-
Notifications
You must be signed in to change notification settings - Fork 53k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Aspeed master v5.15 #883
base: master
Are you sure you want to change the base?
Aspeed master v5.15 #883
Conversation
Enable vclk after video engine initialization to fix memory corruption issue after cold boot Change-Id: I4f972b0f2601958ea4d5557b8569f76452eed2ef
add to support multi-master mode for i2c bus 4 and 13 for MCTP over smbus data transmission. Signed-off-by: Jamin Lin <jamin_lin@aspeedtech.com> Change-Id: If68eb5f18dd4f4c2a37d8d58b77f3f55141419c8
This reverts commit b3fa0ba. Change-Id: I30408b743cc2388e4252f7a660c2dfcaa5a42c1c
Use the driver, drivers/media/platform/aspeed-video.c, instead. Signed-off-by: Jammy Huang <jammy_huang@aspeedtech.com> Change-Id: Ie12daf99c4ebda186cdb52694c202584230c3d8d
Add TPM related configurations on AST2600 platform. Signed-off-by: Chin-Ting Kuo <chin-ting_kuo@aspeedtech.com> Change-Id: Ia681683aa1327f9b75b408933c896562369ca2b6
enable vclk after eclk for Video Engine Change-Id: I4c0eb677dd32908527b2a657d1bf52eb05263c5f
This patch add the HDR function for master to send the ccc or data with HDR mode to the device. Signed-off-by: Billy Tsai <billy_tsai@aspeedtech.com> Change-Id: I98b5b4088d663b4f846e5079cee90eabe25e3afe
This patch implements the HDR transfer function for aspeed i3c master. Currently, the master only support the HDR-DDR mode. Signed-off-by: Billy Tsai <billy_tsai@aspeedtech.com> Change-Id: I8edd56a4b35c24e2e0fe182516f5c211d4f445de
This patch will use HDR-DDR mode to send the i3c data if the device support it and the CONFIG_I3CDEV_XFER_HDR_DDR say 'y'. Signed-off-by: Billy Tsai <billy_tsai@aspeedtech.com> Change-Id: I682405cb460c33fdbcc81962af4076beceebe29c
Add the condition to separate the ibi and response ready interrupt. Before this patch the master will signal the completion of the thread, either ibi or response ready interrupt. However, the completion should only wait for response ready interrupts instead of the ibi. The unexpected wake-up by ibi will result in the process stuck or kernel panic Signed-off-by: Billy Tsai <billy_tsai@aspeedtech.com> Change-Id: I8c27f83dee1779394e131577b2920ea6a508a79a
The user may send the endstate with JTAG_STATE_CURRENT, the driver should replace it with the state store in the drvier. Reported-by: Jim Heck <jim.heck@oracle.com> Signed-off-by: Billy Tsai <billy_tsai@aspeedtech.com> Change-Id: I1afa4cdccd98ecd2e9a8be45f85657247c3cb4de
The default value of the control register is set using the direct mapping information passed to the ->dirmap_create() handler. Dump the mapping range and the SPI memory operation characteristics to analyze how the register value has been computed. spi-aspeed-smc 1e630000.spi: CE0 read dirmap [0x00000000 - 0x04000000] OP 0x6c mode:1.1.1.4 naddr:0x4 ndummies:0x1 ... spi-aspeed-smc 1e630000.spi: CE0 write dirmap [0x00000000 - 0x04000000] OP 0x12 mode:1.1.0.1 naddr:0x4 ndummies:0x0 Reviewed-by: Paul Menzel <pmenzel@molgen.mpg.de> Signed-off-by: Cédric Le Goater <clg@kaod.org> Reviewed-by: Joel Stanley <joel@jms.id.au> Link: https://lore.kernel.org/r/20220622161617.3719096-2-clg@kaod.org Signed-off-by: Mark Brown <broonie@kernel.org> Signed-off-by: Chin-Ting Kuo <chin-ting_kuo@aspeedtech.com> Change-Id: Ia62588a21fa4933b9a17a8740cd85b845ded018e
