* Re: [dpdk-dev] [PATCH 1/3] net/qede/base: fix for FW bug
@ 2019-10-21 18:52 Rasesh Mody
2019-10-21 20:10 ` Kevin Traynor
` (2 more replies)
0 siblings, 3 replies; 10+ messages in thread
From: Rasesh Mody @ 2019-10-21 18:52 UTC (permalink / raw)
To: Kevin Traynor, dev, Jerin Jacob Kollanukkaran, ferruh.yigit
Cc: GR-Everest-DPDK-Dev
Hi Kevin,
>From: Kevin Traynor <ktraynor@redhat.com>
>Sent: Monday, October 21, 2019 3:58 AM
>
>----------------------------------------------------------------------
>On 20/10/2019 06:20, Rasesh Mody wrote:
>> In our testing we have identified a critical FW bug. Performance is
>> degraded significantly for certain packet sizes with 8.40.25.0 FW.
>>
>> This patch updates the FW to version 8.40.33.0. The updated FW has a
>> fix to performance issue.
>>
>> The patch also adds initialization for FW overlay RAM as part of
>> hardware initialization which is required by the new FW.
>>
>> Fixes: 066f0d380a2f ("net/qede/base: update FW to 8.40.25.0")
>>
>
>Hi Rasesh, it should be:
>
>Fixes: 21c959e2cb10 ("net/qede/base: update FW to 8.40.25.0")
>
I could not find the 21c959e2cb10 reference, can you please let me know which tree is it coming from?
>> Signed-off-by: Rasesh Mody <rmody@marvell.com>
>> ---
>> doc/guides/nics/qede.rst | 4 +-
>> drivers/net/qede/base/common_hsi.h | 6 +-
>> drivers/net/qede/base/ecore.h | 4 +-
>> drivers/net/qede/base/ecore_dev.c | 19 ++-
>> drivers/net/qede/base/ecore_hsi_debug_tools.h | 3 +
>> drivers/net/qede/base/ecore_hw.c | 4 +-
>> drivers/net/qede/base/ecore_init_fw_funcs.c | 138
>++++++++++++++++++
>> drivers/net/qede/base/ecore_init_fw_funcs.h | 35 +++++
>> drivers/net/qede/base/ecore_int.c | 14 +-
>> drivers/net/qede/base/ecore_iro_values.h | 122 ++++++++--------
>> drivers/net/qede/qede_ethdev.h | 2 +-
>> drivers/net/qede/qede_main.c | 2 +-
>> 12 files changed, 274 insertions(+), 79 deletions(-)
>>
>> diff --git a/doc/guides/nics/qede.rst b/doc/guides/nics/qede.rst index
>> 471d98014..2f4045795 100644
>> --- a/doc/guides/nics/qede.rst
>> +++ b/doc/guides/nics/qede.rst
>> @@ -78,9 +78,9 @@ Supported QLogic Adapters Prerequisites
>> -------------
>>
>> -- Requires storm firmware version **8.37.7.0**. Firmware may be
>> available
>> +- Requires storm firmware version **8.40.33.0**. Firmware may be
>> +available
>> inbox in certain newer Linux distros under the standard directory
>> - ``E.g. /lib/firmware/qed/qed_init_values-8.37.7.0.bin``.
>> + ``E.g. /lib/firmware/qed/qed_init_values-8.40.33.0.bin``.
>
>Not available in latest Fedora and Ubuntu releases
That is right, it would take some time before its available in latest Fedora/Ubuntu releases.
>
>> If the required firmware files are not available then download it from
>> `linux-firmware git repository
>> <https://urldefense.proofpoint.com/v2/url?u=http-3A__git.kernel.org_pu
>> b_scm_linux_kernel_git_firmware_linux-
>2Dfirmware.git_tree_qed&d=DwIFaQ
>>
>&c=nKjWec2b6R0mOyPaz7xtfQ&r=9aB46H7c7TYTnBun6ODgtnNLQdw3jNiVKH
>bs9eOyBS
>>
>U&m=LjD8kHeZxFHw1ql6Jz5s_XJfxPOIaBqLpyP5FDBNHQc&s=YOhhxcxca9mX
>dFNN7HjG
>> djQF5G22nnsfj1fy4S4cOCM&e= >`_
>
>Not available here either
We are working on getting 8.40.33.0 version into linux-firmware git tree.
Thanks!
-Rasesh
>
>> or `QLogic Driver Download Center
><http://driverdownloads.qlogic.com/QLogicDriverDownloads_UI/DefaultNew
>Search.aspx>`_.
>
>Maybe here? 8.37 is the latest elsewhere.
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [dpdk-dev] [PATCH 1/3] net/qede/base: fix for FW bug
2019-10-21 18:52 [dpdk-dev] [PATCH 1/3] net/qede/base: fix for FW bug Rasesh Mody
@ 2019-10-21 20:10 ` Kevin Traynor
2019-10-22 8:45 ` Ferruh Yigit
2019-10-30 19:15 ` Rasesh Mody
2 siblings, 0 replies; 10+ messages in thread
From: Kevin Traynor @ 2019-10-21 20:10 UTC (permalink / raw)
To: Rasesh Mody, dev, Jerin Jacob Kollanukkaran, ferruh.yigit
Cc: GR-Everest-DPDK-Dev
On 21/10/2019 19:52, Rasesh Mody wrote:
> Hi Kevin,
>
>> From: Kevin Traynor <ktraynor@redhat.com>
>> Sent: Monday, October 21, 2019 3:58 AM
>>
>> ----------------------------------------------------------------------
>> On 20/10/2019 06:20, Rasesh Mody wrote:
>>> In our testing we have identified a critical FW bug. Performance is
>>> degraded significantly for certain packet sizes with 8.40.25.0 FW.
>>>
>>> This patch updates the FW to version 8.40.33.0. The updated FW has a
>>> fix to performance issue.
>>>
>>> The patch also adds initialization for FW overlay RAM as part of
>>> hardware initialization which is required by the new FW.
>>>
>>> Fixes: 066f0d380a2f ("net/qede/base: update FW to 8.40.25.0")
>>>
>>
>> Hi Rasesh, it should be:
>>
>> Fixes: 21c959e2cb10 ("net/qede/base: update FW to 8.40.25.0")
>>
>
> I could not find the 21c959e2cb10 reference, can you please let me know which tree is it coming from?
>
dpdk-next-net and dpdk-next-net-brcm trees. The commit doesn't seem to
be in dpdk tree yet.
>>> Signed-off-by: Rasesh Mody <rmody@marvell.com>
>>> ---
>>> doc/guides/nics/qede.rst | 4 +-
>>> drivers/net/qede/base/common_hsi.h | 6 +-
>>> drivers/net/qede/base/ecore.h | 4 +-
>>> drivers/net/qede/base/ecore_dev.c | 19 ++-
>>> drivers/net/qede/base/ecore_hsi_debug_tools.h | 3 +
>>> drivers/net/qede/base/ecore_hw.c | 4 +-
>>> drivers/net/qede/base/ecore_init_fw_funcs.c | 138
>> ++++++++++++++++++
>>> drivers/net/qede/base/ecore_init_fw_funcs.h | 35 +++++
>>> drivers/net/qede/base/ecore_int.c | 14 +-
>>> drivers/net/qede/base/ecore_iro_values.h | 122 ++++++++--------
>>> drivers/net/qede/qede_ethdev.h | 2 +-
>>> drivers/net/qede/qede_main.c | 2 +-
>>> 12 files changed, 274 insertions(+), 79 deletions(-)
>>>
>>> diff --git a/doc/guides/nics/qede.rst b/doc/guides/nics/qede.rst index
>>> 471d98014..2f4045795 100644
>>> --- a/doc/guides/nics/qede.rst
>>> +++ b/doc/guides/nics/qede.rst
>>> @@ -78,9 +78,9 @@ Supported QLogic Adapters Prerequisites
>>> -------------
>>>
>>> -- Requires storm firmware version **8.37.7.0**. Firmware may be
>>> available
>>> +- Requires storm firmware version **8.40.33.0**. Firmware may be
>>> +available
>>> inbox in certain newer Linux distros under the standard directory
>>> - ``E.g. /lib/firmware/qed/qed_init_values-8.37.7.0.bin``.
