summaryrefslogtreecommitdiff
path: root/drivers
AgeCommit message (Collapse)Author
2022-05-22eth: tg3: silence the GCC 12 array-bounds warningJakub Kicinski
GCC 12 currently generates a rather inconsistent warning: drivers/net/ethernet/broadcom/tg3.c:17795:51: warning: array subscript 5 is above array bounds of ‘struct tg3_napi[5]’ [-Warray-bounds] 17795 | struct tg3_napi *tnapi = &tp->napi[i]; | ~~~~~~~~^~~ i is guaranteed < tp->irq_max which in turn is either 1 or 5. There are more loops like this one in the driver, but strangely GCC 12 dislikes only this single one. Silence this silliness for now. Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-05-22eth: ice: silence the GCC 12 array-bounds warningJakub Kicinski
GCC 12 gets upset because driver allocates partial struct ice_aqc_sw_rules_elem buffers. The writes are within bounds. Silence these warnings for now, our build bot runs GCC 12 so we won't allow any new instances. Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-05-22eth: mtk_eth_soc: silence the GCC 12 array-bounds warningJakub Kicinski
GCC 12 gets upset because in mtk_foe_entry_commit_subflow() this driver allocates a partial structure. The writes are within bounds. Silence these warnings for now, our build bot runs GCC 12 so we won't allow any new instances. Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-05-22dpaa2-eth: unmap the SGT buffer before accessing its contentsIoana Ciornei
DMA unmap the Scatter/Gather table before going through the array to unmap and free each of the header and data chunks. This is so we do not touch the data between the dma_map and dma_unmap calls. Fixes: 3dc709e0cd47 ("dpaa2-eth: add support for software TSO") Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-05-22dpaa2-eth: use the correct software annotation fieldIoana Ciornei
The incorrect software annotation field was being used, swa->sg.sgt_size instead of swa->tso.sgt_size, which meant that the SGT buffer was unmapped with a wrong size. This is also confirmed by the DMA API debug prints which showed the following: [ 38.962434] DMA-API: fsl_dpaa2_eth dpni.2: device driver frees DMA memory with different size [device address=0x0000fffffafba740] [map size=224 bytes] [unmap size=0 bytes] [ 38.980496] WARNING: CPU: 11 PID: 1131 at kernel/dma/debug.c:973 check_unmap+0x58c/0x9b0 [ 38.988586] Modules linked in: [ 38.991631] CPU: 11 PID: 1131 Comm: iperf3 Not tainted 5.18.0-rc7-00117-g59130eeb2b8f #1972 [ 38.999970] Hardware name: NXP Layerscape LX2160ARDB (DT) Fixes: 3dc709e0cd47 ("dpaa2-eth: add support for software TSO") Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-05-22dpaa2-eth: retrieve the virtual address before dma_unmapIoana Ciornei
The TSO header was DMA unmapped before the virtual address was retrieved and then used to free the buffer. This meant that we were actually removing the DMA map and then trying to search for it to help in retrieving the virtual address. This lead to a invalid virtual address being used in the kfree call. Fix this by calling dpaa2_iova_to_virt() prior to the dma_unmap call. [ 487.231819] Unable to handle kernel paging request at virtual address fffffd9807000008 (...) [ 487.354061] Hardware name: SolidRun LX2160A Honeycomb (DT) [ 487.359535] pstate: a0400005 (NzCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) [ 487.366485] pc : kfree+0xac/0x304 [ 487.369799] lr : kfree+0x204/0x304 [ 487.373191] sp : ffff80000c4eb120 [ 487.376493] x29: ffff80000c4eb120 x28: ffff662240c46400 x27: 0000000000000001 [ 487.383621] x26: 0000000000000001 x25: ffff662246da0cc0 x24: ffff66224af78000 [ 487.390748] x23: ffffad184f4ce008 x22: ffffad1850185000 x21: ffffad1838d13cec [ 487.397874] x20: ffff6601c0000000 x19: fffffd9807000000 x18: 0000000000000000 [ 487.405000] x17: ffffb910cdc49000 x16: ffffad184d7d9080 x15: 0000000000004000 [ 487.412126] x14: 0000000000000008 x13: 000000000000ffff x12: 0000000000000000 [ 487.419252] x11: 0000000000000004 x10: 0000000000000001 x9 : ffffad184d7d927c [ 487.426379] x8 : 0000000000000000 x7 : 0000000ffffffd1d x6 : ffff662240a94900 [ 487.433505] x5 : 0000000000000003 x4 : 0000000000000009 x3 : ffffad184f4ce008 [ 487.440632] x2 : ffff662243eec000 x1 : 0000000100000100 x0 : fffffc0000000000 [ 487.447758] Call trace: [ 487.450194] kfree+0xac/0x304 [ 487.453151] dpaa2_eth_free_tx_fd.isra.0+0x33c/0x3e0 [fsl_dpaa2_eth] [ 487.459507] dpaa2_eth_tx_conf+0x100/0x2e0 [fsl_dpaa2_eth] [ 487.464989] dpaa2_eth_poll+0xdc/0x380 [fsl_dpaa2_eth] Fixes: 3dc709e0cd47 ("dpaa2-eth: add support for software TSO") Link: https://bugzilla.kernel.org/show_bug.cgi?id=215886 Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-05-22net: mscc: ocelot: offload tc action "ok" using an empty action vectorVladimir Oltean