When using the normal read operation for data transfers, the dummy bus width is zero. In that case, they are no dummy bytes to transfer and setting the dummy field in the controller register becomes useless. Issue was found on a custom "Bifrost" board based on the AST2500 SoC and using a MX25L51245GMI-08G SPI Flash. Reported-by: Ian Woloschin <ian.woloschin@akamai.com> Reviewed-by: Pratyush Yadav <p.yadav@ti.com> Tested-by: Ian Woloschin <iwolosch@akamai.com> Fixes: 9da06d7 ("spi: aspeed: Add support for direct mapping") Signed-off-by: Cédric Le Goater <clg@kaod.org> Link: https://lore.kernel.org/r/20220622161617.3719096-3-clg@kaod.org Signed-off-by: Mark Brown <broonie@kernel.org> Change-Id: I63c492693e76facad7ef31995587a3606382d1f7 Signed-off-by: Chin-Ting Kuo <chin-ting_kuo@aspeedtech.com>
Both quad SPI TX and RX modes can be supported on AST2600. Correct typo in mode_bits field in both ast2600_fmc_data and ast2600_spi_data structs. OpenBMC-Staging-Count: 1 Signed-off-by: Chin-Ting Kuo <chin-ting_kuo@aspeedtech.com> Reviewed-by: Cédric Le Goater <clg@kaod.org> Link: https://lore.kernel.org/r/20221005083209.222272-1-chin-ting_kuo@aspeedtech.com Signed-off-by: Joel Stanley <joel@jms.id.au> Change-Id: I0c13cc518e71c86a2f1fbab114cf6ab869714234
The offset value of the mapping window in the kernel structure is calculated using the value of the previous window offset. This doesn't reflect how the HW is configured and can lead to erroneous setting of the second flash device (CE1). OpenBMC-Staging-Count: 1 Fixes: e3228ed ("spi: spi-mem: Convert Aspeed SMC driver to spi-mem") Signed-off-by: Cédric Le Goater <clg@kaod.org> Reviewed-by: Joel Stanley <joel@jms.id.au> Link: https://lore.kernel.org/r/20221016155722.3520802-1-clg@kaod.org Signed-off-by: Joel Stanley <joel@jms.id.au> Change-Id: I44bc77de5cfdcb036372bcdda6c9467fea76cee9 Signed-off-by: Chin-Ting Kuo <chin-ting_kuo@aspeedtech.com>
For some flash parts, the original delay timing may be in the margin point. Thus, we extend the delay timing in order to satisfy the reset sequence for the most flash parts. Signed-off-by: Chin-Ting Kuo <chin-ting_kuo@aspeedtech.com> Change-Id: I7cf2175a604a1aa5bb4071fb0c9a4f4c0ed4ca0a
The "chunk >= 0" condition does not work because count is a u32. Also, really we shouldn't enter the loop when "chunk" is zero. Once that condition is fixed then there is no need for the "last" variable. I reversed the "if (chunk <= ep->chunk_max)" as well. The new loop is much simpler. Fixes: 055276c ("usb: gadget: add Aspeed ast2600 udc driver") Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Neal Liu <neal_liu@aspeedtech.com> Change-Id: Ia074b01d1fde7ae3936ef8030bcb521ac432ce7d
The bug is that we should still enter this loop if "tx_len" is zero. After adding the "last" variable, then the "chunk >= 0" condition is no longer required but I left it for readability. Reported-by: Neal Liu <neal_liu@aspeedtech.com> Fixes: c09b1f3 ("usb: gadget: aspeed_udc: cleanup loop in ast_dma_descriptor_setup()") Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Reviewed-by: Neal Liu <neal_liu@aspeedtech.com> Change-Id: I7d169525424de001c6337ac77a16e51d03fbb7ad
Add operation configuration for CRTM enabled scenario. - Use command mode to access flash. Signed-off-by: Chin-Ting Kuo <chin-ting_kuo@aspeedtech.com> Change-Id: I1ce8db4224d047f4e1f7e3a6fa5855916a743698
Problem: CDC NCM low performance issue. BULK IN transfer error - illegal NDP (NCM Datagram Pointers) Signature. Solution: The udc descriptor(DMA) mode allowable transfer length is 4096 bytes. Wrong tx_len will result in decoding NDP16 field error, the iperf testing results would be abnormal. Correct it to fix CDC NCM performance drop issue. Signed-off-by: Neal Liu <neal_liu@aspeedtech.com> Change-Id: Iae60890cd96ec1e338c2e23ae5a228e42ec278cd
