April 2022 - Linux-stable-mirror

FAILED: patch "[PATCH] coredump: Use the vma snapshot in fill_files_note" failed to apply to 5.16-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.16-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 390031c942116d4733310f0684beb8db19885fe6 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" <ebiederm(a)xmission.com> Date: Tue, 8 Mar 2022 13:04:19 -0600 Subject: [PATCH] coredump: Use the vma snapshot in fill_files_note Matthew Wilcox reported that there is a missing mmap_lock in file_files_note that could possibly lead to a user after free. Solve this by using the existing vma snapshot for consistency and to avoid the need to take the mmap_lock anywhere in the coredump code except for dump_vma_snapshot. Update the dump_vma_snapshot to capture vm_pgoff and vm_file that are neeeded by fill_files_note. Add free_vma_snapshot to free the captured values of vm_file. Reported-by: Matthew Wilcox <willy(a)infradead.org> Link: https://lkml.kernel.org/r/20220131153740.2396974-1-willy@infradead.org Cc: stable(a)vger.kernel.org Fixes: a07279c9a8cd ("binfmt_elf, binfmt_elf_fdpic: use a VMA list snapshot") Fixes: 2aa362c49c31 ("coredump: extend core dump note section to contain file names of mapped files") Reviewed-by: Kees Cook <keescook(a)chromium.org> Signed-off-by: "Eric W. Biederman" <ebiederm(a)xmission.com> diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index 7f0c391832cf..ca5296cae979 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -1641,17 +1641,16 @@ static void fill_siginfo_note(struct memelfnote *note, user_siginfo_t *csigdata, * long file_ofs * followed by COUNT filenames in ASCII: "FILE1" NUL "FILE2" NUL... */ -static int fill_files_note(struct memelfnote *note) +static int fill_files_note(struct memelfnote *note, struct coredump_params *cprm) { - struct mm_struct *mm = current->mm; - struct vm_area_struct *vma; unsigned count, size, names_ofs, remaining, n; user_long_t *data; user_long_t *start_end_ofs; char *name_base, *name_curpos; + int i; /* *Estimated* file count and total data size needed */ - count = mm->map_count; + count = cprm->vma_count; if (count > UINT_MAX / 64) return -EINVAL; size = count * 64; @@ -1673,11 +1672,12 @@ static int fill_files_note(struct memelfnote *note) name_base = name_curpos = ((char *)data) + names_ofs; remaining = size - names_ofs; count = 0; - for (vma = mm->mmap; vma != NULL; vma = vma->vm_next) { + for (i = 0; i < cprm->vma_count; i++) { + struct core_vma_metadata *m = &cprm->vma_meta[i]; struct file *file; const char *filename; - file = vma->vm_file; + file = m->file; if (!file) continue; filename = file_path(file, name_curpos, remaining); @@ -1697,9 +1697,9 @@ static int fill_files_note(struct memelfnote *note) memmove(name_curpos, filename, n); name_curpos += n; - *start_end_ofs++ = vma->vm_start; - *start_end_ofs++ = vma->vm_end; - *start_end_ofs++ = vma->vm_pgoff; + *start_end_ofs++ = m->start; + *start_end_ofs++ = m->end; + *start_end_ofs++ = m->pgoff; count++; } @@ -1710,7 +1710,7 @@ static int fill_files_note(struct memelfnote *note) * Count usually is less than mm->map_count, * we need to move filenames down. */ - n = mm->map_count - count; + n = cprm->vma_count - count; if (n != 0) { unsigned shift_bytes = n * 3 * sizeof(data[0]); memmove(name_base - shift_bytes, name_base, @@ -1909,7 +1909,7 @@ static int fill_note_info(struct elfhdr *elf, int phdrs, fill_auxv_note(&info->auxv, current->mm); info->size += notesize(&info->auxv); - if (fill_files_note(&info->files) == 0) + if (fill_files_note(&info->files, cprm) == 0) info->size += notesize(&info->files); return 1; @@ -2098,7 +2098,7 @@ static int fill_note_info(struct elfhdr *elf, int phdrs, fill_auxv_note(info->notes + 3, current->mm); info->numnote = 4; - if (fill_files_note(info->notes + info->numnote) == 0) { + if (fill_files_note(info->notes + info->numnote, cprm) == 0) { info->notes_files = info->notes + info->numnote; info->numnote++; } diff --git a/fs/coredump.c b/fs/coredump.c index 7f100a637264..7ed7d601e5e0 100644 --- a/fs/coredump.c +++ b/fs/coredump.c @@ -55,6 +55,7 @@ #include <trace/events/sched.h> static bool dump_vma_snapshot(struct coredump_params *cprm); +static void free_vma_snapshot(struct coredump_params *cprm); static int core_uses_pid; static unsigned int core_pipe_limit; @@ -765,7 +766,7 @@ void do_coredump(const kernel_siginfo_t *siginfo) dump_emit(&cprm, "", 1); } file_end_write(cprm.file); - kvfree(cprm.vma_meta); + free_vma_snapshot(&cprm); } if (ispipe && core_pipe_limit) wait_for_dump_helpers(cprm.file); @@ -1099,6 +1100,20 @@ static struct vm_area_struct *next_vma(struct vm_area_struct *this_vma, return gate_vma; } +static void free_vma_snapshot(struct coredump_params *cprm) +{ + if (cprm->vma_meta) { + int i; + for (i = 0; i < cprm->vma_count; i++) { + struct file *file = cprm->vma_meta[i].file; + if (file) + fput(file); + } + kvfree(cprm->vma_meta); + cprm->vma_meta = NULL; + } +} + /* * Under the mmap_lock, take a snapshot of relevant information about the task's * VMAs. @@ -1135,6 +1150,11 @@ static bool dump_vma_snapshot(struct coredump_params *cprm) m->end = vma->vm_end; m->flags = vma->vm_flags; m->dump_size = vma_dump_size(vma, cprm->mm_flags); + m->pgoff = vma->vm_pgoff; + + m->file = vma->vm_file; + if (m->file) + get_file(m->file); } mmap_write_unlock(mm); diff --git a/include/linux/coredump.h b/include/linux/coredump.h index 7d05370e555e..08a1d3e7e46d 100644 --- a/include/linux/coredump.h +++ b/include/linux/coredump.h @@ -12,6 +12,8 @@ struct core_vma_metadata { unsigned long start, end; unsigned long flags; unsigned long dump_size; + unsigned long pgoff; + struct file *file; }; struct coredump_params {

2 years, 2 months

1
0
0 0

FAILED: patch "[PATCH] coredump: Use the vma snapshot in fill_files_note" failed to apply to 5.10-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.10-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 390031c942116d4733310f0684beb8db19885fe6 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" <ebiederm(a)xmission.com> Date: Tue, 8 Mar 2022 13:04:19 -0600 Subject: [PATCH] coredump: Use the vma snapshot in fill_files_note Matthew Wilcox reported that there is a missing mmap_lock in file_files_note that could possibly lead to a user after free. Solve this by using the existing vma snapshot for consistency and to avoid the need to take the mmap_lock anywhere in the coredump code except for dump_vma_snapshot. Update the dump_vma_snapshot to capture vm_pgoff and vm_file that are neeeded by fill_files_note. Add free_vma_snapshot to free the captured values of vm_file. Reported-by: Matthew Wilcox <willy(a)infradead.org> Link: https://lkml.kernel.org/r/20220131153740.2396974-1-willy@infradead.org Cc: stable(a)vger.kernel.org Fixes: a07279c9a8cd ("binfmt_elf, binfmt_elf_fdpic: use a VMA list snapshot") Fixes: 2aa362c49c31 ("coredump: extend core dump note section to contain file names of mapped files") Reviewed-by: Kees Cook <keescook(a)chromium.org> Signed-off-by: "Eric W. Biederman" <ebiederm(a)xmission.com> diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index 7f0c391832cf..ca5296cae979 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -1641,17 +1641,16 @@ static void fill_siginfo_note(struct memelfnote *note, user_siginfo_t *csigdata, * long file_ofs * followed by COUNT filenames in ASCII: "FILE1" NUL "FILE2" NUL... */ -static int fill_files_note(struct memelfnote *note) +static int fill_files_note(struct memelfnote *note, struct coredump_params *cprm) { - struct mm_struct *mm = current->mm; - struct vm_area_struct *vma; unsigned count, size, names_ofs, remaining, n; user_long_t *data; user_long_t *start_end_ofs; char *name_base, *name_curpos; + int i; /* *Estimated* file count and total data size needed */ - count = mm->map_count; + count = cprm->vma_count; if (count > UINT_MAX / 64) return -EINVAL; size = count * 64; @@ -1673,11 +1672,12 @@ static int fill_files_note(struct memelfnote *note) name_base = name_curpos = ((char *)data) + names_ofs; remaining = size - names_ofs; count = 0; - for (vma = mm->mmap; vma != NULL; vma = vma->vm_next) { + for (i = 0; i < cprm->vma_count; i++) { + struct core_vma_metadata *m = &cprm->vma_meta[i]; struct file *file; const char *filename; - file = vma->vm_file; + file = m->file; if (!file) continue; filename = file_path(file, name_curpos, remaining); @@ -1697,9 +1697,9 @@ static int fill_files_note(struct memelfnote *note) memmove(name_curpos, filename, n); name_curpos += n; - *start_end_ofs++ = vma->vm_start; - *start_end_ofs++ = vma->vm_end; - *start_end_ofs++ = vma->vm_pgoff; + *start_end_ofs++ = m->start; + *start_end_ofs++ = m->end; + *start_end_ofs++ = m->pgoff; count++; } @@ -1710,7 +1710,7 @@ static int fill_files_note(struct memelfnote *note) * Count usually is less than mm->map_count, * we need to move filenames down. */ - n = mm->map_count - count; + n = cprm->vma_count - count; if (n != 0) { unsigned shift_bytes = n * 3 * sizeof(data[0]); memmove(name_base - shift_bytes, name_base, @@ -1909,7 +1909,7 @@ static int fill_note_info(struct elfhdr *elf, int phdrs, fill_auxv_note(&info->auxv, current->mm); info->size += notesize(&info->auxv); - if (fill_files_note(&info->files) == 0) + if (fill_files_note(&info->files, cprm) == 0) info->size += notesize(&info->files); return 1; @@ -2098,7 +2098,7 @@ static int fill_note_info(struct elfhdr *elf, int phdrs, fill_auxv_note(info->notes + 3, current->mm); info->numnote = 4; - if (fill_files_note(info->notes + info->numnote) == 0) { + if (fill_files_note(info->notes + info->numnote, cprm) == 0) { info->notes_files = info->notes + info->numnote; info->numnote++; } diff --git a/fs/coredump.c b/fs/coredump.c index 7f100a637264..7ed7d601e5e0 100644 --- a/fs/coredump.c +++ b/fs/coredump.c @@ -55,6 +55,7 @@ #include <trace/events/sched.h> static bool dump_vma_snapshot(struct coredump_params *cprm); +static void free_vma_snapshot(struct coredump_params *cprm); static int core_uses_pid; static unsigned int core_pipe_limit; @@ -765,7 +766,7 @@ void do_coredump(const kernel_siginfo_t *siginfo) dump_emit(&cprm, "", 1); } file_end_write(cprm.file); - kvfree(cprm.vma_meta); + free_vma_snapshot(&cprm); } if (ispipe && core_pipe_limit) wait_for_dump_helpers(cprm.file); @@ -1099,6 +1100,20 @@ static struct vm_area_struct *next_vma(struct vm_area_struct *this_vma, return gate_vma; } +static void free_vma_snapshot(struct coredump_params *cprm) +{ + if (cprm->vma_meta) { + int i; + for (i = 0; i < cprm->vma_count; i++) { + struct file *file = cprm->vma_meta[i].file; + if (file) + fput(file); + } + kvfree(cprm->vma_meta); + cprm->vma_meta = NULL; + } +} + /* * Under the mmap_lock, take a snapshot of relevant information about the task's * VMAs. @@ -1135,6 +1150,11 @@ static bool dump_vma_snapshot(struct coredump_params *cprm) m->end = vma->vm_end; m->flags = vma->vm_flags; m->dump_size = vma_dump_size(vma, cprm->mm_flags); + m->pgoff = vma->vm_pgoff; + + m->file = vma->vm_file; + if (m->file) + get_file(m->file); } mmap_write_unlock(mm); diff --git a/include/linux/coredump.h b/include/linux/coredump.h index 7d05370e555e..08a1d3e7e46d 100644 --- a/include/linux/coredump.h +++ b/include/linux/coredump.h @@ -12,6 +12,8 @@ struct core_vma_metadata { unsigned long start, end; unsigned long flags; unsigned long dump_size; + unsigned long pgoff; + struct file *file; }; struct coredump_params {

