c63a1911fae 100644 --- a/arch/x86/kernel/cpu/sgx/encl.c +++ b/arch/x86/kernel/cpu/sgx/encl.c @@ -12,6 +12,30 @@ #include "encls.h" #include "sgx.h" +/* + * Calculate byte offset of a PCMD struct associated with an enclave page. PCMD's + * follow right after the EPC data in the backing storage. In addition to the + * visible enclave pages, there's one extra page slot for SECS, before PCMD + * structs. + */ +static inline pgoff_t sgx_encl_get_backing_page_pcmd_offset(struct sgx_encl *encl, + unsigned long page_index) +{ + pgoff_t epc_end_off = encl->size + sizeof(struct sgx_secs); + + return epc_end_off + page_index * sizeof(struct sgx_pcmd); +} + +/* + * Free a page from the backing storage in the given page index. + */ +static inline void sgx_encl_truncate_backing_page(struct sgx_encl *encl, unsigned long page_index) +{ + struct inode *inode = file_inode(encl->backing); + + shmem_truncate_range(inode, PFN_PHYS(page_index), PFN_PHYS(page_index) + PAGE_SIZE - 1); +} + /* * ELDU: Load an EPC page as unblocked. For more info, see "OS Management of EPC * Pages" in the SDM. @@ -22,9 +46,11 @@ static int __sgx_encl_eldu(struct sgx_encl_page *encl_page, { unsigned long va_offset = encl_page->desc & SGX_ENCL_PAGE_VA_OFFSET_MASK; struct sgx_encl *encl = encl_page->encl; + pgoff_t page_index, page_pcmd_off; struct sgx_pageinfo pginfo; struct sgx_backing b; - pgoff_t page_index; + bool pcmd_page_empty; + u8 *pcmd_page; int ret; if (secs_page) @@ -32,14 +58,16 @@ static int __sgx_encl_eldu(struct sgx_encl_page *encl_page, else page_index = PFN_DOWN(encl->size); + page_pcmd_off = sgx_encl_get_backing_page_pcmd_offset(encl, page_index); + ret = sgx_encl_get_backing(encl, page_index, &b); if (ret) return ret; pginfo.addr = encl_page->desc & PAGE_MASK; pginfo.contents = (unsigned long)kmap_atomic(b.contents); - pginfo.metadata = (unsigned long)kmap_atomic(b.pcmd) + - b.pcmd_offset; + pcmd_page = kmap_atomic(b.pcmd); + pginfo.metadata = (unsigned long)pcmd_page + b.pcmd_offset; if (secs_page) pginfo.secs = (u64)sgx_get_epc_virt_addr(secs_page); @@ -55,11 +83,24 @@ static int __sgx_encl_eldu(struct sgx_encl_page *encl_page, ret = -EFAULT; } - kunmap_atomic((void *)(unsigned long)(pginfo.metadata - b.pcmd_offset)); + memset(pcmd_page + b.pcmd_offset, 0, sizeof(struct sgx_pcmd)); + + /* + * The area for the PCMD in the page was zeroed above. Check if the + * whole page is now empty meaning that all PCMD's have been zeroed: + */ + pcmd_page_empty = !memchr_inv(pcmd_page, 0, PAGE_SIZE); + + kunmap_atomic(pcmd_page); kunmap_atomic((void *)(unsigned long)pginfo.contents); sgx_encl_put_backing(&b, false); + sgx_encl_truncate_backing_page(encl, page_index); + + if (pcmd_page_empty) + sgx_encl_truncate_backing_page(encl, PFN_DOWN(page_pcmd_off)); + return ret; } @@ -579,7 +620,7 @@ static struct page *sgx_encl_get_backing_page(struct sgx_encl *encl, int sgx_encl_get_backing(struct sgx_encl *encl, unsigned long page_index, struct sgx_backing *backing) { - pgoff_t pcmd_index = PFN_DOWN(encl->size) + 1 + (page_index >> 5); + pgoff_t page_pcmd_off = sgx_encl_get_backing_page_pcmd_offset(encl, page_index); struct page *contents; struct page *pcmd; @@ -587,7 +628,7 @@ int sgx_encl_get_backing(struct sgx_encl *encl, unsigned long page_index, if (IS_ERR(contents)) return PTR_ERR(contents); - pcmd = sgx_encl_get_backing_page(encl, pcmd_index); + pcmd = sgx_encl_get_backing_page(encl, PFN_DOWN(page_pcmd_off)); if (IS_ERR(pcmd)) { put_page(contents); return PTR_ERR(pcmd); @@ -596,9 +637,7 @@ int sgx_encl_get_backing(struct sgx_encl *encl, unsigned long page_index, backing->page_index = page_index; backing->contents = contents; backing->pcmd = pcmd; - backing->pcmd_offset = - (page_index & (PAGE_SIZE / sizeof(struct sgx_pcmd) - 1)) * - sizeof(struct sgx_pcmd); + backing->pcmd_offset = page_pcmd_off & (PAGE_SIZE - 1); return 0; } -- cgit