| Age | Commit message (Collapse) | Author |
|---|
|
WQEs must not cross page boundaries, they are padded with NOPs if they
don't fit the page. mlx5e_mpwrq_total_umr_wqebbs doesn't take into
account this padding, risking reserving not enough space.
The padding is not straightforward to add to this calculation, because
WQEs of different sizes may be mixed together in the queue. If each page
ends with a big WQE that doesn't fit and requires at most its size minus
1 WQEBB of padding, the total space can be much bigger than in case when
smaller WQEs take advantage of this padding.
Replace the wrong exact calculation by the following estimation. Each
padding can be at most the size of the maximum WQE used in the queue
minus one WQEBB. Let's call the rest of the page "useful space". If we
divide the total size of all needed WQEs by this useful space, rounding
up, we'll get the number of pages, which is enough to contain all these
WQEs. It's correct, because every WQE that appeared on the boundary
between two blocks of useful space would start in the useful space of
one page and end in the padding of the same page, while our estimation
reserved space for its tail in the next space, making the estimation not
smaller than the real space occupied in the queue.
The code actually uses a looser estimation: instead of taking the
maximum size of all used WQE types minus 1 WQEBB, it takes the maximum
hardware size of a WQE. It's made for simplicity and extensibility.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
UMR MTTs used in striding RQ have certain alignment requirements. While
it's guaranteed to work when UMR pages are aligned to the UMR page size,
in practice it works then UMR pages are aligned to 8 bytes. However,
it's still not enough flexibility for the unaligned mode of XSK. This
patch leverages KSM to map UMR pages without alignment requirements,
when unaligned XSK is active. The downside is that KSM entries are twice
as big as MTTs, which limits the maximum WQE size, so regular RQs and
aligned XSK continue using MTTs.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Some commands use a flexible array after a common header. Add a macro to
safely calculate the total input length of the command, detecting
overflows and printing errors with specific values when such overflows
happen.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Currently, rq->mkey_be keeps a big-endian value of either the PA MKey
(for legacy RQ, no address translation) or MTT MKey (for striding RQ,
direct address translation). Striding RQ stores the same value in
rq->umr_mkey in the native endianness.
The next commit will make striding RQ use KSM MKey (indirect address
translation) for the unaligned mode of XSK, which will require storing
both KSM MKey and PA MKey in the RQ struct. This commit optimizes fields
of mlx5e_rq: umr_mkey is removed (it's redundant), mkey_be always points
to the PA MKey, and mpwqe.umr_mkey_be points to the MTT MKey (or to the
KSM MKey, starting from the next commit).
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
XSK RQs support striding RQ linear mode, but the stride size is always
set to PAGE_SIZE. It may be larger than the XSK frame size,
unnecessarily reducing the useful space in a WQE, but more importantly
causing UMEM data corruption in certain cases.
Normally, stride size bigger than XSK frame size is not a problem if the
hardware enforces the MTU. However, traffic between vports skips the
hardware MTU check, and oversized packets may be received.
If an oversized packet is bigger than the XSK frame but not bigger than
the stride, it will cause overwriting of the adjacent UMEM region. If
the packet takes more than one stride, they can be recycled for reuse
so it's not a problem when the XSK frame size matches the stride size.
To reduce the impact of the above issue, attempt to use the MTT page
size for striding RQ that matches the XSK frame size, allowing to safely
use 2048-byte frames on an up-to-date firmware.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
This commit allows striding RQ to determine MTT page size at runtime,
instead of sticking to the compile-time PAGE_SIZE. This functionality
will be used by a following commit that adjusts the MTT page size to the
XSK frame size.
Stick with PAGE_SIZE for XSK on legacy RQ, as frag_stride is not used in
data path, it only helps calculate how pages are partitioned into
fragments, and PAGE_SIZE will ensure each fragment starts at the
beginning of a new allocation unit (XSK frame).
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Parse new property 'snps,clk-csr' firstly because the new property
is documented in binding file, if failed, fall back to old property
'clk_csr' for legacy case
Signed-off-by: Jianguo Zhang <jianguo.zhang@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
There is a spelling mistake in a devlink_health_report message. Fix it.
Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
There is a spelling mistake in a literal string in the array
bnad_net_stats_strings. Fix it.
Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Implement channel management functions to allow dynamic addition and
removal of transmit queues. The `ethtool --show-channels` and
`ethtool --set-channels` commands can be used to get and set the
number of queues, respectively. Allow the ability to add as many
transmit queues as available processors but never allow more than the
hard maximum of 16. The number of receive queues is one and cannot be
modified.
Depending on whether the requested number of queues is larger or
smaller than the current value, either allocate or free long term
buffers. Since long term buffer construction and destruction can
occur in two different areas, from either channel set requests or
device open/close, define functions for performing this work. If
allocation of a new buffer fails, then attempt to revert back to the
previous number of queues.
Signed-off-by: Nick Child <nnac123@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
The `ndo_start_xmit` function is protected by a spinlock on the tx queue
being used to transmit the skb. Allow concurrent calls to
`ndo_start_xmit` by using more than one tx queue. This allows for
greater throughput when several jobs are trying to transmit data.
Introduce 16 tx queues (leave single rx queue as is) which each
correspond to one DMA mapped long term buffer.
Signed-off-by: Nick Child <nnac123@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Rather than DMA mapping and unmapping every outgoing skb, copy the skb
into a buffer that was mapped during the drivers open function. Copying
the skb and its frags have proven to be more time efficient than
mapping and unmapping. As an effect, performance increases by 3-5
Gbits/s.
Allocate and DMA map one continuous 64KB buffer at `ndo_open`. This
buffer is maintained until `ibmveth_close` is called. This buffer is
large enough to hold the largest possible linnear skb. During
`ndo_start_xmit`, copy the skb and all of it's frags into the continuous
buffer. By manually linnearizing all the socket buffers, time is saved
during memcpy as well as more efficient handling in FW.
As a result, we no longer need to worry about the firmware limitation
of handling a max of 6 frags. So, we only need to maintain 1 descriptor
instead of 6 and can hardcode 0 for the other 5 descriptors during
h_send_logical_lan.
Since, DMA allocation/mapping issues can no longer arise in xmit
functions, we can further reduce code size by removing the need for a
bounce buffer on DMA errors.
Signed-off-by: Nick Child <nnac123@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
There are spelling mistakes in two literal strings. Fix these.
Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
There is a spelling mistake in a netdev_err message. Fix it.
Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Reviewed-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Follow the advice of the Documentation/filesystems/sysfs.rst and show()
should only use sysfs_emit() or sysfs_emit_at() when formatting the value
to be returned to user space.
Signed-off-by: Wang Yufen <wangyufen@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Follow the advice of the Documentation/filesystems/sysfs.rst and show()
should only use sysfs_emit() or sysfs_emit_at() when formatting the value
to be returned to user space.
Signed-off-by: Wang Yufen <wangyufen@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Use page pool for RX buffer handling. Makes RX path more efficient and
is required prework for future XDP support.
Signed-off-by: Gerhard Engleder <gerhard@engleder-embedded.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Received Ethernet frames are assigned to first RX queue per default.
Based on EtherType Ethernet frames can be assigned to other RX queues.
This enables processing of real-time Ethernet protocols on dedicated
RX queues.
Add RX flow classification interface for EtherType based RX queue
assignment.
Signed-off-by: Gerhard Engleder <gerhard@engleder-embedded.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Support additional TX/RX queue pairs if dedicated interrupt is
available. Interrupts are detected by name in device tree.
Signed-off-by: Gerhard Engleder <gerhard@engleder-embedded.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
For multiple queues multiple interrupts shall be used. Therefore, rework
global interrupt to per queue interrupt.
Every interrupt name shall contain interface name and queue information.
To get a valid interface name, the interrupt request needs to by done
during open like in other drivers. Additionally, this allows the removal
of some initialisation checks in the interrupt handler.
Signed-off-by: Gerhard Engleder <gerhard@engleder-embedded.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Commit 26f3a021b37c ("ath11k: allocate smaller chunks of memory for
firmware") and commit f6f92968e1e5 ("ath11k: qmi: try to allocate a
big block of DMA memory first") change ath11k to allocate the memory
chunks for target twice while wlan load. It fails for the 1st time
because of large memory and then changed to allocate many small chunks
for the 2nd time sometimes as below log.
1st time failed:
[10411.640620] ath11k_pci 0000:05:00.0: qmi firmware request memory request
[10411.640625] ath11k_pci 0000:05:00.0: qmi mem seg type 1 size 6881280
[10411.640630] ath11k_pci 0000:05:00.0: qmi mem seg type 4 size 3784704
[10411.640658] ath11k_pci 0000:05:00.0: qmi dma allocation failed (6881280 B type 1), will try later with small size
[10411.640671] ath11k_pci 0000:05:00.0: qmi delays mem_request 2
[10411.640677] ath11k_pci 0000:05:00.0: qmi respond memory request delayed 1
2nd time success:
[10411.642004] ath11k_pci 0000:05:00.0: qmi firmware request memory request
[10411.642008] ath11k_pci 0000:05:00.0: qmi mem seg type 1 size 524288
[10411.642012] ath11k_pci 0000:05:00.0: qmi mem seg type 1 size 524288
[10411.642014] ath11k_pci 0000:05:00.0: qmi mem seg type 1 size 524288
[10411.642016] ath11k_pci 0000:05:00.0: qmi mem seg type 1 size 524288
[10411.642018] ath11k_pci 0000:05:00.0: qmi mem seg type 1 size 524288
[10411.642020] ath11k_pci 0000:05:00.0: qmi mem seg type 1 size 524288
[10411.642022] ath11k_pci 0000:05:00.0: qmi mem seg type 1 size 524288
[10411.642024] ath11k_pci 0000:05:00.0: qmi mem seg type 1 size 524288
[10411.642027] ath11k_pci 0000:05:00.0: qmi mem seg type 1 size 524288
[10411.642029] ath11k_pci 0000:05:00.0: qmi mem seg type 1 size 524288
[10411.642031] ath11k_pci 0000:05:00.0: qmi mem seg type 1 size 458752
[10411.642033] ath11k_pci 0000:05:00.0: qmi mem seg type 1 size 131072
[10411.642035] ath11k_pci 0000:05:00.0: qmi mem seg type 4 size 524288
[10411.642037] ath11k_pci 0000:05:00.0: qmi mem seg type 4 size 524288
[10411.642039] ath11k_pci 0000:05:00.0: qmi mem seg type 4 size 524288
[10411.642041] ath11k_pci 0000:05:00.0: qmi mem seg type 4 size 524288
[10411.642043] ath11k_pci 0000:05:00.0: qmi mem seg type 4 size 524288
[10411.642045] ath11k_pci 0000:05:00.0: qmi mem seg type 4 size 524288
[10411.642047] ath11k_pci 0000:05:00.0: qmi mem seg type 4 size 491520
[10411.642049] ath11k_pci 0000:05:00.0: qmi mem seg type 1 size 524288
And then commit 5962f370ce41 ("ath11k: Reuse the available memory after
firmware reload") skip the ath11k_qmi_free_resource() which frees the
memory chunks while recovery, after that, when run recovery test on
WCN6855, a warning happened every time as below and finally leads fail
for recovery.
[ 159.570318] BUG: Bad page state in process kworker/u16:5 pfn:33300
[ 159.570320] page:0000000096ffdbb9 refcount:1 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x33300
[ 159.570324] flags: 0xfffffc0000000(node=0|zone=1|lastcpupid=0x1fffff)
[ 159.570329] raw: 000fffffc0000000 0000000000000000 dead000000000122 0000000000000000
[ 159.570332] raw: 0000000000000000 0000000000000000 00000001ffffffff 0000000000000000
[ 159.570334] page dumped because: nonzero _refcount
[ 159.570440] firewire_ohci syscopyarea sysfillrect psmouse sdhci_pci ahci sysimgblt firewire_core fb_sys_fops libahci crc_itu_t cqhci drm sdhci e1000e wmi video
[ 159.570460] CPU: 2 PID: 217 Comm: kworker/u16:5 Kdump: loaded Tainted: G B 5.19.0-rc1-wt-ath+ #3
[ 159.570465] Hardware name: LENOVO 418065C/418065C, BIOS 83ET63WW (1.33 ) 07/29/2011
[ 159.570467] Workqueue: qmi_msg_handler qmi_data_ready_work [qmi_helpers]
[ 159.570475] Call Trace:
[ 159.570476] <TASK>
[ 159.570478] dump_stack_lvl+0x49/0x5f
[ 159.570486] dump_stack+0x10/0x12
[ 159.570493] bad_page+0xab/0xf0
[ 159.570502] check_free_page_bad+0x66/0x70
[ 159.570511] __free_pages_ok+0x530/0x9a0
[ 159.570517] ? __dev_printk+0x58/0x6b
[ 159.570525] ? _dev_printk+0x56/0x72