>>> + ``E.g. /lib/firmware/qed/qed_init_values-8.40.33.0.bin``.
>>
>> Not available in latest Fedora and Ubuntu releases
>
> That is right, it would take some time before its available in latest Fedora/Ubuntu releases.
>>
>>> If the required firmware files are not available then download it from
>>> `linux-firmware git repository
>>> <https://urldefense.proofpoint.com/v2/url?u=http-3A__git.kernel.org_pu
>>> b_scm_linux_kernel_git_firmware_linux-
>> 2Dfirmware.git_tree_qed&d=DwIFaQ
>>>
>> &c=nKjWec2b6R0mOyPaz7xtfQ&r=9aB46H7c7TYTnBun6ODgtnNLQdw3jNiVKH
>> bs9eOyBS
>>>
>> U&m=LjD8kHeZxFHw1ql6Jz5s_XJfxPOIaBqLpyP5FDBNHQc&s=YOhhxcxca9mX
>> dFNN7HjG
>>> djQF5G22nnsfj1fy4S4cOCM&e= >`_
>>
>> Not available here either
>
> We are working on getting 8.40.33.0 version into linux-firmware git tree.
>
> Thanks!
> -Rasesh
>>
>>> or `QLogic Driver Download Center
>> <http://driverdownloads.qlogic.com/QLogicDriverDownloads_UI/DefaultNew
>> Search.aspx>`_.
>>
>> Maybe here? 8.37 is the latest elsewhere.
>
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [dpdk-dev] [PATCH 1/3] net/qede/base: fix for FW bug
2019-10-21 18:52 [dpdk-dev] [PATCH 1/3] net/qede/base: fix for FW bug Rasesh Mody
2019-10-21 20:10 ` Kevin Traynor
@ 2019-10-22 8:45 ` Ferruh Yigit
2019-10-22 9:13 ` Kevin Traynor
2019-10-30 19:15 ` Rasesh Mody
2 siblings, 1 reply; 10+ messages in thread
From: Ferruh Yigit @ 2019-10-22 8:45 UTC (permalink / raw)
To: Rasesh Mody, Kevin Traynor, dev, Jerin Jacob Kollanukkaran
Cc: GR-Everest-DPDK-Dev
On 10/21/2019 7:52 PM, Rasesh Mody wrote:
> Hi Kevin,
>
>> From: Kevin Traynor <ktraynor@redhat.com>
>> Sent: Monday, October 21, 2019 3:58 AM
>>
>> ----------------------------------------------------------------------
>> On 20/10/2019 06:20, Rasesh Mody wrote:
>>> In our testing we have identified a critical FW bug. Performance is
>>> degraded significantly for certain packet sizes with 8.40.25.0 FW.
>>>
>>> This patch updates the FW to version 8.40.33.0. The updated FW has a
>>> fix to performance issue.
>>>
>>> The patch also adds initialization for FW overlay RAM as part of
>>> hardware initialization which is required by the new FW.
>>>
>>> Fixes: 066f0d380a2f ("net/qede/base: update FW to 8.40.25.0")
>>>
>>
>> Hi Rasesh, it should be:
>>
>> Fixes: 21c959e2cb10 ("net/qede/base: update FW to 8.40.25.0")
>>
>
> I could not find the 21c959e2cb10 reference, can you please let me know which tree is it coming from?
Hi Kevin,
For similar case, if the fixed commit has been pulled by main tree, I am
updating the fixes tag on the next-net,
if not pulled yet, mostly I am squashing the fix to the original commit in
next-net, if I can't squash for some reason Thomas is fixing the tag while
merging into the main tree. (Our check-git-log commit detected wrong Fixes tag
already.)
So if you pick fixes from main tree only, the fixes tags will be always correct.
If you pick commits from sub-trees, fixes tag won't be only problem, the commits
also will be not fully tested...
>
>>> Signed-off-by: Rasesh Mody <rmody@marvell.com>
>>> ---
>>> doc/guides/nics/qede.rst | 4 +-
>>> drivers/net/qede/base/common_hsi.h | 6 +-
>>> drivers/net/qede/base/ecore.h | 4 +-
>>> drivers/net/qede/base/ecore_dev.c | 19 ++-
>>> drivers/net/qede/base/ecore_hsi_debug_tools.h | 3 +
>>> drivers/net/qede/base/ecore_hw.c | 4 +-
>>> drivers/net/qede/base/ecore_init_fw_funcs.c | 138
>> ++++++++++++++++++
>>> drivers/net/qede/base/ecore_init_fw_funcs.h | 35 +++++
>>> drivers/net/qede/base/ecore_int.c | 14 +-
>>> drivers/net/qede/base/ecore_iro_values.h | 122 ++++++++--------
>>> drivers/net/qede/qede_ethdev.h | 2 +-
>>> drivers/net/qede/qede_main.c | 2 +-
>>> 12 files changed, 274 insertions(+), 79 deletions(-)
>>>
>>> diff --git a/doc/guides/nics/qede.rst b/doc/guides/nics/qede.rst index
>>> 471d98014..2f4045795 100644
>>> --- a/doc/guides/nics/qede.rst
>>> +++ b/doc/guides/nics/qede.rst
>>> @@ -78,9 +78,9 @@ Supported QLogic Adapters Prerequisites
>>> -------------
>>>
>>> -- Requires storm firmware version **8.37.7.0**. Firmware may be
>>> available
>>> +- Requires storm firmware version **8.40.33.0**. Firmware may be
>>> +available
>>> inbox in certain newer Linux distros under the standard directory
>>> - ``E.g. /lib/firmware/qed/qed_init_values-8.37.7.0.bin``.
>>> + ``E.g. /lib/firmware/qed/qed_init_values-8.40.33.0.bin``.
>>
>> Not available in latest Fedora and Ubuntu releases
>
> That is right, it would take some time before its available in latest Fedora/Ubuntu releases.
>>
>>> If the required firmware files are not available then download it from
>>> `linux-firmware git repository
>>> <https://urldefense.proofpoint.com/v2/url?u=http-3A__git.kernel.org_pu
>>> b_scm_linux_kernel_git_firmware_linux-
>> 2Dfirmware.git_tree_qed&d=DwIFaQ
>>>
>> &c=nKjWec2b6R0mOyPaz7xtfQ&r=9aB46H7c7TYTnBun6ODgtnNLQdw3jNiVKH
>> bs9eOyBS
>>>
>> U&m=LjD8kHeZxFHw1ql6Jz5s_XJfxPOIaBqLpyP5FDBNHQc&s=YOhhxcxca9mX
>> dFNN7HjG
>>> djQF5G22nnsfj1fy4S4cOCM&e= >`_
>>
>> Not available here either
>
> We are working on getting 8.40.33.0 version into linux-firmware git tree.
>
> Thanks!
> -Rasesh
>>
>>> or `QLogic Driver Download Center
>> <http://driverdownloads.qlogic.com/QLogicDriverDownloads_UI/DefaultNew
>> Search.aspx>`_.
>>
>> Maybe here? 8.37 is the latest elsewhere.
>
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [dpdk-dev] [PATCH 1/3] net/qede/base: fix for FW bug
2019-10-22 8:45 ` Ferruh Yigit
@ 2019-10-22 9:13 ` Kevin Traynor
0 siblings, 0 replies; 10+ messages in thread
From: Kevin Traynor @ 2019-10-22 9:13 UTC (permalink / raw)
To: Ferruh Yigit, Rasesh Mody, dev, Jerin Jacob Kollanukkaran
Cc: GR-Everest-DPDK-Dev
On 22/10/2019 09:45, Ferruh Yigit wrote:
> On 10/21/2019 7:52 PM, Rasesh Mody wrote:
>> Hi Kevin,
>>
>>> From: Kevin Traynor <ktraynor@redhat.com>
>>> Sent: Monday, October 21, 2019 3:58 AM
>>>
>>> ----------------------------------------------------------------------
>>> On 20/10/2019 06:20, Rasesh Mody wrote:
>>>> In our testing we have identified a critical FW bug. Performance is
>>>> degraded significantly for certain packet sizes with 8.40.25.0 FW.
>>>>
>>>> This patch updates the FW to version 8.40.33.0. The updated FW has a
>>>> fix to performance issue.
