lwn.git/drivers/net/ethernet/intel, branch automarkup

mm: replace all open encodings for NUMA_NO_NODE

2019-03-06T05:07:14+00:00

Patch series "Replace all open encodings for NUMA_NO_NODE", v3. All these places for replacement were found by running the following grep patterns on the entire kernel code. Please let me know if this might have missed some instances. This might also have replaced some false positives. I will appreciate suggestions, inputs and review. 1. git grep "nid == -1" 2. git grep "node == -1" 3. git grep "nid = -1" 4. git grep "node = -1" This patch (of 2): At present there are multiple places where invalid node number is encoded as -1. Even though implicitly understood it is always better to have macros in there. Replace these open encodings for an invalid node number with the global macro NUMA_NO_NODE. This helps remove NUMA related assumptions like 'invalid node' from various places redirecting them to a common definition. Link: http://lkml.kernel.org/r/1545127933-10711-2-git-send-email-anshuman.khandual@arm.com Signed-off-by: Anshuman Khandual Reviewed-by: David Hildenbrand Acked-by: Jeff Kirsher [ixgbe] Acked-by: Jens Axboe [mtip32xx] Acked-by: Vinod Koul [dmaengine.c] Acked-by: Michael Ellerman [powerpc] Acked-by: Doug Ledford [drivers/infiniband] Cc: Joseph Qi Cc: Hans Verkuil Cc: Stephen Rothwell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

ice: fix overlong string, update stats output

2019-02-25T16:56:02+00:00

A test started warning on a string truncation. This led to an unfortunate realization that we are likely not accounting for the stats length correctly before this patch, so fix the issue by putting "port." in front of all the PF stats, instead of magically prepending it at runtime. Signed-off-by: Jesse Brandeburg Signed-off-by: Anirudh Venkataramanan Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher

ice: Fix for FC get rx/tx pause params

2019-02-25T16:56:02+00:00

Ethtool reported pause params based on the currently negotiated link settings instead of current PHY config. User was not able to turn off pause params because ethtool was incorrectly reporting parameters as off when link was down even though PHY was configured to support pause frames. Now pause params are taken from PHY config instead of link status. Signed-off-by: Lukasz Czapnik Signed-off-by: Anirudh Venkataramanan Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher

ice: use absolute vector ID for VFs

2019-02-25T16:56:02+00:00

When the PF driver sets up the VF MSI-X vector allocation, it needs to use the hardware absolute vector ID, not the per-PF vector ID. Without this change we see (apparent) TX hangs when using VFs on multiple PFs. Signed-off-by: Mitch Williams Signed-off-by: Anirudh Venkataramanan Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher

ice: check for a leaf node presence

2019-02-25T16:56:01+00:00

Check for a leaf node presence for a given VSI. This check is required before removing a VSI since VSIs can't be removed with enabled queues (with leaf nodes) from the FW scheduler tree unless its a reset. Signed-off-by: Victor Raj Reviewed-by: Bruce Allan Signed-off-by: Anirudh Venkataramanan Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher

ice: flush Tx pipe on disable queue timeout

2019-02-25T16:56:01+00:00

Set the flush Tx pipe flag instead of getting an EAGAIN error when FW times out in processing the disable Tx queue command. Signed-off-by: Victor Raj Reviewed-by: Bruce Allan Signed-off-by: Anirudh Venkataramanan Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher

ice: clear VF ARQLEN register on reset

2019-02-25T16:56:01+00:00

On older devices like X710 and X722, the VF's ARQLEN register is cleared on reset, so the VF driver uses that register to detect an unannounced reset. Unfortunately, on devices controlled by ice, this register is NOT cleared on reset. This causes the VF to miss resets, and even on properly-announced resets, the VF driver complains that it didn't see the reset. To fix this, we'll do it in software. When we handle a VF reset (whether triggered by software or VFLR), clear this register after the HW reset is complete. Signed-off-by: Mitch Williams Signed-off-by: Anirudh Venkataramanan Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher

ice: don't spam VFs with link messages

2019-02-25T16:56:01+00:00

Don't send a link message to the VFs unless link actually changes state. This avoids a small timing hole in some VF drivers that can cause an apparent TX hang if they receive a link status message at the wrong time. Although we have fixed the timing hole in the current VF driver, there are still lots of drivers in the field that have this timing hole. Let's not fall into it if we can avoid it. Signed-off-by: Mitch Williams Signed-off-by: Anirudh Venkataramanan Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher

ice: only use the VF for ICE_VSI_VF in ice_vsi_release

2019-02-25T16:56:01+00:00

In ice_vsi_release we are always assigning a value to the local VF variable. Change this to only be assigned if the VSI is a VF VSI. Signed-off-by: Brett Creeley Signed-off-by: Anirudh Venkataramanan Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher

ice: fix numeric overflow warning

2019-02-25T16:56:01+00:00

When compiling and analyzing the driver on newer kernels, a static analyzer warns about the following "numeric overflow" issues: "The result of expression: 'budget-1' generates 4-byte type while casting to a bigger size of 8-byte". "The result of expression: '*words-words_read' generates 4-byte type while casting to a bigger size of 8-byte". Fix them both. Signed-off-by: Bruce Allan Signed-off-by: Anirudh Venkataramanan Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher