Commit Graph

1138 Commits

Author SHA1 Message Date
Mohan Thadikamalla
67f9842d5e common: Fix ICD Doxygen Compilation
Issue:
The ICD Doxygen compilation failed for the PPS group.

Fix:
Updated the Doxygen @addtogroup comments for
the PPS group to resolve the compilation issue.

Bug 4585654
Bug 5042311
Jira NET-2169

Change-Id: Ifefad8374b269ede6afa12ab781324a98eef3a8a
Signed-off-by: Mohan Thadikamalla <mohant@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3291256
Reviewed-by: Rakesh Goyal <rgoyal@nvidia.com>
Tested-by: Rakesh Goyal <rgoyal@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Krishna Thota <kthota@nvidia.com>
2025-01-28 20:35:48 -08:00
Bhadram Varka
a3d4368889 osi: frp: Include L4 SRC/DST ports in safety build
Issue: L4 SRC/DST ports FRP configuration is not
working in safety builds because code is removed
from the safety builds.

Fix: Include L4 SRC/DST ports configuration in
safety builds

This change also fix the typo in the code

Bug 4821670

Change-Id: I575d00f199288fab14cf3744943050af914b1d58
Signed-off-by: Bhadram Varka <vbhadram@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3258850
(cherry picked from commit db4982965cff2869d806327878ada17002a91763)
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3286022
Reviewed-by: Krishna Thota <kthota@nvidia.com>
Reviewed-by: svcacv <svcacv@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: Srinivas Ramachandran <srinivasra@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
2025-01-27 09:43:04 -08:00
Bhadram Varka
339336ec1a osi: core: restore HSI configuration
Issue: HSI not enabled in HW after SC7.

Fix: Restore the HSI in resume path.

Bug 4881765

Change-Id: I1c6efbe735ee4da0c2817498b3f71912bc823ff0
Signed-off-by: Bhadram Varka <vbhadram@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3221011
(cherry picked from commit ba0960ad43bf22b3254b0213ee596a54c86c67d4)
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3286021
Reviewed-by: Krishna Thota <kthota@nvidia.com>
Reviewed-by: svcacv <svcacv@nvidia.com>
Reviewed-by: Srinivas Ramachandran <srinivasra@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
2025-01-27 09:42:55 -08:00
Harsukhwinder Singh
2078f0d51e osi: core: Replace osd_usleep_range with osd_usleep
Bug 4921002

Change-Id: Ia12aa1fb94a2b1fbe1afd0e7da3190857479c4f9
Signed-off-by: Harsukhwinder Singh <harsukhwinde@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3268811
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: svcacv <svcacv@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: Sanath Kumar Gampa <sgampa@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Ashutosh Jha <ajha@nvidia.com>
2025-01-22 23:28:47 -08:00
Rakesh Goyal
877664c2ec osi: add support for pps train in digital mode
Issue: Support for PPS train requested

Fix: Added support for PPS train

Bug 4585654
Bug 5042311

Change-Id: I0f94b8b4a5cb72d0084ae7ac14e1843930f6a1e8
Signed-off-by: Rakesh Goyal <rgoyal@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3215524
Reviewed-by: svcacv <svcacv@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: Srinivas Ramachandran <srinivasra@nvidia.com>
Reviewed-by: Bhadram Varka <vbhadram@nvidia.com>
2025-01-22 08:35:13 -08:00
Sanath Kumar Gampa
98f1e40481 osi: core: create separate nvethernet static lib
Create separate static lib for N1Auto so that Tegra specific source is
not included in it

Jira NET-2266

Change-Id: I528513f012102ec18521209ef31709dc6750712c
Signed-off-by: Sanath Kumar Gampa <sgampa@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3281903
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: svcacv <svcacv@nvidia.com>
Reviewed-by: Srinivas Ramachandran <srinivasra@nvidia.com>
2025-01-20 13:27:53 -08:00
Sanath Kumar Gampa
3b9355578b osi: create new lib for T26x EQoS XPCS programming
To support N1Auto platformw hich has different XPCS, seperated out the
T26x EQoS xpcs programming

Bug 4997903

Change-Id: I47d75b66b7c3e7a5b7f2ad2abe8452dd3c2e5656
Signed-off-by: Sanath Kumar Gampa <sgampa@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3268488
Reviewed-by: Srinivas Ramachandran <srinivasra@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
2025-01-20 13:27:49 -08:00
Bhadram Varka
907228f277 mgbe: xpcs: Perform Rx lane power down
Issue:
=====
An Ethernet server crash was observed due to an XPCS reset. A revert of
the XPCS reset changes was created, but this led to UPHY Rx lane
bring-up failures for the TS1 platform across interfaces.