>>>>
>>>> The patch also adds initialization for FW overlay RAM as part of
>>>> hardware initialization which is required by the new FW.
>>>>
>>>> Fixes: 066f0d380a2f ("net/qede/base: update FW to 8.40.25.0")
>>>>
>>>
>>> Hi Rasesh, it should be:
>>>
>>> Fixes: 21c959e2cb10 ("net/qede/base: update FW to 8.40.25.0")
>>>
>>
>> I could not find the 21c959e2cb10 reference, can you please let me know which tree is it coming from?
>
> Hi Kevin,>
Hi Ferruh,
> For similar case, if the fixed commit has been pulled by main tree, I am
> updating the fixes tag on the next-net,
> if not pulled yet, mostly I am squashing the fix to the original commit in
> next-net, if I can't squash for some reason Thomas is fixing the tag while
> merging into the main tree. (Our check-git-log commit detected wrong Fixes tag
> already.)
>
Oh, I see, excellent.
> So if you pick fixes from main tree only, the fixes tags will be always correct.
> If you pick commits from sub-trees, fixes tag won't be only problem, the commits
> also will be not fully tested...
>
Yes, we're only picking from main tree.
I didn't realize you were actively fixing the tags along the way which
is why I made comment about it. Thanks for doing that, it saves manually
checking emails/commit when doing the backports.
Kevin.
>>
>>>> Signed-off-by: Rasesh Mody <rmody@marvell.com>
>>>> ---
>>>> doc/guides/nics/qede.rst | 4 +-
>>>> drivers/net/qede/base/common_hsi.h | 6 +-
>>>> drivers/net/qede/base/ecore.h | 4 +-
>>>> drivers/net/qede/base/ecore_dev.c | 19 ++-
>>>> drivers/net/qede/base/ecore_hsi_debug_tools.h | 3 +
>>>> drivers/net/qede/base/ecore_hw.c | 4 +-
>>>> drivers/net/qede/base/ecore_init_fw_funcs.c | 138
>>> ++++++++++++++++++
>>>> drivers/net/qede/base/ecore_init_fw_funcs.h | 35 +++++
>>>> drivers/net/qede/base/ecore_int.c | 14 +-
>>>> drivers/net/qede/base/ecore_iro_values.h | 122 ++++++++--------
>>>> drivers/net/qede/qede_ethdev.h | 2 +-
>>>> drivers/net/qede/qede_main.c | 2 +-
>>>> 12 files changed, 274 insertions(+), 79 deletions(-)
>>>>
>>>> diff --git a/doc/guides/nics/qede.rst b/doc/guides/nics/qede.rst index
>>>> 471d98014..2f4045795 100644
>>>> --- a/doc/guides/nics/qede.rst
>>>> +++ b/doc/guides/nics/qede.rst
>>>> @@ -78,9 +78,9 @@ Supported QLogic Adapters Prerequisites
>>>> -------------
>>>>
>>>> -- Requires storm firmware version **8.37.7.0**. Firmware may be
>>>> available
>>>> +- Requires storm firmware version **8.40.33.0**. Firmware may be
>>>> +available
>>>> inbox in certain newer Linux distros under the standard directory
>>>> - ``E.g. /lib/firmware/qed/qed_init_values-8.37.7.0.bin``.
>>>> + ``E.g. /lib/firmware/qed/qed_init_values-8.40.33.0.bin``.
>>>
>>> Not available in latest Fedora and Ubuntu releases
>>
>> That is right, it would take some time before its available in latest Fedora/Ubuntu releases.
>>>
>>>> If the required firmware files are not available then download it from
>>>> `linux-firmware git repository
>>>> <https://urldefense.proofpoint.com/v2/url?u=http-3A__git.kernel.org_pu
>>>> b_scm_linux_kernel_git_firmware_linux-
>>> 2Dfirmware.git_tree_qed&d=DwIFaQ
>>>>
>>> &c=nKjWec2b6R0mOyPaz7xtfQ&r=9aB46H7c7TYTnBun6ODgtnNLQdw3jNiVKH
>>> bs9eOyBS
>>>>
>>> U&m=LjD8kHeZxFHw1ql6Jz5s_XJfxPOIaBqLpyP5FDBNHQc&s=YOhhxcxca9mX
>>> dFNN7HjG
>>>> djQF5G22nnsfj1fy4S4cOCM&e= >`_
>>>
>>> Not available here either
>>
>> We are working on getting 8.40.33.0 version into linux-firmware git tree.
>>
>> Thanks!
>> -Rasesh
>>>
>>>> or `QLogic Driver Download Center
>>> <http://driverdownloads.qlogic.com/QLogicDriverDownloads_UI/DefaultNew
>>> Search.aspx>`_.
>>>
>>> Maybe here? 8.37 is the latest elsewhere.
>>
>
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [dpdk-dev] [PATCH 1/3] net/qede/base: fix for FW bug
2019-10-21 18:52 [dpdk-dev] [PATCH 1/3] net/qede/base: fix for FW bug Rasesh Mody
2019-10-21 20:10 ` Kevin Traynor
2019-10-22 8:45 ` Ferruh Yigit
@ 2019-10-30 19:15 ` Rasesh Mody
2 siblings, 0 replies; 10+ messages in thread
From: Rasesh Mody @ 2019-10-30 19:15 UTC (permalink / raw)
To: Kevin Traynor, dev, Jerin Jacob Kollanukkaran, ferruh.yigit
Cc: GR-Everest-DPDK-Dev
Hi Kevin,
>From: Rasesh Mody <rmody@marvell.com>
>Sent: Monday, October 21, 2019 11:53 AM
>
>Hi Kevin,
>
>>From: Kevin Traynor <ktraynor@redhat.com>
>>Sent: Monday, October 21, 2019 3:58 AM
>>
>>On 20/10/2019 06:20, Rasesh Mody wrote:
>>> In our testing we have identified a critical FW bug. Performance is
>>> degraded significantly for certain packet sizes with 8.40.25.0 FW.
>>>
>>> This patch updates the FW to version 8.40.33.0. The updated FW has a
>>> fix to performance issue.
>>>
>>> The patch also adds initialization for FW overlay RAM as part of
>>> hardware initialization which is required by the new FW.
>>>
>>> Fixes: 066f0d380a2f ("net/qede/base: update FW to 8.40.25.0")
>>>
>>
>>Hi Rasesh, it should be:
>>
>>Fixes: 21c959e2cb10 ("net/qede/base: update FW to 8.40.25.0")
>>
>
>I could not find the 21c959e2cb10 reference, can you please let me know
>which tree is it coming from?
>
>>> Signed-off-by: Rasesh Mody <rmody@marvell.com>
>>> ---
>>> doc/guides/nics/qede.rst | 4 +-
>>> drivers/net/qede/base/common_hsi.h | 6 +-
>>> drivers/net/qede/base/ecore.h | 4 +-
>>> drivers/net/qede/base/ecore_dev.c | 19 ++-
>>> drivers/net/qede/base/ecore_hsi_debug_tools.h | 3 +
>>> drivers/net/qede/base/ecore_hw.c | 4 +-
>>> drivers/net/qede/base/ecore_init_fw_funcs.c | 138
>>++++++++++++++++++
>>> drivers/net/qede/base/ecore_init_fw_funcs.h | 35 +++++
>>> drivers/net/qede/base/ecore_int.c | 14 +-
>>> drivers/net/qede/base/ecore_iro_values.h | 122 ++++++++--------
>>> drivers/net/qede/qede_ethdev.h | 2 +-
>>> drivers/net/qede/qede_main.c | 2 +-
>>> 12 files changed, 274 insertions(+), 79 deletions(-)
>>>
>>> diff --git a/doc/guides/nics/qede.rst b/doc/guides/nics/qede.rst
>>> index
>>> 471d98014..2f4045795 100644
>>> --- a/doc/guides/nics/qede.rst
>>> +++ b/doc/guides/nics/qede.rst
>>> @@ -78,9 +78,9 @@ Supported QLogic Adapters Prerequisites
>>> -------------
>>>
>>> -- Requires storm firmware version **8.37.7.0**. Firmware may be
>>> available
>>> +- Requires storm firmware version **8.40.33.0**. Firmware may be
>>> +available
>>> inbox in certain newer Linux distros under the standard directory
>>> - ``E.g. /lib/firmware/qed/qed_init_values-8.37.7.0.bin``.
