- 30 Nov, 2023 40 commits
-
-
Uwe Kleine-König authored
The .remove() callback for a platform driver returns an int which makes many driver authors wrongly assume it's possible to do error handling by returning an error code. However the value returned is ignored (apart from emitting a warning) and this typically results in resource leaks. To improve here there is a quest to make the remove callback return void. In the first step of this quest all drivers are converted to .remove_new(), which already returns void. Eventually after all drivers are converted, .remove_new() will be renamed to .remove(). Trivially convert this driver from always returning zero in the remove callback to the void returning variant. Signed-off-by:
Uwe Kleine-König <u.kleine-koenig@pengutronix.de> Reviewed-by:
Jeff Johnson <quic_jjohnson@quicinc.com> Signed-off-by:
Kalle Valo <kvalo@kernel.org> Link: https://lore.kernel.org/r/20231117093056.873834-13-u.kleine-koenig@pengutronix.de
-
Yang Li authored
./drivers/net/wireless/ralink/rt2x00/rt2800lib.c:1331:47-52: WARNING: conversion to bool not needed here ./drivers/net/wireless/ralink/rt2x00/rt2800lib.c:1332:47-52: WARNING: conversion to bool not needed here Reported-by:
Abaci Robot <abaci@linux.alibaba.com> Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=7531Signed-off-by:
Yang Li <yang.lee@linux.alibaba.com> Signed-off-by:
-
Eric Dumazet authored
syzbot was able to trigger a crash [1] in page_pool_unlist() page_pool_list() only inserts a page pool into a netdev page pool list if a netdev was set in params. Even if the kzalloc() call in page_pool_create happens to initialize pool->user.list, I chose to be more explicit in page_pool_list() adding one INIT_HLIST_NODE(). We could test in page_pool_unlist() if netdev was set, but since netdev can be changed to lo, it seems more robust to check if pool->user.list is hashed before calling hlist_del(). [1] Illegal XDP return value 4294946546 on prog (id 2) dev N/A, expect packet loss! general protection fault, probably for non-canonical address 0xdffffc0000000000: 0000 [#1] PREEMPT SMP KASAN KASAN: null-ptr-deref in range [0x0000000000000000-0x0000000000000007] CPU: 0 PID: 5064 Comm: syz-executor391 Not tainted 6.7.0-rc2-syzkaller-00533-ga3799729 #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 11/10/2023 RIP: 0010:__hlist_del include/linux/list.h:988 [inline] RIP: 0010:hlist_del include/linux/list.h:1002 [inline] RIP: 0010:page_pool_unlist+0xd1/0x170 net/core/page_pool_user.c:342 Code: df 48 89 fa 48 c1 ea 03 80 3c 02 00 0f 85 90 00 00 00 4c 8b a3 f0 06 00 00 48 b8 00 00 00 00 00 fc ff df 4c 89 e2 48 c1 ea 03 <80> 3c 02 00 75 68 48 85 ed 49 89 2c 24 74 24 e8 1b ca 07 f9 48 8d RSP: 0018:ffffc900039ff768 EFLAGS: 00010246 RAX: dffffc0000000000 RBX: ffff88814ae02000 RCX: 0000000000000000 RDX: 0000000000000000 RSI: 0000000000000004 RDI: ffff88814ae026f0 RBP: 0000000000000000 R08: 0000000000000000 R09: fffffbfff1d57fdc R10: ffffffff8eabfee3 R11: ffffffff8aa0008b R12: 0000000000000000 R13: ffff88814ae02000 R14: dffffc0000000000 R15: 0000000000000001 FS: 000055555717a380(0000) GS:ffff8880b9800000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000002555398 CR3: 0000000025044000 CR4: 00000000003506f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: <TASK> __page_pool_destroy net/core/page_pool.c:851 [inline] page_pool_release+0x507/0x6b0 net/core/page_pool.c:891 page_pool_destroy+0x1ac/0x4c0 net/core/page_pool.c:956 xdp_test_run_teardown net/bpf/test_run.c:216 [inline] bpf_test_run_xdp_live+0x1578/0x1af0 net/bpf/test_run.c:388 bpf_prog_test_run_xdp+0x827/0x1530 net/bpf/test_run.c:1254 bpf_prog_test_run kernel/bpf/syscall.c:4041 [inline] __sys_bpf+0x11bf/0x4920 kernel/bpf/syscall.c:5402 __do_sys_bpf kernel/bpf/syscall.c:5488 [inline] __se_sys_bpf kernel/bpf/syscall.c:5486 [inline] __x64_sys_bpf+0x78/0xc0 kernel/bpf/syscall.c:5486 Fixes: 083772c9 ("net: page_pool: record pools per netdev") Reported-and-tested-by: syzbot+f9f8efb58a4db2ca98d0@syzkaller.appspotmail.com Signed-off-by:
Eric Dumazet <edumazet@google.com> Tested-by:
Andrew Lunn <andrew@lunn.ch> Link: https://lore.kernel.org/r/20231130092259.3797753-1-edumazet@google.comSigned-off-by:
Jakub Kicinski <kuba@kernel.org>
-
Paolo Abeni authored
Uwe Kleine-König says: ==================== net: ethernet: Convert to platform remove callback returning void in (implicit) v1 of this series (https://lore.kernel.org/netdev/20231117091655.872426-1-u.kleine-koenig@pengutronix.de) I tried to address the resource leaks in the three cpsw drivers. However this is hard to get right without being able to test the changes. So here comes a series that just converts all drivers below drivers/net/ethernet to use .remove_new() and adds a comment about the potential leaks for someone else to fix the problem. See commit 5c5a7680 ("platform: Provide a remove callback that returns no value") for an extended explanation and the eventual goal. The TL;DR; is to prevent bugs like the three noticed here. Note this series results in no change of behaviour apart from improving the error message for the three cpsw drivers from remove callback returned a non-zero value. This will be ignored. to Failed to resume device (-ESOMETHING) ==================== Link: https://lore.kernel.org/r/20231128173823.867512-1-u.kleine-koenig@pengutronix.deSigned-off-by:
Paolo Abeni <pabeni@redhat.com>
-
Uwe Kleine-König authored
The .remove() callback for a platform driver returns an int which makes many driver authors wrongly assume it's possible to do error handling by returning an error code. However the value returned is ignored (apart from emitting a warning) and this typically results in resource leaks. To improve here there is a quest to make the remove callback return void. In the first step of this quest all drivers are converted to .remove_new(), which already returns void. Eventually after all drivers are converted, .remove_new() will be renamed to .remove(). Trivially convert this driver from always returning zero in the remove callback to the void returning variant. Signed-off-by:
-
Uwe Kleine-König authored
The .remove() callback for a platform driver returns an int which makes many driver authors wrongly assume it's possible to do error handling by returning an error code. However the value returned is ignored (apart from emitting a warning) and this typically results in resource leaks. To improve here there is a quest to make the remove callback return void. In the first step of this quest all drivers are converted to .remove_new(), which already returns void. Eventually after all drivers are converted, .remove_new() will be renamed to .remove(). Replace the error path returning a non-zero value by an error message and a comment that there is more to do. With that this patch results in no change of behaviour in this driver apart from improving the error message. Signed-off-by:
Roger Quadros <rogerq@kernel.org> Signed-off-by:
-
Uwe Kleine-König authored
The .remove() callback for a platform driver returns an int which makes many driver authors wrongly assume it's possible to do error handling by returning an error code. However the value returned is ignored (apart from emitting a warning) and this typically results in resource leaks. To improve here there is a quest to make the remove callback return void. In the first step of this quest all drivers are converted to .remove_new(), which already returns void. Eventually after all drivers are converted, .remove_new() will be renamed to .remove(). Replace the error path returning a non-zero value by an error message and a comment that there is more to do. With that this patch results in no change of behaviour in this driver apart from improving the error message. Signed-off-by:
-
Uwe Kleine-König authored