In failure scenarios, the
PCS_WRAP_UPHY_HW_INIT_CONTROL_0_0[HW_RX_INIT_EN] register is not being
reset properly.

Experiments Conducted:
=====================
UPHY Reset Attempt:

Change-Id: I9ec5096ba88f839b932e001e95de71182a4d07bb
------------------
o Set PCS_WRAP_UPHY_HW_INIT_CONTROL_0_0[UPHY_RESET] = 0 followed by
PCS_WRAP_UPHY_HW_INIT_CONTROL_0_0[UPHY_RESET] = 1 before attempting the
Rx lane bring-up.
o Result: The PCS_WRAP_UPHY_HW_INIT_CONTROL_0_0[HW_RX_INIT_EN] register
remains set to 1.

Explicit Write to Reset Register:
---------------------------------
Explicitly wrote PCS_WRAP_UPHY_HW_INIT_CONTROL_0_0[HW_RX_INIT_EN] = 0.
Result: The register still reads back as 1.

Rx Lane Power Down on Failure Path:
----------------------------------
o Set PCS_WRAP_UPHY_HW_INIT_CONTROL_0_0[HW_RX_P_DN_INIT_EN] = 1 during the
Rx lane bring-up failure path to clear
PCS_WRAP_UPHY_HW_INIT_CONTROL_0_0[HW_RX_INIT_EN] = 0.

o Result: PCS_WRAP_UPHY_HW_INIT_CONTROL_0_0[HW_RX_INIT_EN] cleared
and UPHY Rx lane bring up is successful.

Discussions are ongoing with the hardware team to finalize detailed
steps for the UPHY Rx lane bring-up process.

Fix:
To address the Ethernet server crash and ensure the Ethernet interface
functions as in ToT, a power-down of the Rx lane during the UPHY Rx lane
bring-up failure path has been added along with XPCS reset revert change

Bug 4997856
Bug 5023657
Bug 5026511
Bug 5019079

Change-Id: Ib229d94c5f4672a78c1731f7e1f473a7caa80004
Signed-off-by: Bhadram Varka <vbhadram@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3282289
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Srinivas Ramachandran <srinivasra@nvidia.com>
Reviewed-by: svcacv <svcacv@nvidia.com>
2025-01-18 12:02:59 -08:00
Hareesh Kesireddy
75d7479131 osi: pass > mtu packets to sw without valid flag
- Pass > mtu sized packets to sw driver without setting
  OSI_PKT_CX_VALID flag to let driver discard the packet and
  realloc buffer for descriptor.

Bug 4969187

Change-Id: Icd2b0a35ccb4364147b7c5b5e49410acebace228
Signed-off-by: Hareesh Kesireddy <hkesireddy@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3257267
(cherry picked from commit 9625067d36dffcfd2db2eb2e6fc8a0a2a67f4d17)
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3258734
Reviewed-by: Bhadram Varka <vbhadram@nvidia.com>
Reviewed-by: Aniruddha Paul <anpaul@nvidia.com>
Reviewed-by: Ashutosh Jha <ajha@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
Reviewed-by: Krishna Thota <kthota@nvidia.com>
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-misra <svc-mobile-misra@nvidia.com>
Reviewed-by: svcacv <svcacv@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
2025-01-05 05:01:43 -08:00
Sanath Kumar Gampa
593c8fc44c nvethernetrm: Add DA to SCI LUT with peer MACID
Bug 4754899

Change-Id: I5c15770793d2f64fb54b565fd046ad9f9d7cb4bf
Signed-off-by: Sanath Kumar Gampa <sgampa@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3237499
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-misra <svc-mobile-misra@nvidia.com>
Reviewed-by: Mahesh Patil <maheshp@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Ashutosh Jha <ajha@nvidia.com>
Reviewed-by: svcacv <svcacv@nvidia.com>
2024-12-15 16:23:47 -08:00
kmuthusamy
248fb6375c nvethernetrm: fixes for QNX SDP7.1 compiler error
Bug 4893334

