FIPS 140-3 recommends that all crypto implementations should be tested
before first use. Testmanager performs initial tests based on existing
test vectors. Not all algorithms have defined test vectors, so to improve
this situation, this commit backports recently added test vectors for
some cipher suites.
These vectors were calculated using a software implementation and then
double-checked on Mediatek MT7981 (safexcel) and NXP P2020 (talitos).
Both platforms passed self-tests.
Signed-off-by: Aleksander Jan Bajkowski <olek2@wp.pl>
Link: https://github.com/openwrt/openwrt/pull/23012
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Add patches to improve support for using 3rd-party DSA switches
like MaxLinear MxL862xx with MediaTek's mtk_eth_soc being the
conduit. This involves reorganizing hardware queues to avoid
overlap (currently dp->index is used -- if there is more than one
DSA switch this is problematic), and correctly programming flows
of the non-MTK DSA users ports in the PPE offloading engine.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
LINK_INBAND_ENABLE isn't valid for 5GBase-R/10GBase-R modes which
by definition don't support any in-band an. Correctly report
LINK_INBAND_DISABLE to fix 10G fiber SFP modules no longer working.
While at it also get rid of downstream pn-swap properties in favor
of using the upstream schema.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Currently, the DesignWare PCIe driver cannot configure interrupts on
SoC that do not support MSIX. All MSI interrupts are handled by CPU0.
Backport MSI affinity support for the PCI dwc driver from linux-next,
so now we can adjust MSI interrupts to other CPU cores.
Tested on HINLINK H28K (RK3528) and OrangePi R2S (Ky X1).
Signed-off-by: Chukun Pan <amadeus@jmu.edu.cn>
Link: https://github.com/openwrt/openwrt/pull/21770
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Backport support for the RTL8157. The RTL8157 is a low-cost chipset
designed for USB to 5Gb Ethernet adapters.
Tested on Sabrent NT-C5GA (RTL8157), Wavlink WL-NWU340G (RTL8157) and
UGREEN CM648 (RTL8156BG) adapters.
Signed-off-by: Aleksander Jan Bajkowski <olek2@wp.pl>
Link: https://github.com/openwrt/openwrt/pull/23088
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
At least the XikeStor SKT-2.5G-100M SFP module seems to internally use
MDIO address 0 to access the PHY. This module allows accessing PHY
registers using Rollball protocol on address 0x51, and also provides
read-only C22 access on address 0x56. However, after disabling the
PHYAD0 configuration bit, only 0xffff can be read via both methods
(except for MMD device 30 which can still be accessed).
Since having MDIO address 0 enabled shouldn't do any harm on SFP modules
just leave the configuration bit alone in that case.
Signed-off-by: Jan Hoffmann <jan@3e8.eu>
Link: https://github.com/openwrt/openwrt/pull/23065
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
After the RTL8261N asserts a reset, the MDIO bus becomes temporarily
unavailable during the chip's reinitialization sequence. Any subsequent
read or write issued before the PHY has stabilized will fail.
Add a 30ms delay after triggering the reset to ensure the chip is reachable
via MDIO before resuming communication.
Signed-off-by: Jonas Jelonek <jelonek.jonas@gmail.com>
Signed-off-by: Sven Eckelmann <sven@narfation.org>
Link: https://github.com/openwrt/openwrt/pull/23076
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
In the past, all the configuration of SerDes and PHYs on the realtek
switches were done using u-boot (`rtk init`). But since RTL930x switched
to SerDes configuration under Linux, the SoC side is no longer using the
Realtek-proprietary variant of USXGMII. The communication to the RTL8261N
PHYs on those switches broke because of this incompatibility.
Enabling the full initialization on `CONFIG_MACH_REALTEK_RTL` converts also
the PHY side to the standard USXGMII and therefore ensures that both sides
speak the same dialect.