The .remove() callback for a platform driver returns an int which makes many driver authors wrongly assume it's possible to do error handling by returning an error code. However the value returned is ignored (apart from emitting a warning) and this typically results in resource leaks. To improve here there is a quest to make the remove callback return void. In the first step of this quest all drivers are converted to .remove_new(), which already returns void. Eventually after all drivers are converted, .remove_new() will be renamed to .remove(). Replace the error path returning a non-zero value by an error message and a comment that there is more to do. With that this patch results in no change of behaviour in this driver apart from improving the error message. Signed-off-by:
-
Paolo Abeni authored
Jiri Pirko says: ==================== devlink: warn about existing entities during reload-reinit Recently there has been a couple of attempts from drivers to block devlink reload in certain situations. Turned out, the drivers do not properly tear down ports and related netdevs during reload. To address this, add couple of checks to be done during devlink reload reinit action. Also, extend documentation to be more explicit. ==================== Link: https://lore.kernel.org/r/20231128115255.773377-1-jiri@resnulli.usSigned-off-by:
-
Jiri Pirko authored
During reload-reinit, all entities except for params, resources, regions and health reporter should be removed and re-added. Add a warning to be triggered in case the driver behaves differently. Signed-off-by:
Jiri Pirko <jiri@nvidia.com> Reviewed-by:
Przemek Kitszel <przemyslaw.kitszel@intel.com> Signed-off-by:
-
Jiri Pirko authored
Be more explicit about devlink entities that may stay and that have to be removed during reload reinit action. Signed-off-by:
-
Jakub Kicinski authored
Kuniyuki Iwashima says: ==================== tcp: Clean up and refactor cookie_v[46]_check(). This is a preparation series for upcoming arbitrary SYN Cookie support with BPF. [0] There are slight differences between cookie_v[46]_check(). Such a discrepancy caused an issue in the past, and BPF SYN Cookie support will add more churn. The primary purpose of this series is to clean up and refactor cookie_v[46]_check() to minimise such discrepancies and make the BPF series easier to review. [0]: https://lore.kernel.org/netdev/20231121184245.69569-1-kuniyu@amazon.com/ v2: https://lore.kernel.org/netdev/20231125011638.72056-1-kuniyu@amazon.com/ v1: https://lore.kernel.org/netdev/20231123012521.62841-1-kuniyu@amazon.com/ ==================== Link: https://lore.kernel.org/r/20231129022924.96156-1-kuniyu@amazon.comSigned-off-by:
-
Kuniyuki Iwashima authored
We will support arbitrary SYN Cookie with BPF, and then kfunc at TC will preallocate reqsk and initialise some fields that should not be overwritten later by cookie_v[46]_check(). To simplify the flow in cookie_v[46]_check(), we move such fields' initialisation to cookie_tcp_reqsk_alloc() and factorise non-BPF SYN Cookie handling into cookie_tcp_check(), where we validate the cookie and allocate reqsk, as done by kfunc later. Note that we set ireq->ecn_ok in two steps, the latter of which will be shared by the BPF case. As cookie_ecn_ok() is one-liner, now it's inlined. Signed-off-by:
Kuniyuki Iwashima <kuniyu@amazon.com> Reviewed-by:
Simon Horman <horms@kernel.org> Reviewed-by:
-
Kuniyuki Iwashima authored
We will support arbitrary SYN Cookie with BPF, and then some reqsk fields are initialised in kfunc, and others are done in cookie_v[46]_check(). This patch factorises the common part as cookie_tcp_reqsk_init() and calls it in cookie_tcp_reqsk_alloc() to minimise the discrepancy between cookie_v[46]_check(). Signed-off-by:
-
Kuniyuki Iwashima authored
We initialise treq->af_specific in cookie_tcp_reqsk_alloc() so that we can look up a key later in tcp_create_openreq_child(). Initially, that change was added for MD5 by commit ba5a4fdd ("tcp: make sure treq->af_specific is initialized"), but it has not been used since commit d0f2b7a9 ("tcp: Disable header prediction for MD5 flow."). Now, treq->af_specific is used only by TCP-AO, so, we can move that initialisation into tcp_ao_syncookie(). In addition to that, l3index in cookie_v[46]_check() is only used for tcp_ao_syncookie(), so let's move it as well. While at it, we move down tcp_ao_syncookie() in cookie_v4_check() so that it will be called after security_inet_conn_request() to make functions order consistent with cookie_v6_check(). Signed-off-by:
-
Kuniyuki Iwashima authored
When we create a full socket from SYN Cookie, we initialise tcp_sk(sk)->tsoffset redundantly in tcp_get_cookie_sock() as the field is inherited from tcp_rsk(req)->ts_off. cookie_v[46]_check |- treq->ts_off = 0 `- tcp_get_cookie_sock |- tcp_v[46]_syn_recv_sock | `- tcp_create_openreq_child | `- newtp->tsoffset = treq->ts_off `- tcp_sk(child)->tsoffset = tsoff Let's initialise tcp_rsk(req)->ts_off with the correct offset and remove the second initialisation of tcp_sk(sk)->tsoffset. Signed-off-by: -
Kuniyuki Iwashima authored
tcp_hdr(skb) and SYN Cookie are passed to __cookie_v[46]_check(), but none of the callers passes cookie other than ntohl(th->ack_seq) - 1. Let's fetch it in __cookie_v[46]_check() instead of passing the cookie over and over. Signed-off-by:
-
Kuniyuki Iwashima authored
We will support arbitrary SYN Cookie with BPF, and then reqsk will be preallocated before cookie_v[46]_check(). Depending on how validation fails, we send RST or just drop skb. To make the error handling easier, let's clean up goto labels. Signed-off-by:
-
Kuniyuki Iwashima authored
sock_net(sk) is used repeatedly in cookie_v[46]_check(). Let's cache it in a variable. Signed-off-by:
-
Kuniyuki Iwashima authored
We will grow and cut the xmas tree in cookie_v[46]_check(). This patch cleans it up to make later patches tidy. Signed-off-by:
-
Colin Ian King authored
There is a spelling mistake in struct field hc_tx_err_sqpdid_enforecement. Fix it. Signed-off-by:
Colin Ian King <colin.i.king@gmail.com> Signed-off-by:
Shradha Gupta <shradhagupta@linux.microsoft.com> Link: https://lore.kernel.org/r/20231128095304.515492-1-colin.i.king@gmail.comSigned-off-by:
-
Andy Shevchenko authored
BYTES_PER_KBIT is defined in units.h, use that definition. Signed-off-by:
Andy Shevchenko <andriy.shevchenko@linux.intel.com> Reviewed-by:
Vladimir Oltean <vladimir.oltean@nxp.com> Link: https://lore.kernel.org/r/20231128175027.394754-1-andriy.shevchenko@linux.intel.comSigned-off-by:
-
Jakub Kicinski authored
Mat Martineau says: ==================== mptcp: More selftest coverage and code cleanup for net-next Patches 1-5 and 7-8 add selftest coverage (and an associated subflow counter in the kernel) to validate the recently-updated handling of subflows with ID 0. Patch 6 renames a label in the userspace path manager for clarity. Patches 9-11 and 13-15 factor out common selftest code by moving certain functions to mptcp_lib.sh Patch 12 makes sure the random data file generated for selftest payloads has the intended size. v3: https://lore.kernel.org/r/20231115-send-net-next-2023107-v3-0-1ef58145a882@kernel.org v2: https://lore.kernel.org/r/20231114-send-net-next-2023107-v2-0-b650a477362c@kernel.org v1: https://lore.kernel.org/r/20231027-send-net-next-2023107-v1-0-03eff9452957@kernel.org ==================== Link: https://lore.kernel.org/r/20231128-send-net-next-2023107-v4-0-8d6b94150f6b@kernel.orgSigned-off-by:
-
Geliang Tang authored
To avoid duplicated code in different MPTCP selftests, we can add and use helpers defined in mptcp_lib.sh. wait_local_port_listen() helper is defined in diag.sh, mptcp_connect.sh, mptcp_join.sh and simult_flows.sh, export it into mptcp_lib.sh and rename it with mptcp_lib_ prefix. Use this new helper in all these scripts. Note: We only have IPv4 connections in this helper, not looking at IPv6 (tcp6) but that's OK because we only have IPv4 connections here in diag.sh. Reviewed-by:
Matthieu Baerts <matttbe@kernel.org> Signed-off-by:
Geliang Tang <geliang.tang@suse.com> Signed-off-by:
Mat Martineau <martineau@kernel.org> Link: https://lore.kernel.org/r/20231128-send-net-next-2023107-v4-15-8d6b94150f6b@kernel.orgSigned-off-by:
-
Geliang Tang authored
