summaryrefslogtreecommitdiffstats
path: root/net/ceph/messenger_v2.c
Commit message (Collapse)AuthorAgeFilesLines
* libceph: init the cursor when preparing sparse read in msgr2Xiubo Li2024-03-061-0/+3
| | | | | | | | | | | | | | | The cursor is no longer initialized in the OSD client, causing the sparse read state machine to fall into an infinite loop. The cursor should be initialized in IN_S_PREPARE_SPARSE_DATA state. [ idryomov: use msg instead of con->in_msg, changelog ] Link: https://tracker.ceph.com/issues/64607 Fixes: 8e46a2d068c9 ("libceph: just wait for more data to be available on the socket") Signed-off-by: Xiubo Li <xiubli@redhat.com> Reviewed-by: Ilya Dryomov <idryomov@gmail.com> Tested-by: Luis Henriques <lhenriques@suse.de> Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
* libceph: just wait for more data to be available on the socketXiubo Li2024-02-071-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | A short read may occur while reading the message footer from the socket. Later, when the socket is ready for another read, the messenger invokes all read_partial_*() handlers, including read_partial_sparse_msg_data(). The expectation is that read_partial_sparse_msg_data() would bail, allowing the messenger to invoke read_partial() for the footer and pick up where it left off. However read_partial_sparse_msg_data() violates that and ends up calling into the state machine in the OSD client. The sparse-read state machine assumes that it's a new op and interprets some piece of the footer as the sparse-read header and returns bogus extents/data length, etc. To determine whether read_partial_sparse_msg_data() should bail, let's reuse cursor->total_resid. Because once it reaches to zero that means all the extents and data have been successfully received in last read, else it could break out when partially reading any of the extents and data. And then osd_sparse_read() could continue where it left off. [ idryomov: changelog ] Link: https://tracker.ceph.com/issues/63586 Fixes: d396f89db39a ("libceph: add sparse read support to msgr1") Signed-off-by: Xiubo Li <xiubli@redhat.com> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
* libceph: stop checking crypto_shash_alignmaskEric Biggers2023-10-271-4/+0
| | | | | | | | | | Now that the shash algorithm type does not support nonzero alignmasks, crypto_shash_alignmask() always returns 0 and will be removed. In preparation for this, stop checking crypto_shash_alignmask() in net/ceph/messenger_v2.c. Signed-off-by: Eric Biggers <ebiggers@google.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
* libceph: do not include crypto/algapi.hHerbert Xu2023-08-241-1/+1
| | | | | | | | | The header file crypto/algapi.h is for internal use only. Use the header file crypto/utils.h instead. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Reviewed-by: Ilya Dryomov <idryomov@gmail.com> Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
* libceph: support sparse reads on msgr2 secure codepathJeff Layton2023-08-221-10/+110
| | | | | | | | | | | | | | | | | | | | | | Add a new init_sgs_pages helper that populates the scatterlist from an arbitrary point in an array of pages. Change setup_message_sgs to take an optional pointer to an array of pages. If that's set, then the scatterlist will be set using that array instead of the cursor. When given a sparse read on a secure connection, decrypt the data in-place rather than into the final destination, by passing it the in_enc_pages array. After decrypting, run the sparse_read state machine in a loop, copying data from the decrypted pages until it's complete. Signed-off-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: Xiubo Li <xiubli@redhat.com> Reviewed-and-tested-by: Luís Henriques <lhenriques@suse.de> Reviewed-by: Milind Changire <mchangir@redhat.com> Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
* libceph: new sparse_read op, support sparse reads on msgr2 crc codepathJeff Layton2023-08-221-9/+158