Co-authored-by: Jonas Jelonek <jelonek.jonas@gmail.com>
Signed-off-by: Sven Eckelmann <sven@narfation.org>
Link: https://github.com/openwrt/openwrt/pull/23076
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
The PHY register patch in question is gated by `CONFIG_MACH_REALTEK_RTL`,
has no documented/expected behavior, and is in practice unreachable:
`phy_patch()` is only called from `rtkphy_config_init()`, which is exits
(too) early for `CONFIG_MACH_REALTEK_RTL` builds.
Remove it as a cleanup step before enabling standard USXGMII configuration
for these PHYs.
Fixes: b77fa45d1278 ("kernel: fix rtl8261n driver for realtek")
Co-authored-by: Jonas Jelonek <jelonek.jonas@gmail.com>
Signed-off-by: Sven Eckelmann <sven@narfation.org>
Link: https://github.com/openwrt/openwrt/pull/23076
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Motorcomm YT6801 is a PCIe ethernet controller based on DWMAC4 IP. It
integrates an GbE phy, supporting WOL, VLAN tagging and various types
of offloading. It ships an on-chip eFuse for storing various vendor
configuration, including MAC address.
The PM and plat_data functions were slightly modified to build with
current kernel.
Signed-off-by: Tianling Shen <cnsztl@immortalwrt.org>
Link: https://github.com/openwrt/openwrt/pull/22923
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
These symbols depend on rustc version and is autodetected and set, so make
sure to filter them out.
Signed-off-by: Robert Marko <robert.marko@sartura.hr>
Import pending series introducing support for standalone PCS drivers.
This has previously already been used by the airoha target, and is
also the base for the closer-to-upstream patches for MediaTek MT7988
10G SerDes support.
In order to not having to diverge from upstream also backport series
for standardized handling for PHY and PCS SerDes pair polarity.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Daniel Pawlik figured out that a stale patch hunk breaks one of the
two 10G SerDes PCS ports of MT7988. Remove the hunk to make 10G
Ethernet work on both SerDes PCS with Linux 6.18.
Testing was done using a Aquantia AQR113C SFP+ module.
Signed-off-by: Daniel Pawlik <pawlik.dan@gmail.com>
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Now that 6.12 EMC2305 driver is backported to 7.0 code, lets do the same
for 6.18 so they dont diverge.
Link: https://github.com/openwrt/openwrt/pull/22942
Signed-off-by: Robert Marko <robimarko@gmail.com>
Backport the Microchip EMC2305 driver changes up to current linux-next.
These are required in order to actually be able to use the driver with
DTS support to drive fans as cooling devices.
Replace the bcm27xx RPi patch with the one from RPi kernel 7.0 that was
already adapted to all of these backports.
Link: https://github.com/openwrt/openwrt/pull/22942
Signed-off-by: Robert Marko <robimarko@gmail.com>
Our downstream patch for the new SerDes paths on MT7988 wasn't
correctly ported to Linux 6.18 which already got most of the XPCS
support upstream.
Fix (ie. reduce) the patch to make the built-in 2.5G PHY on MT7988 and
MT7987 work again.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Add missing patch description, so all generic patches can be applied
to a kernel tree using 'git am'.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
devm_gpio_request() was removed since kernel 6.17. Convert it to
devm_gpio_request_one() to fix:
target/linux/generic/files/drivers/net/phy/adm6996.c: In function 'adm6996_gpio_probe':
target/linux/generic/files/drivers/net/phy/adm6996.c:1183:15: error: implicit declaration of function 'devm_gpio_request'; did you mean 'devm_gpio_request_one'? [-Wimplicit-function-declaration]
1183 | ret = devm_gpio_request(&pdev->dev, priv->eecs, "adm_eecs");
| ^~~~~~~~~~~~~~~~~
| devm_gpio_request_one
Link: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=linux-6.18.y&id=a5589313383074c48a1b3751d592a6e084ae0573
Signed-off-by: Shiji Yang <yangshiji66@outlook.com>
Link: https://github.com/openwrt/openwrt/pull/22917
Signed-off-by: Robert Marko <robimarko@gmail.com>
eip93_hmac_setkey() allocates a temporary ahash transform for
computing HMAC ipad/opad key material. The allocation uses the
driver-specific cra_driver_name (e.g. "sha256-eip93") but passes
CRYPTO_ALG_ASYNC as the mask, which excludes async algorithms.
