- Kernel - mailweb.openeuler.org

[PATCH OLK-6.6 0/1] kvm : add coda feature
by l00500167 18 Jul '24

18 Jul '24

From: JunBin Li <lijunbin4(a)huawei.com> virtCCA feature JunBin Li (1): kvm: add coda feature Signed-off-by: JunBin Li <lijunbin4(a)huawei.com> drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h | 14 ++++++++++++++ 1 file changed, 14 insertions(+) -- 2.25.1

2 2

[PATCH OLK-5.10] mips: bmips: BCM6358: make sure CBR is correctly set
by Zhang Changzhong 18 Jul '24

18 Jul '24

From: Christian Marangi <ansuelsmth(a)gmail.com> stable inclusion from stable-v5.10.221 commit 36d771ce6028b886e18a4a8956a5d23688e4e13d category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/IACS75 CVE: CVE-2024-40963 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit ce5cdd3b05216b704a704f466fb4c2dff3778caf ] It was discovered that some device have CBR address set to 0 causing kernel panic when arch_sync_dma_for_cpu_all is called. This was notice in situation where the system is booted from TP1 and BMIPS_GET_CBR() returns 0 instead of a valid address and !!(read_c0_brcm_cmt_local() & (1 << 31)); not failing. The current check whether RAC flush should be disabled or not are not enough hence lets check if CBR is a valid address or not. Fixes: ab327f8acdf8 ("mips: bmips: BCM6358: disable RAC flush for TP1") Signed-off-by: Christian Marangi <ansuelsmth(a)gmail.com> Acked-by: Florian Fainelli <florian.fainelli(a)broadcom.com> Signed-off-by: Thomas Bogendoerfer <tsbogend(a)alpha.franken.de> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Zhang Changzhong <zhangchangzhong(a)huawei.com> --- arch/mips/bmips/setup.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/mips/bmips/setup.c b/arch/mips/bmips/setup.c index 1606308..ac75797 100644 --- a/arch/mips/bmips/setup.c +++ b/arch/mips/bmips/setup.c @@ -110,7 +110,8 @@ static void bcm6358_quirks(void) * RAC flush causes kernel panics on BCM6358 when booting from TP1 * because the bootloader is not initializing it properly. */ - bmips_rac_flush_disable = !!(read_c0_brcm_cmt_local() & (1 << 31)); + bmips_rac_flush_disable = !!(read_c0_brcm_cmt_local() & (1 << 31)) || + !!BMIPS_GET_CBR(); } static void bcm6368_quirks(void) -- 2.9.5

2 1

[PATCH OLK-6.6] mips: bmips: BCM6358: make sure CBR is correctly set
by Zhang Changzhong 18 Jul '24

18 Jul '24

From: Christian Marangi <ansuelsmth(a)gmail.com> stable inclusion from stable-v6.6.36 commit 2cd4854ef14a487bcfb76c7980675980cad27b52 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/IACS75 CVE: CVE-2024-40963 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit ce5cdd3b05216b704a704f466fb4c2dff3778caf ] It was discovered that some device have CBR address set to 0 causing kernel panic when arch_sync_dma_for_cpu_all is called. This was notice in situation where the system is booted from TP1 and BMIPS_GET_CBR() returns 0 instead of a valid address and !!(read_c0_brcm_cmt_local() & (1 << 31)); not failing. The current check whether RAC flush should be disabled or not are not enough hence lets check if CBR is a valid address or not. Fixes: ab327f8acdf8 ("mips: bmips: BCM6358: disable RAC flush for TP1") Signed-off-by: Christian Marangi <ansuelsmth(a)gmail.com> Acked-by: Florian Fainelli <florian.fainelli(a)broadcom.com> Signed-off-by: Thomas Bogendoerfer <tsbogend(a)alpha.franken.de> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Zhang Changzhong <zhangchangzhong(a)huawei.com> --- arch/mips/bmips/setup.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/mips/bmips/setup.c b/arch/mips/bmips/setup.c index ec180ab..66a8ba1 100644 --- a/arch/mips/bmips/setup.c +++ b/arch/mips/bmips/setup.c @@ -110,7 +110,8 @@ static void bcm6358_quirks(void) * RAC flush causes kernel panics on BCM6358 when booting from TP1 * because the bootloader is not initializing it properly. */ - bmips_rac_flush_disable = !!(read_c0_brcm_cmt_local() & (1 << 31)); + bmips_rac_flush_disable = !!(read_c0_brcm_cmt_local() & (1 << 31)) || + !!BMIPS_GET_CBR(); } static void bcm6368_quirks(void) -- 2.9.5