2 years, 2 months

1
0
0 0

FAILED: patch "[PATCH] mmc: core: use sysfs_emit() instead of sprintf()" failed to apply to 5.10-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.10-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From f5d8a5fe77ce933f53eb8f2e22bb7a1a2019ea11 Mon Sep 17 00:00:00 2001 From: Sergey Shtylyov <s.shtylyov(a)omp.ru> Date: Tue, 8 Feb 2022 15:02:15 +0300 Subject: [PATCH] mmc: core: use sysfs_emit() instead of sprintf() sprintf() (still used in the MMC core for the sysfs output) is vulnerable to the buffer overflow. Use the new-fangled sysfs_emit() instead. Found by Linux Verification Center (linuxtesting.org) with the SVACE static analysis tool. Signed-off-by: Sergey Shtylyov <s.shtylyov(a)omp.ru> Cc: stable(a)vger.kernel.org Link: https://lore.kernel.org/r/717729b2-d65b-c72e-9fac-471d28d00b5a@omp.ru Signed-off-by: Ulf Hansson <ulf.hansson(a)linaro.org> diff --git a/drivers/mmc/core/bus.c b/drivers/mmc/core/bus.c index 096ae624be9a..58a60afa650b 100644 --- a/drivers/mmc/core/bus.c +++ b/drivers/mmc/core/bus.c @@ -15,6 +15,7 @@ #include <linux/stat.h> #include <linux/of.h> #include <linux/pm_runtime.h> +#include <linux/sysfs.h> #include <linux/mmc/card.h> #include <linux/mmc/host.h> @@ -34,13 +35,13 @@ static ssize_t type_show(struct device *dev, switch (card->type) { case MMC_TYPE_MMC: - return sprintf(buf, "MMC\n"); + return sysfs_emit(buf, "MMC\n"); case MMC_TYPE_SD: - return sprintf(buf, "SD\n"); + return sysfs_emit(buf, "SD\n"); case MMC_TYPE_SDIO: - return sprintf(buf, "SDIO\n"); + return sysfs_emit(buf, "SDIO\n"); case MMC_TYPE_SD_COMBO: - return sprintf(buf, "SDcombo\n"); + return sysfs_emit(buf, "SDcombo\n"); default: return -EFAULT; } diff --git a/drivers/mmc/core/bus.h b/drivers/mmc/core/bus.h index 8105852c4b62..3996b191b68d 100644 --- a/drivers/mmc/core/bus.h +++ b/drivers/mmc/core/bus.h @@ -9,6 +9,7 @@ #define _MMC_CORE_BUS_H #include <linux/device.h> +#include <linux/sysfs.h> struct mmc_host; struct mmc_card; @@ -17,7 +18,7 @@ struct mmc_card; static ssize_t mmc_##name##_show (struct device *dev, struct device_attribute *attr, char *buf) \ { \ struct mmc_card *card = mmc_dev_to_card(dev); \ - return sprintf(buf, fmt, args); \ + return sysfs_emit(buf, fmt, args); \ } \ static DEVICE_ATTR(name, S_IRUGO, mmc_##name##_show, NULL) diff --git a/drivers/mmc/core/mmc.c b/drivers/mmc/core/mmc.c index bbbbcaf70a59..13abfcd130a5 100644 --- a/drivers/mmc/core/mmc.c +++ b/drivers/mmc/core/mmc.c @@ -12,6 +12,7 @@ #include <linux/slab.h> #include <linux/stat.h> #include <linux/pm_runtime.h> +#include <linux/sysfs.h> #include <linux/mmc/host.h> #include <linux/mmc/card.h> @@ -812,12 +813,11 @@ static ssize_t mmc_fwrev_show(struct device *dev, { struct mmc_card *card = mmc_dev_to_card(dev); - if (card->ext_csd.rev < 7) { - return sprintf(buf, "0x%x\n", card->cid.fwrev); - } else { - return sprintf(buf, "0x%*phN\n", MMC_FIRMWARE_LEN, - card->ext_csd.fwrev); - } + if (card->ext_csd.rev < 7) + return sysfs_emit(buf, "0x%x\n", card->cid.fwrev); + else + return sysfs_emit(buf, "0x%*phN\n", MMC_FIRMWARE_LEN, + card->ext_csd.fwrev); } static DEVICE_ATTR(fwrev, S_IRUGO, mmc_fwrev_show, NULL); @@ -830,10 +830,10 @@ static ssize_t mmc_dsr_show(struct device *dev, struct mmc_host *host = card->host; if (card->csd.dsr_imp && host->dsr_req) - return sprintf(buf, "0x%x\n", host->dsr); + return sysfs_emit(buf, "0x%x\n", host->dsr); else /* return default DSR value */ - return sprintf(buf, "0x%x\n", 0x404); + return sysfs_emit(buf, "0x%x\n", 0x404); } static DEVICE_ATTR(dsr, S_IRUGO, mmc_dsr_show, NULL); diff --git a/drivers/mmc/core/sd.c b/drivers/mmc/core/sd.c index bd87012c220c..24b0418a24bb 100644 --- a/drivers/mmc/core/sd.c +++ b/drivers/mmc/core/sd.c @@ -13,6 +13,7 @@ #include <linux/stat.h> #include <linux/pm_runtime.h> #include <linux/scatterlist.h> +#include <linux/sysfs.h> #include <linux/mmc/host.h> #include <linux/mmc/card.h> @@ -708,18 +709,16 @@ MMC_DEV_ATTR(ocr, "0x%08x\n", card->ocr); MMC_DEV_ATTR(rca, "0x%04x\n", card->rca); -static ssize_t mmc_dsr_show(struct device *dev, - struct device_attribute *attr, - char *buf) +static ssize_t mmc_dsr_show(struct device *dev, struct device_attribute *attr, + char *buf) { - struct mmc_card *card = mmc_dev_to_card(dev); - struct mmc_host *host = card->host; - - if (card->csd.dsr_imp && host->dsr_req) - return sprintf(buf, "0x%x\n", host->dsr); - else - /* return default DSR value */ - return sprintf(buf, "0x%x\n", 0x404); + struct mmc_card *card = mmc_dev_to_card(dev); + struct mmc_host *host = card->host; + + if (card->csd.dsr_imp && host->dsr_req) + return sysfs_emit(buf, "0x%x\n", host->dsr); + /* return default DSR value */ + return sysfs_emit(buf, "0x%x\n", 0x404); } static DEVICE_ATTR(dsr, S_IRUGO, mmc_dsr_show, NULL); @@ -735,9 +734,9 @@ static ssize_t info##num##_show(struct device *dev, struct device_attribute *att \ if (num > card->num_info) \ return -ENODATA; \ - if (!card->info[num-1][0]) \ + if (!card->info[num - 1][0]) \ return 0; \ - return sprintf(buf, "%s\n", card->info[num-1]); \ + return sysfs_emit(buf, "%s\n", card->info[num - 1]); \ } \ static DEVICE_ATTR_RO(info##num) diff --git a/drivers/mmc/core/sdio.c b/drivers/mmc/core/sdio.c index 41164748723d..25799accf8a0 100644 --- a/drivers/mmc/core/sdio.c +++ b/drivers/mmc/core/sdio.c @@ -7,6 +7,7 @@ #include <linux/err.h> #include <linux/pm_runtime.h> +#include <linux/sysfs.h> #include <linux/mmc/host.h> #include <linux/mmc/card.h> @@ -40,9 +41,9 @@ static ssize_t info##num##_show(struct device *dev, struct device_attribute *att \ if (num > card->num_info) \ return -ENODATA; \ - if (!card->info[num-1][0]) \ + if (!card->info[num - 1][0]) \ return 0; \ - return sprintf(buf, "%s\n", card->info[num-1]); \ + return sysfs_emit(buf, "%s\n", card->info[num - 1]); \ } \ static DEVICE_ATTR_RO(info##num) diff --git a/drivers/mmc/core/sdio_bus.c b/drivers/mmc/core/sdio_bus.c index fda03b35c14a..c6268c38c69e 100644 --- a/drivers/mmc/core/sdio_bus.c +++ b/drivers/mmc/core/sdio_bus.c @@ -14,6 +14,7 @@ #include <linux/pm_runtime.h> #include <linux/pm_domain.h> #include <linux/acpi.h> +#include <linux/sysfs.h> #include <linux/mmc/card.h> #include <linux/mmc/host.h> @@ -35,7 +36,7 @@ field##_show(struct device *dev, struct device_attribute *attr, char *buf) \ struct sdio_func *func; \ \ func = dev_to_sdio_func (dev); \ - return sprintf(buf, format_string, args); \ + return sysfs_emit(buf, format_string, args); \ } \ static DEVICE_ATTR_RO(field) @@ -52,9 +53,9 @@ static ssize_t info##num##_show(struct device *dev, struct device_attribute *att \ if (num > func->num_info) \ return -ENODATA; \ - if (!func->info[num-1][0]) \ + if (!func->info[num - 1][0]) \ return 0; \ - return sprintf(buf, "%s\n", func->info[num-1]); \ + return sysfs_emit(buf, "%s\n", func->info[num - 1]); \ } \ static DEVICE_ATTR_RO(info##num)

