From 9b912485e0e74a74e042e4f2dd87f262e46fcdf1 Mon Sep 17 00:00:00 2001 From: Baoquan He Date: Mon, 25 Jun 2018 11:16:55 +0800 Subject: x86/boot/KASLR: Add two new functions for 1GB huge pages handling Introduce two new functions: parse_gb_huge_pages() and process_gb_huge_pages(), which handle a conflict between KASLR and huge pages of 1GB. These two functions will be used in the next patch: - parse_gb_huge_pages() is used to parse kernel command-line to get how many 1GB huge pages have been specified. A static global variable 'max_gb_huge_pages' is added to store the number. - process_gb_huge_pages() is used to skip as many 1GB huge pages as possible from the passed in memory region according to the specified number. Signed-off-by: Baoquan He Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: douly.fnst@cn.fujitsu.com Cc: fanc.fnst@cn.fujitsu.com Cc: indou.takao@jp.fujitsu.com Cc: keescook@chromium.org Cc: lcapitulino@redhat.com Cc: yasu.isimatu@gmail.com Link: http://lkml.kernel.org/r/20180625031656.12443-2-bhe@redhat.com Signed-off-by: Ingo Molnar --- arch/x86/boot/compressed/kaslr.c | 83 ++++++++++++++++++++++++++++++++++++++++ 1 file changed, 83 insertions(+) (limited to 'arch/x86/boot') diff --git a/arch/x86/boot/compressed/kaslr.c b/arch/x86/boot/compressed/kaslr.c index b87a7582853d..d97647b5ffb7 100644 --- a/arch/x86/boot/compressed/kaslr.c +++ b/arch/x86/boot/compressed/kaslr.c @@ -215,6 +215,35 @@ static void mem_avoid_memmap(char *str) memmap_too_large = true; } +/* Store the number of 1GB huge pages which users specified: */ +static unsigned long max_gb_huge_pages; + +static void parse_gb_huge_pages(char *param, char *val) +{ + static bool gbpage_sz; + char *p; + + if (!strcmp(param, "hugepagesz")) { + p = val; + if (memparse(p, &p) != PUD_SIZE) { + gbpage_sz = false; + return; + } + + if (gbpage_sz) + warn("Repeatedly set hugeTLB page size of 1G!\n"); + gbpage_sz = true; + return; + } + + if (!strcmp(param, "hugepages") && gbpage_sz) { + p = val; + max_gb_huge_pages = simple_strtoull(p, &p, 0); + return; + } +} + + static int handle_mem_memmap(void) { char *args = (char *)get_cmd_line_ptr(); @@ -466,6 +495,60 @@ static void store_slot_info(struct mem_vector *region, unsigned long image_size) } } +/* + * Skip as many 1GB huge pages as possible in the passed region + * according to the number which users specified: + */ +static void +process_gb_huge_pages(struct mem_vector *region, unsigned long image_size) +{ + unsigned long addr, size = 0; + struct mem_vector tmp; + int i = 0; + + if (!max_gb_huge_pages) { + store_slot_info(region, image_size); + return; + } + + addr = ALIGN(region->start, PUD_SIZE); + /* Did we raise the address above the passed in memory entry? */ + if (addr < region->start + region->size) + size = region->size - (addr - region->start); + + /* Check how many 1GB huge pages can be filtered out: */ + while (size > PUD_SIZE && max_gb_huge_pages) { + size -= PUD_SIZE; + max_gb_huge_pages--; + i++; + } + + /* No good 1GB huge pages found: */ + if (!i) { + store_slot_info(region, image_size); + return; + } + + /* + * Skip those 'i'*1GB good huge pages, and continue checking and + * processing the remaining head or tail part of the passed region + * if available. + */ + + if (addr >= region->start + image_size) { + tmp.start = region->start; + tmp.size = addr - region->start; + store_slot_info(&tmp, image_size); + } + + size = region->size - (addr - region->start) - i * PUD_SIZE; + if (size >= image_size) { + tmp.start = addr + i * PUD_SIZE; + tmp.size = size; + store_slot_info(&tmp, image_size); + } +} + static unsigned long slots_fetch_random(void) { unsigned long slot; -- cgit v1.2.3 From 747ff6265db4c2b77e8c7384f8054916a0c1eb39 