>>> + ``E.g. /lib/firmware/qed/qed_init_values-8.40.33.0.bin``.
>>
>>Not available in latest Fedora and Ubuntu releases
>
>That is right, it would take some time before its available in latest
>Fedora/Ubuntu releases.
>>
>>> If the required firmware files are not available then download it from
>>> `linux-firmware git repository
>>> <https://urldefense.proofpoint.com/v2/url?u=http-3A__git.kernel.org_p
>>> u
>>> b_scm_linux_kernel_git_firmware_linux-
>>2Dfirmware.git_tree_qed&d=DwIFaQ
>>>
>>&c=nKjWec2b6R0mOyPaz7xtfQ&r=9aB46H7c7TYTnBun6ODgtnNLQdw3jNiVK
>H
>>bs9eOyBS
>>>
>>U&m=LjD8kHeZxFHw1ql6Jz5s_XJfxPOIaBqLpyP5FDBNHQc&s=YOhhxcxca9m
>X
>>dFNN7HjG
>>> djQF5G22nnsfj1fy4S4cOCM&e= >`_
>>
>>Not available here either
>
>We are working on getting 8.40.33.0 version into linux-firmware git tree.
The 8.40.33.0 FW is now available at linux-firmware git repository.
Thanks!
-Rasesh
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [dpdk-dev] [PATCH 1/3] net/qede/base: fix for FW bug
@ 2019-10-21 20:46 Rasesh Mody
2019-10-21 21:41 ` Kevin Traynor
0 siblings, 1 reply; 10+ messages in thread
From: Rasesh Mody @ 2019-10-21 20:46 UTC (permalink / raw)
To: Kevin Traynor, dev, Jerin Jacob Kollanukkaran, ferruh.yigit
Cc: GR-Everest-DPDK-Dev
>From: Kevin Traynor <ktraynor@redhat.com>
>Sent: Monday, October 21, 2019 1:10 PM
>
>----------------------------------------------------------------------
>On 21/10/2019 19:52, Rasesh Mody wrote:
>> Hi Kevin,
>>
>>> From: Kevin Traynor <ktraynor@redhat.com>
>>> Sent: Monday, October 21, 2019 3:58 AM
>>>
>>> ---------------------------------------------------------------------
>>> - On 20/10/2019 06:20, Rasesh Mody wrote:
>>>> In our testing we have identified a critical FW bug. Performance is
>>>> degraded significantly for certain packet sizes with 8.40.25.0 FW.
>>>>
>>>> This patch updates the FW to version 8.40.33.0. The updated FW has a
>>>> fix to performance issue.
>>>>
>>>> The patch also adds initialization for FW overlay RAM as part of
>>>> hardware initialization which is required by the new FW.
>>>>
>>>> Fixes: 066f0d380a2f ("net/qede/base: update FW to 8.40.25.0")
>>>>
>>>
>>> Hi Rasesh, it should be:
>>>
>>> Fixes: 21c959e2cb10 ("net/qede/base: update FW to 8.40.25.0")
>>>
>>
>> I could not find the 21c959e2cb10 reference, can you please let me know
>which tree is it coming from?
>>
>
>dpdk-next-net and dpdk-next-net-brcm trees. The commit doesn't seem to
>be in dpdk tree yet.
dpdk-next-net tree gives a different reference as below.
Fixes: 0cd01921343e ("net/qede/base: update FW to 8.40.25.0")
The one, originally used, is from dpdk-next-net-mrvl tree.
Do you suggest to update the Fixes tag to a commit id from dpdk-next-net tree?
>
>>>> Signed-off-by: Rasesh Mody <rmody@marvell.com>
>>>> ---
>>>> doc/guides/nics/qede.rst | 4 +-
>>>> drivers/net/qede/base/common_hsi.h | 6 +-
>>>> drivers/net/qede/base/ecore.h | 4 +-
>>>> drivers/net/qede/base/ecore_dev.c | 19 ++-
>>>> drivers/net/qede/base/ecore_hsi_debug_tools.h | 3 +
>>>> drivers/net/qede/base/ecore_hw.c | 4 +-
>>>> drivers/net/qede/base/ecore_init_fw_funcs.c | 138
>>> ++++++++++++++++++
>>>> drivers/net/qede/base/ecore_init_fw_funcs.h | 35 +++++
>>>> drivers/net/qede/base/ecore_int.c | 14 +-
>>>> drivers/net/qede/base/ecore_iro_values.h | 122 ++++++++--------
>>>> drivers/net/qede/qede_ethdev.h | 2 +-
>>>> drivers/net/qede/qede_main.c | 2 +-
>>>> 12 files changed, 274 insertions(+), 79 deletions(-)
>>>>
>>>> diff --git a/doc/guides/nics/qede.rst b/doc/guides/nics/qede.rst
>>>> index
>>>> 471d98014..2f4045795 100644
>>>> --- a/doc/guides/nics/qede.rst
>>>> +++ b/doc/guides/nics/qede.rst
>>>> @@ -78,9 +78,9 @@ Supported QLogic Adapters Prerequisites
>>>> -------------
>>>>
>>>> -- Requires storm firmware version **8.37.7.0**. Firmware may be
>>>> available
>>>> +- Requires storm firmware version **8.40.33.0**. Firmware may be
>>>> +available
>>>> inbox in certain newer Linux distros under the standard directory
>>>> - ``E.g. /lib/firmware/qed/qed_init_values-8.37.7.0.bin``.
>>>> + ``E.g. /lib/firmware/qed/qed_init_values-8.40.33.0.bin``.
>>>
>>> Not available in latest Fedora and Ubuntu releases
>>
>> That is right, it would take some time before its available in latest
>Fedora/Ubuntu releases.
>>>
>>>> If the required firmware files are not available then download it from
>>>> `linux-firmware git repository
>>>> <https://urldefense.proofpoint.com/v2/url?u=http-3A__git.kernel.org_
>>>> pu
>>>> b_scm_linux_kernel_git_firmware_linux-
>>> 2Dfirmware.git_tree_qed&d=DwIFaQ
>>>>
>>>
>&c=nKjWec2b6R0mOyPaz7xtfQ&r=9aB46H7c7TYTnBun6ODgtnNLQdw3jNiVKH
>>> bs9eOyBS
>>>>
>>>
>U&m=LjD8kHeZxFHw1ql6Jz5s_XJfxPOIaBqLpyP5FDBNHQc&s=YOhhxcxca9mX
>>> dFNN7HjG
>>>> djQF5G22nnsfj1fy4S4cOCM&e= >`_
>>>
>>> Not available here either
>>
>> We are working on getting 8.40.33.0 version into linux-firmware git tree.
>>
>> Thanks!
>> -Rasesh
>>>
>>>> or `QLogic Driver Download Center
>>>
><http://driverdownloads.qlogic.com/QLogicDriverDownloads_UI/DefaultNe
>>> w
>>> Search.aspx>`_.
>>>
>>> Maybe here? 8.37 is the latest elsewhere.
>>
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [dpdk-dev] [PATCH 1/3] net/qede/base: fix for FW bug
2019-10-21 20:46 Rasesh Mody
@ 2019-10-21 21:41 ` Kevin Traynor
0 siblings, 0 replies; 10+ messages in thread
From: Kevin Traynor @ 2019-10-21 21:41 UTC (permalink / raw)
To: Rasesh Mody, dev, Jerin Jacob Kollanukkaran, ferruh.yigit
Cc: GR-Everest-DPDK-Dev
On 21/10/2019 21:46, Rasesh Mody wrote:
>> From: Kevin Traynor <ktraynor@redhat.com>
>> Sent: Monday, October 21, 2019 1:10 PM
>>
>> ----------------------------------------------------------------------
>> On 21/10/2019 19:52, Rasesh Mody wrote:
>>> Hi Kevin,
>>>
>>>> From: Kevin Traynor <ktraynor@redhat.com>
>>>> Sent: Monday, October 21, 2019 3:58 AM
>>>>
>>>> ---------------------------------------------------------------------
>>>> - On 20/10/2019 06:20, Rasesh Mody wrote:
>>>>> In our testing we have identified a critical FW bug. Performance is
>>>>> degraded significantly for certain packet sizes with 8.40.25.0 FW.