2 1

[PATCH OLK-6.6] fs/9p: fix uninitialized values during inode evict
by Long Li 18 Jul '24

18 Jul '24

From: Eric Van Hensbergen <ericvh(a)kernel.org> mainline inclusion from mainline-v6.9-rc1 commit 6630036b7c228f57c7893ee0403e92c2db2cd21d category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I9U997 CVE: CVE-2024-36923 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- If an iget fails due to not being able to retrieve information from the server then the inode structure is only partially initialized. When the inode gets evicted, references to uninitialized structures (like fscache cookies) were being made. This patch checks for a bad_inode before doing anything other than clearing the inode from the cache. Since the inode is bad, it shouldn't have any state associated with it that needs to be written back (and there really isn't a way to complete those anyways). Reported-by: syzbot+eb83fe1cce5833cd66a0(a)syzkaller.appspotmail.com Signed-off-by: Eric Van Hensbergen <ericvh(a)kernel.org> Conflicts: fs/9p/vfs_inode.c [Context changed since commit 80105ed2fd27 ("9p: Use netfslib read/write_iter") in v9fs_evict_inode] Signed-off-by: Long Li <leo.lilong(a)huawei.com> --- fs/9p/vfs_inode.c | 18 +++++++++++------- 1 file changed, 11 insertions(+), 7 deletions(-) diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c index 853c63b83681..c7c5631fe825 100644 --- a/fs/9p/vfs_inode.c +++ b/fs/9p/vfs_inode.c @@ -374,20 +374,24 @@ void v9fs_evict_inode(struct inode *inode) struct v9fs_inode __maybe_unused *v9inode = V9FS_I(inode); __le32 __maybe_unused version; - truncate_inode_pages_final(&inode->i_data); + if (!is_bad_inode(inode)) { + truncate_inode_pages_final(&inode->i_data); #ifdef CONFIG_9P_FSCACHE - version = cpu_to_le32(v9inode->qid.version); - fscache_clear_inode_writeback(v9fs_inode_cookie(v9inode), inode, - &version); + version = cpu_to_le32(v9inode->qid.version); + fscache_clear_inode_writeback(v9fs_inode_cookie(v9inode), inode, + &version); #endif - clear_inode(inode); - filemap_fdatawrite(&inode->i_data); + clear_inode(inode); + filemap_fdatawrite(&inode->i_data); #ifdef CONFIG_9P_FSCACHE - fscache_relinquish_cookie(v9fs_inode_cookie(v9inode), false); + if (v9fs_inode_cookie(v9inode)) + fscache_relinquish_cookie(v9fs_inode_cookie(v9inode), false); #endif + } else + clear_inode(inode); } static int v9fs_test_inode(struct inode *inode, void *data) -- 2.39.2

2 1

[PATCH openEuler-1.0-LTS 0/2] CVE-2022-48836
by Yuan Can 18 Jul '24

18 Jul '24

Johan Hovold (1): Input: aiptek - fix endpoint sanity check Pavel Skripkin (1): Input: aiptek - properly check endpoint type drivers/input/tablet/aiptek.c | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) -- 2.17.1

2 3

[PATCH openEuler-22.03-LTS-SP1] bpf: Fix overrunning reservations in ringbuf
by Pu Lehui 18 Jul '24

18 Jul '24

From: Daniel Borkmann <daniel(a)iogearbox.net> mainline inclusion from mainline-v6.10-rc6 commit cfa1a2329a691ffd991fcf7248a57d752e712881 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/IADR59 CVE: CVE-2024-41009 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- The BPF ring buffer internally is implemented as a power-of-2 sized circular buffer, with two logical and ever-increasing counters: consumer_pos is the consumer counter to show which logical position the consumer consumed the data, and producer_pos which is the producer counter denoting the amount of data reserved by all producers. Each time a record is reserved, the producer that "owns" the record will successfully advance producer counter. In user space each time a record is read, the consumer of the data advanced the consumer counter once it finished processing. Both counters are stored in separate pages so that from user space, the producer counter is read-only and the consumer counter is read-write. One aspect that simplifies and thus speeds up the implementation of both producers and consumers is how the data area is mapped twice contiguously back-to-back in the virtual memory, allowing to not take any special measures for samples that have to wrap around at the end of the circular buffer data area, because the next page after the last data page would be first data page again, and thus the sample will still appear completely contiguous in virtual memory. Each record has a struct bpf_ringbuf_hdr { u32 len; u32 pg_off; } header for book-keeping the length and offset, and is inaccessible to the BPF program. Helpers like bpf_ringbuf_reserve() return `(void *)hdr + BPF_RINGBUF_HDR_SZ` for the BPF program to use. Bing-Jhong and Muhammad reported that it is however possible to make a second allocated memory chunk overlapping with the first chunk and as a result, the BPF program is now able to edit first chunk's header. For example, consider the creation of a BPF_MAP_TYPE_RINGBUF map with size of 0x4000. Next, the consumer_pos is modified to 0x3000 /before/ a call to bpf_ringbuf_reserve() is made. This will allocate a chunk A, which is in [0x0,0x3008], and the BPF program is able to edit [0x8,0x3008]. Now, lets allocate a chunk B with size 0x3000. This will succeed because consumer_pos was edited ahead of time to pass the `new_prod_pos - cons_pos > rb->mask` check. Chunk B will be in range [0x3008,0x6010], and the BPF program is able to edit [0x3010,0x6010]. Due to the ring buffer memory layout mentioned earlier, the ranges [0x0,0x4000] and [0x4000,0x8000] point to the same data pages. This means that chunk B at [0x4000,0x4008] is chunk A's header. bpf_ringbuf_submit() / bpf_ringbuf_discard() use the header's pg_off to then locate the bpf_ringbuf itself via bpf_ringbuf_restore_from_rec(). Once chunk B modified chunk A's header, then bpf_ringbuf_commit() refers to the wrong page and could cause a crash. Fix it by calculating the oldest pending_pos and check whether the range from the oldest outstanding record to the newest would span beyond the ring buffer size. If that is the case, then reject the request. We've tested with the ring buffer benchmark in BPF selftests (./benchs/run_bench_ringbufs.sh) before/after the fix and while it seems a bit slower on some benchmarks, it is still not significantly enough to matter. Fixes: 457f44363a88 ("bpf: Implement BPF ring buffer and verifier support for it") Reported-by: Bing-Jhong Billy Jheng <billy(a)starlabs.sg> Reported-by: Muhammad Ramdhan <ramdhan(a)starlabs.sg> Co-developed-by: Bing-Jhong Billy Jheng <billy(a)starlabs.sg> Co-developed-by: Andrii Nakryiko <andrii(a)kernel.org> Signed-off-by: Bing-Jhong Billy Jheng <billy(a)starlabs.sg> Signed-off-by: Andrii Nakryiko <andrii(a)kernel.org> Signed-off-by: Daniel Borkmann <daniel(a)iogearbox.net> Signed-off-by: Andrii Nakryiko <andrii(a)kernel.org> Link: https://lore.kernel.org/bpf/20240621140828.18238-1-daniel@iogearbox.net Conflicts: kernel/bpf/ringbuf.c [The Conflicts were due to not merge commit 583c1f420173f] Signed-off-by: Pu Lehui <pulehui(a)huawei.com> --- kernel/bpf/ringbuf.c | 28 +++++++++++++++++++++++----- 1 file changed, 23 insertions(+), 5 deletions(-) diff --git a/kernel/bpf/ringbuf.c b/kernel/bpf/ringbuf.c index d6fbe17432ae..d21d4ba2eb39 100644 --- a/kernel/bpf/ringbuf.c +++ b/kernel/bpf/ringbuf.c @@ -44,6 +44,7 @@ struct bpf_ringbuf { */ unsigned long consumer_pos __aligned(PAGE_SIZE); unsigned long producer_pos __aligned(PAGE_SIZE); + unsigned long pending_pos; char data[] __aligned(PAGE_SIZE); }; @@ -145,6 +146,7 @@ static struct bpf_ringbuf *bpf_ringbuf_alloc(size_t data_sz, int numa_node) rb->mask = data_sz - 1; rb->consumer_pos = 0; rb->producer_pos = 0; + rb->pending_pos = 0; return rb; } @@ -323,9 +325,9 @@ bpf_ringbuf_restore_from_rec(struct bpf_ringbuf_hdr *hdr) static void *__bpf_ringbuf_reserve(struct bpf_ringbuf *rb, u64 size) { - unsigned long cons_pos, prod_pos, new_prod_pos, flags; - u32 len, pg_off; + unsigned long cons_pos, prod_pos, new_prod_pos, pend_pos, flags; struct bpf_ringbuf_hdr *hdr; + u32 len, pg_off, tmp_size, hdr_len; if (unlikely(size > RINGBUF_MAX_RECORD_SZ)) return NULL; @@ -343,13 +345,29 @@ static void *__bpf_ringbuf_reserve(struct bpf_ringbuf *rb, u64 size) spin_lock_irqsave(&rb->spinlock, flags); } + pend_pos = rb->pending_pos; prod_pos = rb->producer_pos; new_prod_pos = prod_pos + len; - /* check for out of ringbuf space by ensuring producer position - * doesn't advance more than (ringbuf_size - 1) ahead + while (pend_pos < prod_pos) { + hdr = (void *)rb->data + (pend_pos & rb->mask); + hdr_len = READ_ONCE(hdr->len); + if (hdr_len & BPF_RINGBUF_BUSY_BIT) + break; + tmp_size = hdr_len & ~BPF_RINGBUF_DISCARD_BIT; + tmp_size = round_up(tmp_size + BPF_RINGBUF_HDR_SZ, 8); + pend_pos += tmp_size; + } + rb->pending_pos = pend_pos; + + /* check for out of ringbuf space: + * - by ensuring producer position doesn't advance more than + * (ringbuf_size - 1) ahead + * - by ensuring oldest not yet committed record until newest + * record does not span more than (ringbuf_size - 1) */ - if (new_prod_pos - cons_pos > rb->mask) { + if (new_prod_pos - cons_pos > rb->mask || + new_prod_pos - pend_pos > rb->mask) { spin_unlock_irqrestore(&rb->spinlock, flags); return NULL; } -- 2.34.1

2 1

[PATCH OLK-5.10] bpf: Fix overrunning reservations in ringbuf
by Pu Lehui 18 Jul '24

18 Jul '24

From: Daniel Borkmann <daniel(a)iogearbox.net> mainline inclusion from mainline-v6.10-rc6 commit cfa1a2329a691ffd991fcf7248a57d752e712881 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/IADR59 CVE: CVE-2024-41009 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- The BPF ring buffer internally is implemented as a power-of-2 sized circular buffer, with two logical and ever-increasing counters: consumer_pos is the consumer counter to show which logical position the consumer consumed the data, and producer_pos which is the producer counter denoting the amount of data reserved by all producers. Each time a record is reserved, the producer that "owns" the record will successfully advance producer counter. In user space each time a record is read, the consumer of the data advanced the consumer counter once it finished processing. Both counters are stored in separate pages so that from user space, the producer counter is read-only and the consumer counter is read-write. One aspect that simplifies and thus speeds up the implementation of both producers and consumers is how the data area is mapped twice contiguously back-to-back in the virtual memory, allowing to not take any special measures for samples that have to wrap around at the end of the circular buffer data area, because the next page after the last data page would be first data page again, and thus the sample will still appear completely contiguous in virtual memory. Each record has a struct bpf_ringbuf_hdr { u32 len; u32 pg_off; } header for book-keeping the length and offset, and is inaccessible to the BPF program. Helpers like bpf_ringbuf_reserve() return `(void *)hdr + BPF_RINGBUF_HDR_SZ` for the BPF program to use. Bing-Jhong and Muhammad reported that it is however possible to make a second allocated memory chunk overlapping with the first chunk and as a result, the BPF program is now able to edit first chunk's header. For example, consider the creation of a BPF_MAP_TYPE_RINGBUF map with size of 0x4000. Next, the consumer_pos is modified to 0x3000 /before/ a call to bpf_ringbuf_reserve() is made. This will allocate a chunk A, which is in [0x0,0x3008], and the BPF program is able to edit [0x8,0x3008]. Now, lets allocate a chunk B with size 0x3000. This will succeed because consumer_pos was edited ahead of time to pass the `new_prod_pos - cons_pos > rb->mask` check. Chunk B will be in range [0x3008,0x6010], and the BPF program is able to edit [0x3010,0x6010]. Due to the ring buffer memory layout mentioned earlier, the ranges [0x0,0x4000] and [0x4000,0x8000] point to the same data pages. This means that chunk B at [0x4000,0x4008] is chunk A's header. bpf_ringbuf_submit() / bpf_ringbuf_discard() use the header's pg_off to then locate the bpf_ringbuf itself via bpf_ringbuf_restore_from_rec(). Once chunk B modified chunk A's header, then bpf_ringbuf_commit() refers to the wrong page and could cause a crash. Fix it by calculating the oldest pending_pos and check whether the range from the oldest outstanding record to the newest would span beyond the ring buffer size. If that is