Mon Sep 17 00:00:00 2001 From: Baoquan He Date: Mon, 25 Jun 2018 11:16:56 +0800 Subject: x86/boot/KASLR: Skip specified number of 1GB huge pages when doing physical randomization (KASLR) When KASLR is enabled then 1GB huge pages allocations might regress sporadically. To reproduce on a KVM guest with 4GB RAM: - add the following options to the kernel command-line: 'default_hugepagesz=1G hugepagesz=1G hugepages=1' - boot the guest and check number of 1GB pages reserved: # grep HugePages_Total /proc/meminfo - sporadically, every couple of bootups the output of this command shows that when booting with "nokaslr" HugePages_Total is always 1, while booting without "nokaslr" sometimes HugePages_Total is set as 0 (that is, reserving the 1GB page failed). Note that you may need to boot a few times to trigger the issue, because it's somewhat non-deterministic. The root cause is that kernel may be put into the only good 1GB huge page in the [0x40000000, 0x7fffffff] physical range randomly. Below is the dmesg output snippet from the KVM guest. We can see that only [0x40000000, 0x7fffffff] region is good 1GB huge page, [0x100000000, 0x13fffffff] will be touched by the memblock top-down allocation: [...] e820: BIOS-provided physical RAM map: [...] BIOS-e820: [mem 0x0000000000000000-0x000000000009fbff] usable [...] BIOS-e820: [mem 0x000000000009fc00-0x000000000009ffff] reserved [...] BIOS-e820: [mem 0x00000000000f0000-0x00000000000fffff] reserved [...] BIOS-e820: [mem 0x0000000000100000-0x00000000bffdffff] usable [...] BIOS-e820: [mem 0x00000000bffe0000-0x00000000bfffffff] reserved [...] BIOS-e820: [mem 0x00000000feffc000-0x00000000feffffff] reserved [...] BIOS-e820: [mem 0x00000000fffc0000-0x00000000ffffffff] reserved [...] BIOS-e820: [mem 0x0000000100000000-0x000000013fffffff] usable Besides, on bare-metal machines with larger memory, one less 1GB huge page might be available with KASLR enabled. That too is because the kernel image might be randomized into those "good" 1GB huge pages. To fix this, firstly parse the kernel command-line to get how many 1GB huge pages are specified. Then try to skip the specified number of 1GB huge pages when decide which memory region kernel can be randomized into. Also change the name of handle_mem_memmap() as handle_mem_options() since it handles not only 'mem=' and 'memmap=', but also 'hugepagesxxx' now. Signed-off-by: Baoquan He Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: douly.fnst@cn.fujitsu.com Cc: fanc.fnst@cn.fujitsu.com Cc: indou.takao@jp.fujitsu.com Cc: keescook@chromium.org Cc: lcapitulino@redhat.com Cc: yasu.isimatu@gmail.com Link: http://lkml.kernel.org/r/20180625031656.12443-3-bhe@redhat.com [ Rewrote the changelog, fixed style problems in the code. ] Signed-off-by: Ingo Molnar --- arch/x86/boot/compressed/kaslr.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) (limited to 'arch/x86/boot') diff --git a/arch/x86/boot/compressed/kaslr.c b/arch/x86/boot/compressed/kaslr.c index d97647b5ffb7..531c9876f573 100644 --- a/arch/x86/boot/compressed/kaslr.c +++ b/arch/x86/boot/compressed/kaslr.c @@ -244,7 +244,7 @@ static void parse_gb_huge_pages(char *param, char *val) } -static int handle_mem_memmap(void) +static int handle_mem_options(void) { char *args = (char *)get_cmd_line_ptr(); size_t len = strlen((char *)args); @@ -252,7 +252,8 @@ static int handle_mem_memmap(void) char *param, *val; u64 mem_size; - if (!strstr(args, "memmap=") && !strstr(args, "mem=")) + if (!strstr(args, "memmap=") && !strstr(args, "mem=") && + !strstr(args, "hugepages")) return 0; tmp_cmdline = malloc(len + 1); @@ -277,6 +278,8 @@ static int handle_mem_memmap(void) if (!strcmp(param, "memmap")) { mem_avoid_memmap(val); + } else