The node names of spd has changed by the patch: commit ee4117e ("ARM: dts: ast2600-evb: use macro to creat I3C SPD child node") Signed-off-by: Billy Tsai <billy_tsai@aspeedtech.com> Change-Id: I97d66570aea3b4ec5e48c63d83e05e650dfb40fa
afm-act will be inside ast1060, and the afm-stg will be used as staging area for image update. afm-rcv is the recovery area for recovering. Signed-off-by: Troy Lee <troy_lee@aspeedtech.com> Change-Id: I4cd52ebdd69dd4fef51a66ec0dde6e8b9694fdf3
- Using the expected SPI clock frequency when user mode is used. - Using the SPI clock frequency in device tree if the better clock frequency is not found. Change-Id: I72490a825d6ec7353565e32571ba8cd939092503 Signed-off-by: Chin-Ting Kuo <chin-ting_kuo@aspeedtech.com>
Clear the rx_warmup flag when any valid packet is found to avoid unexpected runaway rx packet modifying the MCTP's boundary of rx buffer. Signed-off-by: Billy Tsai <billy_tsai@aspeedtech.com> Change-Id: I69dd1e55ff39548f816024e1926a519c980bde4e
The rx length shouldn't bigger than the MTU size. This patch add the warning log to alarm this problem for debugging. Signed-off-by: Billy Tsai <billy_tsai@aspeedtech.com> Change-Id: I687df11ecd0e2025c0f9caa69c0375fe4257c715
Transfer error interrupt will generate one response data, so append this condition to handle the response data for the xfer error. Signed-off-by: Billy Tsai <billy_tsai@aspeedtech.com> Change-Id: I234c846f462e0f4d3c31c4308e8090c674bfe679
Signed-off-by: Steven Lee <steven_lee@aspeedtech.com> Change-Id: I9c55844d9b2fbf1cad07c2a8f1a514ca6e4f561d
Rx runaway will only occur when rx warmup and packets may not have been written to the dram when the driver enter the rx tasklet by the interrupt. If the packet write to the dram after check the packet in sw_ptr, the rx runaway will be detected by mistake and cause the packet sequence to go wrong. So this patch add the condition to avoid the driver to check the runaway packet after getting first valid packet. Failed log: [ 512.642350] aspeed-mctp 1e6e8000.mctp: Runaway RX packet found 79 -> 80 (b204f900) [ 512.650895] aspeed-mctp 1e6e8000.mctp: Hw read ptr = 004f004f [ 512.657339] aspeed-mctp 1e6e8000.mctp: Hw read ptr = 00510051 [ 512.663851] aspeed-mctp 1e6e8000.mctp: Address 84953160 [ 512.669706] aspeed-mctp 1e6e8000.mctp: VDM header: [ 512.675135] aspeed-mctp 1e6e8000.mctp: 72 00 00 10 [ 512.680549] aspeed-mctp 1e6e8000.mctp: 40 00 30 7f [ 512.685918] aspeed-mctp 1e6e8000.mctp: 01 00 1a b4 [ 512.691315] aspeed-mctp 1e6e8000.mctp: 01 0b 09 70 ==> SOM = 0, EOM = 1, seq = 3 ... [ 512.788905] aspeed-mctp 1e6e8000.mctp: Runaway RX packet found 82 -> 79 (b204f8b0) [ 512.797518] aspeed-mctp 1e6e8000.mctp: Hw read ptr = 00520052 [ 512.804060] aspeed-mctp 1e6e8000.mctp: Hw read ptr = 00520052 [ 512.810548] aspeed-mctp 1e6e8000.mctp: Address 84953c60 [ 512.816418] aspeed-mctp 1e6e8000.mctp: VDM header: [ 512.821838] aspeed-mctp 1e6e8000.mctp: 72 00 00 10 [ 512.827252] aspeed-mctp 1e6e8000.mctp: 40 00 00 7f [ 512.832688] aspeed-mctp 1e6e8000.mctp: 01 00 1a b4 [ 512.838075] aspeed-mctp 1e6e8000.mctp: 01 0b 09 20 ==> SOM = 0, EOM = 0, seq = 2 ... [ 513.291771] aspeed-mctp 1e6e8000.mctp: Runaway RX packet found 80 -> 82 (b204f9a0) [ 513.300358] aspeed-mctp 1e6e8000.mctp: Hw read ptr = 00530053 [ 513.306819] aspeed-mctp 1e6e8000.mctp: Hw read ptr = 00530053 [ 513.313341] aspeed-mctp 1e6e8000.mctp: Address 84953420 [ 513.319221] aspeed-mctp 1e6e8000.mctp: VDM header: [ 513.324654] aspeed-mctp 1e6e8000.mctp: 72 00 00 03 [ 513.330066] aspeed-mctp 1e6e8000.mctp: 16 00 20 7f [ 513.335458] aspeed-mctp 1e6e8000.mctp: 01 00 1a b4 [ 513.340883] aspeed-mctp 1e6e8000.mctp: 01 0b 0a c0 ... Signed-off-by: Billy Tsai <billy_tsai@aspeedtech.com> Change-Id: Ie1efad49b2847e172794e0194c8319c1448c728c