To avoid duplicated code in different MPTCP selftests, we can add and use helpers defined in mptcp_lib.sh. check_transfer() and print_file_err() helpers are defined both in mptcp_connect.sh and mptcp_sockopt.sh, export them into mptcp_lib.sh and rename them with mptcp_lib_ prefix. And use them in all scripts. Note: In mptcp_sockopt.sh it is OK to drop 'ret=1' in check_transfer() because it will be set in run_tests() anyway. Reviewed-by:
-
Geliang Tang authored
To avoid duplicated code in different MPTCP selftests, we can add and use helpers defined in mptcp_lib.sh. make_file() helper in mptcp_sockopt.sh and userspace_pm.sh are the same. Export it into mptcp_lib.sh and rename it as mptcp_lib_kill_wait(). Use it in both mptcp_connect.sh and mptcp_join.sh. Reviewed-by:
-
Geliang Tang authored
In mptcp_connect.sh we are missing something like "oflag=append" because this will write "${rem}" bytes at the beginning of the file where there is already some random bytes. It should write that at the end. This patch adds this missing 'oflag=append' flag for 'dd' command in make_file(). Suggested-by: -
Geliang Tang authored
To avoid duplicated code in different MPTCP selftests, we can add and use helpers defined in mptcp_lib.sh. The helper get_counter() in mptcp_join.sh and get_mib_counter() in mptcp_connect.sh have the same functionality, export get_counter() into mptcp_lib.sh and rename it as mptcp_lib_get_counter(). Use this new helper instead of get_counter() and get_mib_counter(). Use this helper in test_prio() in userspace_pm.sh too instead of open-coding. Reviewed-by:
-
Geliang Tang authored
To avoid duplicated code in different MPTCP selftests, we can add and use helpers defined in mptcp_lib.sh. is_v6() helper is defined in mptcp_connect.sh, mptcp_join.sh and mptcp_sockopt.sh, so export it into mptcp_lib.sh and rename it as mptcp_lib_is_v6(). Use this new helper in all scripts. Reviewed-by:
-
Geliang Tang authored
To avoid duplicated code in different MPTCP selftests, we can add and use helpers defined in mptcp_lib.sh. Export kill_wait() helper in userspace_pm.sh into mptcp_lib.sh and rename it as mptcp_lib_kill_wait(). It can be used to instead of kill_wait() in mptcp_join.sh. Use the new helper in both scripts. Reviewed-by:
-
Geliang Tang authored
This patch adds a selftest for userspace PM to remove id 0 address. Use userspace_pm_add_addr() helper to add an id 10 address, then use userspace_pm_rm_addr() helper to remove id 0 address. Reviewed-by:
-
Geliang Tang authored
This patch adds a selftest for userspace PM to remove the initial subflow. Use userspace_pm_add_sf() to add a subflow, and pass initial IP address to userspace_pm_rm_sf() to remove the initial subflow. Reviewed-by:
-
Geliang Tang authored
The value of 'err' will not be only '-EINVAL', but can be '0' in some cases. So it's better to rename the label 'remove_err' to 'out' to avoid confusions. Suggested-by:
-
Geliang Tang authored
This patch adds a selftest to create id 0 subflow. Pass id 0 to the helper userspace_pm_add_sf() to create id 0 subflow. chk_mptcp_info shows one subflow but chk_subflows_total shows two subflows in each namespace. Reviewed-by:
-
Geliang Tang authored
This patch adds a new argument namespace to userspace_pm_add_addr() and userspace_pm_add_sf() to make these two helper more versatile. Add two more versatile helpers for userspace pm remove subflow or address: userspace_pm_rm_addr() and userspace_pm_rm_sf(). The original test helpers userspace_pm_rm_sf_addr_ns1() and userspace_pm_rm_sf_addr_ns2() can be replaced by these new helpers. Reviewed-by:
-
Geliang Tang authored
This patch adds a new helper chk_subflows_total(), in it use the newly added counter mptcpi_subflows_total to get the "correct" amount of subflows, including the initial one. To be compatible with old 'ss' or kernel versions not supporting this counter, get the total subflows by listing TCP connections that are MPTCP subflows: ss -ti state state established state syn-sent state syn-recv | grep -c tcp-ulp-mptcp. Reviewed-by: -
Geliang Tang authored
