From 266d63a7d9d48c6d5dee486378ec0e8c86c4d74a Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Mon, 11 Feb 2019 12:23:51 +0100 Subject: x86/cpufeature: Fix various quality problems in the header Thomas noticed that the new arch/x86/include/asm/cpu_device_id.h header is a train-wreck that didn't incorporate review feedback like not using __u8 in kernel-only headers. While at it also fix all the *other* problems this header has: - Use canonical names for the header guards. It's inexplicable why a non-standard guard was used. - Don't define the header guard to 1. Plus annotate the closing #endif as done absolutely every other header. Again, an inexplicable source of noise. - Move the kernel API calls provided by this header next to each other, there's absolutely no reason to have them spread apart in the header. - Align the INTEL_CPU_DESC() macro initializations vertically, this is easier to read and it's also the canonical style. - Actually name the macro arguments properly: instead of 'mod, step, rev', spell out 'model, stepping, revision' - it's not like we have a lack of characters in this header. - Actually make arguments macro-safe - again it's inexplicable why it wasn't done properly to begin with. Quite amazing how many problems a 41 lines header can contain. This kind of code quality is unacceptable, and it slipped through the review net of 2 developers and 2 maintainers, including myself, until Thomas noticed it. :-/ Reported-by: Thomas Gleixner Cc: Andi Kleen Cc: Kan Liang Cc: Peter Zijlstra (Intel) Cc: Borislav Petkov Cc: Linus Torvalds Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar --- arch/x86/include/asm/cpu_device_id.h | 31 +++++++++++++++---------------- 1 file changed, 15 insertions(+), 16 deletions(-) (limited to 'arch') diff --git a/arch/x86/include/asm/cpu_device_id.h b/arch/x86/include/asm/cpu_device_id.h index 3417110574c1..31c379c1da41 100644 --- a/arch/x86/include/asm/cpu_device_id.h +++ b/arch/x86/include/asm/cpu_device_id.h @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: GPL-2.0 */ -#ifndef _CPU_DEVICE_ID -#define _CPU_DEVICE_ID 1 +#ifndef _ASM_X86_CPU_DEVICE_ID +#define _ASM_X86_CPU_DEVICE_ID /* * Declare drivers belonging to specific x86 CPUs @@ -9,8 +9,6 @@ #include -extern const struct x86_cpu_id *x86_match_cpu(const struct x86_cpu_id *match); - /* * Match specific microcode revisions. * @@ -22,21 +20,22 @@ extern const struct x86_cpu_id *x86_match_cpu(const struct x86_cpu_id *match); */ struct x86_cpu_desc { - __u8 x86_family; - __u8 x86_vendor; - __u8 x86_model; - __u8 x86_stepping; - __u32 x86_microcode_rev; + u8 x86_family; + u8 x86_vendor; + u8 x86_model; + u8 x86_stepping; + u32 x86_microcode_rev; }; -#define INTEL_CPU_DESC(mod, step, rev) { \ - .x86_family = 6, \ - .x86_vendor = X86_VENDOR_INTEL, \ - .x86_model = mod, \ - .x86_stepping = step, \ - .x86_microcode_rev = rev, \ +#define INTEL_CPU_DESC(model, stepping, revision) { \ + .x86_family = 6, \ + .x86_vendor = X86_VENDOR_INTEL, \ + .x86_model = (model), \ + .x86_stepping = (stepping), \ + .x86_microcode_rev = (revision), \ } +extern