crypto: algifaead - Revert to operating out-of-place

In multiple locations, there is a possible way to conduct local privilege escalation due to memory corruption. This could lead to local escalation of privilege with no additional execution privileges needed. User interaction is not needed for exploitation.

Android contains an unspecified vulnerability in the kernel that allows for remote code execution. This vulnerability resides in Linux Kernel and could impact other products, including but not limited to Android OS.

Issue summary: Some non-default TLS server configurations can cause unbounded memory growth when processing TLSv1.3 sessions

Issue summary: A timing side-channel which could potentially allow recovering the private key exists in the ECDSA signature computation.

Impact summary: A timing side-channel in ECDSA signature computations could allow recovering the private key by an attacker. However, measuring the timing would require either local access to the signing application or a very fast network connection with low latency.

There is a timing signal of around 300 nanoseconds when the top word of the inverted ECDSA nonce value is zero. This can happen with significant probability only for some of the supported elliptic curves. In particular the NIST P-521 curve is affected. To be able to measure this leak, the attacker process must either be located in the same physical computer or must have a very fast network connection with low latency. For that reason the severity of this vulnerability is Low.

The FIPS modules in 3.4, 3.3, 3.2, 3.1 and 3.0 are affected by this issue.

Excessive time spent in DH check / generation with large Q parameter value

A type confusion vulnerability was found in OpenSSL when OpenSSL X.400 addresses processing inside an X.509 GeneralName. When CRL checking is enabled (for example, the application sets the X509VFLAGCRLCHECK flag), this vulnerability may allow an attacker to pass arbitrary pointers to a memcmp call, enabling them to read memory contents or cause a denial of service. In most cases, the attack requires the attacker to provide both the certificate chain and CRL, of which neither needs a valid signature. If the attacker only controls one of these inputs, the other input must already contain an X.400 address as a CRL distribution point, which is uncommon. In this case, this vulnerability is likely only to affect applications that have implemented their own functionality for retrieving CRLs over a network.

Issue summary: Processing some specially crafted ASN.1 object identifiers or data containing them may be very slow.

Impact summary: Applications that use OBJobj2txt() directly, or use any of the OpenSSL subsystems OCSP, PKCS7/SMIME, CMS, CMP/CRMF or TS with no message size limit may experience notable to very long delays when processing those messages, which may lead to a Denial of Service.

A use-after-free vulnerability was found in OpenSSL's BIOnewNDEF function. The public API function BIOnewNDEF is a helper function used for streaming ASN.1 data via a BIO. It is primarily used internally by OpenSSL to support the SMIME, CMS, and PKCS7 streaming capabilities, but it may also be called directly by end-user applications. The function receives a BIO from the caller, prepends a new BIOfasn1 filter BIO onto the front of it to form a BIO chain, and then returns the new head of the BIO chain to the caller. Under certain conditions. For example, if a CMS recipient public key is invalid, the new filter BIO is freed, and the function returns a NULL result indicating a failure. However, in this case, the BIO chain is not properly cleaned up, and the BIO passed by the caller still retains internal pointers to the previously freed filter BIO. If the caller then calls BIOpop() on the BIO, a use-after-free will occur, possibly resulting in a crash.

A double-free vulnerability was found in OpenSSL's PEMreadbioex function. The function PEMreadbioex() reads a PEM file from a BIO and parses and decodes the "name" (for example, "CERTIFICATE"), any header data, and the payload data. If the function succeeds, then the "nameout," "header," and "data" arguments are populated with pointers to buffers containing the relevant decoded data. The caller is responsible for freeing those buffers. Constructing a PEM file that results in 0 bytes of payload data is possible. In this case, PEMreadbioex() will return a failure code but will populate the header argument with a pointer to a freed buffer. A double-free will occur if the caller also frees this buffer. This will most likely lead to a crash. This could be exploited by an attacker who can supply malicious PEM files for parsing to achieve a denial of service attack.

A timing based side channel exists in the OpenSSL RSA Decryption implementation which could be sufficient to recover a plaintext across a network in a Bleichenbacher style attack. To achieve a successful decryption an attacker would have to be able to send a very large number of trial messages for decryption. The vulnerability affects all RSA padding modes: PKCS#1 v1.5, RSA-OEAP and RSASVE.