Schedule the rx tacklet when user try to poll/read the packets to avoid the rx tasklet triggered by interrupt not finding a valid packet to improve the performance. Signed-off-by: Billy Tsai <billy_tsai@aspeedtech.com> Change-Id: Ie2fa4938bf716b5dfd9c60e770d9f3ae590b226e
Add the support for the ast10x0 a2 device. Signed-off-by: Billy Tsai <billy_tsai@aspeedtech.com> Change-Id: I58718b135e69d89fedc8c099e73873bc641cac7f
1. use size macro 2. use FIELD_PREP 3. shorten macro naming Signed-off-by: Chia-Wei Wang <chiawei_wang@aspeedtech.com> Change-Id: Ic28226a5d539487c178f892eb6144c2b3a42b200
Support dual and quad IO on fmc_spi.c. Signed-off-by: Chin-Ting Kuo <chin-ting_kuo@aspeedtech.com> Change-Id: I55d88e32f59fdd74b4b3f321eb2c35908dbeb526
Remove unnecessary SPI bus-width settings. Signed-off-by: Chin-Ting Kuo <chin-ting_kuo@aspeedtech.com> Change-Id: Icb710e395b2a3bb3f071c7941a3e88857f786844
Rplace the ADC "vref = <2500>;" property with "aspeed,int-vref-microvolt = <2500000>;" Signed-off-by: Billy Tsai <billy_tsai@aspeedtech.com> Change-Id: Icc3f4d9f9dc407f3eeda2d0170b3a770f9e2ae05
Use IDA to get a unique ID for device node indexing if the 'kcs-channel' property is not specified. Signed-off-by: Chia-Wei Wang <chiawei_wang@aspeedtech.com> Change-Id: I1103d6299c7251ff5cb6544c613ae74041486e9a
1. Use IDA to get a unique ID for device node indexing if the 'bt-channel' property is not specified. 2. Add upstream SerIRQ configuration support via the 'bt-upstream-serirq' property 3. Add IO base address configuration support via the 'bt-io-addr' property. Signed-off-by: Chia-Wei Wang <chiawei_wang@aspeedtech.com> Change-Id: I8ca10401a24374e464764adda503db70b21a7186
Only IRQ_TYPE_LEVEL_LOW/HIGH are supported. Signed-off-by: Chia-Wei Wang <chiawei_wang@aspeedtech.com> Change-Id: I01602a0c6fed767d2b499412af2ade2179da5ebd
Refine dev_info() message upon successful probing. Signed-off-by: Chia-Wei Wang <chiawei_wang@aspeedtech.com> Change-Id: I0b4612136dc078fe8eb9e8d13f43f11047874b12
Add dual node support and refactor coding style. Signed-off-by: Chia-Wei Wang <chiawei_wang@aspeedtech.com> Change-Id: Ie219aa261e0be7ffe7519ae706d2c812cf1335b6
Add dual node support. Signed-off-by: Chia-Wei Wang <chiawei_wang@aspeedtech.com> Change-Id: Ie8927674e228191aff540fe9bb4b5d35058242cd
Add dual node support. Signed-off-by: Chia-Wei Wang <chiawei_wang@aspeedtech.com> Change-Id: I89e69552bebf98e28afe8a77ff6c56619294179c
fix update for index 1~4 Signed-off-by: Ryan Chen <ryan_chen@aspeedtech.com> Change-Id: I01c5111ae2c4832319595fcd03d180b9ede9fcd8 (cherry picked from commit 476bb03)
Change-Id: I8906d03d85870e19e77ff27e0562f77cadf6f489 Signed-off-by: Kevin Chen <kevin_chen@aspeedtech.com>
Change-Id: I3e3895bc0d3754239aa6d995d279061ae2a50414 Signed-off-by: Kevin Chen <kevin_chen@aspeedtech.com>
The dma_alloc_coherent will automatically clear the allocated memory, causing a time gap when the MCTP controller receives the packet after BMC reboot or kernel module reinstall. ast2500: The MCTP rx can't be stopped by the software or WDT. ast2600: The MCTP rx can be stopped by the WDT, but still cannot be stopped by the software. This may result in using '0' as the rx data destination address, leading to memory pollution. Instead, this patch uses ioremap and phys_to_dma to obtain the CPU address and DMA address to avoid it. Signed-off-by: Billy Tsai <billy_tsai@aspeedtech.com> Change-Id: I26300a35e4e91e0527b64e88e96adf77a26f7c55