const struct x86_cpu_id *x86_match_cpu(const struct x86_cpu_id *match); extern bool x86_cpu_has_min_microcode_rev(const struct x86_cpu_desc *table); -#endif +#endif /* _ASM_X86_CPU_DEVICE_ID */ -- cgit v1.2.3-58-ga151 From a3151724437f54076cc10bc02b1c4f0003ae36cd Mon Sep 17 00:00:00 2001 From: Matteo Croce Date: Mon, 18 Mar 2019 22:24:03 +0100 Subject: x86/mm: Don't leak kernel addresses Since commit: ad67b74d2469d9b8 ("printk: hash addresses printed with %p") at boot "____ptrval____" is printed instead of actual addresses: found SMP MP-table at [mem 0x000f5cc0-0x000f5ccf] mapped at [(____ptrval____)] Instead of changing the print to "%px", and leaking a kernel addresses, just remove the print completely, like in: 071929dbdd865f77 ("arm64: Stop printing the virtual memory layout"). Signed-off-by: Matteo Croce Cc: Borislav Petkov Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Signed-off-by: Ingo Molnar --- arch/x86/kernel/mpparse.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch') diff --git a/arch/x86/kernel/mpparse.c b/arch/x86/kernel/mpparse.c index 3482460d984d..1bfe5c6e6cfe 100644 --- a/arch/x86/kernel/mpparse.c +++ b/arch/x86/kernel/mpparse.c @@ -598,8 +598,8 @@ static int __init smp_scan_config(unsigned long base, unsigned long length) mpf_base = base; mpf_found = true; - pr_info("found SMP MP-table at [mem %#010lx-%#010lx] mapped at [%p]\n", - base, base + sizeof(*mpf) - 1, mpf); + pr_info("found SMP MP-table at [mem %#010lx-%#010lx]\n", + base, base + sizeof(*mpf) - 1); memblock_reserve(base, sizeof(*mpf)); if (mpf->physptr) -- cgit v1.2.3-58-ga151 From a9c640ac96e19b3966357ec9bb586edd2e1e74de Mon Sep 17 00:00:00 2001 From: Nick Desaulniers Date: Thu, 14 Mar 2019 15:14:57 -0700 Subject: x86/boot: Restrict header scope to make Clang happy The inclusion of was causing issue as the definition of __arch_hweight64 from arch/x86/include/asm/arch_hweight.h eventually gets included. The definition is problematic when compiled with -m16 (all code in arch/x86/boot/ is) as the "D" inline assembly constraint is rejected by both compilers when passed an argument of type long long (regardless of signedness, anything smaller is fine). Because GCC performs inlining before semantic analysis, and __arch_hweight64 is dead in this translation unit, GCC does not report any issues at compile time. Clang does the semantic analysis in the front end, before inlining (run in the middle) can determine the code is dead. I consider this another case of PR33587, which I think we can do more work to solve. It turns out that arch/x86/boot/string.c doesn't actually need linux/kernel.h, simply linux/limits.h and linux/compiler.h. Suggested-by: Stephen Rothwell Signed-off-by: Nick Desaulniers Signed-off-by: Thomas Gleixner Tested-by: Nathan Chancellor Reviewed-by: Nathan Chancellor Cc: bp@alien8.de Cc: niravd@google.com Cc: "H. Peter Anvin" Cc: Chao Fan Cc: Uros Bizjak Link: https://bugs.llvm.org/show_bug.cgi?id=33587 Link: https://github.com/ClangBuiltLinux/linux/issues/347 Link: https://lkml.kernel.org/r/20190314221458.83047-1-ndesaulniers@google.com --- arch/x86/boot/string.