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add support for a new sparse_read ceph_connection operation. The idea is that the client driver can define this operation use it to do special handling for incoming reads. The alloc_msg routine will look at the request and determine whether the reply is expected to be sparse. If it is, then we'll dispatch to a different set of state machine states that will repeatedly call the driver's sparse_read op to get length and placement info for reading the extent map, and the extents themselves. This necessitates adding some new field to some other structs: - The msg gets a new bool to track whether it's a sparse_read request. - A new field is added to the cursor to track the amount remaining in the current extent. This is used to cap the read from the socket into the msg_data - Handing a revoke with all of this is particularly difficult, so I've added a new data_len_remain field to the v2 connection info, and then use that to skip that much on a revoke. We may want to expand the use of that to the normal read path as well, just for consistency's sake. Signed-off-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: Xiubo Li <xiubli@redhat.com> Reviewed-and-tested-by: Luís Henriques <lhenriques@suse.de> Reviewed-by: Milind Changire <mchangir@redhat.com> Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
* libceph: harden msgr2.1 frame segment length checksIlya Dryomov2023-07-131-15/+26
| | | | | | | | | | | | | | | | | ceph_frame_desc::fd_lens is an int array. decode_preamble() thus effectively casts u32 -> int but the checks for segment lengths are written as if on unsigned values. While reading in HELLO or one of the AUTH frames (before authentication is completed), arithmetic in head_onwire_len() can get duped by negative ctrl_len and produce head_len which is less than CEPH_PREAMBLE_LEN but still positive. This would lead to a buffer overrun in prepare_read_control() as the preamble gets copied to the newly allocated buffer of size head_len. Cc: stable@vger.kernel.org Fixes: cd1a677cad99 ("libceph, ceph: implement msgr2.1 protocol (crc and secure modes)") Reported-by: Thelford Williams <thelford@google.com> Signed-off-by: Ilya Dryomov <idryomov@gmail.com> Reviewed-by: Xiubo Li <xiubli@redhat.com>
* libceph: Partially revert changes to support MSG_SPLICE_PAGESDavid Howells2023-06-271-18/+70
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fix the mishandling of MSG_DONTWAIT and also reinstates the per-page checking of the source pages (which might have come from a DIO write by userspace) by partially reverting the changes to support MSG_SPLICE_PAGES and doing things a little differently. In messenger_v1: (1) The ceph_tcp_sendpage() is resurrected and the callers reverted to use that. (2) The callers now pass MSG_MORE unconditionally. Previously, they were passing in MSG_MORE|MSG_SENDPAGE_NOTLAST and then degrading that to just MSG_MORE on the last call to ->sendpage(). (3) Make ceph_tcp_sendpage() a wrapper around sendmsg() rather than sendpage(), setting MSG_SPLICE_PAGES if sendpage_ok() returns true on the page. In messenger_v2: (4) Bring back do_try_sendpage() and make the callers use that. (5) Make do_try_sendpage() use sendmsg() for both cases and set MSG_SPLICE_PAGES if sendpage_ok() is set. Fixes: 40a8c17aa770 ("ceph: Use sendmsg(MSG_SPLICE_PAGES) rather than sendpage") Fixes: fa094ccae1e7 ("ceph: Use sendmsg(MSG_SPLICE_PAGES) rather than sendpage()") Reported-by: Ilya Dryomov <idryomov@gmail.com> Link: https://lore.kernel.org/r/CAOi1vP9vjLfk3W+AJFeexC93jqPaPUn2dD_4NrzxwoZTbYfOnw@mail.gmail.com/ Link: https://lore.kernel.org/r/CAOi1vP_Bn918j24S94MuGyn+Gxk212btw7yWeDrRcW1U8pc_BA@mail.gmail.com/ Signed-off-by: David Howells <dhowells@redhat.com> cc: Xiubo Li <xiubli@redhat.com> cc: Jeff Layton <jlayton@kernel.org> cc: Jens Axboe <axboe@kernel.dk> cc: Matthew Wilcox <willy@infradead.org> Link: https://lore.kernel.org/r/3101881.1687801973@warthog.procyon.org.uk/ # v1 Link: https://lore.kernel.org/r/3111635.1687813501@warthog.procyon.org.uk/ # v2 Reviewed-by: Ilya Dryomov <idryomov@gmail.com> Link: https://lore.kernel.org/r/3199652.1687873788@warthog.procyon.org.uk Signed-off-by: Jakub Kicinski <kuba@kernel.org>
* ceph: Use sendmsg(MSG_SPLICE_PAGES) rather than sendpage()David Howells2023-06-241-72/+19
| | | | | | | | | | | | | | | | | Use sendmsg() and MSG_SPLICE_PAGES rather than sendpage in ceph when transmitting data. For the moment, this can only transmit one page at a time because of the architecture of net/ceph/, but if write_partial_message_data() can be given a bvec[] at a time by the iteration code, this would allow pages to be sent in a batch. Signed-off-by: David Howells <dhowells@redhat.com> cc: Ilya Dryomov <idryomov@gmail.com> cc: Xiubo Li <xiubli@redhat.com> cc: Jeff Layton <jlayton@kernel.org> cc: Jens Axboe <axboe@kernel.dk> cc: Matthew Wilcox <willy@infradead.org> Link: https://lore.kernel.org/r/20230623225513.2732256-5-dhowells@redhat.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