Signed-off-by: Karuna Muthusamy <kmuthusamy@nvidia.com>
Change-Id: I625ecf8bc2b9b40c76b575a41ff5f838aa295b82
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3262293
Reviewed-by: Hareesh Kesireddy <hkesireddy@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Srinivas Ramachandran <srinivasra@nvidia.com>
Reviewed-by: Bhadram Varka <vbhadram@nvidia.com>
Reviewed-by: svcacv <svcacv@nvidia.com>
2024-12-10 14:39:28 -08:00
Hareesh Kesireddy
6fa36e885d osi: use udelay in timestamp poll paths
- Use udelay() for timestamp register poll paths
  to avoid linux crash issues.
- Reduce reg polling time to 10msec instead of 1sec.

Bug 4951493

Change-Id: Id86a05c91f5e54b48912409225e7d7a850b097a3
Signed-off-by: Hareesh Kesireddy <hkesireddy@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3247162
(cherry picked from commit 3ccfb43c977bfdd5e38ed32eafda234e9972d888)
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3250080
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: svcacv <svcacv@nvidia.com>
Reviewed-by: Srinivas Ramachandran <srinivasra@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
2024-12-04 08:06:12 -08:00
Srinivas Ramachandran
1e8dfafa84 osi: Use osd_usleep vs osd_udelay appropriately
Issue: osd_udelay is used irrespective of the
       duration of the delay/sleep. In certain
       environments like HVRTOS, udelay is a
       busy loop and it starves other tasks on the CPU.
Fix:   Use udelay only for small tight checks, and use
       usleep for larger delays.

Bug 4676601

Change-Id: I59d9a403f34d46c6e2d17ca6f7e8a277d5283db5
Signed-off-by: Srinivas Ramachandran <srinivasra@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3176350
(cherry picked from commit c2abe16a34af853f86fcaa4bb91b7036e2a8eb79)
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3250083
Reviewed-by: Hareesh Kesireddy <hkesireddy@nvidia.com>
Reviewed-by: svcacv <svcacv@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Tested-by: Hareesh Kesireddy <hkesireddy@nvidia.com>
2024-12-04 08:06:03 -08:00
Mohan Thadikamalla
8b85eca5f2 OSI: Fix giza compilation Issues
Issue:
Doxygen compilation issues were observed
due to changes associated with 4569357 bug.

Fix:
Updated Doxygen comments in the OSI public
header files to resolve the compilation errors.

Bug 4569357

Change-Id: I4b87a5a498e956678f25f2674e7d84aefac17767
Signed-off-by: Mohan Thadikamalla <mohant@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3257876
Reviewed-by: Aniruddha Paul <anpaul@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Ashutosh Jha <ajha@nvidia.com>
Reviewed-by: svcacv <svcacv@nvidia.com>
2024-11-28 12:55:14 -08:00
Sanath Kumar Gampa
35eda10b40 osi: Segregate the MACSEC HSI code
Move MACSEC HSI code to MACSEC h/w specific lib

Bug 4942473

Change-Id: I8f04526f26774736d975fad340de3bfcbbc45dd1
Signed-off-by: Sanath Kumar Gampa <sgampa@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3243835
Reviewed-by: Bhadram Varka <vbhadram@nvidia.com>
Reviewed-by: svcacv <svcacv@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Krishna Thota <kthota@nvidia.com>
2024-11-28 12:39:46 -08:00
Aniruddha Paul
fc6e9530b6 OSI: Set Tx IOC for every fixed number of descs
A new field 'use_tx_descs' gets programmed by OSD layer.
Tx ring accounts current desc counts and when the current desc
count increases above use_tx_descs and the last bit is set
in the desc, we set the IOC bit.