2 years, 2 months

1
0
0 0

FAILED: patch "[PATCH] mmc: core: use sysfs_emit() instead of sprintf()" failed to apply to 5.15-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.15-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From f5d8a5fe77ce933f53eb8f2e22bb7a1a2019ea11 Mon Sep 17 00:00:00 2001 From: Sergey Shtylyov <s.shtylyov(a)omp.ru> Date: Tue, 8 Feb 2022 15:02:15 +0300 Subject: [PATCH] mmc: core: use sysfs_emit() instead of sprintf() sprintf() (still used in the MMC core for the sysfs output) is vulnerable to the buffer overflow. Use the new-fangled sysfs_emit() instead. Found by Linux Verification Center (linuxtesting.org) with the SVACE static analysis tool. Signed-off-by: Sergey Shtylyov <s.shtylyov(a)omp.ru> Cc: stable(a)vger.kernel.org Link: https://lore.kernel.org/r/717729b2-d65b-c72e-9fac-471d28d00b5a@omp.ru Signed-off-by: Ulf Hansson <ulf.hansson(a)linaro.org> diff --git a/drivers/mmc/core/bus.c b/drivers/mmc/core/bus.c index 096ae624be9a..58a60afa650b 100644 --- a/drivers/mmc/core/bus.c +++ b/drivers/mmc/core/bus.c @@ -15,6 +15,7 @@ #include <linux/stat.h> #include <linux/of.h> #include <linux/pm_runtime.h> +#include <linux/sysfs.h> #include <linux/mmc/card.h> #include <linux/mmc/host.h> @@ -34,13 +35,13 @@ static ssize_t type_show(struct device *dev, switch (card->type) { case MMC_TYPE_MMC: - return sprintf(buf, "MMC\n"); + return sysfs_emit(buf, "MMC\n"); case MMC_TYPE_SD: - return sprintf(buf, "SD\n"); + return sysfs_emit(buf, "SD\n"); case MMC_TYPE_SDIO: - return sprintf(buf, "SDIO\n"); + return sysfs_emit(buf, "SDIO\n"); case MMC_TYPE_SD_COMBO: - return sprintf(buf, "SDcombo\n"); + return sysfs_emit(buf, "SDcombo\n"); default: return -EFAULT; } diff --git a/drivers/mmc/core/bus.h b/drivers/mmc/core/bus.h index 8105852c4b62..3996b191b68d 100644 --- a/drivers/mmc/core/bus.h +++ b/drivers/mmc/core/bus.h @@ -9,6 +9,7 @@ #define _MMC_CORE_BUS_H #include <linux/device.h> +#include <linux/sysfs.h> struct mmc_host; struct mmc_card; @@ -17,7 +18,7 @@ struct mmc_card; static ssize_t mmc_##name##_show (struct device *dev, struct device_attribute *attr, char *buf) \ { \ struct mmc_card *card = mmc_dev_to_card(dev); \ - return sprintf(buf, fmt, args); \ + return sysfs_emit(buf, fmt, args); \ } \ static DEVICE_ATTR(name, S_IRUGO, mmc_##name##_show, NULL) diff --git a/drivers/mmc/core/mmc.c b/drivers/mmc/core/mmc.c index bbbbcaf70a59..13abfcd130a5 100644 --- a/drivers/mmc/core/mmc.c +++ b/drivers/mmc/core/mmc.c @@ -12,6 +12,7 @@ #include <linux/slab.h> #include <linux/stat.h> #include <linux/pm_runtime.h> +#include <linux/sysfs.h> #include <linux/mmc/host.h> #include <linux/mmc/card.h> @@ -812,12 +813,11 @@ static ssize_t mmc_fwrev_show(struct device *dev, { struct mmc_card *card = mmc_dev_to_card(dev); - if (card->ext_csd.rev < 7) { - return sprintf(buf, "0x%x\n", card->cid.fwrev); - } else { - return sprintf(buf, "0x%*phN\n", MMC_FIRMWARE_LEN, - card->ext_csd.fwrev); - } + if (card->ext_csd.rev < 7) + return sysfs_emit(buf, "0x%x\n", card->cid.fwrev); + else + return sysfs_emit(buf, "0x%*phN\n", MMC_FIRMWARE_LEN, + card->ext_csd.fwrev); } static DEVICE_ATTR(fwrev, S_IRUGO, mmc_fwrev_show, NULL); @@ -830,10 +830,10 @@ static ssize_t mmc_dsr_show(struct device *dev, struct mmc_host *host = card->host; if (card->csd.dsr_imp && host->dsr_req) - return sprintf(buf, "0x%x\n", host->dsr); + return sysfs_emit(buf, "0x%x\n", host->dsr); else /* return default DSR value */ - return sprintf(buf, "0x%x\n", 0x404); + return sysfs_emit(buf, "0x%x\n", 0x404); } static DEVICE_ATTR(dsr, S_IRUGO, mmc_dsr_show, NULL); diff --git a/drivers/mmc/core/sd.c b/drivers/mmc/core/sd.c index bd87012c220c..24b0418a24bb 100644 --- a/drivers/mmc/core/sd.c +++ b/drivers/mmc/core/sd.c @@ -13,6 +13,7 @@ #include <linux/stat.h> #include <linux/pm_runtime.h> #include <linux/scatterlist.h> +#include <linux/sysfs.h> #include <linux/mmc/host.h> #include <linux/mmc/card.h> @@ -708,18 +709,16 @@ MMC_DEV_ATTR(ocr, "0x%08x\n", card->ocr); MMC_DEV_ATTR(rca, "0x%04x\n", card->rca); -static ssize_t mmc_dsr_show(struct device *dev, - struct device_attribute *attr, - char *buf) +static ssize_t mmc_dsr_show(struct device *dev, struct device_attribute *attr, + char *buf) { - struct mmc_card *card = mmc_dev_to_card(dev); - struct mmc_host *host = card->host; - - if (card->csd.dsr_imp && host->dsr_req) - return sprintf(buf, "0x%x\n", host->dsr); - else - /* return default DSR value */ - return sprintf(buf, "0x%x\n", 0x404); + struct mmc_card *card = mmc_dev_to_card(dev); + struct mmc_host *host = card->host; + + if (card->csd.dsr_imp && host->dsr_req) + return sysfs_emit(buf, "0x%x\n", host->dsr); + /* return default DSR value */ + return sysfs_emit(buf, "0x%x\n", 0x404); } static DEVICE_ATTR(dsr, S_IRUGO, mmc_dsr_show, NULL); @@ -735,9 +734,9 @@ static ssize_t info##num##_show(struct device *dev, struct device_attribute *att \ if (num > card->num_info) \ return -ENODATA; \ - if (!card->info[num-1][0]) \ + if (!card->info[num - 1][0]) \ return 0; \ - return sprintf(buf, "%s\n", card->info[num-1]); \ + return sysfs_emit(buf, "%s\n", card->info[num - 1]); \ } \ static DEVICE_ATTR_RO(info##num) diff --git a/drivers/mmc/core/sdio.c b/drivers/mmc/core/sdio.c index 41164748723d..25799accf8a0 100644 --- a/drivers/mmc/core/sdio.c +++ b/drivers/mmc/core/sdio.c @@ -7,6 +7,7 @@ #include <linux/err.h> #include <linux/pm_runtime.h> +#include <linux/sysfs.h> #include <linux/mmc/host.h> #include <linux/mmc/card.h> @@ -40,9 +41,9 @@ static ssize_t info##num##_show(struct device *dev, struct device_attribute *att \ if (num > card->num_info) \ return -ENODATA; \ - if (!card->info[num-1][0]) \ + if (!card->info[num - 1][0]) \ return 0; \ - return sprintf(buf, "%s\n", card->info[num-1]); \ + return sysfs_emit(buf, "%s\n", card->info[num - 1]); \ } \ static DEVICE_ATTR_RO(info##num) diff --git a/drivers/mmc/core/sdio_bus.c b/drivers/mmc/core/sdio_bus.c index fda03b35c14a..c6268c38c69e 100644 --- a/drivers/mmc/core/sdio_bus.c +++ b/drivers/mmc/core/sdio_bus.c @@ -14,6 +14,7 @@ #include <linux/pm_runtime.h> #include <linux/pm_domain.h> #include <linux/acpi.h> +#include <linux/sysfs.h> #include <linux/mmc/card.h> #include <linux/mmc/host.h> @@ -35,7 +36,7 @@ field##_show(struct device *dev, struct device_attribute *attr, char *buf) \ struct sdio_func *func; \ \ func = dev_to_sdio_func (dev); \ - return sprintf(buf, format_string, args); \ + return sysfs_emit(buf, format_string, args); \ } \ static DEVICE_ATTR_RO(field) @@ -52,9 +53,9 @@ static ssize_t info##num##_show(struct device *dev, struct device_attribute *att \ if (num > func->num_info) \ return -ENODATA; \ - if (!func->info[num-1][0]) \ + if (!func->info[num - 1][0]) \ return 0; \ - return sprintf(buf, "%s\n", func->info[num-1]); \ + return sysfs_emit(buf, "%s\n", func->info[num - 1]); \ } \ static DEVICE_ATTR_RO(info##num)