The "ok" tc action is useful when placed in front of a more generic filter to exclude some more specific rules from matching it. The ocelot switches can offload this tc action by creating an empty action vector (no _ENA fields set to 1). This makes sense for all of VCAP IS1, IS2 and ES0 (but not for PSFP). Add support for this action. Note that this makes the gact_drop_and_ok_test() selftest pass, where "action ok" is used in front of an "action drop" rule, both offloaded to VCAP IS2. Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-05-22Merge power-supply 'fixes' branchSebastian Reichel
Merge power-supply fixes, that missed the v5.18 merge window into power-supply's for-next branch. Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
2022-05-22Merge tag 'intel-pinctrl-v5.19-2' of ↵Linus Walleij
gitolite.kernel.org:pub/scm/linux/kernel/git/pinctrl/intel into devel intel-pinctrl for v5.19-2 * Fix immutable IRQ chip examples in the GPIO documentation * Make use of immutable IRQ chip in Intel pin control drivers * Add module alias for Intel Apollo Lake The following is an automated git shortlog grouped by driver: baytrail: - make irq_chip immutable broxton: - Add module alias for Intel Apollo Lake cherryview: - Use GPIO chip pointer in chv_gpio_irq_mask_unmask() - make irq_chip immutable Documentation: - gpio: Advertise irqd_to_hwirq() helper in the examples - gpio: Fix IRQ mask and unmask examples intel: - Fix kernel doc format, i.e. add return sections - Drop unused irqchip member in struct intel_pinctrl - make irq_chip immutable lynxpoint: - make irq_chip immutable
2022-05-22net: fec: Do proper error checking for enet_out clkUwe Kleine-König
An error code returned by devm_clk_get() might have other meanings than "This clock doesn't exist". So use devm_clk_get_optional() and handle all remaining errors as fatal. Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-05-22hinic: Avoid some over memory allocationChristophe JAILLET
'prod_idx' (atomic_t) is larger than 'shadow_idx' (u16), so some memory is over-allocated. Fixes: b15a9f37be2b ("net-next/hinic: Add wq") Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-05-22net: fec: Do proper error checking for optional clksUwe Kleine-König
An error code returned by devm_clk_get() might have other meanings than "This clock doesn't exist". So use devm_clk_get_optional() and handle all remaining errors as fatal. Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-05-22net: phy: DP83822: enable rgmii mode if phy_interface_is_rgmiiTommaso Merciai
RGMII mode can be enable from dp83822 straps, and also writing bit 9 of register 0x17 - RMII and Status Register (RCSR). When phy_interface_is_rgmii rgmii mode must be enabled, same for contrary, this prevents malconfigurations of hw straps References: - https://www.ti.com/lit/gpn/dp83822i p66 Signed-off-by: Tommaso Merciai <tommaso.merciai@amarulasolutions.com> Co-developed-by: Michael Trimarchi <michael@amarulasolutions.com> Suggested-by: Alberto Bianchi <alberto.bianchi@amarulasolutions.com> Tested-by: Tommaso Merciai <tommaso.merciai@amarulasolutions.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-05-22pinctrl: berlin: bg4ct: Use devm_platform_*ioremap_resource() APIszhaoxiao
Use devm_platform_get_and_ioremap_resource() and devm_platform_ioremap_resource() APIs instead of their open coded analogues. Signed-off-by: zhaoxiao <zhaoxiao@uniontech.com> Link: https://lore.kernel.org/r/20220520052021.25631-1-zhaoxiao@uniontech.com Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2022-05-22random: check for signals after page of pool writesJason A. Donenfeld
get_random_bytes_user() checks for signals after producing a PAGE_SIZE worth of output, just like /dev/zero does. write_pool() is doing basically the same work (actually, slightly more expensive), and so should stop to check for signals in the same way. Let's also name it write_pool_user() to match get_random_bytes_user(), so this won't be misused in the future. Before this patch, massive writes to /dev/urandom would tie up the process for an extremely long time and make it unterminatable. After, it can be successfully interrupted. The following test program can be used to see this works as intended: #include <unistd.h> #include <fcntl.h> #include <signal.h> #include <stdio.h> static unsigned char x[~0U]; static void handle(int) { } int main(int argc, char *argv[]) { pid_t pid = getpid(), child; int fd; signal(SIGUSR1, handle); if (!(child = fork())) { for (;;) kill(pid, SIGUSR1); } fd = open("/dev/urandom", O_WRONLY); pause(); printf("interrupted after writing %zd bytes\n", write(fd, x, sizeof(x))); close(fd); kill(child, SIGTERM); return 0; } Result before: "interrupted after writing 2147479552 bytes" Result after: "interrupted after writing 4096 bytes" Cc: Dominik Brodowski <linux@dominikbrodowski.net> Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2022-05-22gpio: sim: Use correct order for the parameters of devm_kcalloc()Christophe JAILLET