Bug 4569357

Change-Id: Ida2c7b84e0096007b874e79e3b7502c997f71980
Signed-off-by: Aniruddha Paul <anpaul@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3207582
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: svcacv <svcacv@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Srinivas Ramachandran <srinivasra@nvidia.com>
2024-11-20 21:53:02 -08:00
Aniruddha Paul
22a75ac576 osi: Remove duplicate macros
The below macros were duplicated in the respective
components of osi/core and osi/dma.
1. MGBE_DMA_CHX_MAX_PBL
2. MGBE_DMA_CHX_PBL_16
3. MGBE_DMA_CHX_PBL_8
4. MGBE_DMA_CHX_PBL_4
5. MGBE_DMA_CHX_PBL_1
6. osi_valid_pbl_value()
7. osi_memset()

Move these common macros and APIs which are used by both
osi/core and osi/dma to include/osi_common.h

This also statically assigns TxPBL=16 for Orin.

Bug 4569357

Change-Id: I390e0ad9c0bfda47a1a7f9dd94cf5f7f45d96b9c
Signed-off-by: Aniruddha Paul <anpaul@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3211481
Reviewed-by: Mahesh Patil <maheshp@nvidia.com>
Reviewed-by: Ashutosh Jha <ajha@nvidia.com>
Reviewed-by: Srinivas Ramachandran <srinivasra@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
2024-11-20 21:52:52 -08:00
Mohan Thadikamalla
8cf1ea5184 osi: Fix ICD doxygen compilation issues
Issue:
ICD Doxygen compilation issues have been
observed with Docker version 1.9.0.

Fix:
Update the Doxygen comments in the OSI
public header files to resolve
the Doxygen compilation issues.

Jira NET-1801

Change-Id: I15f305cbdb1bb1a88132498886da353d2bbb714f
Signed-off-by: Mohan Thadikamalla <mohant@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3242062
Reviewed-by: svcacv <svcacv@nvidia.com>
Reviewed-by: Srinivas Ramachandran <srinivasra@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Bhadram Varka <vbhadram@nvidia.com>
2024-11-07 16:22:45 -08:00
Sanath Kumar Gampa
25a07d7228 osi_core: macsec: Reorg MACSec code for DOS-SHR-10437
Reorganize MACSec source code to accommodate N1Auto MACSec HW IP OSI
driver.

Separated MACSec hardware-specific code into a different static lib

Bug 4874880

Change-Id: I74e4cca8ba615def283ec938cc94985d32099190
Signed-off-by: Sanath Kumar Gampa <sgampa@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3214907
Reviewed-by: Bhadram Varka <vbhadram@nvidia.com>
Reviewed-by: Mahesh Patil <maheshp@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Ashutosh Jha <ajha@nvidia.com>
2024-10-29 08:12:49 -07:00
Mahesh Patil
8ebeeaafe3 osi: Add osi dma/core release functions
- Adding osi dma and osi core release function to release resources

Bug 4791340

Change-Id: I031d11ca34a31d2f7c5c8460251509714cba5509
Signed-off-by: Mahesh Patil <maheshp@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3214558
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
2024-10-16 19:47:53 -07:00
Mahesh Patil
45f67fb52f osi: Add PCS BASE-R FEC setting
Adding PCS BASE-R FEC setting using on
sysfs node

Bug 4674473

Change-Id: Icdc2300705d11ce2c96e2c8e1663599a3bb4fadd
Signed-off-by: Mahesh Patil <maheshp@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3201974
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: Nagaraj Annaiah <nannaiah@nvidia.com>
Reviewed-by: Srinivas Ramachandran <srinivasra@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
Reviewed-by: svc-mobile-misra <svc-mobile-misra@nvidia.com>
2024-10-05 04:20:43 -07:00
Rakesh Goyal
cc7aae9570 osi: core: correct SSINC value for thor EQOS
Issue: SSINC and ptp_ref_clk are wongly
programmed
Fix: SSINC should be 4 as per HW guideline
ptp_ref_clk should be 312.5 MHz

Bug 4747430

Change-Id: Ibbcbfa072c91ccb0cf0271169a52fa09869b5038
Signed-off-by: Rakesh Goyal <rgoyal@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3198414
Reviewed-by: Bhadram Varka <vbhadram@nvidia.com>
Reviewed-by: Hareesh Kesireddy <hkesireddy@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Krishna Thota <kthota@nvidia.com>
2024-10-05 04:19:16 -07:00
Mahesh Patil
c8bea5c22f osi: Properly set the dma chan bits
- Properly set the dma chan bits for chan > 31 in receive chan list
  bitmap when filters are added
- Fix delete condition for T264 when filter is deleted