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'arch') diff --git a/arch/x86/boot/string.c b/arch/x86/boot/string.c index 315a67b8896b..90154df8f125 100644 --- a/arch/x86/boot/string.c +++ b/arch/x86/boot/string.c @@ -13,8 +13,9 @@ */ #include -#include +#include #include +#include #include #include "ctype.h" #include "string.h" -- cgit v1.2.3-58-ga151 From 725e29db8cb9058976559bc3239c97ef7db40eea Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Thu, 14 Mar 2019 23:08:38 +0000 Subject: x86/lib: Fix indentation issue, remove extra tab The increment of buff is indented one level too deeply, clean this up by removing a tab. Signed-off-by: Colin Ian King Signed-off-by: Thomas Gleixner Cc: Borislav Petkov Cc: "H . Peter Anvin" Cc: kernel-janitors@vger.kernel.org Link: https://lkml.kernel.org/r/20190314230838.18256-1-colin.king@canonical.com --- arch/x86/lib/csum-partial_64.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch') diff --git a/arch/x86/lib/csum-partial_64.c b/arch/x86/lib/csum-partial_64.c index 9baca3e054be..e7925d668b68 100644 --- a/arch/x86/lib/csum-partial_64.c +++ b/arch/x86/lib/csum-partial_64.c @@ -94,7 +94,7 @@ static unsigned do_csum(const unsigned char *buff, unsigned len) : "m" (*(unsigned long *)buff), "r" (zero), "0" (result)); --count; - buff += 8; + buff += 8; } result = add32_with_carry(result>>32, result&0xffffffff); -- cgit v1.2.3-58-ga151 From 2e84f116afca3719c9d0a1a78b47b48f75fd5724 Mon Sep 17 00:00:00 2001 From: Aditya Pakki Date: Mon, 18 Mar 2019 21:19:56 -0500 Subject: x86/hpet: Prevent potential NULL pointer dereference hpet_virt_address may be NULL when ioremap_nocache fail, but the code lacks a check. Add a check to prevent NULL pointer dereference. Signed-off-by: Aditya Pakki Signed-off-by: Thomas Gleixner Cc: kjlu@umn.edu Cc: Borislav Petkov Cc: "H. Peter Anvin" Cc: Kees Cook Cc: Joe Perches Cc: Nicolai Stange Cc: Roland Dreier Link: https://lkml.kernel.org/r/20190319021958.17275-1-pakki001@umn.edu --- arch/x86/kernel/hpet.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'arch') diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c index dfd3aca82c61..fb32925a2e62 100644 --- a/arch/x86/kernel/hpet.c +++ b/arch/x86/kernel/hpet.c @@ -905,6 +905,8 @@ int __init hpet_enable(void) return 0; hpet_set_mapping(); + if (!hpet_virt_address) + return 0; /* * Read the period and check for a sane value: -- cgit v1.2.3-58-ga151 From 534c89c22e26b183d838294f0937ee092c82ad3a Mon Sep 17 00:00:00 2001 From: Kangjie Lu Date: Thu, 14 Mar 2019 00:46:51 -0500 Subject: x86/hyperv: Prevent potential NULL pointer dereference The page allocation in hv_cpu_init() can fail, but the code does not have a check for that. Add a check and return -ENOMEM when the allocation fails. [ tglx: Massaged changelog ] Signed-off-by: Kangjie Lu Signed-off-by: Thomas Gleixner Reviewed-by: Mukesh Ojha Acked-by: "K. Y. Srinivasan" Cc: pakki001@umn.edu Cc: Haiyang Zhang Cc: Stephen Hemminger Cc: Sasha Levin Cc: Borislav Petkov Cc: "H. Peter Anvin" Cc: linux-hyperv@vger.kernel.org Link: https://lkml.kernel.org/r/20190314054651.1315-1-kjlu@umn.edu --- arch/x86/hyperv/hv_init.