2 years, 2 months

1
0
0 0

FAILED: patch "[PATCH] ASoC: SOF: Intel: Fix NULL ptr dereference when ENOMEM" failed to apply to 5.4-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.4-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From b7fb0ae09009d076964afe4c1a2bde1ee2bd88a9 Mon Sep 17 00:00:00 2001 From: Ammar Faizi <ammarfaizi2(a)gnuweeb.org> Date: Fri, 25 Feb 2022 01:58:36 +0700 Subject: [PATCH] ASoC: SOF: Intel: Fix NULL ptr dereference when ENOMEM Do not call snd_dma_free_pages() when snd_dma_alloc_pages() returns -ENOMEM because it leads to a NULL pointer dereference bug. The dmesg says: [ T1387] sof-audio-pci-intel-tgl 0000:00:1f.3: error: memory alloc failed: -12 [ T1387] BUG: kernel NULL pointer dereference, address: 0000000000000000 [ T1387] #PF: supervisor read access in kernel mode [ T1387] #PF: error_code(0x0000) - not-present page [ T1387] PGD 0 P4D 0 [ T1387] Oops: 0000 [#1] PREEMPT SMP NOPTI [ T1387] CPU: 6 PID: 1387 Comm: alsa-sink-HDA A Tainted: G W 5.17.0-rc4-superb-owl-00055-g80d47f5de5e3 [ T1387] Hardware name: HP HP Laptop 14s-dq2xxx/87FD, BIOS F.15 09/15/2021 [ T1387] RIP: 0010:dma_free_noncontiguous+0x37/0x80 [ T1387] Code: [... snip ...] [ T1387] RSP: 0000:ffffc90002b87770 EFLAGS: 00010246 [ T1387] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000 [ T1387] RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffff888101db30d0 [ T1387] RBP: 00000000fffffff4 R08: 0000000000000000 R09: 0000000000000000 [ T1387] R10: 0000000000000000 R11: ffffc90002b874d0 R12: 0000000000000001 [ T1387] R13: 0000000000058000 R14: ffff888105260c68 R15: ffff888105260828 [ T1387] FS: 00007f42e2ffd640(0000) GS:ffff888466b80000(0000) knlGS:0000000000000000 [ T1387] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ T1387] CR2: 0000000000000000 CR3: 000000014acf0003 CR4: 0000000000770ee0 [ T1387] PKRU: 55555554 [ T1387] Call Trace: [ T1387] <TASK> [ T1387] cl_stream_prepare+0x10a/0x120 [snd_sof_intel_hda_common 146addf995b9279ae7f509621078cccbe4f875e1] [... snip ...] [ T1387] </TASK> Cc: Daniel Baluta <daniel.baluta(a)nxp.com> Cc: Jaroslav Kysela <perex(a)perex.cz> Cc: Kai Vehmanen <kai.vehmanen(a)linux.intel.com> Cc: Keyon Jie <yang.jie(a)linux.intel.com> Cc: Liam Girdwood <lgirdwood(a)gmail.com> Cc: Mark Brown <broonie(a)kernel.org> Cc: Rander Wang <rander.wang(a)intel.com> Cc: Ranjani Sridharan <ranjani.sridharan(a)linux.intel.com> Cc: Takashi Iwai <tiwai(a)suse.com> Cc: sound-open-firmware(a)alsa-project.org Cc: alsa-devel(a)alsa-project.org Cc: linux-kernel(a)vger.kernel.org Cc: stable(a)vger.kernel.org # v5.2+ Fixes: d16046ffa6de040bf580a64d5f4d0aa18258a854 ("ASoC: SOF: Intel: Add Intel specific HDA firmware loader") Link: https://lore.kernel.org/lkml/20220224145124.15985-1-ammarfaizi2@gnuweeb.org/ # v1 Link: https://lore.kernel.org/lkml/20220224180850.34592-1-ammarfaizi2@gnuweeb.org/ # v2 Link: https://lore.kernel.org/lkml/20220224182818.40301-1-ammarfaizi2@gnuweeb.org/ # v3 Reviewed-by: Peter Ujfalusi <peter.ujfalusi(a)linux.intel.com> Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart(a)linux.intel.com> Signed-off-by: Ammar Faizi <ammarfaizi2(a)gnuweeb.org> Link: https://lore.kernel.org/r/20220224185836.44907-1-ammarfaizi2@gnuweeb.org Signed-off-by: Mark Brown <broonie(a)kernel.org> diff --git a/sound/soc/sof/intel/hda-loader.c b/sound/soc/sof/intel/hda-loader.c index 33306d2023a7..9bbfdab8009d 100644 --- a/sound/soc/sof/intel/hda-loader.c +++ b/sound/soc/sof/intel/hda-loader.c @@ -47,7 +47,7 @@ static struct hdac_ext_stream *cl_stream_prepare(struct snd_sof_dev *sdev, unsig ret = snd_dma_alloc_pages(SNDRV_DMA_TYPE_DEV_SG, &pci->dev, size, dmab); if (ret < 0) { dev_err(sdev->dev, "error: memory alloc failed: %d\n", ret); - goto error; + goto out_put; } hstream->period_bytes = 0;/* initialize period_bytes */ @@ -58,22 +58,23 @@ static struct hdac_ext_stream *cl_stream_prepare(struct snd_sof_dev *sdev, unsig ret = hda_dsp_iccmax_stream_hw_params(sdev, dsp_stream, dmab, NULL); if (ret < 0) { dev_err(sdev->dev, "error: iccmax stream prepare failed: %d\n", ret); - goto error; + goto out_free; } } else { ret = hda_dsp_stream_hw_params(sdev, dsp_stream, dmab, NULL); if (ret < 0) { dev_err(sdev->dev, "error: hdac prepare failed: %d\n", ret); - goto error; + goto out_free; } hda_dsp_stream_spib_config(sdev, dsp_stream, HDA_DSP_SPIB_ENABLE, size); } return dsp_stream; -error: - hda_dsp_stream_put(sdev, direction, hstream->stream_tag); +out_free: snd_dma_free_pages(dmab); +out_put: + hda_dsp_stream_put(sdev, direction, hstream->stream_tag); return ERR_PTR(ret); }