[ 159.570534] ? qmi_decode+0x119/0x470 [qmi_helpers]
[ 159.570543] __free_pages+0x91/0xd0
[ 159.570548] dma_free_contiguous+0x50/0x60
[ 159.570556] dma_direct_free+0xe5/0x140
[ 159.570564] dma_free_attrs+0x35/0x50
[ 159.570570] ath11k_qmi_msg_mem_request_cb+0x2ae/0x3c0 [ath11k]
[ 159.570620] qmi_invoke_handler+0xac/0xe0 [qmi_helpers]
[ 159.570630] qmi_handle_message+0x6d/0x180 [qmi_helpers]
[ 159.570643] qmi_data_ready_work+0x2ca/0x440 [qmi_helpers]
[ 159.570656] process_one_work+0x227/0x440
[ 159.570667] worker_thread+0x31/0x3d0
[ 159.570676] ? process_one_work+0x440/0x440
[ 159.570685] kthread+0xfe/0x130
[ 159.570692] ? kthread_complete_and_exit+0x20/0x20
[ 159.570701] ret_from_fork+0x22/0x30
[ 159.570712] </TASK>
The reason is because when wlan start to recovery, the type, size and
count is not same for the 1st and 2nd QMI_WLFW_REQUEST_MEM_IND message,
Then it leads the parameter size is not correct for the dma_free_coherent().
For the chunk[1], the actual dma size is 524288 which allocate in the
2nd time of the initial wlan load phase, and the size which pass to
dma_free_coherent() is 3784704 which is got in the 1st time of recovery
phase, then warning above happened.
Change to use prev_size of struct target_mem_chunk for the paramter of
dma_free_coherent() since prev_size is the real size of last load/recovery.
Also change to check both type and size of struct target_mem_chunk to
reuse the memory to avoid mismatch buffer size for target. Then the
warning disappear and recovery success. When the 1st QMI_WLFW_REQUEST_MEM_IND
for recovery arrived, the trunk[0] is freed in ath11k_qmi_alloc_target_mem_chunk()
and then dma_alloc_coherent() failed caused by large size, and then
trunk[1] is freed in ath11k_qmi_free_target_mem_chunk(), the left 18
trunks will be reuse for the 2nd QMI_WLFW_REQUEST_MEM_IND message.
Tested-on: WCN6855 hw2.0 PCI WLAN.HSP.1.1-03125-QCAHSPSWPL_V1_V2_SILICONZ_LITE-3
Fixes: 5962f370ce41 ("ath11k: Reuse the available memory after firmware reload")
Signed-off-by: Wen Gong <quic_wgong@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20220928073832.16251-1-quic_wgong@quicinc.com
|
|
Currently, ath11k_pcic_read() returns an error if wakeup()
fails, this makes firmware crash debug quite hard because we can
get nothing.
Change to go ahead on wakeup failure, in that case we still may
get something valid to check. There should be no mislead due
to incorrect content because we are aware of the failure with the
log printed.
Tested-on: WCN6855 hw2.0 PCI WLAN.HSP.1.1-01720.1-QCAHSPSWPL_V1_V2_SILICONZ_LITE-1
Signed-off-by: Baochen Qiang <quic_bqiang@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20220928015140.5431-1-quic_bqiang@quicinc.com
|
|
Zero-length arrays are deprecated and we are moving towards adopting
C99 flexible-array members, instead. So, replace zero-length arrays
declarations in anonymous union with the new DECLARE_FLEX_ARRAY()
helper macro.
This helper allows for flexible-array members in unions.
Link: https://github.com/KSPP/linux/issues/193
Link: https://github.com/KSPP/linux/issues/215
Link: https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html
Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/YzIdWc8QSdZFHBYg@work
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/next-queue
Tony Nguyen says:
====================
Intel Wired LAN Driver Updates 2022-09-28 (ice)
Arkadiusz implements a single pin initialization function, checking feature
bits, instead of having separate device functions and updates sub-device
IDs for recognizing E810T devices.
