kernel/drivers/net/ethernet/intel/e1000e/mac.c, branch linux-6.18.y

e1000e: Fix real-time violations on link up

2025-02-10T18:43:48Z

Link down and up triggers update of MTA table. This update executes many PCIe writes and a final flush. Thus, PCIe will be blocked until all writes are flushed. As a result, DMA transfers of other targets suffer from delay in the range of 50us. This results in timing violations on real-time systems during link down and up of e1000e in combination with an Intel i3-2310E Sandy Bridge CPU. The i3-2310E is quite old. Launched 2011 by Intel but still in use as robot controller. The exact root cause of the problem is unclear and this situation won't change as Intel support for this CPU has ended years ago. Our experience is that the number of posted PCIe writes needs to be limited at least for real-time systems. With posted PCIe writes a much higher throughput can be generated than with PCIe reads which cannot be posted. Thus, the load on the interconnect is much higher. Additionally, a PCIe read waits until all posted PCIe writes are done. Therefore, the PCIe read can block the CPU for much more than 10us if a lot of PCIe writes were posted before. Both issues are the reason why we are limiting the number of posted PCIe writes in row in general for our real-time systems, not only for this driver. A flush after a low enough number of posted PCIe writes eliminates the delay but also increases the time needed for MTA table update. The following measurements were done on i3-2310E with e1000e for 128 MTA table entries: Single flush after all writes: 106us Flush after every write: 429us Flush after every 2nd write: 266us Flush after every 4th write: 180us Flush after every 8th write: 141us Flush after every 16th write: 121us A flush after every 8th write delays the link up by 35us and the negative impact to DMA transfers of other targets is still tolerable. Execute a flush after every 8th write. This prevents overloading the interconnect with posted writes. Signed-off-by: Gerhard Engleder Link: https://lore.kernel.org/netdev/f8fe665a-5e6c-4f95-b47a-2f3281aa0e6c@lunn.ch/T/ CC: Vitaly Lifshits Reviewed-by: Przemek Kitszel Tested-by: Avigail Dahan Reviewed-by: Vitaly Lifshits Reviewed-by: Simon Horman Signed-off-by: Tony Nguyen

intel: legacy: field get conversion

2023-12-18T19:20:43Z

Refactor several older Intel drivers to use FIELD_GET(), which reduces lines of code and adds clarity of intent. This code was generated by the following coccinelle/spatch script and then manually repaired. @get@ constant shift,mask; type T; expression a; @@ ( -((T)((a) & mask) >> shift) +FIELD_GET(mask, a) and applied via: spatch --sp-file field_prep.cocci --in-place --dir \ drivers/net/ethernet/intel/ Cc: Julia Lawall CC: Alexander Lobakin Reviewed-by: Marcin Szycik Reviewed-by: Simon Horman Signed-off-by: Jesse Brandeburg Tested-by: Pucha Himasekhar Reddy (A Contingent worker at Intel) Signed-off-by: Tony Nguyen

e1000e: Use pcie_capability_read_word() for reading LNKSTA

2023-12-12T20:21:29Z

Use pcie_capability_read_word() for reading LNKSTA and remove the custom define that matches to PCI_EXP_LNKSTA. As only single user for cap_offset remains, replace it with a call to pci_pcie_cap(). Instead of e1000_adapter, make local variable out of pci_dev because both users are interested in it. Signed-off-by: Ilpo Järvinen Reviewed-by: Jonathan Cameron Reviewed-by: Simon Horman Tested-by: Naama Meir Signed-off-by: Tony Nguyen

e1000e: Use PCI_EXP_LNKSTA_NLW & FIELD_GET() instead of custom defines/code

2023-12-12T18:55:22Z

e1000e has own copy of PCI Negotiated Link Width field defines. Use the ones from include/uapi/linux/pci_regs.h instead of the custom ones and remove the custom ones and convert to FIELD_GET(). Suggested-by: Jonathan Cameron Signed-off-by: Ilpo Järvinen Reviewed-by: Jonathan Cameron Reviewed-by: Simon Horman Tested-by: Naama Meir Signed-off-by: Tony Nguyen

intel/e1000e:fix repeated words in comments

2022-06-30T17:49:06Z

Delete the redundant word 'frames'. Signed-off-by: Jilin Yuan Signed-off-by: Tony Nguyen

net: intel: e1000e: fix possible sleep-in-atomic-context bugs in e1000e_get_hw_semaphore()