2 years, 2 months

1
0
0 0

FAILED: patch "[PATCH] arm64: Do not defer reserve_crashkernel() for platforms with" failed to apply to 5.4-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.4-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 031495635b4668f94e964e037ca93d0d38bfde58 Mon Sep 17 00:00:00 2001 From: Vijay Balakrishna <vijayb(a)linux.microsoft.com> Date: Wed, 2 Mar 2022 09:38:09 -0800 Subject: [PATCH] arm64: Do not defer reserve_crashkernel() for platforms with no DMA memory zones The following patches resulted in deferring crash kernel reservation to mem_init(), mainly aimed at platforms with DMA memory zones (no IOMMU), in particular Raspberry Pi 4. commit 1a8e1cef7603 ("arm64: use both ZONE_DMA and ZONE_DMA32") commit 8424ecdde7df ("arm64: mm: Set ZONE_DMA size based on devicetree's dma-ranges") commit 0a30c53573b0 ("arm64: mm: Move reserve_crashkernel() into mem_init()") commit 2687275a5843 ("arm64: Force NO_BLOCK_MAPPINGS if crashkernel reservation is required") Above changes introduced boot slowdown due to linear map creation for all the memory banks with NO_BLOCK_MAPPINGS, see discussion[1]. The proposed changes restore crash kernel reservation to earlier behavior thus avoids slow boot, particularly for platforms with IOMMU (no DMA memory zones). Tested changes to confirm no ~150ms boot slowdown on our SoC with IOMMU and 8GB memory. Also tested with ZONE_DMA and/or ZONE_DMA32 configs to confirm no regression to deferring scheme of crash kernel memory reservation. In both cases successfully collected kernel crash dump. [1] https://lore.kernel.org/all/9436d033-579b-55fa-9b00-6f4b661c2dd7@linux.micr… Signed-off-by: Vijay Balakrishna <vijayb(a)linux.microsoft.com> Cc: stable(a)vger.kernel.org Reviewed-by: Pasha Tatashin <pasha.tatashin(a)soleen.com> Link: https://lore.kernel.org/r/1646242689-20744-1-git-send-email-vijayb@linux.mi… [will: Add #ifdef CONFIG_KEXEC_CORE guards to fix 'crashk_res' references in allnoconfig build] Signed-off-by: Will Deacon <will(a)kernel.org> diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c index db63cc885771..919be440494f 100644 --- a/arch/arm64/mm/init.c +++ b/arch/arm64/mm/init.c @@ -61,8 +61,34 @@ EXPORT_SYMBOL(memstart_addr); * unless restricted on specific platforms (e.g. 30-bit on Raspberry Pi 4). * In such case, ZONE_DMA32 covers the rest of the 32-bit addressable memory, * otherwise it is empty. + * + * Memory reservation for crash kernel either done early or deferred + * depending on DMA memory zones configs (ZONE_DMA) -- + * + * In absence of ZONE_DMA configs arm64_dma_phys_limit initialized + * here instead of max_zone_phys(). This lets early reservation of + * crash kernel memory which has a dependency on arm64_dma_phys_limit. + * Reserving memory early for crash kernel allows linear creation of block + * mappings (greater than page-granularity) for all the memory bank rangs. + * In this scheme a comparatively quicker boot is observed. + * + * If ZONE_DMA configs are defined, crash kernel memory reservation + * is delayed until DMA zone memory range size initilazation performed in + * zone_sizes_init(). The defer is necessary to steer clear of DMA zone + * memory range to avoid overlap allocation. So crash kernel memory boundaries + * are not known when mapping all bank memory ranges, which otherwise means + * not possible to exclude crash kernel range from creating block mappings + * so page-granularity mappings are created for the entire memory range. + * Hence a slightly slower boot is observed. + * + * Note: Page-granularity mapppings are necessary for crash kernel memory + * range for shrinking its size via /sys/kernel/kexec_crash_size interface. */ -phys_addr_t arm64_dma_phys_limit __ro_after_init; +#if IS_ENABLED(CONFIG_ZONE_DMA) || IS_ENABLED(CONFIG_ZONE_DMA32) +phys_addr_t __ro_after_init arm64_dma_phys_limit; +#else +const phys_addr_t arm64_dma_phys_limit = PHYS_MASK + 1; +#endif #ifdef CONFIG_KEXEC_CORE /* @@ -153,8 +179,6 @@ static void __init zone_sizes_init(unsigned long min, unsigned long max) if (!arm64_dma_phys_limit) arm64_dma_phys_limit = dma32_phys_limit; #endif - if (!arm64_dma_phys_limit) - arm64_dma_phys_limit = PHYS_MASK + 1; max_zone_pfns[ZONE_NORMAL] = max; free_area_init(max_zone_pfns); @@ -315,6 +339,9 @@ void __init arm64_memblock_init(void) early_init_fdt_scan_reserved_mem(); + if (!IS_ENABLED(CONFIG_ZONE_DMA) && !IS_ENABLED(CONFIG_ZONE_DMA32)) + reserve_crashkernel(); + high_memory = __va(memblock_end_of_DRAM() - 1) + 1; } @@ -361,7 +388,8 @@ void __init bootmem_init(void) * request_standard_resources() depends on crashkernel's memory being * reserved, so do it here. */ - reserve_crashkernel(); + if (IS_ENABLED(CONFIG_ZONE_DMA) || IS_ENABLED(CONFIG_ZONE_DMA32)) + reserve_crashkernel(); memblock_dump_all(); } diff --git a/arch/arm64/mm/mmu.c b/arch/arm64/mm/mmu.c index acfae9b41cc8..ed21bf83d0b7 100644 --- a/arch/arm64/mm/mmu.c +++ b/arch/arm64/mm/mmu.c @@ -517,7 +517,7 @@ static void __init map_mem(pgd_t *pgdp) */ BUILD_BUG_ON(pgd_index(direct_map_end - 1) == pgd_index(direct_map_end)); - if (can_set_direct_map() || crash_mem_map || IS_ENABLED(CONFIG_KFENCE)) + if (can_set_direct_map() || IS_ENABLED(CONFIG_KFENCE)) flags |= NO_BLOCK_MAPPINGS | NO_CONT_MAPPINGS; /* @@ -528,6 +528,17 @@ static void __init map_mem(pgd_t *pgdp) */ memblock_mark_nomap(kernel_start, kernel_end - kernel_start); +#ifdef CONFIG_KEXEC_CORE + if (crash_mem_map) { + if (IS_ENABLED(CONFIG_ZONE_DMA) || + IS_ENABLED(CONFIG_ZONE_DMA32)) + flags |= NO_BLOCK_MAPPINGS | NO_CONT_MAPPINGS; + else if (crashk_res.end) + memblock_mark_nomap(crashk_res.start, + resource_size(&crashk_res)); + } +#endif + /* map all the memory banks */ for_each_mem_range(i, &start, &end) { if (start >= end) @@ -554,6 +565,25 @@ static void __init map_mem(pgd_t *pgdp) __map_memblock(pgdp, kernel_start, kernel_end, PAGE_KERNEL, NO_CONT_MAPPINGS); memblock_clear_nomap(kernel_start, kernel_end - kernel_start); + + /* + * Use page-level mappings here so that we can shrink the region + * in page granularity and put back unused memory to buddy system + * through /sys/kernel/kexec_crash_size interface. + */ +#ifdef CONFIG_KEXEC_CORE + if (crash_mem_map && + !IS_ENABLED(CONFIG_ZONE_DMA) && !IS_ENABLED(CONFIG_ZONE_DMA32)) { + if (crashk_res.end) { + __map_memblock(pgdp, crashk_res.start, + crashk_res.end + 1, + PAGE_KERNEL, + NO_BLOCK_MAPPINGS | NO_CONT_MAPPINGS); + memblock_clear_nomap(crashk_res.start, + resource_size(&crashk_res)); + } + } +#endif } void mark_rodata_ro(void)