Martyna adds support for switchdev filters on VLAN priority field.
* '100GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/next-queue:
ice: Add support for VLAN priority filters in switchdev
ice: support features on new E810T variants
ice: Merge pin initialization of E810 and E810T adapters
====================
Link: https://lore.kernel.org/r/20220928203217.411078-1-anthony.l.nguyen@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Zbynek reports that alx trips an rtnl assertion on resume:
RTNL: assertion failed at net/core/dev.c (2891)
RIP: 0010:netif_set_real_num_tx_queues+0x1ac/0x1c0
Call Trace:
<TASK>
__alx_open+0x230/0x570 [alx]
alx_resume+0x54/0x80 [alx]
? pci_legacy_resume+0x80/0x80
dpm_run_callback+0x4a/0x150
device_resume+0x8b/0x190
async_resume+0x19/0x30
async_run_entry_fn+0x30/0x130
process_one_work+0x1e5/0x3b0
indeed the driver does not hold rtnl_lock during its internal close
and re-open functions during suspend/resume. Note that this is not
a huge bug as the driver implements its own locking, and does not
implement changing the number of queues, but we need to silence
the splat.
Fixes: 4a5fe57e7751 ("alx: use fine-grained locking instead of RTNL")
Reported-and-tested-by: Zbynek Michl <zbynek.michl@gmail.com>
Reviewed-by: Niels Dossche <dossche.niels@gmail.com>
Link: https://lore.kernel.org/r/20220928181236.1053043-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Follow the advice of the Documentation/filesystems/sysfs.rst and show()
should only use sysfs_emit() or sysfs_emit_at() when formatting the value
to be returned to user space.
Signed-off-by: Wang Yufen <wangyufen@huawei.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/1664364860-29153-1-git-send-email-wangyufen@huawei.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
The driver currently sets the PTCMSDUR register statically to the max
MTU supported by the interface. Keep this logic if tc-taprio is absent
or if the max_sdu for a traffic class is 0, and follow the requested max
SDU size otherwise.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
The Port Time Gating Control Register (PTGCR) and Port Time Gating
Capability Register (PTGCAPR) have definitions in the driver which
aren't in line with the other registers. Rename these.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
The &priv->si->hw construct dereferences 2 pointers and makes lines
longer than they need to be, in turn making the code harder to read.
Replace &priv->si->hw accesses with a "hw" variable when there are 2 or
more accesses within a function that dereference this. This includes
loops, since &priv->si->hw is a loop invariant.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Add support for configuring the max SDU per priority and per port. If not
specified, keep the default.
Signed-off-by: Kurt Kanzenbach <kurt@linutronix.de>
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
The following patch will need to make this function also respond to
TC_QUERY_BASE, so make the processing more structured around the
tc_setup_type.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Kurt Kanzenbach <kurt@linutronix.de>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Our current vsc9959_tas_guard_bands_update() algorithm has a limitation
imposed by the hardware design. To avoid packet overruns between one
gate interval and the next (which would add jitter for scheduled traffic
in the next gate), we configure the switch to use guard bands. These are
as large as the largest packet which is possible to be transmitted.
The problem is that at tc-taprio intervals of sizes comparable to a
guard band, there isn't an obvious place in which to split the interval
between the useful portion (for scheduling) and the guard band portion
(where scheduling is blocked).
For example, a 10 us interval at 1Gbps allows 1225 octets to be
transmitted. We currently split the interval between the bare minimum of
33 ns useful time (required to schedule a single packet) and the rest as
guard band.
But 33 ns of useful scheduling time will only allow a single packet to
be sent, be that packet 1200 octets in size, or 60 octets in size. It is
impossible to send 2 60 octets frames in the 10 us window. Except that
if we reduced the guard band (and therefore the maximum allowable SDU
size) to 5 us, the useful time for scheduling is now also 5 us, so more
packets could be scheduled.
The hardware inflexibility of not scheduling according to individual
packet lengths must unfortunately propagate to the user, who needs to
tune the queueMaxSDU values if he wants to fit more small packets into a
10 us interval, rather than one large packet.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
No conflicts.