the case, then reject the request. We've tested with the ring buffer benchmark in BPF selftests (./benchs/run_bench_ringbufs.sh) before/after the fix and while it seems a bit slower on some benchmarks, it is still not significantly enough to matter. Fixes: 457f44363a88 ("bpf: Implement BPF ring buffer and verifier support for it") Reported-by: Bing-Jhong Billy Jheng <billy(a)starlabs.sg> Reported-by: Muhammad Ramdhan <ramdhan(a)starlabs.sg> Co-developed-by: Bing-Jhong Billy Jheng <billy(a)starlabs.sg> Co-developed-by: Andrii Nakryiko <andrii(a)kernel.org> Signed-off-by: Bing-Jhong Billy Jheng <billy(a)starlabs.sg> Signed-off-by: Andrii Nakryiko <andrii(a)kernel.org> Signed-off-by: Daniel Borkmann <daniel(a)iogearbox.net> Signed-off-by: Andrii Nakryiko <andrii(a)kernel.org> Link: https://lore.kernel.org/bpf/20240621140828.18238-1-daniel@iogearbox.net Conflicts: kernel/bpf/ringbuf.c [The Conflicts were due to not merge commit 583c1f420173f] Signed-off-by: Pu Lehui <pulehui(a)huawei.com> --- kernel/bpf/ringbuf.c | 28 +++++++++++++++++++++++----- 1 file changed, 23 insertions(+), 5 deletions(-) diff --git a/kernel/bpf/ringbuf.c b/kernel/bpf/ringbuf.c index d6fbe17432ae..d21d4ba2eb39 100644 --- a/kernel/bpf/ringbuf.c +++ b/kernel/bpf/ringbuf.c @@ -44,6 +44,7 @@ struct bpf_ringbuf { */ unsigned long consumer_pos __aligned(PAGE_SIZE); unsigned long producer_pos __aligned(PAGE_SIZE); + unsigned long pending_pos; char data[] __aligned(PAGE_SIZE); }; @@ -145,6 +146,7 @@ static struct bpf_ringbuf *bpf_ringbuf_alloc(size_t data_sz, int numa_node) rb->mask = data_sz - 1; rb->consumer_pos = 0; rb->producer_pos = 0; + rb->pending_pos = 0; return rb; } @@ -323,9 +325,9 @@ bpf_ringbuf_restore_from_rec(struct bpf_ringbuf_hdr *hdr) static void *__bpf_ringbuf_reserve(struct bpf_ringbuf *rb, u64 size) { - unsigned long cons_pos, prod_pos, new_prod_pos, flags; - u32 len, pg_off; + unsigned long cons_pos, prod_pos, new_prod_pos, pend_pos, flags; struct bpf_ringbuf_hdr *hdr; + u32 len, pg_off, tmp_size, hdr_len; if (unlikely(size > RINGBUF_MAX_RECORD_SZ)) return NULL; @@ -343,13 +345,29 @@ static void *__bpf_ringbuf_reserve(struct bpf_ringbuf *rb, u64 size) spin_lock_irqsave(&rb->spinlock, flags); } + pend_pos = rb->pending_pos; prod_pos = rb->producer_pos; new_prod_pos = prod_pos + len; - /* check for out of ringbuf space by ensuring producer position - * doesn't advance more than (ringbuf_size - 1) ahead + while (pend_pos < prod_pos) { + hdr = (void *)rb->data + (pend_pos & rb->mask); + hdr_len = READ_ONCE(hdr->len); + if (hdr_len & BPF_RINGBUF_BUSY_BIT) + break; + tmp_size = hdr_len & ~BPF_RINGBUF_DISCARD_BIT; + tmp_size = round_up(tmp_size + BPF_RINGBUF_HDR_SZ, 8); + pend_pos += tmp_size; + } + rb->pending_pos = pend_pos; + + /* check for out of ringbuf space: + * - by ensuring producer position doesn't advance more than + * (ringbuf_size - 1) ahead + * - by ensuring oldest not yet committed record until newest + * record does not span more than (ringbuf_size - 1) */ - if (new_prod_pos - cons_pos > rb->mask) { + if (new_prod_pos - cons_pos > rb->mask || + new_prod_pos - pend_pos > rb->mask) { spin_unlock_irqrestore(&rb->spinlock, flags); return NULL; } -- 2.34.1