We should have 'n', then 'size', not the opposite. This is harmless because the 2 values are just multiplied, but having the correct order silence a (unpublished yet) smatch warning. Fixes: cb8c474e79be ("gpio: sim: new testing module") Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr> Signed-off-by: Bartosz Golaszewski <brgl@bgdev.pl>
2022-05-22net: ipa: use data space for command opcodesAlex Elder
The 64-bit data field in a transaction is not used for commands. And the opcode array is *only* used for commands. They're (currently) the same size; save a little space in the transaction structure by enclosing the two fields in a union. Signed-off-by: Alex Elder <elder@linaro.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-05-22net: ipa: remove command info poolAlex Elder
The ipa_cmd_info structure now contains only one field, and it's an enumerated type whose values all fit in 8 bits. Currently we'll never use more than 8 TREs in a command transaction, and we can represent that number of command opcodes in the same space as a 64 bit pointer to an ipa_cmd_info structure. Define IPA_COMMAND_TRANS_TRE_MAX as the maximum number of TREs that can be in a command transaction. Replace the info pointer in a transaction with a fixed-size array named cmd_opcode[] of that many bytes. Store the opcode in this array when adding a command TRE to a transaction, as was done previously for the info array. This makes the ipa_cmd_info unused, so get rid of it. When committing an immediate command transaction, use the channel's Boolean command flag to determine whether to fill in the opcode, which will be taken (as before) from the array in the transaction. This makes the command info pool unnecessary, so get rid of it. Signed-off-by: Alex Elder <elder@linaro.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-05-22net: ipa: remove command direction argumentAlex Elder
We no longer use the direction argument for gsi_trans_cmd_add(), so get rid of it in its definition, and in its seven callers. Signed-off-by: Alex Elder <elder@linaro.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-05-22net: ipa: get rid of ipa_cmd_info->directionAlex Elder
The direction field of the ipa_cmd_info structure is set, but never used. It seems it might have been used for the DMA_SHARED_MEM immediate command, but the DIRECTION flag is set based on the value of the passed-in direction flag there. Anyway, remove this unused field from the ipa_cmd_info structure. This is done as a separate patch to make it very obvious that it's not required. Signed-off-by: Alex Elder <elder@linaro.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-05-22net: ipa: count the number of modem TX endpointsAlex Elder
In ipa_endpoint_modem_exception_reset_all(), a high estimate was made of the number of endpoints that need their status register updated. We only used what was needed, so the high estimate didn't matter much. However the next few patches are going to limit the number of commands in a single transaction, and the overestimate would exceed that. So count the number of modem TX endpoints at initialization time, and use it in ipa_endpoint_modem_exception_reset_all(). Signed-off-by: Alex Elder <elder@linaro.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-05-22net: ipa: kill gsi_trans_commit_wait_timeout()Alex Elder
Since the beginning gsi_trans_commit_wait_timeout() has existed to provide a way to allow waiting a limited time for a transaction to complete. But that function has never been used. In fact, there is no use for this function, because a transaction committed to hardware should *always* complete. The only reason it might not complete is if there were a hardware failure, or perhaps a system configuration error. Furthermore, if a timeout ever did occur, the IPA hardware would be in an indeterminate state, from which there is no recovery. It would require some sort of complete IPA reset, and would require the participation of the modem, and at this time there is no such sequence defined. So get rid of the definition of gsi_trans_commit_wait_timeout(), and update a few comments accordingly. Signed-off-by: Alex Elder <elder@linaro.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-05-22net: ipa: specify RX aggregation time limit in config dataAlex Elder