Since the EIP93 hash algorithms are the only ones registered
under those driver names and they are inherently async, the
lookup is self-contradictory and always fails with -ENOENT.
When called from the AEAD setkey path, this failure leaves the
SA record partially initialized with zeroed digest fields. A
subsequent crypto operation then dereferences a NULL pointer in
the request context, resulting in a kernel panic:
```
pc : eip93_aead_handle_result+0xc8c/0x1240 [crypto_hw_eip93]
lr : eip93_aead_handle_result+0xbec/0x1240 [crypto_hw_eip93]
sp : ffffffc082feb820
x29: ffffffc082feb820 x28: ffffff8011043980 x27: 0000000000000000
x26: 0000000000000000 x25: ffffffc078da0bc8 x24: 0000000091043980
x23: ffffff8004d59e50 x22: ffffff8004d59410 x21: ffffff8004d593c0
x20: ffffff8004d593c0 x19: ffffff8004d4f300 x18: 0000000000000000
x17: 0000000000000000 x16: 0000000000000000 x15: 0000007fda7aa498
x14: 0000000000000000 x13: 0000000000000000 x12: 0000000000000000
x11: 0000000000000000 x10: fffffffff8127a80 x9 : 0000000000000000
x8 : ffffff8004d4f380 x7 : 0000000000000000 x6 : 000000000000003f
x5 : 0000000000000040 x4 : 0000000000000008 x3 : 0000000000000009
x2 : 0000000000000008 x1 : 0000000028000003 x0 : ffffff8004d388c0
Code: 910142b6 f94012e0 f9002aa0 f90006d3 (f9400740)
```
The reported symbol eip93_aead_handle_result+0xc8c is a
resolution artifact from static functions being merged under
the nearest exported symbol. Decoding the faulting sequence:
```
910142b6 ADD X22, X21, #0x50
f94012e0 LDR X0, [X23, #0x20]
f9002aa0 STR X0, [X21, #0x50]
f90006d3 STR X19, [X22, #0x8]
f9400740 LDR X0, [X26, #0x8]
```
The faulting LDR at [X26, #0x8] is loading ctx->flags
(offset 8 in eip93_hash_ctx), where ctx has been resolved
to NULL from a partially initialized or unreachable
transform context following the failed setkey.
Fix this by dropping the CRYPTO_ALG_ASYNC mask from the
crypto_alloc_ahash() call. The code already handles async
completion correctly via crypto_wait_req(), so there is no
requirement to restrict the lookup to synchronous algorithms.
Note that hashing a single 64-byte block through the hardware
is likely slower than doing it in software due to the DMA
round-trip overhead, but offloading it may still spare CPU
cycles on the slower embedded cores where this IP is found.
Fixes: 9739f5f93b78 ("crypto: eip93 - Add Inside Secure SafeXcel EIP-93 crypto engine support")
Signed-off-by: Aleksander Jan Bajkowski <olek2@wp.pl>
[Detailed investigation report of this bug]
Signed-off-by: Kenneth Kasilag <kenneth@kasilag.me>
Link: https://github.com/openwrt/openwrt/pull/22886
Signed-off-by: Robert Marko <robimarko@gmail.com>
Some new DRM kernel modules appeared in the 6.18 kernel and
are now prompting for selection when enabling DRM in the kernel.
The Gemini D-Link DIR-685 is always the first to run into this
problem hence let's fix it before someone else tries to enable
DRM.
Link: https://github.com/openwrt/openwrt/pull/22875
Signed-off-by: Linus Walleij <linusw@kernel.org>
This patch enables for the configuration of the sync_time parameter.
The value of the sync_time needs to be specified in nanoseconds.
The default value is 200us (200 000 ns).
This is especially helpful to get feedback from users about suitable
values. Once a solid solution is found this patch can be removed.
Signed-off-by: Jonas Köppeler <j.koeppeler@tu-berlin.de>
Link: https://github.com/openwrt/openwrt/pull/22380
[Add patch to kernel 6.18 too]
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
This symbol is selected by CONFIG_BPF, which was already enabled
on generic config-6.18.
Signed-off-by: Shiji Yang <yangshiji66@outlook.com>
Link: https://github.com/openwrt/openwrt/pull/22730
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
This symbol is selected by CONFIG_BPF, which was already enabled
on generic config-6.12.
Signed-off-by: Shiji Yang <yangshiji66@outlook.com>
Link: https://github.com/openwrt/openwrt/pull/22730
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
This symbol depends on the specific CPU type, so we should not
set it globally.
Signed-off-by: Shiji Yang <yangshiji66@outlook.com>
Link: https://github.com/openwrt/openwrt/pull/22730
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
This symbol is selected by CONFIG_BUILTIN_DTB. However, we didn't
enable it in generic config-6.18.
Signed-off-by: Shiji Yang <yangshiji66@outlook.com>
Link: https://github.com/openwrt/openwrt/pull/22730
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
GCC generates the same code for 24kc and 34kc. Since we have
converted all 34kc targets to the 24kc, it's better to switch
kernel -mtune to 24kc to avoid confusing developers.
Signed-off-by: Shiji Yang <yangshiji66@outlook.com>
Link: https://github.com/openwrt/openwrt/pull/22703
Signed-off-by: Robert Marko <robimarko@gmail.com>
We patch DMA_BUF to make it tristate, so once ALL_KMODS is selected it will
be built as a module even if previously disabled in the config.
So, since IO_URING_ZCRX does not depend on DMA_BUF linking will fail with:
aarch64-openwrt-linux-musl-ld: Unexpected GOT/PLT entries detected!
aarch64-openwrt-linux-musl-ld: Unexpected run-time procedure linkages detected!
aarch64-openwrt-linux-musl-ld: io_uring/zcrx.o: in function `io_release_dmabuf':
io_uring/zcrx.c:94:(.text+0x20): undefined reference to `dma_buf_unmap_attachment_unlocked'
aarch64-openwrt-linux-musl-ld: io_uring/zcrx.c:97:(.text+0x30): undefined reference to `dma_buf_detach'
aarch64-openwrt-linux-musl-ld: io_uring/zcrx.c:99:(.text+0x3c): undefined reference to `dma_buf_put'
aarch64-openwrt-linux-musl-ld: io_uring/zcrx.o: in function `io_import_dmabuf':
io_uring/zcrx.c:125:(.text+0x1b20): undefined reference to `dma_buf_get'
aarch64-openwrt-linux-musl-ld: io_uring/zcrx.c:132:(.text+0x1b34): undefined reference to `dma_buf_attach'
aarch64-openwrt-linux-musl-ld: io_uring/zcrx.c:139:(.text+0x1b48): undefined reference to `dma_buf_map_attachment_unlocked'
make[6]: *** [scripts/Makefile.vmlinux:72: vmlinux.unstripped] Error 1
So, lets use IS_REACHABLE() to check for CONFIG_DMA_SHARED_BUFFER instead
to avoid adding a dependency to CONFIG_DMA_SHARED_BUFFER.
Signed-off-by: Robert Marko <robert.marko@sartura.hr>
All targets using ar8327 use OF.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/22069
Signed-off-by: Robert Marko <robimarko@gmail.com>
Introduce fix which prevents the kernel from crashing in case the mxl862xx
driver fails to probe due to outdated firmware running on the switch.
Cancel all pending work and prevent rescheduling of counter polling in case
the driver errors out during probe.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
CONFIG_TOOLS_SUPPORT_RELR is set during runtime by the kernel so it should
not end up in our static configs.
So, filter it out by default.
Signed-off-by: Robert Marko <robimarko@gmail.com>
If DSA support is enabled, then these will pop up.
So instead of ending up in target configs, add them to the generic configs.
Signed-off-by: Robert Marko <robimarko@gmail.com>