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Add __init/__exit annotations to module init/exit funcs
Signed-off-by: ruanjinjie <ruanjinjie@huawei.com>
Acked-by: Florian Fainelli <f.fainelli@gmail.com>
Link: https://lore.kernel.org/r/20220928031708.89120-1-ruanjinjie@huawei.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
The 'mem' returned by platform_get_resource() has been checked in probe
function, so it is no need do this check in remove function.
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Link: https://lore.kernel.org/r/20220927151406.797800-1-yangyingliang@huawei.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
Use skb_put_data() instead of skb_put() and memcpy(), which is clear.
Signed-off-by: Shang XiaoJing <shangxiaojing@huawei.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Niklas Söderlund <niklas.soderlund@corigine.com>
Link: https://lore.kernel.org/r/20220927141835.19221-1-shangxiaojing@huawei.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
After commit 6870957ed5bc("liquidio: make soft command calls synchronous"), no
one use struct lio_trusted_vf_ctx, so remove it.
Signed-off-by: Yuan Can <yuancan@huawei.com>
Link: https://lore.kernel.org/r/20220927133940.104181-1-yuancan@huawei.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
Use DEFINE_SHOW_ATTRIBUTE helper macro to simplify the code.
No functional change.
Signed-off-by: Liu Shixin <liushixin2@huawei.com>
Link: https://lore.kernel.org/r/20220927111925.2424100-1-liushixin2@huawei.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
Use skb_put_data() instead of skb_put() and memcpy(), which is clear.
Signed-off-by: Shang XiaoJing <shangxiaojing@huawei.com>
Link: https://lore.kernel.org/r/20220927024511.14665-1-shangxiaojing@huawei.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
Use skb_put_data() instead of skb_put() and memcpy(), which is clear.
Signed-off-by: Shang XiaoJing <shangxiaojing@huawei.com>
Link: https://lore.kernel.org/r/20220927023043.17769-1-shangxiaojing@huawei.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
Use skb_put_data() instead of skb_put() and memcpy(), which is shorter
and clear. Drop the tmp variable that is not needed any more.
Signed-off-by: Shang XiaoJing <shangxiaojing@huawei.com>
Link: https://lore.kernel.org/r/20220927022802.16050-1-shangxiaojing@huawei.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
Remove the unused ieee80211_hw* parameter, and pass ieee80211_hdr*
instead of relying on skb->data having the right value at the time
the function is called.
This doesn't change the functionality at all.
Fixes: 26f1fad29ad9 ("New driver: rtl8xxxu (mac80211)")
Signed-off-by: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Acked-by: Jes Sorensen <jes@trained-monkey.org>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/2af44c28-1c12-46b9-85b9-011560bf7f7e@gmail.com
|
|
ieee80211_tx_queue_params.aifs is not supposed to be written directly
to the REG_EDCA_*_PARAM registers. Instead process it like the vendor
drivers do. It's kinda hacky but it works.
This change boosts the download speed and makes it more stable.
Tested with RTL8188FU but all the other supported chips should also
benefit.
Fixes: 26f1fad29ad9 ("New driver: rtl8xxxu (mac80211)")
Signed-off-by: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Acked-by: Jes Sorensen <jes@trained-monkey.org>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/038cc03f-3567-77ba-a7bd-c4930e3b2fad@gmail.com
|
|
The module parameter ht40_2g was supposed to enable 40 MHz operation,
but it didn't.
Tell the firmware about the channel width when updating the rate mask.
This makes it work with my gen 2 chip RTL8188FU.
I'm not sure if anything needs to be done for the gen 1 chips, if 40
MHz channel width already works or not. They update the rate mask with
a different structure which doesn't have a field for the channel width.
Also set the channel width correctly for sta_statistics.
Fixes: f653e69009c6 ("rtl8xxxu: Implement basic 8723b specific update_rate_mask() function")
Fixes: bd917b3d28c9 ("rtl8xxxu: fill up txrate info for gen1 chips")
Signed-off-by: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Acked-by: Jes Sorensen <jes@trained-monkey.org>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/3a950997-7580-8a6b-97a0-e0a81a135456@gmail.com
|
|
Some of the parameters of striding RQ are compile-time constants, but
they are going to become dynamically calculated at runtime in a
following commit. This commit prepares the datapath to take cached
runtime parameters, prefilled at queue creation.