Don't assume that a 500 microsecond time limit should be used for all receive endpoints that support aggregation. Instead, specify the time limit to use in the configuration data. Set a 500 microsecond limit for all existing RX endpoints, as before. Checking for overflow for the time limit field is a bit complicated. Rather than duplicate a lot of code in ipa_endpoint_data_valid_one(), call WARN() if any value is found to be too large when encoding it. Signed-off-by: Alex Elder <elder@linaro.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-05-22net: ipa: support hard aggregation limitsAlex Elder
Add a new flag for AP receive endpoints that indicates whether a "hard limit" is used as a criterion for closing aggregation. Add comments explaining the difference between "hard" and "soft" aggregation limits. Pass a flag to ipa_aggr_size_kb() so it computes the proper aggregation size value whether using hard or soft limits. Move that function earlier in "ipa_endpoint.c" so it can be used without a forward-reference. Update ipa_endpoint_data_valid_one() so it validates endpoints whose data indicate a hard aggregation limit is used, and so it reports set aggregation flags for endpoints without aggregation enabled. Signed-off-by: Alex Elder <elder@linaro.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-05-22net: ipa: make endpoint HOLB drop configurableAlex Elder
Add a new Boolean flag for RX endpoints defining whether HOLB drop is initially enabled or disabled for the endpoint. All existing AP endpoints should have HOLB drop disabled. Signed-off-by: Alex Elder <elder@linaro.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-05-22qed: fix typos in commentsJulia Lawall
Spelling mistakes (triple letters) in comments. Detected with the help of Coccinelle. Signed-off-by: Julia Lawall <Julia.Lawall@inria.fr> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-05-22nfp: flower: fix typo in commentJulia Lawall
Spelling mistake (triple letters) in comment. Detected with the help of Coccinelle. Signed-off-by: Julia Lawall <Julia.Lawall@inria.fr> Reviewed-by: Niklas Söderlund <niklas.soderlund@corigine.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-05-22net: marvell: prestera: fix typo in commentJulia Lawall
Spelling mistake (triple letters) in comment. Detected with the help of Coccinelle. Signed-off-by: Julia Lawall <Julia.Lawall@inria.fr> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-05-22cirrus: cs89x0: fix typo in commentJulia Lawall
Spelling mistake (triple letters) in comment. Detected with the help of Coccinelle. Signed-off-by: Julia Lawall <Julia.Lawall@inria.fr> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-05-22net/mlx5: fix typo in commentJulia Lawall
Spelling mistake (triple letters) in comment. Detected with the help of Coccinelle. Signed-off-by: Julia Lawall <Julia.Lawall@inria.fr> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-05-22net: mvpp2: fix typo in commentJulia Lawall
Spelling mistake (triple letters) in comment. Detected with the help of Coccinelle. Signed-off-by: Julia Lawall <Julia.Lawall@inria.fr> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-05-22net: sparx5: switchdev: fix typo in commentJulia Lawall
Spelling mistake (triple letters) in comment. Detected with the help of Coccinelle. Signed-off-by: Julia Lawall <Julia.Lawall@inria.fr> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-05-22hwmon: (aquacomputer_d5next) Fix an error handling path in aqc_probe()Christophe JAILLET
If no memory can be allocated, some resources still need to be released as already done in the other error handling paths. Fixes: 752b927951ea ("hwmon: (aquacomputer_d5next) Add support for Aquacomputer Octo") Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr> Link: https://lore.kernel.org/r/be6b955d50de140fcc96bd116150b435021bf567.1653225250.git.christophe.jaillet@wanadoo.fr Reviewed-by: Aleksa Savic <savicaleksa83@gmail.com> Signed-off-by: Guenter Roeck <linux@roeck-us.net>
2022-05-22habanalabs: use separate structure info for each error collect dataTal Cohen