2 1

[PATCH OLK-6.6] bpf: Fix overrunning reservations in ringbuf
by Pu Lehui 18 Jul '24

18 Jul '24

From: Daniel Borkmann <daniel(a)iogearbox.net> stable inclusion from stable-v6.6.37 commit 511804ab701c0503b72eac08217eabfd366ba069 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/IADR59 CVE: CVE-2024-41009 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit cfa1a2329a691ffd991fcf7248a57d752e712881 ] The BPF ring buffer internally is implemented as a power-of-2 sized circular buffer, with two logical and ever-increasing counters: consumer_pos is the consumer counter to show which logical position the consumer consumed the data, and producer_pos which is the producer counter denoting the amount of data reserved by all producers. Each time a record is reserved, the producer that "owns" the record will successfully advance producer counter. In user space each time a record is read, the consumer of the data advanced the consumer counter once it finished processing. Both counters are stored in separate pages so that from user space, the producer counter is read-only and the consumer counter is read-write. One aspect that simplifies and thus speeds up the implementation of both producers and consumers is how the data area is mapped twice contiguously back-to-back in the virtual memory, allowing to not take any special measures for samples that have to wrap around at the end of the circular buffer data area, because the next page after the last data page would be first data page again, and thus the sample will still appear completely contiguous in virtual memory. Each record has a struct bpf_ringbuf_hdr { u32 len; u32 pg_off; } header for book-keeping the length and offset, and is inaccessible to the BPF program. Helpers like bpf_ringbuf_reserve() return `(void *)hdr + BPF_RINGBUF_HDR_SZ` for the BPF program to use. Bing-Jhong and Muhammad reported that it is however possible to make a second allocated memory chunk overlapping with the first chunk and as a result, the BPF program is now able to edit first chunk's header. For example, consider the creation of a BPF_MAP_TYPE_RINGBUF map with size of 0x4000. Next, the consumer_pos is modified to 0x3000 /before/ a call to bpf_ringbuf_reserve() is made. This will allocate a chunk A, which is in [0x0,0x3008], and the BPF program is able to edit [0x8,0x3008]. Now, lets allocate a chunk B with size 0x3000. This will succeed because consumer_pos was edited ahead of time to pass the `new_prod_pos - cons_pos > rb->mask` check. Chunk B will be in range [0x3008,0x6010], and the BPF program is able to edit [0x3010,0x6010]. Due to the ring buffer memory layout mentioned earlier, the ranges [0x0,0x4000] and [0x4000,0x8000] point to the same data pages. This means that chunk B at [0x4000,0x4008] is chunk A's header. bpf_ringbuf_submit() / bpf_ringbuf_discard() use the header's pg_off to then locate the bpf_ringbuf itself via bpf_ringbuf_restore_from_rec(). Once chunk B modified chunk A's header, then bpf_ringbuf_commit() refers to the wrong page and could cause a crash. Fix it by calculating the oldest pending_pos and check whether the range from the oldest outstanding record to the newest would span beyond the ring buffer size. If that is the case, then reject the request. We've tested with the ring buffer benchmark in BPF selftests (./benchs/run_bench_ringbufs.sh) before/after the fix and while it seems a bit slower on some benchmarks, it is still not