>>>>>
>>>>> This patch updates the FW to version 8.40.33.0. The updated FW has a
>>>>> fix to performance issue.
>>>>>
>>>>> The patch also adds initialization for FW overlay RAM as part of
>>>>> hardware initialization which is required by the new FW.
>>>>>
>>>>> Fixes: 066f0d380a2f ("net/qede/base: update FW to 8.40.25.0")
>>>>>
>>>> Hi Rasesh, it should be:
>>>>
>>>> Fixes: 21c959e2cb10 ("net/qede/base: update FW to 8.40.25.0")
>>>>
>>> I could not find the 21c959e2cb10 reference, can you please let me know
>> which tree is it coming from?
>> dpdk-next-net and dpdk-next-net-brcm trees. The commit doesn't seem to
>> be in dpdk tree yet.
> dpdk-next-net tree gives a different reference as below.
> Fixes: 0cd01921343e ("net/qede/base: update FW to 8.40.25.0")
>
ah ok, I figured out it is a problem on my side - got back in sync with
the remote and I see the same commit-id. Sorry for the noise.
> The one, originally used, is from dpdk-next-net-mrvl tree.
>
> Do you suggest to update the Fixes tag to a commit id from dpdk-next-net tree?
>
The best would be the dpdk tree, but you can't do that as it's not there
yet. I'd probably go with the highest tree that it has been merged into
(dpdk-next-net) but i'm not sure it really matters. From a stable view,
we won't find the sha1 in the dpdk tree, but we'd see that there's no
stable tag and we'd find the commit through the title.
^ permalink raw reply [flat|nested] 10+ messages in thread
* [dpdk-dev] [PATCH 1/3] net/qede/base: fix for FW bug
@ 2019-10-20 5:20 Rasesh Mody
2019-10-21 10:58 ` Kevin Traynor
0 siblings, 1 reply; 10+ messages in thread
From: Rasesh Mody @ 2019-10-20 5:20 UTC (permalink / raw)
To: dev, jerinj, ferruh.yigit; +Cc: Rasesh Mody, GR-Everest-DPDK-Dev
In our testing we have identified a critical FW bug. Performance
is degraded significantly for certain packet sizes with 8.40.25.0 FW.
This patch updates the FW to version 8.40.33.0. The updated FW has a
fix to performance issue.
The patch also adds initialization for FW overlay RAM as part of
hardware initialization which is required by the new FW.
Fixes: 066f0d380a2f ("net/qede/base: update FW to 8.40.25.0")
Signed-off-by: Rasesh Mody <rmody@marvell.com>
---
doc/guides/nics/qede.rst | 4 +-
drivers/net/qede/base/common_hsi.h | 6 +-
drivers/net/qede/base/ecore.h | 4 +-
drivers/net/qede/base/ecore_dev.c | 19 ++-
drivers/net/qede/base/ecore_hsi_debug_tools.h | 3 +
drivers/net/qede/base/ecore_hw.c | 4 +-
drivers/net/qede/base/ecore_init_fw_funcs.c | 138 ++++++++++++++++++
drivers/net/qede/base/ecore_init_fw_funcs.h | 35 +++++
drivers/net/qede/base/ecore_int.c | 14 +-
drivers/net/qede/base/ecore_iro_values.h | 122 ++++++++--------
drivers/net/qede/qede_ethdev.h | 2 +-
drivers/net/qede/qede_main.c | 2 +-
12 files changed, 274 insertions(+), 79 deletions(-)
diff --git a/doc/guides/nics/qede.rst b/doc/guides/nics/qede.rst
index 471d98014..2f4045795 100644
--- a/doc/guides/nics/qede.rst
+++ b/doc/guides/nics/qede.rst
@@ -78,9 +78,9 @@ Supported QLogic Adapters
Prerequisites
-------------
-- Requires storm firmware version **8.37.7.0**. Firmware may be available
+- Requires storm firmware version **8.40.33.0**. Firmware may be available
inbox in certain newer Linux distros under the standard directory
- ``E.g. /lib/firmware/qed/qed_init_values-8.37.7.0.bin``.
+ ``E.g. /lib/firmware/qed/qed_init_values-8.40.33.0.bin``.
If the required firmware files are not available then download it from
`linux-firmware git repository <http://git.kernel.org/pub/scm/linux/kernel/git/firmware/linux-firmware.git/tree/qed>`_
or `QLogic Driver Download Center <http://driverdownloads.qlogic.com/QLogicDriverDownloads_UI/DefaultNewSearch.aspx>`_.
diff --git a/drivers/net/qede/base/common_hsi.h b/drivers/net/qede/base/common_hsi.h
index 74afed1ec..e230fe5ac 100644
--- a/drivers/net/qede/base/common_hsi.h
+++ b/drivers/net/qede/base/common_hsi.h
@@ -92,10 +92,10 @@
/****************************************************************************/
-#define FW_MAJOR_VERSION 8
+#define FW_MAJOR_VERSION 8
#define FW_MINOR_VERSION 40
-#define FW_REVISION_VERSION 25
-#define FW_ENGINEERING_VERSION 0
+#define FW_REVISION_VERSION 33
+#define FW_ENGINEERING_VERSION 0
/***********************/
/* COMMON HW CONSTANTS */
diff --git a/drivers/net/qede/base/ecore.h b/drivers/net/qede/base/ecore.h
index 925b75cb9..b2077bc46 100644
--- a/drivers/net/qede/base/ecore.h
+++ b/drivers/net/qede/base/ecore.h
@@ -29,7 +29,7 @@
#define ECORE_MAJOR_VERSION 8
#define ECORE_MINOR_VERSION 40
-#define ECORE_REVISION_VERSION 18
+#define ECORE_REVISION_VERSION 26
#define ECORE_ENGINEERING_VERSION 0
#define ECORE_VERSION \
@@ -699,6 +699,8 @@ struct ecore_hwfn {
*/
bool b_en_pacing;
+ struct phys_mem_desc *fw_overlay_mem;
+
/* @DPDK */
struct ecore_ptt *p_arfs_ptt;
};
diff --git a/drivers/net/qede/base/ecore_dev.c b/drivers/net/qede/base/ecore_dev.c
index 2c47aba48..9d1db1459 100644
--- a/drivers/net/qede/base/ecore_dev.c
+++ b/drivers/net/qede/base/ecore_dev.c
@@ -1653,6 +1653,7 @@ void ecore_resc_free(struct ecore_dev *p_dev)
ecore_dmae_info_free(p_hwfn);
ecore_dcbx_info_free(p_hwfn);
ecore_dbg_user_data_free(p_hwfn);
+ ecore_fw_overlay_mem_free(p_hwfn, p_hwfn->fw_overlay_mem);
/* @@@TBD Flush work-queue ? */
/* destroy doorbell recovery mechanism */