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'arch') diff --git a/arch/x86/hyperv/hv_init.c b/arch/x86/hyperv/hv_init.c index 6461a16b4559..e4ba467a9fc6 100644 --- a/arch/x86/hyperv/hv_init.c +++ b/arch/x86/hyperv/hv_init.c @@ -103,9 +103,13 @@ static int hv_cpu_init(unsigned int cpu) u64 msr_vp_index; struct hv_vp_assist_page **hvp = &hv_vp_assist_page[smp_processor_id()]; void **input_arg; + struct page *pg; input_arg = (void **)this_cpu_ptr(hyperv_pcpu_input_arg); - *input_arg = page_address(alloc_page(GFP_KERNEL)); + pg = alloc_page(GFP_KERNEL); + if (unlikely(!pg)) + return -ENOMEM; + *input_arg = page_address(pg); hv_get_vp_index(msr_vp_index); -- cgit v1.2.3-58-ga151 From 9bd681251b7c1db1c6cfe29a72c5ea1b1c0ba022 Mon Sep 17 00:00:00 2001 From: Borislav Petkov Date: Wed, 13 Mar 2019 12:00:22 +0100 Subject: x86/microcode: Announce reload operation's completion By popular demand, issue a single line to dmesg after the reload operation completes to let the user know that a reload has at least been attempted. Signed-off-by: Borislav Petkov Signed-off-by: Thomas Gleixner Link: https://lkml.kernel.org/r/20190313110022.8229-1-bp@alien8.de --- arch/x86/kernel/cpu/microcode/core.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'arch') diff --git a/arch/x86/kernel/cpu/microcode/core.c b/arch/x86/kernel/cpu/microcode/core.c index 97f9ada9ceda..5260185cbf7b 100644 --- a/arch/x86/kernel/cpu/microcode/core.c +++ b/arch/x86/kernel/cpu/microcode/core.c @@ -608,6 +608,8 @@ static int microcode_reload_late(void) if (ret > 0) microcode_check(); + pr_info("Reload completed, microcode revision: 0x%x\n", boot_cpu_data.microcode); + return ret; } -- cgit v1.2.3-58-ga151 From 18fb053f9b827bd98cfc64f2a35df8ab19745a1d Mon Sep 17 00:00:00 2001 From: Matthew Whitehead Date: Thu, 14 Mar 2019 16:46:00 -0400 Subject: x86/cpu/cyrix: Use correct macros for Cyrix calls on Geode processors There are comments in processor-cyrix.h advising you to _not_ make calls using the deprecated macros in this style: setCx86_old(CX86_CCR4, getCx86_old(CX86_CCR4) | 0x80); This is because it expands the macro into a non-functioning calling sequence. The calling order must be: outb(CX86_CCR2, 0x22); inb(0x23); From the comments: * When using the old macros a line like * setCx86(CX86_CCR2, getCx86(CX86_CCR2) | 0x88); * gets expanded to: * do { * outb((CX86_CCR2), 0x22); * outb((({ * outb((CX86_CCR2), 0x22); * inb(0x23); * }) | 0x88), 0x23); * } while (0); The new macros fix this problem, so use them instead. Tested on an actual Geode processor. Signed-off-by: Matthew Whitehead Signed-off-by: Thomas Gleixner Cc: luto@kernel.org Link: https://lkml.kernel.org/r/1552596361-8967-2-git-send-email-tedheadster@gmail.com --- arch/x86/kernel/cpu/cyrix.