significantly enough to matter. Fixes: 457f44363a88 ("bpf: Implement BPF ring buffer and verifier support for it") Reported-by: Bing-Jhong Billy Jheng <billy(a)starlabs.sg> Reported-by: Muhammad Ramdhan <ramdhan(a)starlabs.sg> Co-developed-by: Bing-Jhong Billy Jheng <billy(a)starlabs.sg> Co-developed-by: Andrii Nakryiko <andrii(a)kernel.org> Signed-off-by: Bing-Jhong Billy Jheng <billy(a)starlabs.sg> Signed-off-by: Andrii Nakryiko <andrii(a)kernel.org> Signed-off-by: Daniel Borkmann <daniel(a)iogearbox.net> Signed-off-by: Andrii Nakryiko <andrii(a)kernel.org> Link: https://lore.kernel.org/bpf/20240621140828.18238-1-daniel@iogearbox.net Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Pu Lehui <pulehui(a)huawei.com> --- kernel/bpf/ringbuf.c | 31 +++++++++++++++++++++++++------ 1 file changed, 25 insertions(+), 6 deletions(-) diff --git a/kernel/bpf/ringbuf.c b/kernel/bpf/ringbuf.c index f045fde632e5..238d9b206bbd 100644 --- a/kernel/bpf/ringbuf.c +++ b/kernel/bpf/ringbuf.c @@ -51,7 +51,8 @@ struct bpf_ringbuf { * This prevents a user-space application from modifying the * position and ruining in-kernel tracking. The permissions of the * pages depend on who is producing samples: user-space or the - * kernel. + * kernel. Note that the pending counter is placed in the same + * page as the producer, so that it shares the same cache line. * * Kernel-producer * --------------- @@ -70,6 +71,7 @@ struct bpf_ringbuf { */ unsigned long consumer_pos __aligned(PAGE_SIZE); unsigned long producer_pos __aligned(PAGE_SIZE); + unsigned long pending_pos; char data[] __aligned(PAGE_SIZE); }; @@ -179,6 +181,7 @@ static struct bpf_ringbuf *bpf_ringbuf_alloc(size_t data_sz, int numa_node) rb->mask = data_sz - 1; rb->consumer_pos = 0; rb->producer_pos = 0; + rb->pending_pos = 0; return rb; } @@ -404,9 +407,9 @@ bpf_ringbuf_restore_from_rec(struct bpf_ringbuf_hdr *hdr) static void *__bpf_ringbuf_reserve(struct bpf_ringbuf *rb, u64 size) { - unsigned long cons_pos, prod_pos, new_prod_pos, flags; - u32 len, pg_off; + unsigned long cons_pos, prod_pos, new_prod_pos, pend_pos, flags; struct bpf_ringbuf_hdr *hdr; + u32 len, pg_off, tmp_size, hdr_len; if (unlikely(size > RINGBUF_MAX_RECORD_SZ)) return NULL; @@ -424,13 +427,29 @@ static void *__bpf_ringbuf_reserve(struct bpf_ringbuf *rb, u64 size) spin_lock_irqsave(&rb->spinlock, flags); } + pend_pos = rb->pending_pos; prod_pos = rb->producer_pos; new_prod_pos = prod_pos + len; - /* check for out of ringbuf space by ensuring producer position - * doesn't advance more than (ringbuf_size - 1) ahead + while (pend_pos < prod_pos) { + hdr = (void *)rb->data + (pend_pos & rb->mask); + hdr_len = READ_ONCE(hdr->len); + if (hdr_len & BPF_RINGBUF_BUSY_BIT) + break; + tmp_size = hdr_len & ~BPF_RINGBUF_DISCARD_BIT; + tmp_size = round_up(tmp_size + BPF_RINGBUF_HDR_SZ, 8); + pend_pos += tmp_size; + } + rb->pending_pos = pend_pos; + + /* check for out of ringbuf space: + * - by ensuring producer position doesn't advance more than + * (ringbuf_size - 1) ahead + * - by ensuring oldest not yet committed record until newest + * record does not span more than (ringbuf_size - 1) */ - if (new_prod_pos - cons_pos > rb->mask) { + if (new_prod_pos - cons_pos > rb->mask || + new_prod_pos - pend_pos > rb->mask) { spin_unlock_irqrestore(&rb->spinlock, flags); return NULL; } -- 2.34.1

2 1

[openeuler:openEuler-1.0-LTS] BUILD SUCCESS 728c3850425c43379b7236298ae61810bc81b3ce
by kernel test robot 18 Jul '24