For example, in a TLS connection, RSA is commonly used by a client to send an encrypted pre-master secret to the server. An attacker that had observed a genuine connection between a client and a server could use this flaw to send trial messages to the server and record the time taken to process them. After a sufficiently large number of messages the attacker could recover the pre-master secret used for the original connection and thus be able to decrypt the application data sent over that connection.

A use after free can have a range of potential consequences such as the corruption of valid data, crashes or execution of arbitrary code. However, only applications that directly call the SSLfreebuffers function are affected by this issue. Applications that do not call this function are not vulnerable. Our investigations indicate that this function is rarely used by applications.

The SSLfreebuffers function is used to free the internal OpenSSL buffer used when processing an incoming record from the network. The call is only expected to succeed if the buffer is not currently in use. However, two scenarios have been identified where the buffer is freed even when still in use.

The first scenario occurs where a record header has been received from the network and processed by OpenSSL, but the full record body has not yet arrived. In this case calling SSLfreebuffers will succeed even though a record has only been partially processed and the buffer is still in use.

The second scenario occurs where a full record containing application data has been received and processed by OpenSSL but the application has only read part of this data. Again a call to SSLfreebuffers will succeed even though the buffer is still in use.

While these scenarios could occur accidentally during normal operation a malicious attacker could attempt to engineer a stituation where this occurs. We are not aware of this issue being actively exploited.

The FIPS modules in 3.3, 3.2, 3.1 and 3.0 are not affected by this issue.

OpenSSL 1.0.2 is also not affected by this issue.

OpenSSL 3.3, 3.2, 3.1, 3.0 and 1.1.1 are vulnerable to this issue.

OpenSSL 3.3 users should upgrade to OpenSSL 3.3.1 once it is released.

OpenSSL 3.2 users should upgrade to OpenSSL 3.2.2 once it is released.

OpenSSL 3.1 users should upgrade to OpenSSL 3.1.6 once it is released.

OpenSSL 3.0 users should upgrade to OpenSSL 3.0.14 once it is released.

OpenSSL 1.1.1 users should upgrade to OpenSSL 1.1.1y once it is released (premium support customers only).

Due to the low severity of this issue we are not issuing new releases of OpenSSL at this time. The fix will be included in the next releases when they become available. The fix is also available in commit e5093133c3 (for 3.3), commit c88c3de510 (for 3.2), commit 704f725b96 (for 3.1) and commit b3f0eb0a29 (for 3.0) in the OpenSSL git repository. It is available to premium support customers in commit f7a045f314 (for 1.1.1).

This issue was reported on 10th April 2024 by William Ahern (Akamai). The fix was developed by Matt Caswell and Watson Ladd (Akamai).

In the Linux kernel, the following vulnerability has been resolved:

tls: separate no-async decryption request handling from async

If we're not doing async, the handling is much simpler. There's no reference counting, we just need to wait for the completion to wake us up and return its result.

We should preferably also use a separate cryptowait. I'm not seeing a UAF as I did in the past, I think aec7961916f3 ("tls: fix race between async notify and socket close") took care of it.

This will make the next fix easier.

In the Linux kernel, the following vulnerability has been resolved:

perf/aux: Fix AUX buffer serialization

Ole reported that event->mmapmutex is strictly insufficient to serialize the AUX buffer, add a per RB mutex to fully serialize it.

Note that in the lock order comment the perfevent::mmapmutex order was already wrong, that is, it nesting under mmaplock is not new with this patch.

A denial-of-service (DoS) attack was found in the mlx5 driver in the Linux kernel. A KVM guest VM using virtio-net can crash the host by sending a short packet (i.e. size < ETHHLEN). The packet may traverse through vhost-net, macvtap and vlan without any validation/drop. When this packet is presented to mlx5 driver on the host side, the kernel panic happens since mlx5core assumes the frame size is always >= ETHHLEN.

This vulnerability affects both drivers/net/tun.c and drivers/net/tap.c. CVE-2024-41091 has been assigned to the TUN side of the issue.