* libceph: use bvec_set_page to initialize bvecsChristoph Hellwig2023-02-031-17/+11
| | | | | | | | | Use the bvec_set_page helper to initialize bvecs. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Ilya Dryomov <idryomov@gmail.com> Link: https://lore.kernel.org/r/20230203150634.3199647-24-hch@lst.de Signed-off-by: Jens Axboe <axboe@kernel.dk>
* use less confusing names for iov_iter direction initializersAl Viro2022-11-251-7/+7
| | | | | | | | | | | | | READ/WRITE proved to be actively confusing - the meanings are "data destination, as used with read(2)" and "data source, as used with write(2)", but people keep interpreting those as "we read data from it" and "we write data to it", i.e. exactly the wrong way. Call them ITER_DEST and ITER_SOURCE - at least that is harder to misinterpret... Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
* libceph: drop last_piece flag from ceph_msg_data_cursorJeff Layton2022-10-041-1/+1
| | | | | | | | | | | ceph_msg_data_next is always passed a NULL pointer for this field. Some of the "next" operations look at it in order to determine the length, but we can just take the min of the data on the page or cursor->resid. Signed-off-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: Xiubo Li <xiubli@redhat.com> Reviewed-by: Ilya Dryomov <idryomov@gmail.com> Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
* libceph: drop else branches in prepare_read_data{,_cont}Jeff Layton2022-03-011-6/+2
| | | | | | | | Just call set_in_bvec in the non-conditional part. Signed-off-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: Ilya Dryomov <idryomov@gmail.com> Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
* libceph: optionally use bounce buffer on recv path in crc modeIlya Dryomov2022-02-021-14/+44
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Both msgr1 and msgr2 in crc mode are zero copy in the sense that message data is read from the socket directly into the destination buffer. We assume that the destination buffer is stable (i.e. remains unchanged while it is being read to) though. Otherwise, CRC errors ensue: libceph: read_partial_message 0000000048edf8ad data crc 1063286393 != exp. 228122706 libceph: osd1 (1)192.168.122.1:6843 bad crc/signature libceph: bad data crc, calculated 57958023, expected 1805382778 libceph: osd2 (2)192.168.122.1:6876 integrity error, bad crc Introduce rxbounce option to enable use of a bounce buffer when receiving message data. In particular this is needed if a mapped image is a Windows VM disk, passed to QEMU. Windows has a system-wide "dummy" page that may be mapped into the destination buffer (potentially more than once into the same buffer) by the Windows Memory Manager in an effort to generate a single large I/O [1][2]. QEMU makes a point of preserving overlap relationships when cloning I/O vectors, so krbd gets exposed to this behaviour. [1] "What Is Really in That MDL?" https://docs.microsoft.com/en-us/previous-versions/windows/hardware/design/dn614012(v=vs.85) [2] https://blogs.msmvps.com/kernelmustard/2005/05/04/dummy-pages/ URL: https://bugzilla.redhat.com/show_bug.cgi?id=1973317 Signed-off-by: Ilya Dryomov <idryomov@gmail.com> Reviewed-by: Jeff Layton <jlayton@kernel.org>
* libceph: make recv path in secure mode work the same as send pathIlya Dryomov2022-02-021-62/+154
| | | | | | | | | | | | | | | | | | The recv path of secure mode is intertwined with that of crc mode. While it's slightly more efficient that way (the ciphertext is read into the destination buffer and decrypted in place, thus avoiding two potentially heavy memory allocations for the bounce buffer and the corresponding sg array), it isn't really amenable to changes. Sacrifice that edge and align with the send path which always uses a full-sized bounce buffer (currently there is no other way -- if the kernel crypto API ever grows support for streaming (piecewise) en/decryption for GCM [1], we would be able to easily take advantage of that on both sides). [1] https://lore.kernel.org/all/20141225202830.GA18794@gondor.apana.org.au/ Signed-off-by: Ilya Dryomov <idryomov@gmail.com> Reviewed-by: Jeff Layton <jlayton@kernel.org>