2 years, 2 months

1
0
0 0

FAILED: patch "[PATCH] arm64: Do not defer reserve_crashkernel() for platforms with" failed to apply to 5.10-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.10-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 031495635b4668f94e964e037ca93d0d38bfde58 Mon Sep 17 00:00:00 2001 From: Vijay Balakrishna <vijayb(a)linux.microsoft.com> Date: Wed, 2 Mar 2022 09:38:09 -0800 Subject: [PATCH] arm64: Do not defer reserve_crashkernel() for platforms with no DMA memory zones The following patches resulted in deferring crash kernel reservation to mem_init(), mainly aimed at platforms with DMA memory zones (no IOMMU), in particular Raspberry Pi 4. commit 1a8e1cef7603 ("arm64: use both ZONE_DMA and ZONE_DMA32") commit 8424ecdde7df ("arm64: mm: Set ZONE_DMA size based on devicetree's dma-ranges") commit 0a30c53573b0 ("arm64: mm: Move reserve_crashkernel() into mem_init()") commit 2687275a5843 ("arm64: Force NO_BLOCK_MAPPINGS if crashkernel reservation is required") Above changes introduced boot slowdown due to linear map creation for all the memory banks with NO_BLOCK_MAPPINGS, see discussion[1]. The proposed changes restore crash kernel reservation to earlier behavior thus avoids slow boot, particularly for platforms with IOMMU (no DMA memory zones). Tested changes to confirm no ~150ms boot slowdown on our SoC with IOMMU and 8GB memory. Also tested with ZONE_DMA and/or ZONE_DMA32 configs to confirm no regression to deferring scheme of crash kernel memory reservation. In both cases successfully collected kernel crash dump. [1] https://lore.kernel.org/all/9436d033-579b-55fa-9b00-6f4b661c2dd7@linux.micr… Signed-off-by: Vijay Balakrishna <vijayb(a)linux.microsoft.com> Cc: stable(a)vger.kernel.org Reviewed-by: Pasha Tatashin <pasha.tatashin(a)soleen.com> Link: https://lore.kernel.org/r/1646242689-20744-1-git-send-email-vijayb@linux.mi… [will: Add #ifdef CONFIG_KEXEC_CORE guards to fix 'crashk_res' references in allnoconfig build] Signed-off-by: Will Deacon <will(a)kernel.org> diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c index db63cc885771..919be440494f 100644 --- a/arch/arm64/mm/init.c +++ b/arch/arm64/mm/init.c @@ -61,8 +61,34 @@ EXPORT_SYMBOL(memstart_addr); * unless restricted on specific platforms (e.g. 30-bit on Raspberry Pi 4). * In such case, ZONE_DMA32 covers the rest of the 32-bit addressable memory, * otherwise it is empty. + * + * Memory reservation for crash kernel either done early or deferred + * depending on DMA memory zones configs (ZONE_DMA) -- + * + * In absence of ZONE_DMA configs arm64_dma_phys_limit initialized + * here instead of max_zone_phys(). This lets early reservation of + * crash kernel memory which has a dependency on arm64_dma_phys_limit. + * Reserving memory early for crash kernel allows linear creation of block + * mappings (greater than page-granularity) for all the memory bank rangs. + * In this scheme a comparatively quicker boot is observed. + * + * If ZONE_DMA configs are defined, crash kernel memory reservation + * is delayed until DMA zone memory range size initilazation performed in + * zone_sizes_init(). The defer is necessary to steer clear of DMA zone + * memory range to avoid overlap allocation. So crash kernel memory boundaries + * are not known when mapping all bank memory ranges, which otherwise means + * not possible to exclude crash kernel range from creating block mappings + * so page-granularity mappings are created for the entire memory range. + * Hence a slightly slower boot is observed. + * + * Note: Page-granularity mapppings are necessary for crash kernel memory + * range for shrinking its size via /sys/kernel/kexec_crash_size interface. */ -phys_addr_t arm64_dma_phys_limit __ro_after_init; +#if IS_ENABLED(CONFIG_ZONE_DMA) || IS_ENABLED(CONFIG_ZONE_DMA32) +phys_addr_t __ro_after_init arm64_dma_phys_limit; +#else +const phys_addr_t arm64_dma_phys_limit = PHYS_MASK + 1; +#endif #ifdef CONFIG_KEXEC_CORE /* @@ -153,8 +179,6 @@ static void __init zone_sizes_init(unsigned long min, unsigned long max) if (!arm64_dma_phys_limit) arm64_dma_phys_limit = dma32_phys_limit; #endif - if (!arm64_dma_phys_limit) - arm64_dma_phys_limit = PHYS_MASK + 1; max_zone_pfns[ZONE_NORMAL] = max; free_area_init(max_zone_pfns); @@ -315,6 +339,9 @@ void __init arm64_memblock_init(void) early_init_fdt_scan_reserved_mem(); + if (!IS_ENABLED(CONFIG_ZONE_DMA) && !IS_ENABLED(CONFIG_ZONE_DMA32)) + reserve_crashkernel(); + high_memory = __va(memblock_end_of_DRAM() - 1) + 1; } @@ -361,7 +388,8 @@ void __init bootmem_init(void) * request_standard_resources() depends on crashkernel's memory being * reserved, so do it here. */ - reserve_crashkernel(); + if (IS_ENABLED(CONFIG_ZONE_DMA) || IS_ENABLED(CONFIG_ZONE_DMA32)) + reserve_crashkernel(); memblock_dump_all(); } diff --git a/arch/arm64/mm/mmu.c b/arch/arm64/mm/mmu.c index acfae9b41cc8..ed21bf83d0b7 100644 --- a/arch/arm64/mm/mmu.c +++ b/arch/arm64/mm/mmu.c @@ -517,7 +517,7 @@ static void __init map_mem(pgd_t *pgdp) */ BUILD_BUG_ON(pgd_index(direct_map_end - 1) == pgd_index(direct_map_end)); - if (can_set_direct_map() || crash_mem_map || IS_ENABLED(CONFIG_KFENCE)) + if (can_set_direct_map() || IS_ENABLED(CONFIG_KFENCE)) flags |= NO_BLOCK_MAPPINGS | NO_CONT_MAPPINGS; /* @@ -528,6 +528,17 @@ static void __init map_mem(pgd_t *pgdp) */ memblock_mark_nomap(kernel_start, kernel_end - kernel_start); +#ifdef CONFIG_KEXEC_CORE + if (crash_mem_map) { + if (IS_ENABLED(CONFIG_ZONE_DMA) || + IS_ENABLED(CONFIG_ZONE_DMA32)) + flags |= NO_BLOCK_MAPPINGS | NO_CONT_MAPPINGS; + else if (crashk_res.end) + memblock_mark_nomap(crashk_res.start, + resource_size(&crashk_res)); + } +#endif + /* map all the memory banks */ for_each_mem_range(i, &start, &end) { if (start >= end) @@ -554,6 +565,25 @@ static void __init map_mem(pgd_t *pgdp) __map_memblock(pgdp, kernel_start, kernel_end, PAGE_KERNEL, NO_CONT_MAPPINGS); memblock_clear_nomap(kernel_start, kernel_end - kernel_start); + + /* + * Use page-level mappings here so that we can shrink the region + * in page granularity and put back unused memory to buddy system + * through /sys/kernel/kexec_crash_size interface. + */ +#ifdef CONFIG_KEXEC_CORE + if (crash_mem_map && + !IS_ENABLED(CONFIG_ZONE_DMA) && !IS_ENABLED(CONFIG_ZONE_DMA32)) { + if (crashk_res.end) { + __map_memblock(pgdp, crashk_res.start, + crashk_res.end + 1, + PAGE_KERNEL, + NO_BLOCK_MAPPINGS | NO_CONT_MAPPINGS); + memblock_clear_nomap(crashk_res.start, + resource_size(&crashk_res)); + } + } +#endif } void mark_rodata_ro(void)

2 years, 2 months

1
0
0 0

[PATCH for v5.15+] PCI: fu740: Force 2.5GT/s for initial device probe

by Dimitri John Ledkov

From: Ben Dooks <ben.dooks(a)codethink.co.uk> commit a382c757ec5ef83137a86125f43a4c43dc2ab50b upstream. The fu740 PCIe core does not probe any devices on the SiFive Unmatched board without this fix (or having U-Boot explicitly start the PCIe via either boot-script or user command). The fix is to start the link at 2.5GT/s speeds and once the link is up then change the maximum speed back to the default. The U-Boot driver claims to set the link-speed to 2.5GT/s to get the probe to work (and U-Boot does print link up at 2.5GT/s) in the following code: https://source.denx.de/u-boot/u-boot/-/blob/master/drivers/pci/pcie_dw_sifi… Link: https://lore.kernel.org/r/20220318152430.526320-1-ben.dooks@codethink.co.uk Signed-off-by: Ben Dooks <ben.dooks(a)codethink.co.uk> Signed-off-by: Bjorn Helgaas <bhelgaas(a)google.com> Acked-by: Palmer Dabbelt <palmer(a)rivosinc.com> Signed-off-by: Dimitri John Ledkov <dimitri.ledkov(a)canonical.com> --- Please apply this patch to v5.15+ stable trees which fixes PCIe on the very popular SiFive Unmatched RISC-V board. drivers/pci/controller/dwc/pcie-fu740.c | 51 ++++++++++++++++++++++++- 1 file changed, 50 insertions(+), 1 deletion(-) diff --git a/drivers/pci/controller/dwc/pcie-fu740.c b/drivers/pci/controller/dwc/pcie-fu740.c index 00cde9a248b5..78d002be4f82 100644 --- a/drivers/pci/controller/dwc/pcie-fu740.c +++ b/drivers/pci/controller/dwc/pcie-fu740.c @@ -181,10 +181,59 @@ static int fu740_pcie_start_link(struct dw_pcie *pci) { struct device *dev = pci->dev; struct fu740_pcie *afp = dev_get_drvdata(dev); + u8 cap_exp = dw_pcie_find_capability(pci, PCI_CAP_ID_EXP); + int ret; + u32 orig, tmp; + + /* + * Force 2.5GT/s when starting the link, due to some devices not + * probing at higher speeds. This happens with the PCIe switch + * on the Unmatched board when U-Boot has not initialised the PCIe. + * The fix in U-Boot is to force 2.5GT/s, which then gets cleared + * by the soft reset done by this driver. + */ + dev_dbg(dev, "cap_exp at %x\n", cap_exp); + dw_pcie_dbi_ro_wr_en(pci); + + tmp = dw_pcie_readl_dbi(pci, cap_exp + PCI_EXP_LNKCAP); + orig = tmp & PCI_EXP_LNKCAP_SLS; + tmp &= ~PCI_EXP_LNKCAP_SLS; + tmp |= PCI_EXP_LNKCAP_SLS_2_5GB; + dw_pcie_writel_dbi(pci, cap_exp + PCI_EXP_LNKCAP, tmp); /* Enable LTSSM */ writel_relaxed(0x1, afp->mgmt_base + PCIEX8MGMT_APP_LTSSM_ENABLE); - return 0; + + ret = dw_pcie_wait_for_link(pci); + if (ret) { + dev_err(dev, "error: link did not start\n"); + goto err; + } + + tmp = dw_pcie_readl_dbi(pci, cap_exp + PCI_EXP_LNKCAP); + if ((tmp & PCI_EXP_LNKCAP_SLS) != orig) { + dev_dbg(dev, "changing speed back to original\n"); + + tmp &= ~PCI_EXP_LNKCAP_SLS; + tmp |= orig; + dw_pcie_writel_dbi(pci, cap_exp + PCI_EXP_LNKCAP, tmp); + + tmp = dw_pcie_readl_dbi(pci, PCIE_LINK_WIDTH_SPEED_CONTROL); + tmp |= PORT_LOGIC_SPEED_CHANGE; + dw_pcie_writel_dbi(pci, PCIE_LINK_WIDTH_SPEED_CONTROL, tmp); + + ret = dw_pcie_wait_for_link(pci); + if (ret) { + dev_err(dev, "error: link did not start at new speed\n"); + goto err; + } + } + + ret = 0; +err: + WARN_ON(ret); /* we assume that errors will be very rare */ + dw_pcie_dbi_ro_wr_dis(pci); + return ret; } static int fu740_pcie_host_init(struct pcie_port *pp) -- 2.32.0