dma_alloc_coherent cannot alloc the memory bigger than 32MB in linux-5.15 in ast2600. Use the ioremap by the physical to get the virtual address for release_firmware usage. Change-Id: I3ac532836a1b47044d4e156e4430cf1030f5ccf2
Add XT25W512B and XT25W01GB flash parts support. Signed-off-by: Chin-Ting Kuo <chin-ting_kuo@aspeedtech.com> Change-Id: I86b08cc2c401453780ed69ed59aa4151c7a5d8c4
HOST_CONTROL2 register will be cleared after top reset, it needs to be saved/resotred while reset. Signed-off-by: Cool Lee <cool_lee@aspeedtech.com> Change-Id: Ie97acac87eb81d73fa20294c777776dbacb377e0 (cherry picked from commit 0c69253)
Change-Id: I4b61d289e68370e3149aab1eae8ac332809be46c Signed-off-by: Kevin Chen <kevin_chen@aspeedtech.com>
Change the SMBus support just on i2c buffer mode. Signed-off-by: Tommy Huang <tommy_huang@aspeedtech.com> Change-Id: I0d225e28f94beced2e3a14f914847d449fd5766b
Fix the typo when applying `dev_grp->mask.clr`. Signed-off-by: Dylan Hung <dylan_hung@aspeedtech.com> Change-Id: I35b2fe0c5f8abc1e4020a8f5b3c24e535813c467
When the device nak occur, the DMA map would not be released. Add behavior to check if the DMA need to be released before leaving master transfer function. Signed-off-by: Tommy Huang <tommy_huang@aspeedtech.com> Change-Id: Ic83674df33e373e5861f30f3c263f172cc086bec (cherry picked from commit 465f886)
The hardware timed reset pattern cannot be sent if the SDA is held low by the target device. This commit implements it via software force mode. Signed-off-by: Dylan Hung <dylan_hung@aspeedtech.com> Change-Id: Ifaa4dc96e88a95427cc2acd30074e68aa2f94c64
The bdf value is only valid when pcie link-up. Signed-off-by: Billy Tsai <billy_tsai@aspeedtech.com> Change-Id: I4667075c88ceee295088e80164765794eeaadd10
The register offset to disable the internal pull-down of GPIOR~T is 0x630 instead of 0x620, as specified in the Ast2600 datasheet v15 The datasheet can download from the official Aspeed website. Fixes: 15711ba ("pinctrl: aspeed-g6: Add AST2600 pinconf support") Signed-off-by: Billy Tsai <billy_tsai@aspeedtech.com> Reviewed-by: Paul Menzel <pmenzel@molgen.mpg.de> Change-Id: Iaa8e09a7d357c84c10798a9b7342e065ef374b0b
Toggle the SCL pulses (up tp 19 pulses) when failed to transfer and the SDA stuck low. Signed-off-by: Dylan Hung <dylan_hung@aspeedtech.com> Change-Id: I6d40994e5e7b5fe6005df42380d1b0a80620b075
Try to prevent using a software-forced timed reset, as this approach doesn't take care of ongoing transfers and IBIs. The new implementation lets the controller enter a halt state first to ensure that ongoing transfers are completed. Then, it recovers the bus to an idle state if the current SDA level is tied low by the target devices. If the SDA remains low, it uses a software-forced mode; otherwise, it uses the hardware mode. Signed-off-by: Dylan Hung <dylan_hung@aspeedtech.com> Change-Id: Id2ad10ded18d7dfe7c032803ddde76dcf1d0a775
The interrupt controller node needs the #interrupt-cells property to inform the consumer about how to specify its parameters. This patch adds it to complete the interrupt property for sgpiom as an interrupt controller Signed-off-by: Billy Tsai <billy_tsai@aspeedtech.com> Change-Id: I63621548a0e39b4ee7f7e4784579c3d8d9bf59eb
Reset the i2c controller when an i2c transfer timeout occurs. The remaining interrupts and device should be reset to avoid unpredictable controller behavior. Signed-off-by: Tommy Huang <tommy_huang@aspeedtech.com> Change-Id: I21531d4eef586e6d1464a341f54761c6ddcd1966