This patch adds a new helper get_info_value(), using 'sed' command to parse the value of the given item name in the line with the given keyword, to make chk_mptcp_info() and pedit_action_pkts() more readable. Also add another helper evts_get_info() to use get_info_value() to parse the output of 'pm_nl_ctl events' command, to make all the userspace pm selftests more readable, both in mptcp_join.sh and userspace_pm.sh. Reviewed-by:
-
Geliang Tang authored
If the initial subflow has been removed, we cannot know without checking other counters, e.g. ss -ti <filter> | grep -c tcp-ulp-mptcp or getsockopt(SOL_MPTCP, MPTCP_FULL_INFO, ...) (or others except MPTCP_INFO of course) and then check mptcp_subflow_data->num_subflows to get the total amount of subflows. This patch adds a new counter mptcpi_subflows_total in mptcpi_flags to store the total amount of subflows, including the initial one. A new helper __mptcp_has_initial_subflow() is added to check whether the initial subflow has been removed or not. With this helper, we can then compute the total amount of subflows from mptcp_info by doing something like: mptcpi_subflows_total = mptcpi_subflows + __mptcp_has_initial_subflow(msk). Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/428Reviewed-by: -
Jakub Kicinski authored
Petr Machata says: ==================== mlxsw: Support CFF flood mode The registers to configure to initialize a flood table differ between the controlled and CFF flood modes. In therefore needs to be an op. Add it, hook up the current init to the existing families, and invoke the op. PGT is an in-HW table that maps addresses to sets of ports. Then when some HW process needs a set of ports as an argument, instead of embedding the actual set in the dynamic configuration, what gets configured is the address referencing the set. The HW then works with the appropriate PGT entry. Among other allocations, the PGT currently contains two large blocks for bridge flooding: one for 802.1q and one for 802.1d. Within each of these blocks are three tables, for unknown-unicast, multicast and broadcast flooding: . . . | 802.1q | 802.1d | . . . | UC | MC | BC | UC | MC | BC | \______ _____/ \_____ ______/ v v FID flood vectors Thus each FID (which corresponds to an 802.1d bridge or one VLAN in an 802.1q bridge) uses three flood vectors spread across a fairly large region of PGT. This way of organizing the flood table (called "controlled") is not very flexible. E.g. to decrease a bridge scale and store more IP MC vectors, one would need to completely rewrite the bridge PGT blocks, or resort to hacks such as storing individual MC flood vectors into unused part of the bridge table. In order to address these shortcomings, Spectrum-2 and above support what is called CFF flood mode, for Compressed FID Flooding. In CFF flood mode, each FID has a little table of its own, with three entries adjacent to each other, one for unknown-UC, one for MC, one for BC. This allows for a much more fine-grained approach to PGT management, where bits of it are allocated on demand. . . . | FID | FID | FID | FID | FID | . . . |U|M|B|U|M|B|U|M|B|U|M|B|U|M|B| \_____________ _____________/ v FID flood vectors Besides the FID table organization, the CFF flood mode also impacts Router Subport (RSP) table. This table contains flood vectors for rFIDs, which are FIDs that reference front panel ports or LAGs. The RSP table contains two entries per front panel port and LAG, one for unknown-UC traffic, and one for everything else. Currently, the FW allocates and manages the table in its own part of PGT. rFIDs are marked with flood_rsp bit and managed specially. In CFF mode, rFIDs are managed as all other FIDs. The driver therefore has to allocate and maintain the flood vectors. Like with bridge FIDs, this is more work, but increases flexibility of the system. The FW currently supports both the controlled and CFF flood modes. To shed complexity, in the future it should only support CFF flood mode. Hence this patchset, which adds CFF flood mode support to mlxsw. Since mlxsw needs to maintain both