@@ -3467,6 +3468,8 @@ ecore_hw_init_pf(struct ecore_hwfn *p_hwfn, struct ecore_ptt *p_ptt,
if (rc)
return rc;
+ ecore_fw_overlay_init_ram(p_hwfn, p_ptt, p_hwfn->fw_overlay_mem);
+
/* Pure runtime initializations - directly to the HW */
ecore_int_igu_init_pure_rt(p_hwfn, p_ptt, true, true);
@@ -3494,7 +3497,7 @@ ecore_hw_init_pf(struct ecore_hwfn *p_hwfn, struct ecore_ptt *p_ptt,
if (IS_LEAD_HWFN(p_hwfn)) {
rc = ecore_llh_hw_init_pf(p_hwfn, p_ptt,
p_params->avoid_eng_affin);
- if (rc)
+ if (rc != ECORE_SUCCESS)
return rc;
}
@@ -3510,7 +3513,6 @@ ecore_hw_init_pf(struct ecore_hwfn *p_hwfn, struct ecore_ptt *p_ptt,
if (rc) {
DP_NOTICE(p_hwfn, true,
"Function start ramrod failed\n");
- } else {
return rc;
}
prs_reg = ecore_rd(p_hwfn, p_ptt, PRS_REG_SEARCH_TAG1);
@@ -3677,6 +3679,8 @@ enum _ecore_status_t ecore_hw_init(struct ecore_dev *p_dev,
u32 load_code, resp, param, drv_mb_param;
bool b_default_mtu = true;
struct ecore_hwfn *p_hwfn;
+ const u32 *fw_overlays;
+ u32 fw_overlays_len;
enum _ecore_status_t rc = ECORE_SUCCESS;
u16 ether_type;
int i;
@@ -3815,6 +3819,17 @@ enum _ecore_status_t ecore_hw_init(struct ecore_dev *p_dev,
*/
ecore_pglueb_clear_err(p_hwfn, p_hwfn->p_main_ptt);
+ fw_overlays = p_dev->fw_data->fw_overlays;
+ fw_overlays_len = p_dev->fw_data->fw_overlays_len;
+ p_hwfn->fw_overlay_mem =
+ ecore_fw_overlay_mem_alloc(p_hwfn, fw_overlays,
+ fw_overlays_len);
+ if (!p_hwfn->fw_overlay_mem) {
+ DP_NOTICE(p_hwfn, false,
+ "Failed to allocate fw overlay memory\n");
+ goto load_err;
+ }
+
switch (load_code) {
case FW_MSG_CODE_DRV_LOAD_ENGINE:
rc = ecore_hw_init_common(p_hwfn, p_hwfn->p_main_ptt,
diff --git a/drivers/net/qede/base/ecore_hsi_debug_tools.h b/drivers/net/qede/base/ecore_hsi_debug_tools.h
index a959aeea7..eb72e93cf 100644
--- a/drivers/net/qede/base/ecore_hsi_debug_tools.h
+++ b/drivers/net/qede/base/ecore_hsi_debug_tools.h
@@ -917,6 +917,9 @@ enum dbg_grc_params {
DBG_GRC_PARAM_NO_FW_VER /* dont read FW/MFW version (0/1) */,
DBG_GRC_PARAM_RESERVED3 /* reserved */,
DBG_GRC_PARAM_DUMP_MCP_HW_DUMP /* dump MCP HW Dump (0/1) */,
+ DBG_GRC_PARAM_DUMP_ILT_CDUC /* dump ILT CDUC client (0/1) */,
+ DBG_GRC_PARAM_DUMP_ILT_CDUT /* dump ILT CDUT client (0/1) */,
+ DBG_GRC_PARAM_DUMP_CAU_EXT /* dump CAU extended memories (0/1) */,
MAX_DBG_GRC_PARAMS
};
diff --git a/drivers/net/qede/base/ecore_hw.c b/drivers/net/qede/base/ecore_hw.c
index 0aed043bb..1db39d6a3 100644
--- a/drivers/net/qede/base/ecore_hw.c
+++ b/drivers/net/qede/base/ecore_hw.c
@@ -712,7 +712,7 @@ static enum _ecore_status_t ecore_dmae_operation_wait(struct ecore_hwfn *p_hwfn)
while (*p_hwfn->dmae_info.p_completion_word != DMAE_COMPLETION_VAL) {
OSAL_UDELAY(DMAE_MIN_WAIT_TIME);
if (++wait_cnt > wait_cnt_limit) {
- DP_NOTICE(p_hwfn->p_dev, ECORE_MSG_HW,
+ DP_NOTICE(p_hwfn->p_dev, false,
"Timed-out waiting for operation to"
" complete. Completion word is 0x%08x"
" expected 0x%08x.\n",
@@ -805,7 +805,7 @@ ecore_dmae_execute_sub_operation(struct ecore_hwfn *p_hwfn,
length_dw * sizeof(u32), true);
if (ecore_status != ECORE_SUCCESS) {
- DP_NOTICE(p_hwfn, ECORE_MSG_HW,
+ DP_NOTICE(p_hwfn, false,
"Wait Failed. source_addr 0x%lx, grc_addr 0x%lx, size_in_dwords 0x%x, intermediate buffer 0x%lx.\n",
(unsigned long)src_addr, (unsigned long)dst_addr,
length_dw,
diff --git a/drivers/net/qede/base/ecore_init_fw_funcs.c b/drivers/net/qede/base/ecore_init_fw_funcs.c
index fa1e89a86..6a52f32cc 100644
--- a/drivers/net/qede/base/ecore_init_fw_funcs.c
+++ b/drivers/net/qede/base/ecore_init_fw_funcs.c
@@ -2058,3 +2058,141 @@ void ecore_enable_context_validation(struct ecore_hwfn *p_hwfn,
ctx_validation = CDU_CONTEXT_VALIDATION_DEFAULT_CFG << 8;
ecore_wr(p_hwfn, p_ptt, CDU_REG_TCFC_CTX_VALID0, ctx_validation);
}
+
+#define PHYS_ADDR_DWORDS DIV_ROUND_UP(sizeof(dma_addr_t), 4)
+#define OVERLAY_HDR_SIZE_DWORDS (sizeof(struct fw_overlay_buf_hdr) / 4)
+
+static u32 ecore_get_overlay_addr_ram_addr(struct ecore_hwfn *p_hwfn,
+ u8 storm_id)
+{
+ switch (storm_id) {
+ case 0: return TSEM_REG_FAST_MEMORY + SEM_FAST_REG_INT_RAM +
+ TSTORM_OVERLAY_BUF_ADDR_OFFSET;
+ case 1: return MSEM_REG_FAST_MEMORY + SEM_FAST_REG_INT_RAM +
+ MSTORM_OVERLAY_BUF_ADDR_OFFSET;
+ case 2: return USEM_REG_FAST_MEMORY + SEM_FAST_REG_INT_RAM +
+ USTORM_OVERLAY_BUF_ADDR_OFFSET;
+ case 3: return XSEM_REG_FAST_MEMORY + SEM_FAST_REG_INT_RAM +
+ XSTORM_OVERLAY_BUF_ADDR_OFFSET;
+ case 4: return YSEM_REG_FAST_MEMORY + SEM_FAST_REG_INT_RAM +
+ YSTORM_OVERLAY_BUF_ADDR_OFFSET;
+ case 5: return PSEM_REG_FAST_MEMORY + SEM_FAST_REG_INT_RAM +
+ PSTORM_OVERLAY_BUF_ADDR_OFFSET;
+
+ default: return 0;
+ }
+}
+
+struct phys_mem_desc *ecore_fw_overlay_mem_alloc(struct ecore_hwfn *p_hwfn,
+ const u32 *const fw_overlay_in_buf,
+ u32 buf_size_in_bytes)
+{
+ u32 buf_size = buf_size_in_bytes / sizeof(u32), buf_offset = 0;
+ struct phys_mem_desc *allocated_mem;
+
+ if (!buf_size)
+ return OSAL_NULL;
+
+ allocated_mem = (struct phys_mem_desc *)OSAL_ZALLOC(p_hwfn->p_dev,
+ GFP_KERNEL,
+ NUM_STORMS *
+ sizeof(struct phys_mem_desc));
+ if (!allocated_mem)
+ return OSAL_NULL;
+
+ OSAL_MEMSET(allocated_mem, 0, NUM_STORMS *
+ sizeof(struct phys_mem_desc));
+
+ /* For each Storm, set physical address in RAM */
+ while (buf_offset < buf_size) {
+ struct phys_mem_desc *storm_mem_desc;