2020-02-19T23:25:59Z

The driver may sleep while holding a spinlock. The function call path (from bottom to top) in Linux 4.19 is: drivers/net/ethernet/intel/e1000e/mac.c, 1366: usleep_range in e1000e_get_hw_semaphore drivers/net/ethernet/intel/e1000e/80003es2lan.c, 322: e1000e_get_hw_semaphore in e1000_release_swfw_sync_80003es2lan drivers/net/ethernet/intel/e1000e/80003es2lan.c, 197: e1000_release_swfw_sync_80003es2lan in e1000_release_phy_80003es2lan drivers/net/ethernet/intel/e1000e/netdev.c, 4883: (FUNC_PTR) e1000_release_phy_80003es2lan in e1000e_update_phy_stats drivers/net/ethernet/intel/e1000e/netdev.c, 4917: e1000e_update_phy_stats in e1000e_update_stats drivers/net/ethernet/intel/e1000e/netdev.c, 5945: e1000e_update_stats in e1000e_get_stats64 drivers/net/ethernet/intel/e1000e/netdev.c, 5944: spin_lock in e1000e_get_stats64 drivers/net/ethernet/intel/e1000e/mac.c, 1384: usleep_range in e1000e_get_hw_semaphore drivers/net/ethernet/intel/e1000e/80003es2lan.c, 322: e1000e_get_hw_semaphore in e1000_release_swfw_sync_80003es2lan drivers/net/ethernet/intel/e1000e/80003es2lan.c, 197: e1000_release_swfw_sync_80003es2lan in e1000_release_phy_80003es2lan drivers/net/ethernet/intel/e1000e/netdev.c, 4883: (FUNC_PTR) e1000_release_phy_80003es2lan in e1000e_update_phy_stats drivers/net/ethernet/intel/e1000e/netdev.c, 4917: e1000e_update_phy_stats in e1000e_update_stats drivers/net/ethernet/intel/e1000e/netdev.c, 5945: e1000e_update_stats in e1000e_get_stats64 drivers/net/ethernet/intel/e1000e/netdev.c, 5944: spin_lock in e1000e_get_stats64 (FUNC_PTR) means a function pointer is called. To fix these bugs, usleep_range() is replaced with udelay(). These bugs are found by a static analysis tool STCheck written by myself. Signed-off-by: Jia-Ju Bai Tested-by: Aaron Brown Signed-off-by: Jeff Kirsher

e1000e: Reduce boot time by tightening sleep ranges

2019-06-28T22:59:52Z

The e1000e driver is a great user of the usleep_range() API, and has nice ranges that in principle help power management. However the ranges that are used only during system startup are very long (and can add easily 100 msec to the boot time) while the power savings of such long ranges is irrelevant due to the one-off, boot only, nature of these functions. This patch shrinks some of the longest ranges to be shorter (while still using a power friendly 1 msec range); this saves 100msec+ of boot time on my BDW NUCs Signed-off-by: Arjan van de Ven Signed-off-by: Paul Menzel Tested-by: Aaron Brown Signed-off-by: Jeff Kirsher

net: intel: Cleanup the copyright/license headers

2018-04-27T18:00:04Z

After many years of having a ~30 line copyright and license header to our source files, we are finally able to reduce that to one line with the advent of the SPDX identifier. Also caught a few files missing the SPDX license identifier, so fixed them up. Signed-off-by: Jeff Kirsher Acked-by: Shannon Nelson Acked-by: Richard Cochran Tested-by: Andrew Bowers Signed-off-by: David S. Miller

intel: add SPDX identifiers to all the Intel drivers

2018-03-23T16:18:21Z

Add the SPDX identifiers to all the Intel wired LAN driver files, as outlined in Documentation/process/license-rules.rst. Signed-off-by: Jeff Kirsher Tested-by: Aaron Brown Signed-off-by: David S. Miller

e1000e: Fix link check race condition

2018-03-12T19:05:39Z

Alex reported the following race condition: /* link goes up... interrupt... schedule watchdog */ \ e1000_watchdog_task \ e1000e_has_link \ hw->mac.ops.check_for_link() === e1000e_check_for_copper_link \ e1000e_phy_has_link_generic(..., &link) link = true /* link goes down... interrupt */ \ e1000_msix_other hw->mac.get_link_status = true /* link is up */ mac->get_link_status = false link_active = true /* link_active is true, wrongly, and stays so because * get_link_status is false */ Avoid this problem by making sure that we don't set get_link_status = false after having checked the link. It seems this problem has been present since the introduction of e1000e. Link: https://lkml.org/lkml/2018/1/29/338 Reported-by: Alexander Duyck Signed-off-by: Benjamin Poirier Acked-by: Alexander Duyck Tested-by: Aaron Brown Signed-off-by: Jeff Kirsher