Bug 4844451

Change-Id: Ia9f9e7a99b5b3d10813d10aded55061754ed1824
Signed-off-by: Mahesh Patil <maheshp@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3209203
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
Reviewed-by: svcacv <svcacv@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Ashutosh Jha <ajha@nvidia.com>
Reviewed-by: svc-mobile-misra <svc-mobile-misra@nvidia.com>
Reviewed-by: Rakesh Goyal <rgoyal@nvidia.com>
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
2024-09-23 12:50:54 -07:00
Mahesh Patil
7afef3637f osi: Enable all MGBE 48 VDMA channels
Bug 4746911

Change-Id: Ieff8f483df47083722b85ade5d591217c9a27380
Signed-off-by: Mahesh Patil <maheshp@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3178611
Reviewed-by: Ashutosh Jha <ajha@nvidia.com>
Reviewed-by: Nagaraj Annaiah <nannaiah@nvidia.com>
Reviewed-by: Michael Hsu <mhsu@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: svcacv <svcacv@nvidia.com>
2024-09-07 11:00:30 -07:00
Bhadram Varka
2e3848b279 mgbe: pcs: Fix offset for T26X_XPCS_WRAP_INTERRUPT_CONTROL
Fix the offset for T26X_XPCS_WRAP_INTERRUPT_CONTROL
and removes unused macro.

Bug 4778785

Change-Id: I3f2ac54c8119727f05fc77348d95b12e9ac3f525
Signed-off-by: Bhadram Varka <vbhadram@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3199177
Reviewed-by: Narayana Reddy P <narayanr@nvidia.com>
Reviewed-by: Krishna Thota <kthota@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
2024-08-28 16:59:22 -07:00
Narayan Reddy
a81ae285f2 osi: core: mgbe: Program ANP value for 25G
Issue: Seeing an increment in the ping RTT when
FRP rule is enabled in 25G.

Fix: by default ANP value is 5, which is recommended
for 10G. For 25G we need to change it to 3 as per HW
team suggestion.

Bug 4745125

Signed-off-by: Narayan Reddy <narayanr@nvidia.com>
Change-Id: I2e2cc0f49199908fb5919d0b29da2850030e6be2
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3197679
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
2024-08-22 13:35:28 -07:00
Narayan Reddy
c172c2201f mgbe: enable DRCHM for mmc rx counters
Issue: MMC Rx counters are not getting incremented

Fix: Enable DRCHM MMC_Control register for the
Rx counters to get incremented

Bug 4742494

Change-Id: I2dec776f12a0ecaeb4e3bfd1f4e040e9d7490e45
Signed-off-by: Narayan Reddy <narayanr@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3180193
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Srinivas Ramachandran <srinivasra@nvidia.com>
Reviewed-by: Bhadram Varka <vbhadram@nvidia.com>
2024-08-22 13:18:36 -07:00
Sanath Kumar Gampa
27ab898c1c osi: update delays for uPHY XLGPS init for 25G
Update the uPHY delays for 25G speed per manual

Bug 4790491

Change-Id: I79dc2d94db204f4e59a19f9763d53fbf7c99c80e
Signed-off-by: Sanath Kumar Gampa <sgampa@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3191324
Reviewed-by: Ashutosh Jha <ajha@nvidia.com>
Reviewed-by: Narayana Reddy P <narayanr@nvidia.com>
Reviewed-by: Mahesh Patil <maheshp@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
2024-08-22 04:58:28 -07:00
Sanath Kumar Gampa
e8bb74d0c1 osi: core: xpcs: Program RxEQ for 25G
Issue: Observed CRC failures when running perf with 25G speed.