Hi @PennixRv! Thanks for your contribution to the Linux kernel! Linux kernel development happens on mailing lists, rather than on GitHub - this GitHub repository is a read-only mirror that isn't used for accepting contributions. So that your change can become part of Linux, please email it to us as a patch. Sending patches isn't quite as simple as sending a pull request, but fortunately it is a well documented process. Here's what to do:
How do I format my contribution?The Linux kernel community is notoriously picky about how contributions are formatted and sent. Fortunately, they have documented their expectations. Firstly, all contributions need to be formatted as patches. A patch is a plain text document showing the change you want to make to the code, and documenting why it is a good idea. You can create patches with Secondly, patches need 'commit messages', which is the human-friendly documentation explaining what the change is and why it's necessary. Thirdly, changes have some technical requirements. There is a Linux kernel coding style, and there are licensing requirements you need to comply with. Both of these are documented in the Submitting Patches documentation that is part of the kernel. Note that you will almost certainly have to modify your existing git commits to satisfy these requirements. Don't worry: there are many guides on the internet for doing this. Where do I send my contribution?The Linux kernel is composed of a number of subsystems. These subsystems are maintained by different people, and have different mailing lists where they discuss proposed changes. If you don't already know what subsystem your change belongs to, the
Make sure that your list of recipients includes a mailing list. If you can't find a more specific mailing list, then LKML - the Linux Kernel Mailing List - is the place to send your patches. It's not usually necessary to subscribe to the mailing list before you send the patches, but if you're interested in kernel development, subscribing to a subsystem mailing list is a good idea. (At this point, you probably don't need to subscribe to LKML - it is a very high traffic list with about a thousand messages per day, which is often not useful for beginners.) How do I send my contribution?Use For more information about using How do I get help if I'm stuck?Firstly, don't get discouraged! There are an enormous number of resources on the internet, and many kernel developers who would like to see you succeed. Many issues - especially about how to use certain tools - can be resolved by using your favourite internet search engine. If you can't find an answer, there are a few places you can turn:
If you get really, really stuck, you could try the owners of this bot, @daxtens and @ajdlinux. Please be aware that we do have full-time jobs, so we are almost certainly the slowest way to get answers! I sent my patch - now what?You wait. You can check that your email has been received by checking the mailing list archives for the mailing list you sent your patch to. Messages may not be received instantly, so be patient. Kernel developers are generally very busy people, so it may take a few weeks before your patch is looked at. Then, you keep waiting. Three things may happen:
Further information
Happy hacking! This message was posted by a bot - if you have any questions or suggestions, please talk to my owners, @ajdlinux and @daxtens, or raise an issue at https://github.com/ajdlinux/KernelPRBot. |
No description provided.