18 Jul '24

tree/branch: https://gitee.com/openeuler/kernel.git openEuler-1.0-LTS branch HEAD: 728c3850425c43379b7236298ae61810bc81b3ce !10033 ipv6: prevent possible NULL dereference in rt6_probe() Warning ids grouped by kconfigs: recent_errors `-- x86_64-buildonly-randconfig-002-20240717 `-- include-asm-generic-atomic-instrumented.h:warning:wait_req-may-be-used-uninitialized elapsed time: 1445m configs tested: 35 configs skipped: 134 The following configs have been built successfully. More configs may be tested in the coming days. tested configs: arm64 allnoconfig gcc-14.1.0 arm64 defconfig gcc-14.1.0 arm64 randconfig-001-20240717 gcc-14.1.0 arm64 randconfig-002-20240717 gcc-14.1.0 arm64 randconfig-003-20240717 gcc-14.1.0 arm64 randconfig-004-20240717 gcc-14.1.0 x86_64 alldefconfig gcc-13 x86_64 allnoconfig clang-18 x86_64 allyesconfig clang-18 x86_64 buildonly-randconfig-001-20240717 gcc-13 x86_64 buildonly-randconfig-002-20240717 gcc-13 x86_64 buildonly-randconfig-003-20240717 gcc-13 x86_64 buildonly-randconfig-004-20240717 clang-18 x86_64 buildonly-randconfig-005-20240717 clang-18 x86_64 buildonly-randconfig-006-20240717 gcc-13 x86_64 defconfig gcc-13 x86_64 randconfig-001-20240717 gcc-7 x86_64 randconfig-002-20240717 gcc-9 x86_64 randconfig-003-20240717 clang-18 x86_64 randconfig-004-20240717 gcc-9 x86_64 randconfig-005-20240717 clang-18 x86_64 randconfig-006-20240717 gcc-11 x86_64 randconfig-011-20240717 gcc-8 x86_64 randconfig-012-20240717 clang-18 x86_64 randconfig-013-20240717 gcc-12 x86_64 randconfig-014-20240717 gcc-10 x86_64 randconfig-015-20240717 clang-18 x86_64 randconfig-016-20240717 clang-18 x86_64 randconfig-071-20240717 gcc-12 x86_64 randconfig-072-20240717 clang-18 x86_64 randconfig-073-20240717 gcc-10 x86_64 randconfig-074-20240717 gcc-13 x86_64 randconfig-075-20240717 gcc-13 x86_64 randconfig-076-20240717 gcc-12 x86_64 rhel-8.3-rust clang-18 -- 0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests/wiki

1 0

[PATCH OLK-6.6] mm/sharepool: Fix kabi problems
by Yuan Can 18 Jul '24

18 Jul '24

From: Wang Wensheng <wangwensheng4(a)huawei.com> hulk inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I8LNGH -------------------------------- Use kabi_reserve in vm_area_struct and mm_struct. This is used for sharepool. Signed-off-by: Wang Wensheng <wangwensheng4(a)huawei.com> --- include/linux/mm_types.h | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index b99610d53256..bc46f518d77d 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -703,9 +703,10 @@ struct vm_area_struct { #endif struct vm_userfaultfd_ctx vm_userfaultfd_ctx; #ifdef CONFIG_SHARE_POOL - struct sp_area *spa; -#endif + KABI_USE(1, struct sp_area *spa) +#else KABI_RESERVE(1) +#endif KABI_RESERVE(2) KABI_RESERVE(3) KABI_RESERVE(4) @@ -967,9 +968,6 @@ struct mm_struct { #endif } lru_gen; #endif /* CONFIG_LRU_GEN */ -#ifdef CONFIG_SHARE_POOL - struct sp_group_master *sp_group_master; -#endif #ifdef CONFIG_MEMORY_RELIABLE /* total used reliable pages */ atomic_long_t reliable_nr_page; @@ -979,7 +977,11 @@ struct mm_struct { #endif } __randomize_layout; +#ifdef CONFIG_SHARE_POOL + KABI_USE(1, struct sp_group_master *sp_group_master) +#else KABI_RESERVE(1) +#endif KABI_RESERVE(2) KABI_RESERVE(3) KABI_RESERVE(4) -- 2.17.1

2 1