the controlled mode as well as CFF mode support, we will keep the layout as compatible as possible. The bridge tables will stay in the same overall shape, just their inner organization will change from flood mode -> FID to FID -> flood mode. Likewise will RSP be kept as a contiguous block of PGT memory, as was the case when the FW maintained it. - The way FIDs get configured under the CFF flood mode differs from the currently used controlled mode. The simple approach of having several globally visible arrays for spectrum.c to statically choose from no longer works. Patch #1 thus privatizes all FID initialization and finalization logic, and exposes it as ops instead. - Patch #2 renames the ops that are specific to the controlled mode, to make room in the namespace for the CFF variants. Patch #3 extracts a helper to compute flood table base out of mlxsw_sp_fid_flood_table_mid(). - The op fid_setup configured fid_offset, i.e. the number of this FID within its family. For rFIDs in CFF mode, to determine this number, the driver will need to do fallible queries. Thus in patch #4, make the FID setup operation fallible as well. - Flood mode initialization routine differs between the controlled and CFF flood modes. The controlled mode needs to configure flood table layout, which the CFF mode does not need to do. In patch #5, move mlxsw_sp_fid_flood_table_init() up so that the following patch can make use of it. In patch #6, add an op to be invoked per table (if defined). - The current way of determining PGT allocation size depends on the number of FIDs and number of flood tables. RFIDs however have PGT footprint depending not on number of FIDs, but on number of ports and LAGs, because which ports an rFID should flood to does not depend on the FID itself, but on the port or LAG that it references. Therefore in patch #7, add FID family ops for determining PGT allocation size. - As elaborated above, layout of PGT will differ between controlled and CFF flood modes. In CFF mode, it will further differ between rFIDs and other FIDs (as described at previous patch). The way to pack the SFMR register to configure a FID will likewise differ from controlled to CFF. Thus in patches #8 and #9 add FID family ops to determine PGT base address for a FID and to pack SFMR. - Patches #10 and #11 add more bits for RSP support. In patch #10, add a new traffic type enumerator, for non-UC traffic. This is a combination of BC and MC traffic, but the way that mlxsw maps these mnemonic names to actual traffic type configurations requires that we have a new name to describe this class of traffic. Patch #11 then adds hooks necessary for RSP table maintenance. As ports come and go, and join and leave LAGs, it is necessary to update flood vectors that the rFIDs use. These new hooks will make that possible. - Patches #12, #13 and #14 introduce flood profiles. These have been implicit so far, but the way that CFF flood mode works with profile IDs requires that we make them explicit. Thus in patch #12, introduce flood profile objects as a set of flood tables that FID families then refer to. The FID code currently only uses a single flood profile. In patch #13, add a flood profile ID to flood profile objects. In patch #14, when in CFF mode, configure SFFP according to the existing flood profiles (or the one that exists as of that point). - Patches #15 and #16 add code to implement, respectively, bridge FIDs and RSP FIDs in CFF mode. - In patch #17, toggle flood_mode_prefer_cff on Spectrum-2 and above, which makes the newly-added code live. ==================== Link: https://lore.kernel.org/r/cover.1701183891.git.petrm@nvidia.comSigned-off-by: -
Petr Machata authored
Mark all Spectrum>2 systems as preferring CFF flood mode if supported by the firmware. Signed-off-by:
Petr Machata <petrm@nvidia.com> Reviewed-by:
Amit Cohen <amcohen@nvidia.com> Reviewed-by:
Ido Schimmel <idosch@nvidia.com> Link: https://lore.kernel.org/r/8a3d2ad96b943f7e3f53f998bd333a14e19cd641.1701183892.git.petrm@nvidia.comSigned-off-by:
-