Fix: RxEQ is enabled to avoid Rx CRC errors in perf runs

Bug 4756650

Change-Id: Idc531585e3fac686ca5518b56587980e779125ed
Signed-off-by: Sanath Kumar Gampa <sgampa@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3187277
Reviewed-by: Krishna Thota <kthota@nvidia.com>
Reviewed-by: svcacv <svcacv@nvidia.com>
Reviewed-by: Bhadram Varka <vbhadram@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
2024-08-09 21:58:19 -07:00
Sanath Kumar Gampa
35d0eb1f7d osi: MACSec: Update IPG changes for EQoS MACSec
Issue: IPG is not updated when MACsec is enabled for Thor

Fix: Removed the check to program IPG only if MAC type is Orin

Bug 4768403

Change-Id: I68ddac02d241a94e0a270d15d3c441fb7fc945a7
Signed-off-by: Sanath Kumar Gampa <sgampa@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3188748
Reviewed-by: Krishna Thota <kthota@nvidia.com>
Reviewed-by: svcacv <svcacv@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Mahesh Patil <maheshp@nvidia.com>
2024-08-09 13:01:29 -07:00
Mahesh Patil
26c551ceba app: FSI EQOS driver bring up
- Fix GVS build errors

Bug 4552441

Change-Id: Ib9a25b34ad4980d966806238aadf27dc13395044
Signed-off-by: Mahesh Patil <maheshp@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3185939
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Michael Hsu <mhsu@nvidia.com>
Reviewed-by: Ashutosh Jha <ajha@nvidia.com>
2024-08-06 15:33:57 -07:00
Rakesh Goyal
4a0972a389 osi: core: fix tsn hw initialization
Issue: TSN init is not correct for mgbe
Fix: Fix integration error from side branch to
     dev-main by adding missing code. Now TSN
     init is correct.

Bug 4742997
Bug 4436888

Change-Id: I70d4ca4cad0901a367731e699c7d641a4842d442
Signed-off-by: Rakesh Goyal <rgoyal@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3173951
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Krishna Thota <kthota@nvidia.com>
Reviewed-by: svcacv <svcacv@nvidia.com>
Reviewed-by: Mahesh Patil <maheshp@nvidia.com>
2024-08-05 12:21:34 -07:00
Rakesh Goyal
eeeb693843 osi: fix ptp tx timestamp for vdma
Issue: PTP time sync not working for vdma other
       than 0.
Fix: Fix integration error from side branch to
     dev-main by adding missing code. Now PTP time
     sync works for any vdma

Bug 4768411

Change-Id: I7f5727f5a0ba66de9baf8983707c54e02e7ef25e
Signed-off-by: Rakesh Goyal <rgoyal@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3182405
Reviewed-by: Mahesh Patil <maheshp@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-misra <svc-mobile-misra@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Krishna Thota <kthota@nvidia.com>
Reviewed-by: Bhadram Varka <vbhadram@nvidia.com>
2024-08-01 06:04:41 -07:00
Sanath Kumar Gampa
f386d8cba7 osi: Update MGBE MACSec Tx SOT value
Update the MGBE MACSec Tx SOT value as per MACSec IAS

Bug 4753677

Change-Id: I352debc77f916abd190e47394e46ec1b410c5460
Signed-off-by: Sanath Kumar Gampa <sgampa@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3180861
Reviewed-by: svc-mobile-misra <svc-mobile-misra@nvidia.com>
Reviewed-by: Mahesh Patil <maheshp@nvidia.com>
Reviewed-by: Srinivas Ramachandran <srinivasra@nvidia.com>
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: Ashutosh Jha <ajha@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
2024-08-01 06:01:00 -07:00
Mahesh Patil
cfdfccb488 osi: FSI EQOS driver bring up
Bug 4552441

Change-Id: I490f9680cc7a71754a64035fb7c63ba56d0287ba
Signed-off-by: Mahesh Patil <maheshp@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3166031
Reviewed-by: Ashutosh Jha <ajha@nvidia.com>
Reviewed-by: Michael Hsu <mhsu@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
2024-07-28 21:35:56 -07:00
Narayan Reddy
e6c71e09d0 osi: Adding WAR changes for MGBE 25G
Adding MGBE 25G WAR changes to match HW script sequence

Bug 4709627
Bug 4713751

Signed-off-by: Narayan Reddy <narayanr@nvidia.com>
Change-Id: Ie7aaf1c8c8b9bfbe62d4aa404fc65fe62dbf8efe
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3168031
Reviewed-by: Bhadram Varka <vbhadram@nvidia.com>
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-misra <svc-mobile-misra@nvidia.com>
Reviewed-by: Ashutosh Jha <ajha@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
2024-07-08 15:56:04 -07:00
Rakesh Goyal
b535351c41 osi: core: add support for new ethernet HW
Issue: kernel crash when set PTP time using phc_ctl
Fix: add register PTP register offset for new
ethernet HW