Create separate info structure for each error type. The structures shall be used inside the large structure that contains the last session error. This is more scalable for adding more errors in the future. Signed-off-by: Tal Cohen <talcohen@habana.ai> Reviewed-by: Oded Gabbay <ogabbay@kernel.org> Signed-off-by: Oded Gabbay <ogabbay@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2022-05-22habanalabs: fix missing handle shift during mmapYuri Nudelman
During mmap operation on the unified memory manager buffer, the vma page offset is shifted to extract the handle value. Due to a typo, it was not shifted back at the end. That could cause the offset to be modified after mmap operation, that may affect subsequent operations. In addition, in allocation flow, in case of out of memory error, idr would not be correctly destroyed, again because of a missing shift. Signed-off-by: Yuri Nudelman <ynudelman@habana.ai> Reviewed-by: Oded Gabbay <ogabbay@kernel.org> Signed-off-by: Oded Gabbay <ogabbay@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2022-05-22habanalabs: remove hdev from hl_ctx_get argsOhad Sharabi
This argument is unused by the function. Signed-off-by: Ohad Sharabi <osharabi@habana.ai> Reviewed-by: Oded Gabbay <ogabbay@kernel.org> Signed-off-by: Oded Gabbay <ogabbay@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2022-05-22habanalabs: do MMU prefetch as deferred workOhad Sharabi
When user requests to prefetch the MMU translations, the driver will not block the user until prefetch is done. Instead, the prefetch work will be delegated to a WQ which will do it in the background. This way, the prefetch may progress without blocking the user at all. Signed-off-by: Ohad Sharabi <osharabi@habana.ai> Reviewed-by: Oded Gabbay <ogabbay@kernel.org> Signed-off-by: Oded Gabbay <ogabbay@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2022-05-22habanalabs: order memory manager messagesYuri Nudelman
Changing format of memory manager messages to make it more readable. In addition, reducing the priority of a warning on missing handle during put. This scenario is not an indication of a problem and may happen in a legal flow, when handle is put from multiple flows. For example, in timeout and completion. Signed-off-by: Yuri Nudelman <ynudelman@habana.ai> Reviewed-by: Oded Gabbay <ogabbay@kernel.org> Signed-off-by: Oded Gabbay <ogabbay@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2022-05-22habanalabs: return -EFAULT on copy_to_user errorOded Gabbay
If copy_to_user failed in info ioctl, we always return -EFAULT so the user will know there was an error. Reported-by: kernel test robot <lkp@intel.com> Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Oded Gabbay <ogabbay@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2022-05-22habanalabs: use NULL for eventfdOded Gabbay
eventfd is pointer. As such, it should be initialized to NULL, not to 0. In addition, no need to initialize it after creation because the entire structure is zeroed-out. Also, no need to initialize it before release because the entire structure is freed. Reported-by: kernel test robot <lkp@intel.com> Signed-off-by: Oded Gabbay <ogabbay@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2022-05-22habanalabs: update firmware headerOded Gabbay
Update cpucp_if.h to latest version. Signed-off-by: Oded Gabbay <ogabbay@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2022-05-22habanalabs: add support for notification via eventfdTal Cohen
The driver will be able to send notification events towards a user process, using user's registered event file descriptor. The driver uses the notification mechanism to inform the user about an occurred event. A user thread can wait until a notification is received from the driver. The driver stores the occurred event until the user reads it, using HL_INFO_GET_EVENTS - new ioctl opcode in the INFO ioctl. Gaudi specific implementation includes sending a notification on a TPC assertion event that is received from f/w. Signed-off-by: Tal Cohen <talcohen@habana.ai> Reviewed-by: Oded Gabbay <ogabbay@kernel.org> Signed-off-by: Oded Gabbay <ogabbay@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2022-05-22habanalabs: add topic to memory manager bufferYuri Nudelman
Currently, buffers from multiple flows pass through the same infra. This way, in logs, we are unable to distinguish between buffers that came from separate flows. To address this problem, add a "topic" to buffer behavior descriptor - a string identifier that will be used to identify in logs the flow this buffer relates to. Signed-off-by: Yuri Nudelman <ynudelman@habana.ai> Reviewed-by: Oded Gabbay <ogabbay@kernel.org> Signed-off-by: Oded Gabbay <ogabbay@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2022-05-22habanalabs: handle race in driver finiDani Liberman
