kernel/linux.git, branch v5.10.126

Linux 5.10.126

2022-06-27T07:41:01+00:00

Signed-off-by: Greg Kroah-Hartman

io_uring: use separate list entry for iopoll requests

2022-06-27T07:41:01+00:00

A previous commit ended up enabling file tracking for iopoll requests, which conflicts with both of them using the same list entry for tracking. Add a separate list entry just for iopoll requests, avoid this issue. No upstream commit exists for this issue. Reported-by: Greg Thelen Fixes: df3f3bb5059d ("io_uring: add missing item types for various requests") Signed-off-by: Jens Axboe Signed-off-by: Greg Kroah-Hartman

Linux 5.10.125

2022-06-25T13:16:09+00:00

Link: https://lore.kernel.org/r/20220623164322.296526800@linuxfoundation.org Tested-by: Salvatore Bonaccorso Tested-by: Pavel Machek (CIP) Tested-by: Florian Fainelli Tested-by: Shuah Khan Tested-by: Hulk Robot Tested-by: Jon Hunter Tested-by: Sudip Mukherjee Tested-by: Guenter Roeck Signed-off-by: Greg Kroah-Hartman

io_uring: add missing item types for various requests

2022-06-25T13:16:09+00:00

Any read/write should grab current->nsproxy, denoted by IO_WQ_WORK_FILES as it refers to current->files as well, and connect and recv/recvmsg, send/sendmsg should grab current->fs which is denoted by IO_WQ_WORK_FS. No upstream commit exists for this issue. Reported-by: Bing-Jhong Billy Jheng Signed-off-by: Jens Axboe Signed-off-by: Greg Kroah-Hartman

arm64: mm: Don't invalidate FROM_DEVICE buffers at start of DMA transfer

2022-06-25T13:16:09+00:00

commit c50f11c6196f45c92ca48b16a5071615d4ae0572 upstream. Invalidating the buffer memory in arch_sync_dma_for_device() for FROM_DEVICE transfers When using the streaming DMA API to map a buffer prior to inbound non-coherent DMA (i.e. DMA_FROM_DEVICE), we invalidate any dirty CPU cachelines so that they will not be written back during the transfer and corrupt the buffer contents written by the DMA. This, however, poses two potential problems: (1) If the DMA transfer does not write to every byte in the buffer, then the unwritten bytes will contain stale data once the transfer has completed. (2) If the buffer has a virtual alias in userspace, then stale data may be visible via this alias during the period between performing the cache invalidation and the DMA writes landing in memory. Address both of these issues by cleaning (aka writing-back) the dirty lines in arch_sync_dma_for_device(DMA_FROM_DEVICE) instead of discarding them using invalidation. Cc: Ard Biesheuvel Cc: Christoph Hellwig Cc: Robin Murphy Cc: Russell King Cc: Link: https://lore.kernel.org/r/20220606152150.GA31568@willie-the-truck Signed-off-by: Will Deacon Reviewed-by: Ard Biesheuvel Link: https://lore.kernel.org/r/20220610151228.4562-2-will@kernel.org Signed-off-by: Catalin Marinas Signed-off-by: Greg Kroah-Hartman

serial: core: Initialize rs485 RTS polarity already on probe

2022-06-25T13:16:09+00:00

commit 2dd8a74fddd21b95dcc60a2d3c9eaec993419d69 upstream. RTS polarity of rs485-enabled ports is currently initialized on uart open via: tty_port_open() tty_port_block_til_ready() tty_port_raise_dtr_rts() # if (C_BAUD(tty)) uart_dtr_rts() uart_port_dtr_rts() There's at least three problems here: First, if no baud rate is set, RTS polarity is not initialized. That's the right thing to do for rs232, but not for rs485, which requires that RTS is deasserted unconditionally. Second, if the DeviceTree property "linux,rs485-enabled-at-boot-time" is present, RTS should be deasserted as early as possible, i.e. on probe. Otherwise it may remain asserted until first open. Third, even though RTS is deasserted on open and close, it may subsequently be asserted by uart_throttle(), uart_unthrottle() or uart_set_termios() because those functions aren't rs485-aware. (Only uart_tiocmset() is.) To address these issues, move RTS initialization from uart_port_dtr_rts() to uart_configure_port(). Prevent subsequent modification of RTS polarity by moving the existing rs485 check from uart_tiocmget() to uart_update_mctrl(). That way, RTS is initialized on probe and then remains unmodified unless the uart transmits data. If rs485 is enabled at runtime (instead of at boot) through a TIOCSRS485 ioctl(), RTS is initialized by the uart driver's ->rs485_config() callback and then likewise remains unmodified. The PL011 driver initializes RTS on uart open and prevents subsequent modification in its ->set_mctrl() callback. That code is obsoleted by the present commit, so drop it. Cc: Jan Kiszka Cc: Su Bao Cheng Signed-off-by: Lukas Wunner Link: https://lore.kernel.org/r/2d2acaf3a69e89b7bf687c912022b11fd29dfa1e.1642909284.git.lukas@wunner.de Signed-off-by: Greg Kroah-Hartman

tcp: drop the hash_32() part from the index calculation

2022-06-25T13:16:09+00:00

commit e8161345ddbb66e449abde10d2fdce93f867eba9 upstream. In commit 190cc82489f4 ("tcp: change source port randomizarion at connect() time"), the table_perturb[] array was introduced and an index was taken from the port_offset via hash_32(). But it turns out that hash_32() performs a multiplication while the input here comes from the output of SipHash in secure_seq, that is well distributed enough to avoid the need for yet another hash. Suggested-by: Amit Klein Reviewed-by: Eric Dumazet Signed-off-by: Willy Tarreau Signed-off-by: Jakub Kicinski Cc: Ben Hutchings Signed-off-by: Greg Kroah-Hartman

tcp: increase source port perturb table to 2^16

2022-06-25T13:16:09+00:00

commit 4c2c8f03a5ab7cb04ec64724d7d176d00bcc91e5 upstream. Moshe Kol, Amit Klein, and Yossi Gilad reported being able to accurately identify a client by forcing it to emit only 40 times more connections than there are entries in the table_perturb[] table. The previous two improvements consisting in resalting the secret every 10s and adding randomness to each port selection only slightly improved the situation, and the current value of 2^8 was too small as it's not very difficult to make a client emit 10k connections in less than 10 seconds. Thus we're increasing the perturb table from 2^8 to 2^16 so that the same precision now requires 2.6M connections, which is more difficult in this time frame and harder to hide as a background activity. The impact is that the table now uses 256 kB instead of 1 kB, which could mostly affect devices making frequent outgoing connections. However such components usually target a small set of destinations (load balancers, database clients, perf assessment tools), and in practice only a few entries will be visited, like before. A live test at 1 million connections per second showed no performance difference from the previous value. Reported-by: Moshe Kol Reported-by: Yossi Gilad Reported-by: Amit Klein Reviewed-by: Eric Dumazet Signed-off-by: Willy Tarreau Signed-off-by: Jakub Kicinski Cc: Ben Hutchings Signed-off-by: Greg Kroah-Hartman

tcp: dynamically allocate the perturb table used by source ports

2022-06-25T13:16:08+00:00

commit e9261476184be1abd486c9434164b2acbe0ed6c2 upstream. We'll need to further increase the size of this table and it's likely that at some point its size will not be suitable anymore for a static table. Let's allocate it on boot from inet_hashinfo2_init(), which is called from tcp_init(). Cc: Moshe Kol Cc: Yossi Gilad Cc: Amit Klein Reviewed-by: Eric Dumazet Signed-off-by: Willy Tarreau Signed-off-by: Jakub Kicinski Cc: Ben Hutchings Signed-off-by: Greg Kroah-Hartman

tcp: add small random increments to the source port

2022-06-25T13:16:08+00:00

commit ca7af0402550f9a0b3316d5f1c30904e42ed257d upstream. Here we're randomly adding between 0 and 7 random increments to the selected source port in order to add some noise in the source port selection that will make the next port less predictable. With the default port range of 32768-60999 this means a worst case reuse scenario of 14116/8=1764 connections between two consecutive uses of the same port, with an average of 14116/4.5=3137. This code was stressed at more than 800000 connections per second to a fixed target with all connections closed by the client using RSTs (worst condition) and only 2 connections failed among 13 billion, despite the hash being reseeded every 10 seconds, indicating a perfectly safe situation. Cc: Moshe Kol Cc: Yossi Gilad Cc: Amit Klein Reviewed-by: Eric Dumazet Signed-off-by: Willy Tarreau Signed-off-by: Jakub Kicinski Cc: Ben Hutchings Signed-off-by: Greg Kroah-Hartman