Bug 4727541

Change-Id: I60cee4e669dce470d19ff780efab08e7e3288ec2
Signed-off-by: Rakesh Goyal <rgoyal@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3167812
Reviewed-by: Ashutosh Jha <ajha@nvidia.com>
Reviewed-by: Narayana Reddy P <narayanr@nvidia.com>
Reviewed-by: Mahesh Patil <maheshp@nvidia.com>
Reviewed-by: svcacv <svcacv@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
2024-07-07 12:51:24 -07:00
Sanath Kumar Gampa
ca0db16125 osi: macsec: Update MACSEC_CONTROL1 per h/w
1) Stop MAC transmit while enabling MACSEC
2) Disallow FPE and MACSEC co-existence
3) Update SOT for T264 MACSEC
4) 10M and 100M speeds are not supported by MACSec

Bug 4588266
Bug 4456073

Change-Id: I3bc598bae8d2786104c2018bac18cc49092adcdd
Signed-off-by: Sanath Kumar Gampa <sgampa@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3163257
Reviewed-by: Ashutosh Jha <ajha@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Mahesh Patil <maheshp@nvidia.com>
2024-07-02 05:08:44 -07:00
Sanath Kumar Gampa
c8272c3000 osi: fix mmc counters issue
Reduce the ivc message size to lower than 2048 to fix a build error. To
reduce the same macsec mmc counters are updated with lower 4 bytes of
the register values

Bug 4703442

Change-Id: I72e936622ca9740ac1fe28cb19129437d5a67d2c
Signed-off-by: Sanath Kumar Gampa <sgampa@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3163253
Reviewed-by: Mahesh Patil <maheshp@nvidia.com>
Reviewed-by: Ashutosh Jha <ajha@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
2024-07-02 05:07:46 -07:00
Sathyam Panda
ba3950c623 osi: core: handle mtu update with disbaled MACSEC
In situation when MACSEC is disabled in DT, macsec_ops
are uninitialized. And an attempt to update the MTU
in this state using OSI command OS_CMD_MAC_MTU can cause
kernel panic.
Add aditional checks to avoid this and exit quietly when
MACSEC is disbaled in DT.

Bug 4646570

Change-Id: If87f3e05f0d24f685c5eb99faeea65d1857d5b43
Signed-off-by: Sathyam Panda <sathyamp@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3138409
(cherry picked from commit d6897400d2d989c4b5868cb69c31ef54504b69b4)
Reviewed-by: Hardik Shah <hardikts@nvidia.com>
Reviewed-by: Abhijit . <abhijit@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3158143
Tested-by: Revanth Kumar Uppala <ruppala@nvidia.com>
Reviewed-by: Jon Hunter <jonathanh@nvidia.com>
Reviewed-by: Revanth Kumar Uppala <ruppala@nvidia.com>
2024-06-28 10:49:07 -07:00
Mahesh Patil
da071420b2 osi: T264 SLT EQOS and MGBE changes
- Adding WAR changes to match HW scripts during
 SLT EQOS and MGBE bringup

Bug 4709627

Change-Id: I2bb15c66a037a0a42ef6f40b2d07d716b7d97008
Signed-off-by: Mahesh Patil <maheshp@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3162307
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
2024-06-26 22:56:48 -07:00
Nagaraj Annaiah
53fa8d800b xpcs_uphy_lane_bring_up: Increase the retry to 1000 for T264
Issue: Uphy lane bring up fails on T264

Fix: Increase the retry to 1000 for T264 and this is temporary fix.

Bug 4709627

Change-Id: Ia974e439de3a099ef666803033a3b0155f6dfa61
Signed-off-by: Nagaraj Annaiah <nannaiah@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3160679
Tested-by: Mahesh Patil <maheshp@nvidia.com>
Reviewed-by: Bhadram Varka <vbhadram@nvidia.com>
Reviewed-by: Krishna Thota <kthota@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Mahesh Patil <maheshp@nvidia.com>
2024-06-26 22:50:23 -07:00
Bhadram Varka
a89360f8db osi: eqos: Keep the reset value for T264 EQOS
Issue: Ethernet driver is programming the ASID - 0x3
for EQOS IP which is different from the value passed from DT.
It resulted SMMU/memory issues while performing data transfers

Fix: Keep the on reset values for T264 EQOS since lower
8-bits of ASID (0x160) needs to program in ASID registers of EQOS IP.