Scenario: 1. During hard reset, driver executes device_kill_open_processes. 2. Drivers file descriptor is not closed yet (user process is alive), hence we are starting loop on all open file descriptors. 3. Just before getting task struct of user process, according to pid, SIGKILL is sent to the user process, hence get_pid_task fails, driver prints a warning and device_kill_open_processes returns an error. 4. Returned error causing driver fini do disable the device object of the process which causes a kernel crash. The fix is to handle this case not as an error and continue fini flow as normal, since the killed process (by the SIGKILL) will release its resources just like it will do when the driver sends him the sigkill. Signed-off-by: Dani Liberman <dliberman@habana.ai> Reviewed-by: Oded Gabbay <ogabbay@kernel.org> Signed-off-by: Oded Gabbay <ogabbay@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2022-05-22habanalabs: add device memory scrub ability through debugfsDafna Hirschfeld
Add the ability to scrub the device memory with a given value. Add file 'dram_mem_scrub_val' to set the value and a file 'dram_mem_scrub' to scrub the dram. This is very important to help during automated tests, when you want the CI system to randomize the memory before training certain DL topologies. Signed-off-by: Dafna Hirschfeld <dhirschfeld@habana.ai> Reviewed-by: Oded Gabbay <ogabbay@kernel.org> Signed-off-by: Oded Gabbay <ogabbay@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2022-05-22habanalabs: use unified memory manager for CB flowYuri Nudelman
With the new code required for the flow added, we can now switch to using the new memory manager infrastructure, removing the old code. Signed-off-by: Yuri Nudelman <ynudelman@habana.ai> Reviewed-by: Oded Gabbay <ogabbay@kernel.org> Signed-off-by: Oded Gabbay <ogabbay@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2022-05-22habanalabs: unified memory manager new code for CB flowYuri Nudelman
This commit adds the new code needed for command buffer flow using the new unified memory manager, without changing the actual functionality. Signed-off-by: Yuri Nudelman <ynudelman@habana.ai> Reviewed-by: Oded Gabbay <ogabbay@kernel.org> Signed-off-by: Oded Gabbay <ogabbay@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2022-05-22habanalabs/gaudi: set arbitration timeout to a high valueOfir Bitton
In certain workloads, arbitration timeout might expire although no actual issue present. Hence, we set timeout to a very high value. Signed-off-by: Ofir Bitton <obitton@habana.ai> Reviewed-by: Oded Gabbay <ogabbay@kernel.org> Signed-off-by: Oded Gabbay <ogabbay@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2022-05-22habanalabs: add put by handle method to memory managerYuri Nudelman
Putting object by its handle and not by object pointer is useful in some finalization flows that do not have object pointer available. It eliminates the need to first get the object and then perform put twice. Signed-off-by: Yuri Nudelman <ynudelman@habana.ai> Reviewed-by: Oded Gabbay <ogabbay@kernel.org> Signed-off-by: Oded Gabbay <ogabbay@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2022-05-22habanalabs: hide memory manager page shiftYuri Nudelman
The new unified memory manager uses page offset to pass buffer handle during the mmap operation. One problem with this approach is that it requires the handle to always be divisible by the page size, else, the user would not be able to pass it correctly as an argument to the mmap system call. Previously, this was achieved by shifting the handle left after alloc operation, and shifting it right before get operation. This was done in the user code. This creates code duplication, and, what's worse, requires some knowledge from the user regarding the handle internal structure, hurting the encapsulation. This patch encloses all the page shifts inside memory manager functions. This way, the user can take the handle as a black box, and simply use it, without any concert about how it actually works. Signed-off-by: Yuri Nudelman <ynudelman@habana.ai> Reviewed-by: Oded Gabbay <ogabbay@kernel.org> Signed-off-by: Oded Gabbay <ogabbay@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
generated by cgit (git 2.34.1) at 2025-09-13 13:22:05 +0000