2 years, 2 months

2
1
0 0

[PATCH stable v5.10-v5.17] can: isotp: sanitize CAN ID checks in isotp_bind()

by Oliver Hartkopp

From: Oliver Hartkopp <socketcan(a)hartkopp.net> commit 3ea566422cbde9610c2734980d1286ab681bb40e upstream. Syzbot created an environment that lead to a state machine status that can not be reached with a compliant CAN ID address configuration. The provided address information consisted of CAN ID 0x6000001 and 0xC28001 which both boil down to 11 bit CAN IDs 0x001 in sending and receiving. Sanitize the SFF/EFF CAN ID values before performing the address checks. Cc: stable(a)vger.kernel.org # v5.10 - v5.17 Fixes: e057dd3fc20f ("can: add ISO 15765-2:2016 transport protocol") Link: https://lore.kernel.org/all/20220316164258.54155-1-socketcan@hartkopp.net Reported-by: syzbot+2339c27f5c66c652843e(a)syzkaller.appspotmail.com Signed-off-by: Oliver Hartkopp <socketcan(a)hartkopp.net> Signed-off-by: Marc Kleine-Budde <mkl(a)pengutronix.de> --- net/can/isotp.c | 38 ++++++++++++++++++++------------------ 1 file changed, 20 insertions(+), 18 deletions(-) diff --git a/net/can/isotp.c b/net/can/isotp.c index d2a430b6a13b..f8e3aeb79e3f 100644 --- a/net/can/isotp.c +++ b/net/can/isotp.c @@ -1102,19 +1102,30 @@ static int isotp_bind(struct socket *sock, struct sockaddr *uaddr, int len) struct sock *sk = sock->sk; struct isotp_sock *so = isotp_sk(sk); struct net *net = sock_net(sk); int ifindex; struct net_device *dev; + canid_t tx_id, rx_id; int err = 0; int notify_enetdown = 0; int do_rx_reg = 1; if (len < ISOTP_MIN_NAMELEN) return -EINVAL; - if (addr->can_addr.tp.tx_id & (CAN_ERR_FLAG | CAN_RTR_FLAG)) - return -EADDRNOTAVAIL; + /* sanitize tx/rx CAN identifiers */ + tx_id = addr->can_addr.tp.tx_id; + if (tx_id & CAN_EFF_FLAG) + tx_id &= (CAN_EFF_FLAG | CAN_EFF_MASK); + else + tx_id &= CAN_SFF_MASK; + + rx_id = addr->can_addr.tp.rx_id; + if (rx_id & CAN_EFF_FLAG) + rx_id &= (CAN_EFF_FLAG | CAN_EFF_MASK); + else + rx_id &= CAN_SFF_MASK; if (!addr->can_ifindex) return -ENODEV; lock_sock(sk); @@ -1122,25 +1133,17 @@ static int isotp_bind(struct socket *sock, struct sockaddr *uaddr, int len) /* do not register frame reception for functional addressing */ if (so->opt.flags & CAN_ISOTP_SF_BROADCAST) do_rx_reg = 0; /* do not validate rx address for functional addressing */ - if (do_rx_reg) { - if (addr->can_addr.tp.rx_id == addr->can_addr.tp.tx_id) { - err = -EADDRNOTAVAIL; - goto out; - } - - if (addr->can_addr.tp.rx_id & (CAN_ERR_FLAG | CAN_RTR_FLAG)) { - err = -EADDRNOTAVAIL; - goto out; - } + if (do_rx_reg && rx_id == tx_id) { + err = -EADDRNOTAVAIL; + goto out; } if (so->bound && addr->can_ifindex == so->ifindex && - addr->can_addr.tp.rx_id == so->rxid && - addr->can_addr.tp.tx_id == so->txid) + rx_id == so->rxid && tx_id == so->txid) goto out; dev = dev_get_by_index(net, addr->can_ifindex); if (!dev) { err = -ENODEV; @@ -1160,12 +1163,11 @@ static int isotp_bind(struct socket *sock, struct sockaddr *uaddr, int len) notify_enetdown = 1; ifindex = dev->ifindex; if (do_rx_reg) - can_rx_register(net, dev, addr->can_addr.tp.rx_id, - SINGLE_MASK(addr->can_addr.tp.rx_id), + can_rx_register(net, dev, rx_id, SINGLE_MASK(rx_id), isotp_rcv, sk, "isotp", sk); dev_put(dev); if (so->bound && do_rx_reg) { @@ -1181,12 +1183,12 @@ static int isotp_bind(struct socket *sock, struct sockaddr *uaddr, int len) } } /* switch to new settings */ so->ifindex = ifindex; - so->rxid = addr->can_addr.tp.rx_id; - so->txid = addr->can_addr.tp.tx_id; + so->rxid = rx_id; + so->txid = tx_id; so->bound = 1; out: release_sock(sk); -- 2.30.2

2 years, 2 months

2
1
0 0

FAILED: patch "[PATCH] mm: fix race between MADV_FREE reclaim and blkdev direct IO" failed to apply to 4.19-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 4.19-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 6c8e2a256915a223f6289f651d6b926cd7135c9e Mon Sep 17 00:00:00 2001 From: Mauricio Faria de Oliveira <mfo(a)canonical.com> Date: Thu, 24 Mar 2022 18:14:09 -0700 Subject: [PATCH] mm: fix race between MADV_FREE reclaim and blkdev direct IO read Problem: ======= Userspace might read the zero-page instead of actual data from a direct IO read on a block device if the buffers have been called madvise(MADV_FREE) on earlier (this is discussed below) due to a race between page reclaim on MADV_FREE and blkdev direct IO read. - Race condition: ============== During page reclaim, the MADV_FREE page check in try_to_unmap_one() checks if the page is not dirty, then discards its rmap PTE(s) (vs. remap back if the page is dirty). However, after try_to_unmap_one() returns to shrink_page_list(), it might keep the page _anyway_ if page_ref_freeze() fails (it expects exactly _one_ page reference, from the isolation for page reclaim). Well, blkdev_direct_IO() gets references for all pages, and on READ operations it only sets them dirty _later_. So, if MADV_FREE'd pages (i.e., not dirty) are used as buffers for direct IO read from block devices, and page reclaim happens during __blkdev_direct_IO[_simple]() exactly AFTER bio_iov_iter_get_pages() returns, but BEFORE the pages are set dirty, the situation happens. The direct IO read eventually completes. Now, when userspace reads the buffers, the PTE is no longer there and the page fault handler do_anonymous_page() services that with the zero-page, NOT the data! A synthetic reproducer is provided. - Page faults: =========== If page reclaim happens BEFORE bio_iov_iter_get_pages() the issue doesn't happen, because that faults-in all pages as writeable, so do_anonymous_page() sets up a new page/rmap/PTE, and that is used by direct IO. The userspace reads don't fault as the PTE is there (thus zero-page is not used/setup). But if page reclaim happens AFTER it / BEFORE setting pages dirty, the PTE is no longer there; the subsequent page faults can't help: The data-read from the block device probably won't generate faults due to DMA (no MMU) but even in the case it wouldn't use DMA, that happens on different virtual addresses (not user-mapped addresses) because `struct bio_vec` stores `struct page` to figure addresses out (which are different from user-mapped addresses) for the read. Thus userspace reads (to user-mapped addresses) still fault, then do_anonymous_page() gets another `struct page` that would address/ map to other memory than the `struct page` used by `struct bio_vec` for the read. (The original `struct page` is not available, since it wasn't freed, as page_ref_freeze() failed due to more page refs. And even if it were available, its data cannot be trusted anymore.) Solution: ======== One solution is to check for the expected page reference count in try_to_unmap_one(). There should be one reference from the isolation (that is also checked in shrink_page_list() with page_ref_freeze()) plus one or more references from page mapping(s) (put in discard: label). Further references mean that rmap/PTE cannot be unmapped/nuked. (Note: there might be more than one reference from mapping due to fork()/clone() without CLONE_VM, which use the same `struct page` for references, until the copy-on-write page gets copied.) So, additional page references (e.g., from direct IO read) now prevent the rmap/PTE from being unmapped/dropped; similarly to the page is not freed per shrink_page_list()/page_ref_freeze()). - Races and Barriers: ================== The new check in try_to_unmap_one() should be safe in races with bio_iov_iter_get_pages() in get_user_pages() fast and slow paths, as it's done under the PTE lock. The fast path doesn't take the lock, but it checks if the PTE has changed and if so, it drops the reference and leaves the page for the slow path (which does take that lock). The fast path requires synchronization w/ full memory barrier: it writes the page reference count first then it reads the PTE later, while try_to_unmap() writes PTE first then it reads page refcount. And a second barrier is needed, as the page dirty flag should not be read before the page reference count (as in __remove_mapping()). (This can be a load memory barrier only; no writes are involved.) Call stack/comments: - try_to_unmap_one() - page_vma_mapped_walk() - map_pte() # see pte_offset_map_lock(): pte_offset_map() spin_lock() - ptep_get_and_clear() # write PTE - smp_mb() # (new barrier) GUP fast path - page_ref_count() # (new check) read refcount - page_vma_mapped_walk_done() # see pte_unmap_unlock(): pte_unmap() spin_unlock() - bio_iov_iter_get_pages() - __bio_iov_iter_get_pages() - iov_iter_get_pages() - get_user_pages_fast() - internal_get_user_pages_fast() # fast path - lockless_pages_from_mm() - gup_{pgd,p4d,pud,pmd,pte}_range() ptep = pte_offset_map() # not _lock() pte = ptep_get_lockless(ptep) page = pte_page(pte) try_grab_compound_head(page) # inc refcount # (RMW/barrier # on success) if (pte_val(pte) != pte_val(*ptep)) # read PTE put_compound_head(page) # dec refcount # go slow path # slow path - __gup_longterm_unlocked() - get_user_pages_unlocked() - __get_user_pages_locked() - __get_user_pages() - follow_{page,p4d,pud,pmd}_mask() - follow_page_pte() ptep = pte_offset_map_lock() pte = *ptep page = vm_normal_page(pte) try_grab_page(page) # inc refcount pte_unmap_unlock() - Huge Pages: ========== Regarding transparent hugepages, that logic shouldn't change, as MADV_FREE (aka lazyfree) pages are PageAnon() && !PageSwapBacked() (madvise_free_pte_range() -> mark_page_lazyfree() -> lru_lazyfree_fn()) thus should reach shrink_page_list() -> split_huge_page_to_list() before try_to_unmap[_one](), so it deals with normal pages only. (And in case unlikely/TTU_SPLIT_HUGE_PMD/split_huge_pmd_address() happens, which should not or be rare, the page refcount should be greater than mapcount: the head page is referenced by tail pages. That also prevents checking the head `page` then incorrectly call page_remove_rmap(subpage) for a tail page, that isn't even in the shrink_page_list()'s page_list (an effect of split huge pmd/pmvw), as it might happen today in this unlikely scenario.) MADV_FREE'd buffers: =================== So, back to the "if MADV_FREE pages are used as buffers" note. The case is arguable, and subject to multiple interpretations. The madvise(2) manual page on the MADV_FREE advice value says: 1) 'After a successful MADV_FREE ... data will be lost when the kernel frees the pages.' 2) 'the free operation will be canceled if the caller writes into the page' / 'subsequent writes ... will succeed and then [the] kernel cannot free those dirtied pages' 3) 'If there is no subsequent write, the kernel can free the pages at any time.' Thoughts, questions, considerations... respectively: 1) Since the kernel didn't actually free the page (page_ref_freeze() failed), should the data not have been lost? (on userspace read.) 2) Should writes performed by the direct IO read be able to cancel the free operation? - Should the direct IO read be considered as 'the caller' too, as it's been requested by 'the caller'? - Should the bio technique to dirty pages on return to userspace (bio_check_pages_dirty() is called/used by __blkdev_direct_IO()) be considered in another/special way here? 3) Should an upcoming write from a previously requested direct IO read be considered as a subsequent write, so the kernel should not free the pages? (as it's known at the time of page reclaim.) And lastly: Technically, the last point would seem a reasonable consideration and balance, as the madvise(2) manual page apparently (and fairly) seem to assume that 'writes' are memory access from the userspace process (not explicitly considering writes from the kernel or its corner cases; again, fairly).. plus the kernel fix implementation for the corner case of the largely 'non-atomic write' encompassed by a direct IO read operation, is relatively simple; and it helps. Reproducer: ========== @ test.c (simplified, but works) #define _GNU_SOURCE #include <fcntl.h> #include <stdio.h> #include <unistd.h> #include <sys/mman.h> int main() { int fd, i; char *buf; fd = open(DEV, O_RDONLY | O_DIRECT); buf = mmap(NULL, BUF_SIZE, PROT_READ | PROT_WRITE, MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); for (i = 0; i < BUF_SIZE; i += PAGE_SIZE) buf[i] = 1; // init to non-zero madvise(buf, BUF_SIZE, MADV_FREE); read(fd, buf, BUF_SIZE); for (i = 0; i < BUF_SIZE; i += PAGE_SIZE) printf("%p: 0x%x\n", &buf[i], buf[i]); return 0; } @ block/fops.c (formerly fs/block_dev.c) +#include <linux/swap.h> ... ... __blkdev_direct_IO[_simple](...) { ... + if (!strcmp(current->comm, "good")) + shrink_all_memory(ULONG_MAX); + ret = bio_iov_iter_get_pages(...); + + if (!strcmp(current->comm, "bad")) + shrink_all_memory(ULONG_MAX); ... } @ shell # NUM_PAGES=4 # PAGE_SIZE=$(getconf PAGE_SIZE) # yes | dd of=test.img bs=${PAGE_SIZE} count=${NUM_PAGES} # DEV=$(losetup -f --show test.img) # gcc -DDEV=\"$DEV\" \ -DBUF_SIZE=$((PAGE_SIZE * NUM_PAGES)) \ -DPAGE_SIZE=${PAGE_SIZE} \ test.c -o test # od -tx1 $DEV 0000000 79 0a 79 0a 79 0a 79 0a 79 0a 79 0a 79 0a 79 0a * 0040000 # mv test good # ./good 0x7f7c10418000: 0x79 0x7f7c10419000: 0x79 0x7f7c1041a000: 0x79 0x7f7c1041b000: 0x79 # mv good bad # ./bad 0x7fa1b8050000: 0x0 0x7fa1b8051000: 0x0 0x7fa1b8052000: 0x0 0x7fa1b8053000: 0x0 Note: the issue is consistent on v5.17-rc3, but it's intermittent with the support of MADV_FREE on v4.5 (60%-70% error; needs swap). [wrap do_direct_IO() in do_blockdev_direct_IO() @ fs/direct-io.c]. - v5.17-rc3: # for i in {1..1000}; do ./good; done \ | cut -d: -f2 | sort | uniq -c 4000 0x79 # mv good bad # for i in {1..1000}; do ./bad; done \ | cut -d: -f2 | sort | uniq -c 4000 0x0 # free | grep Swap Swap: 0 0 0 - v4.5: # for i in {1..1000}; do ./good; done \ | cut -d: -f2 | sort | uniq -c 4000 0x79 # mv good bad # for i in {1..1000}; do ./bad; done \ | cut -d: -f2 | sort | uniq -c 2702 0x0 1298 0x79 # swapoff -av swapoff /swap # for i in {1..1000}; do ./bad; done \ | cut -d: -f2 | sort | uniq -c 4000 0x79 Ceph/TCMalloc: ============= For documentation purposes, the use case driving the analysis/fix is Ceph on Ubuntu 18.04, as the TCMalloc library there still uses MADV_FREE to release unused memory to the system from the mmap'ed page heap (might be committed back/used again; it's not munmap'ed.) - PageHeap::DecommitSpan() -> TCMalloc_SystemRelease() -> madvise() - PageHeap::CommitSpan() -> TCMalloc_SystemCommit() -> do nothing. Note: TCMalloc switched back to MADV_DONTNEED a few commits after the release in Ubuntu 18.04 (google-perftools/gperftools 2.5), so the issue just 'disappeared' on Ceph on later Ubuntu releases but is still present in the kernel, and can be hit by other use cases. The observed issue seems to be the old Ceph bug #22464 [1], where checksum mismatches are observed (and instrumentation with buffer dumps shows zero-pages read from mmap'ed/MADV_FREE'd page ranges). The issue in Ceph was reasonably deemed a kernel bug (comment #50) and mostly worked around with a retry mechanism, but other parts of Ceph could still hit that (rocksdb). Anyway, it's less likely to be hit again as TCMalloc switched out of MADV_FREE by default. (Some kernel versions/reports from the Ceph bug, and relation with the MADV_FREE introduction/changes; TCMalloc versions not checked.) - 4.4 good - 4.5 (madv_free: introduction) - 4.9 bad - 4.10 good? maybe a swapless system - 4.12 (madv_free: no longer free instantly on swapless systems) - 4.13 bad [1] https://tracker.ceph.com/issues/22464 Thanks: ====== Several people contributed to analysis/discussions/tests/reproducers in the first stages when drilling down on ceph/tcmalloc/linux kernel: - Dan Hill - Dan Streetman - Dongdong Tao - Gavin Guo - Gerald Yang - Heitor Alves de Siqueira - Ioanna Alifieraki - Jay Vosburgh - Matthew Ruffell - Ponnuvel Palaniyappan Reviews, suggestions, corrections, comments: - Minchan Kim - Yu Zhao - Huang, Ying - John Hubbard - Christoph Hellwig [mfo(a)canonical.com: v4] Link: https://lkml.kernel.org/r/20220209202659.183418-1-mfo@canonical.comLink: https://lkml.kernel.org/r/20220131230255.789059-1-mfo@canonical.com Fixes: 802a3a92ad7a ("mm: reclaim MADV_FREE pages") Signed-off-by: Mauricio Faria de Oliveira <mfo(a)canonical.com> Reviewed-by: "Huang, Ying" <ying.huang(a)intel.com> Cc: Minchan Kim <minchan(a)kernel.org> Cc: Yu Zhao <yuzhao(a)google.com> Cc: Yang Shi <shy828301(a)gmail.com> Cc: Miaohe Lin <linmiaohe(a)huawei.com> Cc: Dan Hill <daniel.hill(a)canonical.com> Cc: Dan Streetman <dan.streetman(a)canonical.com> Cc: Dongdong Tao <dongdong.tao(a)canonical.com> Cc: Gavin Guo <gavin.guo(a)canonical.com> Cc: Gerald Yang <gerald.yang(a)canonical.com> Cc: Heitor Alves de Siqueira <halves(a)canonical.com> Cc: Ioanna Alifieraki <ioanna-maria.alifieraki(a)canonical.com> Cc: Jay Vosburgh <jay.vosburgh(a)canonical.com> Cc: Matthew Ruffell <matthew.ruffell(a)canonical.com> Cc: Ponnuvel Palaniyappan <ponnuvel.palaniyappan(a)canonical.com> Cc: <stable(a)vger.kernel.org> Cc: Christoph Hellwig <hch(a)infradead.org> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds(a)linux-foundation.org> diff --git a/mm/rmap.c b/mm/rmap.c index bfcc8e3d412f..5cb970d51f0a 100644 --- a/mm/rmap.c +++ b/mm/rmap.c @@ -1588,7 +1588,30 @@ static bool try_to_unmap_one(struct folio *folio, struct vm_area_struct *vma, /* MADV_FREE page check */ if (!folio_test_swapbacked(folio)) { - if (!folio_test_dirty(folio)) { + int ref_count, map_count; + + /* + * Synchronize with gup_pte_range(): + * - clear PTE; barrier; read refcount + * - inc refcount; barrier; read PTE + */ + smp_mb(); + + ref_count = folio_ref_count(folio); + map_count = folio_mapcount(folio); + + /* + * Order reads for page refcount and dirty flag + * (see comments in __remove_mapping()). + */ + smp_rmb(); + + /* + * The only page refs must be one from isolation + * plus the rmap(s) (dropped by discard:). + */ + if (ref_count == 1 + map_count && + !folio_test_dirty(folio)) { /* Invalidate as we cleared the pte */ mmu_notifier_invalidate_range(mm, address, address + PAGE_SIZE);

2 years, 2 months

1
0
0 0

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror April 2022