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) (limited to 'arch') diff --git a/arch/x86/kernel/cpu/cyrix.c b/arch/x86/kernel/cpu/cyrix.c index d12226f60168..1d9b8aaea06c 100644 --- a/arch/x86/kernel/cpu/cyrix.c +++ b/arch/x86/kernel/cpu/cyrix.c @@ -124,7 +124,7 @@ static void set_cx86_reorder(void) setCx86(CX86_CCR3, (ccr3 & 0x0f) | 0x10); /* enable MAPEN */ /* Load/Store Serialize to mem access disable (=reorder it) */ - setCx86_old(CX86_PCR0, getCx86_old(CX86_PCR0) & ~0x80); + setCx86(CX86_PCR0, getCx86(CX86_PCR0) & ~0x80); /* set load/store serialize from 1GB to 4GB */ ccr3 |= 0xe0; setCx86(CX86_CCR3, ccr3); @@ -135,11 +135,11 @@ static void set_cx86_memwb(void) pr_info("Enable Memory-Write-back mode on Cyrix/NSC processor.\n"); /* CCR2 bit 2: unlock NW bit */ - setCx86_old(CX86_CCR2, getCx86_old(CX86_CCR2) & ~0x04); + setCx86(CX86_CCR2, getCx86(CX86_CCR2) & ~0x04); /* set 'Not Write-through' */ write_cr0(read_cr0() | X86_CR0_NW); /* CCR2 bit 2: lock NW bit and set WT1 */ - setCx86_old(CX86_CCR2, getCx86_old(CX86_CCR2) | 0x14); + setCx86(CX86_CCR2, getCx86(CX86_CCR2) | 0x14); } /* @@ -153,14 +153,14 @@ static void geode_configure(void) local_irq_save(flags); /* Suspend on halt power saving and enable #SUSP pin */ - setCx86_old(CX86_CCR2, getCx86_old(CX86_CCR2) | 0x88); + setCx86(CX86_CCR2, getCx86(CX86_CCR2) | 0x88); ccr3 = getCx86(CX86_CCR3); setCx86(CX86_CCR3, (ccr3 & 0x0f) | 0x10); /* enable MAPEN */ /* FPU fast, DTE cache, Mem bypass */ - setCx86_old(CX86_CCR4, getCx86_old(CX86_CCR4) | 0x38); + setCx86(CX86_CCR4, getCx86(CX86_CCR4) | 0x38); setCx86(CX86_CCR3, ccr3); /* disable MAPEN */ set_cx86_memwb(); @@ -296,7 +296,7 @@ static void init_cyrix(struct cpuinfo_x86 *c) /* GXm supports extended cpuid levels 'ala' AMD */ if (c->cpuid_level == 2) { /* Enable cxMMX extensions (GX1 Datasheet 54) */ - setCx86_old(CX86_CCR7, getCx86_old(CX86_CCR7) | 1); + setCx86(CX86_CCR7, getCx86(CX86_CCR7) | 1); /* * GXm : 0x30 ... 0x5f GXm datasheet 51 @@ -319,7 +319,7 @@ static void init_cyrix(struct cpuinfo_x86 *c) if (dir1 > 7) { dir0_msn++; /* M II */ /* Enable MMX extensions (App note 108) */ - setCx86_old(CX86_CCR7, getCx86_old(CX86_CCR7)|1); + setCx86(CX86_CCR7, getCx86(CX86_CCR7)|1); } else { /* A 6x86MX - it has the bug. */ set_cpu_bug(c, X86_BUG_COMA); -- cgit v1.2.3-58-ga151 From 0f4d3aa761b71cd6984330baca1e18bf0590e441 Mon Sep 17 00:00:00 2001 From: Matthew Whitehead Date: Thu, 14 Mar 2019 16:46:01 -0400 Subject: x86/cpu/cyrix: Remove {get,set}Cx86_old macros used for Cyrix processors The getCx86_old() and setCx86_old() macros have been replaced with correctly working getCx86() and setCx86(), so remove these unused macros. Signed-off-by: Matthew Whitehead Signed-off-by: Thomas Gleixner Cc: luto@kernel.org Link: https://lkml.kernel.org/r/1552596361-8967-3-git-send-email-tedheadster@gmail.com --- arch/x86/include/asm/processor-cyrix.h | 21 --------------------- 1 file changed, 21 deletions(-) (limited to 'arch') diff --git a/arch/x86/include/asm/processor-cyrix.h b/arch/x86/include/asm/processor-cyrix.h index aaedd73ea2c6..df700a6cc869 100644 --- a/arch/x86/include/asm/processor-cyrix.h +++ b/arch/x86/include/asm/processor-cyrix.h @@ -3,19 +3,6 @@ * NSC/Cyrix CPU indexed register access. Must be inlined instead of * macros to ensure correct access ordering * Access order is always 0x22 (=offset), 0x23 (=value) - * - * When using the old macros a line like - * setCx86(CX86_CCR2, getCx86(CX86_CCR2) | 0x88); - * gets expanded to: - * do { - * outb((CX86_CCR2), 0x22); - * outb((({ - * outb((CX86_CCR2), 0x22); - * inb(0x23); - * }) | 0x88), 0x23); - * } while (0); - * - * which in fact violates the access order (= 0x22, 0x22, 0x23, 0x23). */ static inline u8 getCx86(u8 reg) @@ -29,11 +16,3 @@ static inline void setCx86(u8 reg, u8 data) outb(reg, 0x22); outb(data, 0x23); } - -#define getCx86_old(reg) ({ outb((reg), 0x22); inb(0x23); }) - -#define setCx86_old(reg, data) do { \ - outb((reg), 0x22); \ - outb((data), 0x23); \ -} while (0) - -- cgit v1.2.3-58-ga151 From 4fe64a62e04cfb2dc1daab0d8f05d212aa014161 Mon Sep 17 00:00:00 2001 From: Valdis Kletnieks Date: Tue, 12 Mar 2019 03:47:53 -0400 Subject: x86/mm/pti: Make local symbols static With 'make C=2 W=1', sparse and gcc both complain: CHECK arch/x86/mm/pti.c arch/x86/mm/pti.c:84:3: warning: symbol 'pti_mode' was not declared. Should it be static? arch/x86/mm/pti.c:605:6: warning: symbol 'pti_set_kernel_image_nonglobal' was not declared. Should it be static? CC arch/x86/mm/pti.o arch/x86/mm/pti.c:605:6: warning: no previous prototype for 'pti_set_kernel_image_nonglobal' [-Wmissing-prototypes] 605 | void pti_set_kernel_image_nonglobal(void) | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ pti_set_kernel_image_nonglobal() is only used locally. 'pti_mode' exists in drivers/hwtracing/intel_th/pti.c as well, but it's a completely unrelated local (static) symbol. Make both static. Signed-off-by: Valdis Kletnieks Signed-off-by: Thomas Gleixner Cc: Dave Hansen Cc: Andy Lutomirski Cc: Peter Zijlstra Link: https://lkml.kernel.org/r/27680.1552376873@turing-police --- arch/x86/mm/pti.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch') diff --git a/arch/x86/mm/pti.c b/arch/x86/mm/pti.c index 4fee5c3003ed..139b28a01ce4 100644 --- a/arch/x86/mm/pti.c +++ b/arch/x86/mm/pti.c @@ -77,7 +77,7 @@ static void __init pti_print_if_secure(const char *reason) pr_info("%s\n", reason); } -enum pti_mode { +static enum pti_mode { PTI_AUTO = 0, PTI_FORCE_OFF, PTI_FORCE_ON @@ -602,7 +602,7 @@ static void pti_clone_kernel_text(void) set_memory_global(start, (end_global - start) >> PAGE_SHIFT); } -void pti_set_kernel_image_nonglobal(void) +static void pti_set_kernel_image_nonglobal(void) { /* * The identity map is created with PMDs, regardless of the -- cgit v1.2.3-58-ga151 From e898e69d6b9475bf123f99b3c5d1a67bb7cb2361 Mon Sep 17 00:00:00 2001 From: Nathan Chancellor Date: Thu, 7 Mar 2019 14:27:56 -0700 Subject: x86/hw_breakpoints: Make default case in hw_breakpoint_arch_parse() return an error When building with -Wsometimes-uninitialized, Clang warns: arch/x86/kernel/hw_breakpoint.c:355:2: warning: variable 'align' is used uninitialized whenever switch default is taken [-Wsometimes-uninitialized] The default cannot be reached because arch_build_bp_info() initializes hw->len to one of the specified cases. Nevertheless the warning is valid and returning -EINVAL makes sure that this cannot be broken by future modifications. Suggested-by: Nick Desaulniers Signed-off-by: Nathan Chancellor Signed-off-by: Thomas Gleixner Reviewed-by: Nick Desaulniers Cc: Borislav Petkov Cc: "H. Peter Anvin" Cc: clang-built-linux@googlegroups.com Link: https://github.com/ClangBuiltLinux/linux/issues/392 Link: https://lkml.kernel.org/r/20190307212756.4648-1-natechancellor@gmail.com --- arch/x86/kernel/hw_breakpoint.c | 1 + 1 file changed, 1 insertion(+) (limited to 'arch') diff --git a/arch/x86/kernel/hw_breakpoint.c b/arch/x86/kernel/hw_breakpoint.c index ff9bfd40429e..d73083021002 100644 --- a/arch/x86/kernel/hw_breakpoint.c +++ b/arch/x86/kernel/hw_breakpoint.c @@ -354,6 +354,7 @@ int hw_breakpoint_arch_parse(struct perf_event *bp, #endif default: WARN_ON_ONCE(1); + return -EINVAL; } /* -- cgit v1.2.3-58-ga151 From ffc8599aa9763f39f6736a79da4d1575e7006f9a Mon Sep 17 00:00:00 2001 From: Kairui Song Date: Fri, 8 Mar 2019 11:05:08 +0800 Subject: x86/gart: Exclude GART aperture from kcore On machines where the GART aperture is mapped over physical RAM, /proc/kcore contains the GART aperture range. Accessing the GART range via /proc/kcore results in a kernel crash. vmcore used to have the same issue, until it was fixed with commit 2a3e83c6f96c ("x86/gart: Exclude GART aperture from vmcore")', leveraging existing hook infrastructure in vmcore to let /proc/vmcore return zeroes when attempting to read the aperture region, and so it won't read from the actual memory. Apply the same workaround for kcore. First implement the same hook infrastructure for kcore, then reuse the hook functions introduced in the previous vmcore fix. Just with some minor adjustment, rename some functions for more general usage, and simplify the hook infrastructure a bit as there is no module usage yet. Suggested-by: Baoquan He Signed-off-by: Kairui Song Signed-off-by: Thomas Gleixner Reviewed-by: Jiri Bohac Acked-by: Baoquan He Cc: Borislav Petkov Cc: "H. Peter Anvin" Cc: Alexey Dobriyan Cc: Andrew Morton Cc: Omar Sandoval Cc: Dave Young Link: https://lkml.kernel.org/r/20190308030508.13548-1-kasong@redhat.com --- arch/x86/kernel/aperture_64.c | 20 +++++++++++++------- fs/proc/kcore.c | 27 +++++++++++++++++++++++++++ include/linux/kcore.h | 2 ++ 3 files changed, 42 insertions(+), 7 deletions(-) (limited to 'arch') diff --git a/arch/x86/kernel/aperture_64.c b/arch/x86/kernel/aperture_64.c index 58176b56354e..294ed4392a0e 100644 --- a/arch/x86/kernel/aperture_64.c +++ b/arch/x86/kernel/aperture_64.c @@ -14,6 +14,7 @@ #define pr_fmt(fmt) "AGP: " fmt #include +#include #include #include #include @@ -57,7 +58,7 @@ int fallback_aper_force __initdata; int fix_aperture __initdata = 1; -#ifdef CONFIG_PROC_VMCORE +#if defined(CONFIG_PROC_VMCORE) || defined(CONFIG_PROC_KCORE) /* * If the first kernel maps the aperture over e820 RAM, the kdump kernel will * use the same range because it will remain configured in the northbridge. @@ -66,20 +67,25 @@ int fix_aperture __initdata = 1; */ static unsigned long aperture_pfn_start, aperture_page_count; -static int gart_oldmem_pfn_is_ram(unsigned long pfn) +static int gart_mem_pfn_is_ram(unsigned