Bug 4701860

Change-Id: I2c799554d054103bea4c8a05a7bdd0961be81683
Signed-off-by: Bhadram Varka <vbhadram@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3159696
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Tested-by: Seema Khowala <seemaj@nvidia.com>
Reviewed-by: svcacv <svcacv@nvidia.com>
Reviewed-by: Narayana Reddy P <narayanr@nvidia.com>
2024-06-19 21:47:06 -07:00
Mahesh Patil
c479ba1650 osi: T26x EQOS changes
1) Added T26X EQOS bring up changes
2) Fixes added during bring up

Bug 4639097

Change-Id: I0036a12ad08d690bb62a655df6f4efd26a0bf585
Signed-off-by: Mahesh Patil <maheshp@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3152936
Tested-by: Seema Khowala <seemaj@nvidia.com>
Reviewed-by: Narayana Reddy P <narayanr@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
2024-06-18 11:37:38 -07:00
Mahesh Patil
887b450741 osi: T26x EQOS XPCS bring up changes
Bug 4639097

Change-Id: I02256885285d86d57764ff1f64a44abd98622075
Signed-off-by: Mahesh Patil <maheshp@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/kernel/nvethernetrm/+/3153395
Reviewed-by: Narayana Reddy P <narayanr@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Tested-by: Seema Khowala <seemaj@nvidia.com>
2024-06-18 11:37:33 -07:00
Bhadram Varka
42f3fbeca7 Merge remote-tracking branch 'origin/dev/t264-ethernet' into dev-t264
Bug 4687787

Change-Id: Ie3411c626b31d1ed4b536954c7e5a875aab29e86
Signed-off-by: Bhadram Varka <vbhadram@nvidia.com>
2024-06-07 15:11:49 +00:00
Bhadram Varka
12bdb2f1ea osi: core: WAR Werror=frame-larger-than and ivc issue
Issue:
1. compilation issue:
error: the frame size of 3232 bytes is larger than 2048 bytes
[-Werror=frame-larger-than=]

Because of increase in the value of OSI_MACSEC_SC_INDEX_MAX
it resulted in increase in the size of ivc_msg_common
more than 2048 bytes.

2. IVC communication breaking because IVC buffer size
increased more than 2048 because of #1.

WAR: Revert back to the orin value for OSI_MACSEC_SC_INDEX_MAX

Bug 4687787

Change-Id: I5262cd772acf6b4275cc3f47696f0072b70916e0
Signed-off-by: Bhadram Varka <vbhadram@nvidia.com>
2024-06-07 13:16:02 +00:00
Mahesh Patil
ed305924de osi: Add support for XLGPCS for 25G speed
Ported from -
https://git-master.nvidia.com/r/c/nvethernet-docs/+/2953123

Bug 4599341

Change-Id: I2d0f8b871a081b4c30f586854ab0f56b84e9bfe3
Signed-off-by: Mahesh Patil <maheshp@nvidia.com>
Signed-off-by: Michael Hsu <mhsu@nvidia.com>
2024-06-06 00:37:38 +00:00
Mahesh Patil
b236911c75 osi: T26x CAR changes
Ported from -
https://git-master.nvidia.com/r/c/nvethernet-docs/+/3126091

Bug 4228185

Change-Id: Ia9a7eede0230e25366a37dc8ce7f1a4635bd11cf
Signed-off-by: Mahesh Patil <maheshp@nvidia.com>
Signed-off-by: Michael Hsu <mhsu@nvidia.com>
2024-06-06 00:37:38 +00:00
Mahesh Patil
13e42ef1d2 osi: update T264 mgbe perf configs
Update MGBE perf configs as per IAS recomendations

Ported from -
https://git-master.nvidia.com/r/c/nvethernet-docs/+/2984047

Bug 4228205

Change-Id: I76f55f2f613417728def92a997957a2f6d18099b
Signed-off-by: Mahesh Patil <maheshp@nvidia.com>
Signed-off-by: Michael Hsu <mhsu@nvidia.com>
2024-06-06 00:37:38 +00:00