* mm: allow !GFP_KERNEL allocations for kvmallocMichal Hocko2022-01-151-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | Support for GFP_NO{FS,IO} and __GFP_NOFAIL has been implemented by previous patches so we can allow the support for kvmalloc. This will allow some external users to simplify or completely remove their helpers. GFP_NOWAIT semantic hasn't been supported so far but it hasn't been explicitly documented so let's add a note about that. ceph_kvmalloc is the first helper to be dropped and changed to kvmalloc. Link: https://lkml.kernel.org/r/20211122153233.9924-5-mhocko@kernel.org Signed-off-by: Michal Hocko <mhocko@suse.com> Reviewed-by: Uladzislau Rezki (Sony) <urezki@gmail.com> Acked-by: Vlastimil Babka <vbabka@suse.cz> Cc: Christoph Hellwig <hch@lst.de> Cc: Dave Chinner <david@fromorbit.com> Cc: Ilya Dryomov <idryomov@gmail.com> Cc: Jeff Layton <jlayton@kernel.org> Cc: Neil Brown <neilb@suse.de> Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* libceph: zero out session key and connection secretIlya Dryomov2021-01-041-19/+26
| | | | | | | | Try and avoid leaving bits and pieces of session key and connection secret (gets split into GCM key and a pair of GCM IVs) around. Signed-off-by: Ilya Dryomov <idryomov@gmail.com> Reviewed-by: Jeff Layton <jlayton@kernel.org>
* libceph: align session_key and con_secret to 16 bytesIlya Dryomov2020-12-281-2/+10
| | | | | | | | | | | | | crypto_shash_setkey() and crypto_aead_setkey() will do a (small) GFP_ATOMIC allocation to align the key if it isn't suitably aligned. It's not a big deal, but at the same time easy to avoid. The actual alignment requirement is dynamic, queryable with crypto_shash_alignmask() and crypto_aead_alignmask(), but shouldn't be stricter than 16 bytes for our algorithms. Fixes: cd1a677cad99 ("libceph, ceph: implement msgr2.1 protocol (crc and secure modes)") Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
* libceph: fix auth_signature buffer allocation in secure modeIlya Dryomov2020-12-281-1/+2
| | | | | | | | | | | | auth_signature frame is 68 bytes in plain mode and 96 bytes in secure mode but we are requesting 68 bytes in both modes. By luck, this doesn't actually result in any invalid memory accesses because the allocation is satisfied out of kmalloc-96 slab and so exactly 96 bytes are allocated, but KASAN rightfully complains. Fixes: cd1a677cad99 ("libceph, ceph: implement msgr2.1 protocol (crc and secure modes)") Reported-by: Luis Henriques <lhenriques@suse.de> Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
* Merge tag 'ceph-for-5.11-rc1' of git://github.com/ceph/ceph-clientLinus Torvalds2020-12-171-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Pull ceph updates from Ilya Dryomov: "The big ticket item here is support for msgr2 on-wire protocol, which adds the option of full in-transit encryption using AES-GCM algorithm (myself). On top of that we have a series to avoid intermittent errors during recovery with recover_session=clean and some MDS request encoding work from Jeff, a cap handling fix and assorted observability improvements from Luis and Xiubo and a good number of cleanups. Luis also ran into a corner case with quotas which sadly means that we are back to denying cross-quota-realm renames" * tag 'ceph-for-5.11-rc1' of git://github.com/ceph/ceph-client: (59 commits) libceph: drop ceph_auth_{create,update}_authorizer() libceph, ceph: make use of __ceph_auth_get_authorizer() in msgr1 libceph, ceph: implement msgr2.1 protocol (crc and secure modes) libceph: introduce connection modes and ms_mode option libceph, rbd: ignore addr->type while comparing in some cases libceph, ceph: get and handle cluster maps with addrvecs libceph: factor out finish_auth() libceph: drop ac->ops->name field libceph: amend cephx init_protocol() and build_request() libceph, ceph: incorporate nautilus cephx changes libceph: safer en/decoding of cephx requests and replies libceph: more insight into ticket expiry and invalidation libceph: move msgr1 protocol specific fields to its own struct libceph: move msgr1 protocol implementation to its own file libceph: separate msgr1 protocol implementation libceph: export remaining protocol independent infrastructure libceph: export zero_page libceph: rename and export con->flags bits libceph: rename and export con->state states libceph: make con->state an int ...
* libceph, ceph: implement msgr2.1 protocol (crc and secure modes)Ilya Dryomov2020-12-141-0/+3443
Implement msgr2.1 wire protocol, available since nautilus 14.2.11 and octopus 15.2.5. msgr2.0 wire protocol is not implemented -- it has several security, integrity and robustness issues and therefore considered deprecated. Signed-off-by: Ilya Dryomov <idryomov@gmail.com>