long pfn) { return likely((pfn < aperture_pfn_start) || (pfn >= aperture_pfn_start + aperture_page_count)); } -static void exclude_from_vmcore(u64 aper_base, u32 aper_order) +static void __init exclude_from_core(u64 aper_base, u32 aper_order) { aperture_pfn_start = aper_base >> PAGE_SHIFT; aperture_page_count = (32 * 1024 * 1024) << aper_order >> PAGE_SHIFT; - WARN_ON(register_oldmem_pfn_is_ram(&gart_oldmem_pfn_is_ram)); +#ifdef CONFIG_PROC_VMCORE + WARN_ON(register_oldmem_pfn_is_ram(&gart_mem_pfn_is_ram)); +#endif +#ifdef CONFIG_PROC_KCORE + WARN_ON(register_mem_pfn_is_ram(&gart_mem_pfn_is_ram)); +#endif } #else -static void exclude_from_vmcore(u64 aper_base, u32 aper_order) +static void exclude_from_core(u64 aper_base, u32 aper_order) { } #endif @@ -474,7 +480,7 @@ out: * may have allocated the range over its e820 RAM * and fixed up the northbridge */ - exclude_from_vmcore(last_aper_base, last_aper_order); + exclude_from_core(last_aper_base, last_aper_order); return 1; } @@ -520,7 +526,7 @@ out: * overlap with the first kernel's memory. We can't access the * range through vmcore even though it should be part of the dump. */ - exclude_from_vmcore(aper_alloc, aper_order); + exclude_from_core(aper_alloc, aper_order); /* Fix up the north bridges */ for (i = 0; i < amd_nb_bus_dev_ranges[i].dev_limit; i++) { diff --git a/fs/proc/kcore.c b/fs/proc/kcore.c index bbcc185062bb..d29d869abec1 100644 --- a/fs/proc/kcore.c +++ b/fs/proc/kcore.c @@ -54,6 +54,28 @@ static LIST_HEAD(kclist_head); static DECLARE_RWSEM(kclist_lock); static int kcore_need_update = 1; +/* + * Returns > 0 for RAM pages, 0 for non-RAM pages, < 0 on error + * Same as oldmem_pfn_is_ram in vmcore + */ +static int (*mem_pfn_is_ram)(unsigned long pfn); + +int __init register_mem_pfn_is_ram(int (*fn)(unsigned long pfn)) +{ + if (mem_pfn_is_ram) + return -EBUSY; + mem_pfn_is_ram = fn; + return 0; +} + +static int pfn_is_ram(unsigned long pfn) +{ + if (mem_pfn_is_ram) + return mem_pfn_is_ram(pfn); + else + return 1; +} + /* This doesn't grab kclist_lock, so it should only be used at init time. */ void __init kclist_add(struct kcore_list *new, void *addr, size_t size, int type) @@ -465,6 +487,11 @@ read_kcore(struct file *file, char __user *buffer, size_t buflen, loff_t *fpos) goto out; } m = NULL; /* skip the list anchor */ + } else if (!pfn_is_ram(__pa(start) >> PAGE_SHIFT)) { + if (clear_user(buffer, tsz)) { + ret = -EFAULT; + goto out; + } } else if (m->type == KCORE_VMALLOC) { vread(buf, (char *)start, tsz); /* we have to zero-fill user buffer even if no read */ diff --git a/include/linux/kcore.h b/include/linux/kcore.h index 8c3f8c14eeaa..c843f4a9c512 100644 --- a/include/linux/kcore.h +++ b/include/linux/kcore.h @@ -44,6 +44,8 @@ void kclist_add_remap(struct kcore_list *m, void *addr, void *vaddr, size_t sz) m->vaddr = (unsigned long)vaddr; kclist_add(m, addr, sz, KCORE_REMAP); } + +extern int __init register_mem_pfn_is_ram(int (*fn)(unsigned long pfn)); #else static inline void kclist_add(struct kcore_list *new, void *addr, size_t size, int type) -- cgit v1.2.3-58-ga151