+ struct fw_overlay_buf_hdr *hdr;
+ u32 storm_buf_size;
+ u8 storm_id;
+
+ hdr =
+ (struct fw_overlay_buf_hdr *)&fw_overlay_in_buf[buf_offset];
+ storm_buf_size = GET_FIELD(hdr->data,
+ FW_OVERLAY_BUF_HDR_BUF_SIZE);
+ storm_id = GET_FIELD(hdr->data, FW_OVERLAY_BUF_HDR_STORM_ID);
+ storm_mem_desc = allocated_mem + storm_id;
+ storm_mem_desc->size = storm_buf_size * sizeof(u32);
+
+ /* Allocate physical memory for Storm's overlays buffer */
+ storm_mem_desc->virt_addr =
+ OSAL_DMA_ALLOC_COHERENT(p_hwfn->p_dev,
+ &storm_mem_desc->phys_addr,
+ storm_mem_desc->size);
+ if (!storm_mem_desc->virt_addr)
+ break;
+
+ /* Skip overlays buffer header */
+ buf_offset += OVERLAY_HDR_SIZE_DWORDS;
+
+ /* Copy Storm's overlays buffer to allocated memory */
+ OSAL_MEMCPY(storm_mem_desc->virt_addr,
+ &fw_overlay_in_buf[buf_offset],
+ storm_mem_desc->size);
+
+ /* Advance to next Storm */
+ buf_offset += storm_buf_size;
+ }
+
+ /* If memory allocation has failed, free all allocated memory */
+ if (buf_offset < buf_size) {
+ ecore_fw_overlay_mem_free(p_hwfn, allocated_mem);
+ return OSAL_NULL;
+ }
+
+ return allocated_mem;
+}
+
+void ecore_fw_overlay_init_ram(struct ecore_hwfn *p_hwfn,
+ struct ecore_ptt *p_ptt,
+ struct phys_mem_desc *fw_overlay_mem)
+{
+ u8 storm_id;
+
+ for (storm_id = 0; storm_id < NUM_STORMS; storm_id++) {
+ struct phys_mem_desc *storm_mem_desc =
+ (struct phys_mem_desc *)fw_overlay_mem + storm_id;
+ u32 ram_addr, i;
+
+ /* Skip Storms with no FW overlays */
+ if (!storm_mem_desc->virt_addr)
+ continue;
+
+ /* Calculate overlay RAM GRC address of current PF */
+ ram_addr = ecore_get_overlay_addr_ram_addr(p_hwfn, storm_id) +
+ sizeof(dma_addr_t) * p_hwfn->rel_pf_id;
+
+ /* Write Storm's overlay physical address to RAM */
+ for (i = 0; i < PHYS_ADDR_DWORDS; i++, ram_addr += sizeof(u32))
+ ecore_wr(p_hwfn, p_ptt, ram_addr,
+ ((u32 *)&storm_mem_desc->phys_addr)[i]);
+ }
+}
+
+void ecore_fw_overlay_mem_free(struct ecore_hwfn *p_hwfn,
+ struct phys_mem_desc *fw_overlay_mem)
+{
+ u8 storm_id;
+
+ if (!fw_overlay_mem)
+ return;
+
+ for (storm_id = 0; storm_id < NUM_STORMS; storm_id++) {
+ struct phys_mem_desc *storm_mem_desc =
+ (struct phys_mem_desc *)fw_overlay_mem + storm_id;
+
+ /* Free Storm's physical memory */
+ if (storm_mem_desc->virt_addr)
+ OSAL_DMA_FREE_COHERENT(p_hwfn->p_dev,
+ storm_mem_desc->virt_addr,
+ storm_mem_desc->phys_addr,
+ storm_mem_desc->size);
+ }
+
+ /* Free allocated virtual memory */
+ OSAL_FREE(p_hwfn->p_dev, fw_overlay_mem);
+}
diff --git a/drivers/net/qede/base/ecore_init_fw_funcs.h b/drivers/net/qede/base/ecore_init_fw_funcs.h
index 1d1b107c4..912451662 100644
--- a/drivers/net/qede/base/ecore_init_fw_funcs.h
+++ b/drivers/net/qede/base/ecore_init_fw_funcs.h
@@ -553,5 +553,40 @@ void ecore_set_rdma_error_level(struct ecore_hwfn *p_hwfn,
struct ecore_ptt *p_ptt,
u8 assert_level[NUM_STORMS]);
+/**
+ * @brief ecore_fw_overlay_mem_alloc - Allocates and fills the FW overlay memory
+ *
+ * @param p_hwfn - HW device data
+ * @param fw_overlay_in_buf - the input FW overlay buffer.
+ * @param buf_size - the size of the input FW overlay buffer in bytes.
+ * must be aligned to dwords.
+ * @param fw_overlay_out_mem - OUT: a pointer to the allocated overlays memory.
+ *
+ * @return a pointer to the allocated overlays memory, or OSAL_NULL in case of
+ * failures.
+ */
+struct phys_mem_desc *ecore_fw_overlay_mem_alloc(struct ecore_hwfn *p_hwfn,
+ const u32 *const fw_overlay_in_buf,
+ u32 buf_size_in_bytes);
+
+/**
+ * @brief ecore_fw_overlay_init_ram - Initializes the FW overlay RAM.
+ *
+ * @param p_hwfn - HW device data.
+ * @param p_ptt - ptt window used for writing the registers.
+ * @param fw_overlay_mem - the allocated FW overlay memory.
+ */
+void ecore_fw_overlay_init_ram(struct ecore_hwfn *p_hwfn,
+ struct ecore_ptt *p_ptt,
+ struct phys_mem_desc *fw_overlay_mem);
+
+/**
+ * @brief ecore_fw_overlay_mem_free - Frees the FW overlay memory.
+ *
+ * @param p_hwfn - HW device data.
+ * @param fw_overlay_mem - the allocated FW overlay memory to free.
+ */
+void ecore_fw_overlay_mem_free(struct ecore_hwfn *p_hwfn,
+ struct phys_mem_desc *fw_overlay_mem);
#endif
diff --git a/drivers/net/qede/base/ecore_int.c b/drivers/net/qede/base/ecore_int.c
index b1e127849..4207b1853 100644
--- a/drivers/net/qede/base/ecore_int.c
+++ b/drivers/net/qede/base/ecore_int.c
@@ -192,7 +192,10 @@ static enum _ecore_status_t ecore_pswhst_attn_cb(struct ecore_hwfn *p_hwfn)
return ECORE_SUCCESS;
}
-#define ECORE_GRC_ATTENTION_VALID_BIT (1 << 0)
+/* Register GRC_REG_TIMEOUT_ATTN_ACCESS_VALID */
+#define ECORE_GRC_ATTENTION_VALID_BIT_MASK (0x1)
+#define ECORE_GRC_ATTENTION_VALID_BIT_SHIFT (0)
+
#define ECORE_GRC_ATTENTION_ADDRESS_MASK (0x7fffff << 0)
#define ECORE_GRC_ATTENTION_RDWR_BIT (1 << 23)
#define ECORE_GRC_ATTENTION_MASTER_MASK (0xf << 24)
@@ -237,15 +240,14 @@ static enum _ecore_status_t ecore_grc_attn_cb(struct ecore_hwfn *p_hwfn)
u32 tmp, tmp2;
/* We've already cleared the timeout interrupt register, so we learn
- * of interrupts via the validity register.
- * Any attention which is not for a timeout event is treated as fatal.
+ * of interrupts via the validity register. If it is not a timeout do
+ * nothing. It is too late at this stage to differentiate spurious
+ * interrupt from fatal grc attention.