New fields added to struct mlx5e_rq fit into an existing 7-byte hole.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
This commit moves the dma_info array to the end of struct mlx5e_mpw_info
to make it a flexible array. It also removes the intermediate struct
mlx5e_umr_dma_info, which used to contain only this array. The
flexibility of dma_info will allow to choose its size dynamically in a
following commit.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Normally, the MTU change requires reopening the channels, but it can be
skipped if the new MTU doesn't change any of the queue parameters and if
MTU is not used in the data path.
The shortcut is applicable to the non-linear mode of striding RQ,
because the only thing affected by MTU is the queue length. As ethtool
sets the queue length in packets, but striding RQ length is defined in
strides or bytes, we estimate the RQ length to be at least as big as the
requested number of MTU-sized packets, that's why it depends on MTU.
Improve the shortcut by actually checking whether the RQ length stayed
the same, instead of an intermediate step in the calculation.
As MTU also affects the SHAMPO parameters, skip the shortcut if SHAMPO
is in use.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
In a typical scenario, if an XSK socket is opened first, then an XDP
program is attached, mlx5e_validate_xsk_param will be called twice:
first on XSK bind, second on channel restart caused by enabling XDP. The
validation includes a call to mlx5e_rx_is_linear_skb, which checks the
presence of the XDP program.
The above means that mlx5e_rx_is_linear_skb might return true the first
time, but false the second time, as mlx5e_rx_get_linear_sz_skb's return
value will increase, because of a different headroom used with XDP.
As XSK RQs never exist without XDP, it would make sense to trick
mlx5e_rx_get_linear_sz_skb into thinking XDP is enabled at the first
check as well. This way, if MTU is too big, it would be detected on XSK
bind, without giving false hope to the userspace application.
However, it turns out that this check is too restrictive in the first
place. SKBs created on XDP_PASS on XSK RQs don't have any headroom. That
means that big MTUs filtered out on the first and the second checks
might actually work.
So, address this issue in the proper way, but taking into account the
absence of the SKB headroom on XSK RQs, when calculating the buffer
size.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
One of the checks in mlx5e_rx_is_linear_skb verifies that the RX buffer
fits into the XSK frame size. Remove the duplicating check from
mlx5e_validate_xsk_param. It allows to make mlx5e_rx_get_min_frag_sz
static.
Remove mlx5e_rx_is_xdp altogether, as its only usage is located in a
branch where xsk == NULL.
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Linear RX buffers must be big enough to fit the MTU-sized packet along
with the headroom. On the other hand, they must be small enough to fit
into a page (or into an XSK frame). A straightforward way to check
whether the linear mode is possible would be comparing the required
buffer size to PAGE_SIZE or XSK frame size.
Stride size in the linear mode is defined by the following constraints:
1. A stride is at least as big as the buffer size, and it's a power of
two.
2. If non-XSK XDP is enabled, the stride size is PAGE_SIZE, because
mlx5e requires each packet to be in its own page when XDP is in use. The
previous constraint is automatically fulfilled, because buffer size
can't be bigger than PAGE_SIZE.
3. XSK uses stride size equal to PAGE_SIZE, but the following commits
will allow it to use roundup_pow_of_two(XSK frame size), by allowing the
NIC's MMU to use page sizes not equal to the CPU page size.
This commit puts the above requirements and constraints straight to the
code in an attempt to simplify it and to prepare it for changes made in
the next patches.
For the reference, the old code uses an equivalent, but trickier
calculation (high-level simplified pseudocode):
if XDP or XSK:
mlx5e_rx_get_linear_frag_sz := max(buffer size, PAGE_SIZE)
else:
mlx5e_rx_get_linear_frag_sz := buffer size
mlx5e_rx_is_linear_skb := mlx5e_rx_get_linear_frag_sz <= PAGE_SIZE
stride size := roundup_pow_of_two(mlx5e_rx_get_linear_frag_sz)
The new code effectively removes mlx5e_rx_get_linear_frag_sz that used
to return either buffer size or stride size, depending on the situation,
making it hard to work with and to make changes:
if XDP or XSK:
mlx5e_rx_get_linear_stride_sz := PAGE_SIZE
else
mlx5e_rx_get_linear_stride_sz := roundup_pow_of_two(buffer size)
mlx5e_rx_is_linear_skb := buffer size <= (PAGE_SIZE or XSK frame sz)
stride size := mlx5e_rx_get_linear_stride_sz
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