Reference: https://www.openwall.com/lists/oss-security/2024/07/24/4

A denial-of-service (DoS) attack was found in the mlx5 driver in the Linux kernel. A KVM guest VM using virtio-net can crash the host by sending a short packet (i.e. size < ETHHLEN). The packet may traverse through vhost-net, macvtap and vlan without any validation/drop. When this packet is presented to mlx5 driver on the host side, the kernel panic happens, since mlx5core assumes the frame size is always >= ETHHLEN.

This vulnerability affects both drivers/net/tun.c and drivers/net/tap.c. CVE-2024-41090 has been assigned to the TAP side of the issue.

Reference: https://www.openwall.com/lists/oss-security/2024/07/24/4

In the Linux kernel, the following vulnerability has been resolved:

netfilter: tproxy: bail out if IP has been disabled on the device

syzbot reports: general protection fault, probably for non-canonical address 0xdffffc0000000003: 0000 [#1] PREEMPT SMP KASAN PTI KASAN: null-ptr-deref in range [0x0000000000000018-0x000000000000001f] [..] RIP: 0010:nftproxyladdr4+0xb7/0x340 net/ipv4/netfilter/nftproxyipv4.c:62 Call Trace: nfttproxyevalv4 net/netfilter/nfttproxy.c:56 [inline] nfttproxyeval+0xa9a/0x1a00 net/netfilter/nfttproxy.c:168

indevgetrcu() can return NULL, so check for this.

A Speculative Race Condition (SRC) vulnerability that impacts modern CPU architectures supporting speculative execution (related to Spectre V1) has been disclosed. An unauthenticated attacker can exploit this vulnerability to disclose arbitrary data from the CPU using race conditions to access the speculative executable code paths.

Dear Linux Developers,

We're reaching out to you as part of the disclosure process of our research.

In our research, which we will present at IEEE Security & Privacy in May 2024, we found that attackers can not only create TCP-spoofed connections (which was already known), but can also reliably transmit IP-spoofed data over such connections. This has security implications for applications that rely on TCP endpoint IP addresses, such as firewalling or host-based authentication (e.g., SMTP/SPF, of DBs).

We basically discovered two TCP spoofing primitives. First, attackers can bruteforce the server-chosen send window by acknowledging data that was never sent (what we call "ghost ACKs"; see Figure 3 in the paper). Second, we show that there are side channels that allow the attacker to leak the otherwise-secret server-chosen initial sequence number (ISN). One of these side channels leverages TCP SYN cookies.

We believe that the TCP/IP stack can take countermeasures to prevent such attacks, or at least make them harder. For example, we think that TCP endpoints should ignore ghost ACKs, and have some ideas to randomize the TCP backlog queue to prevent the SYN cookie side channel.

At the same time, we have disclosed our findings to the IETF folks and hope that they have helpful feedback for us.

In the Linux kernel, the following vulnerability has been resolved:

tty: ngsm: require CAPNETADMIN to attach NGSM0710 ldisc

Any unprivileged user can attach NGSM0710 ldisc, but it requires CAPNETADMIN to create a GSM network anyway.

Require initial namespace CAPNETADMIN to do that.

A flaw in the Linux Kernel found. Use after free in the net/sched classifiers (clsfw, clsu32 and clsroute) can happen because of mainline/net/sched/clsfw.c incorrect handling of the existing filter in .change method that leads to an extra unbindtcf call for the associated class and that allows that class to be removed while it's still used. These bugs can be used for a local privilege escalation.

Upstream patch: https://lore.kernel.org/netdev/193d6cdf-d6c9-f9be-c36a-b2a7551d5fb6@mojatatu.com/

A flaw in the Linux Kernel found. An out-of-bounds write vulnerability in the Linux kernel's net/sched: schqfq component can be exploited to achieve local privilege escalation. The qfqchangeagg() function in net/sched/schqfq.c allows an out-of-bounds write because lmax is updated according to packet sizes without bounds checks.

Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=3e337087c3b5805fe0b8a46ba622a962880b5d64

A flaw in the Linux Kernel found. Use after free in the net/sched classifiers (clsfw, clsu32 and clsroute) can happen because of mainline/net/sched/clsfw.c incorrect handling of the existing filter in .change method that leads to an extra unbindtcf call for the associated class and that allows that class to be removed while it's still used. These bugs can be used for a local privilege escalation.

Upstream patch: https://lore.kernel.org/netdev/193d6cdf-d6c9-f9be-c36a-b2a7551d5fb6@mojatatu.com/

A use-after-free vulnerability in the Linux kernel's net/sched: clsu32 component can be exploited to achieve local privilege escalation.

When u32change() is called on an existing filter, the whole tcfresult struct is always copied into the new instance of the filter. This causes a problem when updating a filter bound to a class, as tcfunbindfilter() is always called on the old instance in the success path, decreasing filtercnt of the still referenced class and allowing it to be deleted, leading to a use-after-free.

We recommend upgrading past commit 3044b16e7c6fe5d24b1cdbcf1bd0a9d92d1ebd81.

A flaw in the Linux Kernel found. A use-after-free vulnerability in the Linux kernel's net/sched: clsfw component can be exploited to achieve local privilege escalation. If tcfchangeindev() fails, fwsetparms() will immediately return an error after incrementing or decrementing the reference counter in tcfbindfilter(). If an attacker can control the reference counter and set it to zero, they can cause the reference to be freed, leading to a use-after-free vulnerability.

Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit?id=0323bce598eea038714f941ce2b22541c46d488f

A use-after-free flaw was found in route4change in the net/sched/clsroute.c filter implementation in the Linux kernel. This flaw allows a local user to crash the system and possibly lead to a local privilege escalation problem.

i40e: fix idx validation in config queues msg

In the Linux kernel, the following vulnerability has been resolved:

sched/membarrier: reduce the ability to hammer on sysmembarrier

On some systems, sysmembarrier can be very expensive, causing overall slowdowns for everything. So put a lock on the path in order to serialize the accesses to prevent the ability for this to be called at too high of a frequency and saturate the machine.

In the Linux kernel, the following vulnerability has been resolved:

mm: fix zswap writeback race condition

The zswap writeback mechanism can cause a race condition resulting in memory corruption, where a swapped out page gets swapped in with data that was written to a different page.

The race unfolds like this: 1. a page with data A and swap offset X is stored in zswap 2. page A is removed off the LRU by zpool driver for writeback in zswap-shrink work, data for A is mapped by zpool driver 3. user space program faults and invalidates page entry A, offset X is considered free 4. kswapd stores page B at offset X in zswap (zswap could also be full, if so, page B would then be IOed to X, then skip step 5.) 5. entry A is replaced by B in tree->rbroot, this doesn't affect the local reference held by zswap-shrink work 6. zswap-shrink work writes back A at X, and frees zswap entry A 7. swapin of slot X brings A in memory instead of B

The fix: Once the swap page cache has been allocated (case ZSWAPSWAPCACHENEW), zswap-shrink work just checks that the local zswapentry reference is still the same as the one in the tree. If it's not the same it means that it's either been invalidated or replaced, in both cases the writeback is aborted because the local entry contains stale data.

Reproducer: I originally found this by running stress overnight to validate my work on the zswap writeback mechanism, it manifested after hours on my test machine. The key to make it happen is having zswap writebacks, so whatever setup pumps /sys/kernel/debug/zswap/writtenbackpages should do the trick.

In order to reproduce this faster on a vm, I setup a system with ~100M of available memory and a 500M swap file, then running stress --vm 1 --vm-bytes 300000000 --vm-stride 4000 makes it happen in matter of tens of minutes. One can speed things up even more by swinging /sys/module/zswap/parameters/maxpoolpercent up and down between, say, 20 and 1; this makes it reproduce in tens of seconds. It's crucial to set --vm-stride to something other than 4096 otherwise stress won't realize that memory has been corrupted because all pages would have the same data.

Prior to Apache HTTP Server 2.4.55, a malicious backend can cause the response headers to be truncated early, resulting in some headers being incorporated into the response body. If the later headers have any security purpose, they will not be interpreted by the client.