if (strstr(param, "hugepages")) { + parse_gb_huge_pages(param, val); } else if (!strcmp(param, "mem")) { char *p = val; @@ -416,7 +419,7 @@ static void mem_avoid_init(unsigned long input, unsigned long input_size, /* We don't need to set a mapping for setup_data. */ /* Mark the memmap regions we need to avoid */ - handle_mem_memmap(); + handle_mem_options(); #ifdef CONFIG_X86_VERBOSE_BOOTUP /* Make sure video RAM can be used. */ @@ -629,7 +632,7 @@ static void process_mem_region(struct mem_vector *entry, /* If nothing overlaps, store the region and return. */ if (!mem_avoid_overlap(®ion, &overlap)) { - store_slot_info(®ion, image_size); + process_gb_huge_pages(®ion, image_size); return; } @@ -639,7 +642,7 @@ static void process_mem_region(struct mem_vector *entry, beginning.start = region.start; beginning.size = overlap.start - region.start; - store_slot_info(&beginning, image_size); + process_gb_huge_pages(&beginning, image_size); } /* Return if overlap extends to or past end of region. */ -- cgit v1.2.3 From 5db1b1e1ee34871b1965b3f890e3ccbdb185fa52 Mon Sep 17 00:00:00 2001 From: zhong jiang Date: Mon, 30 Jul 2018 21:44:33 +0800 Subject: x86/boot/KASLR: Make local variable mem_limit static Fix the following sparse warning: arch/x86/boot/compressed/kaslr.c:102:20: warning: symbol 'mem_limit' was not declared. Should it be static? Signed-off-by: zhong jiang Signed-off-by: Thomas Gleixner Cc: Link: https://lkml.kernel.org/r/1532958273-47725-1-git-send-email-zhongjiang@huawei.com --- arch/x86/boot/compressed/kaslr.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86/boot') diff --git a/arch/x86/boot/compressed/kaslr.c b/arch/x86/boot/compressed/kaslr.c index 531c9876f573..302517929932 100644 --- a/arch/x86/boot/compressed/kaslr.c +++ b/arch/x86/boot/compressed/kaslr.c @@ -102,7 +102,7 @@ static bool memmap_too_large; /* Store memory limit specified by "mem=nn[KMG]" or "memmap=nn[KMG]" */ -unsigned long long mem_limit = ULLONG_MAX; +static unsigned long long mem_limit = ULLONG_MAX; enum mem_avoid_index { -- cgit v1.2.3 From 216a37202f10b7d78f2f98e26a6681f367165f05 Mon Sep 17 00:00:00 2001 From: Uros Bizjak Date: Fri, 29 Jun 2018 16:28:44 +0200 Subject: x86/boot: Use CC_SET()/CC_OUT() instead of open coding it Remove open-coded uses of set instructions with CC_SET()/CC_OUT(). Signed-off-by: Uros Bizjak Signed-off-by: Thomas Gleixner Link: https://lkml.kernel.org/r/20180629142844.15200-1-ubizjak@gmail.com --- arch/x86/boot/bitops.h | 3 ++- arch/x86/boot/string.c | 5 +++-- 2 files changed, 5 insertions(+), 3 deletions(-) (limited to 'arch/x86/boot') diff --git a/arch/x86/boot/bitops.h b/arch/x86/boot/bitops.h index 0d41d68131cc..2e1382486e91 100644 --- a/arch/x86/boot/bitops.h +++ b/arch/x86/boot/bitops.h @@ -17,6 +17,7 @@ #define _LINUX_BITOPS_H /* Inhibit inclusion of */ #include +#include static inline bool constant_test_bit(int nr, const void *addr) { @@ -28,7 +29,7 @@ static inline bool variable_test_bit(int nr, const void *addr) bool v; const u32 *p = (const u32 *)addr; - asm("btl %2,%1; setc %0" : "=qm" (v) : "m" (*p), "Ir" (nr)); + asm("btl %2,%1" CC_SET(c) : CC_OUT(c) (v) : "m" (*p), "Ir" (nr)); return v; } diff --git a/arch/x86/boot/string.c b/arch/x86/boot/string.c index 16f49123d747..c4428a176973 100644 --- a/arch/x86/boot/string.c +++ b/arch/x86/boot/string.c @@ -13,6 +13,7 @@ */ #include +#include #include "ctype.h" #include "string.h" @@ -28,8 +29,8 @@ int memcmp(const void *s1, const void *s2, size_t len) { bool diff; - asm("repe; cmpsb; setnz %0" - : "=qm" (diff), "+D" (s1), "+S" (s2), "+c" (len)); + asm("repe; cmpsb" CC_SET(nz) + : CC_OUT(nz) (diff), "+D" (s1), "+S" (s2), "+c" (len)); return diff; } -- cgit v1.2.3