*/
tmp = ecore_rd(p_hwfn, p_hwfn->p_dpc_ptt,
GRC_REG_TIMEOUT_ATTN_ACCESS_VALID);
- if (!(tmp & ECORE_GRC_ATTENTION_VALID_BIT)) {
- rc = ECORE_INVAL;
+ if (!(GET_FIELD(tmp, ECORE_GRC_ATTENTION_VALID_BIT)))
goto out;
- }
/* Read the GRC timeout information */
tmp = ecore_rd(p_hwfn, p_hwfn->p_dpc_ptt,
diff --git a/drivers/net/qede/base/ecore_iro_values.h b/drivers/net/qede/base/ecore_iro_values.h
index 6442057ac..dd7349778 100644
--- a/drivers/net/qede/base/ecore_iro_values.h
+++ b/drivers/net/qede/base/ecore_iro_values.h
@@ -17,8 +17,8 @@ static const u32 iro_arr[] = {
/* offset=0x0, size=0x8 */
0x00000000, 0x00000000, 0x00080000,
/* TSTORM_PORT_STAT_OFFSET(port_id), */
- /* offset=0x3908, mult1=0x88, size=0x88 */
- 0x00003908, 0x00000088, 0x00880000,
+ /* offset=0x3288, mult1=0x88, size=0x88 */
+ 0x00003288, 0x00000088, 0x00880000,
/* TSTORM_LL2_PORT_STAT_OFFSET(port_id), */
/* offset=0x58f0, mult1=0x20, size=0x20 */
0x000058f0, 0x00000020, 0x00200000,
@@ -38,38 +38,38 @@ static const u32 iro_arr[] = {
/* offset=0x84, mult1=0x8, size=0x2 */
0x00000084, 0x00000008, 0x00020000,
/* XSTORM_PQ_INFO_OFFSET(pq_id), */
- /* offset=0x5618, mult1=0x4, size=0x4 */
- 0x00005618, 0x00000004, 0x00040000,
- /* XSTORM_INTEG_TEST_DATA_OFFSET */
- /* offset=0x4cd0, size=0x78 */
- 0x00004cd0, 0x00000000, 0x00780000,
+ /* offset=0x5718, mult1=0x4, size=0x4 */
+ 0x00005718, 0x00000004, 0x00040000,
+ /* XSTORM_INTEG_TEST_DATA_OFFSET, */
+ /* offset=0x4dd0, size=0x78 */
+ 0x00004dd0, 0x00000000, 0x00780000,
/* YSTORM_INTEG_TEST_DATA_OFFSET */
/* offset=0x3e40, size=0x78 */
0x00003e40, 0x00000000, 0x00780000,
- /* PSTORM_INTEG_TEST_DATA_OFFSET */
- /* offset=0x3e00, size=0x78 */
- 0x00003e00, 0x00000000, 0x00780000,
- /* TSTORM_INTEG_TEST_DATA_OFFSET */
- /* offset=0x3890, size=0x78 */
- 0x00003890, 0x00000000, 0x00780000,
+ /* PSTORM_INTEG_TEST_DATA_OFFSET, */
+ /* offset=0x4480, size=0x78 */
+ 0x00004480, 0x00000000, 0x00780000,
+ /* TSTORM_INTEG_TEST_DATA_OFFSET, */
+ /* offset=0x3210, size=0x78 */
+ 0x00003210, 0x00000000, 0x00780000,
/* MSTORM_INTEG_TEST_DATA_OFFSET */
/* offset=0x3b50, size=0x78 */
0x00003b50, 0x00000000, 0x00780000,
/* USTORM_INTEG_TEST_DATA_OFFSET */
/* offset=0x7f58, size=0x78 */
0x00007f58, 0x00000000, 0x00780000,
- /* XSTORM_OVERLAY_BUF_ADDR_OFFSET */
- /* offset=0x5e58, size=0x8 */
- 0x00005e58, 0x00000000, 0x00080000,
+ /* XSTORM_OVERLAY_BUF_ADDR_OFFSET, */
+ /* offset=0x5f58, size=0x8 */
+ 0x00005f58, 0x00000000, 0x00080000,
/* YSTORM_OVERLAY_BUF_ADDR_OFFSET */
/* offset=0x7100, size=0x8 */
0x00007100, 0x00000000, 0x00080000,
- /* PSTORM_OVERLAY_BUF_ADDR_OFFSET */
- /* offset=0xa820, size=0x8 */
- 0x0000a820, 0x00000000, 0x00080000,
- /* TSTORM_OVERLAY_BUF_ADDR_OFFSET */
- /* offset=0x4a18, size=0x8 */
- 0x00004a18, 0x00000000, 0x00080000,
+ /* PSTORM_OVERLAY_BUF_ADDR_OFFSET, */
+ /* offset=0xaea0, size=0x8 */
+ 0x0000aea0, 0x00000000, 0x00080000,
+ /* TSTORM_OVERLAY_BUF_ADDR_OFFSET, */
+ /* offset=0x4398, size=0x8 */
+ 0x00004398, 0x00000000, 0x00080000,
/* MSTORM_OVERLAY_BUF_ADDR_OFFSET */
/* offset=0xa5a0, size=0x8 */
0x0000a5a0, 0x00000000, 0x00080000,
@@ -86,8 +86,8 @@ static const u32 iro_arr[] = {
/* offset=0xc210, mult1=0x30, size=0x30 */
0x0000c210, 0x00000030, 0x00300000,
/* CORE_LL2_PSTORM_PER_QUEUE_STAT_OFFSET(core_tx_stats_id), */
- /* offset=0xaa08, mult1=0x38, size=0x38 */
- 0x0000aa08, 0x00000038, 0x00380000,
+ /* offset=0xb088, mult1=0x38, size=0x38 */
+ 0x0000b088, 0x00000038, 0x00380000,
/* MSTORM_QUEUE_STAT_OFFSET(stat_counter_id), */
/* offset=0x3d20, mult1=0x80, size=0x40 */
0x00003d20, 0x00000080, 0x00400000,
@@ -110,20 +110,20 @@ static const u32 iro_arr[] = {
/* offset=0xe840, mult1=0x60, size=0x60 */
0x0000e840, 0x00000060, 0x00600000,
/* PSTORM_QUEUE_STAT_OFFSET(stat_counter_id), */
- /* offset=0x3f98, mult1=0x80, size=0x38 */
- 0x00003f98, 0x00000080, 0x00380000,
+ /* offset=0x4618, mult1=0x80, size=0x38 */
+ 0x00004618, 0x00000080, 0x00380000,
/* PSTORM_ETH_PF_STAT_OFFSET(pf_id), */
- /* offset=0x100b8, mult1=0xc0, size=0xc0 */
- 0x000100b8, 0x000000c0, 0x00c00000,
+ /* offset=0x10738, mult1=0xc0, size=0xc0 */
+ 0x00010738, 0x000000c0, 0x00c00000,
/* PSTORM_CTL_FRAME_ETHTYPE_OFFSET(ethType_id), */
/* offset=0x1f8, mult1=0x2, size=0x2 */
0x000001f8, 0x00000002, 0x00020000,
- /* TSTORM_ETH_PRS_INPUT_OFFSET */
- /* offset=0xa2a0, size=0x108 */
- 0x0000a2a0, 0x00000000, 0x01080000,
+ /* TSTORM_ETH_PRS_INPUT_OFFSET, */
+ /* offset=0xa2a8, size=0x108 */
+ 0x0000a2a8, 0x00000000, 0x01080000,
/* ETH_RX_RATE_LIMIT_OFFSET(pf_id), */
- /* offset=0xa3a8, mult1=0x8, size=0x8 */
- 0x0000a3a8, 0x00000008, 0x00080000,
+ /* offset=0xa3b0, mult1=0x8, size=0x8 */
+ 0x0000a3b0, 0x00000008, 0x00080000,
/* TSTORM_ETH_RSS_UPDATE_OFFSET(pf_id), */
/* offset=0x1c0, mult1=0x8, size=0x8 */
0x000001c0, 0x00000008, 0x00080000,
@@ -149,8 +149,8 @@ static const u32 iro_arr[] = {
/* offset=0xb78, mult1=0x18, mult2=0x8, size=0x2 */
0x00000b78, 0x00080018, 0x00020000,
/* TSTORM_ISCSI_RX_STATS_OFFSET(storage_func_id), */
- /* offset=0xc640, mult1=0x50, size=0x3c */
- 0x0000c640, 0x00000050, 0x003c0000,
+ /* offset=0xc648, mult1=0x50, size=0x3c */
+ 0x0000c648, 0x00000050, 0x003c0000,
/* MSTORM_ISCSI_RX_STATS_OFFSET(storage_func_id), */
/* offset=0x12038, mult1=0x18, size=0x10 */
0x00012038, 0x00000018, 0x00100000,
@@ -158,38 +158,38 @@ static const u32 iro_arr[] = {
/* offset=0x11b00, mult1=0x40, size=0x18 */
0x00011b00, 0x00000040, 0x00180000,
/* XSTORM_ISCSI_TX_STATS_OFFSET(storage_func_id), */
- /* offset=0x94d0, mult1=0x50, size=0x20 */
- 0x000094d0, 0x00000050, 0x00200000,
+ /* offset=0x95d0, mult1=0x50, size=0x20 */
+ 0x000095d0, 0x00000050, 0x00200000,
/* YSTORM_ISCSI_TX_STATS_OFFSET(storage_func_id), */
/* offset=0x8b10, mult1=0x40, size=0x28 */
0x00008b10, 0x00000040, 0x00280000,
/* PSTORM_ISCSI_TX_STATS_OFFSET(storage_func_id), */
- /* offset=0x10fc0, mult1=0x18, size=0x10 */
- 0x00010fc0, 0x00000018, 0x00100000,
+ /* offset=0x11640, mult1=0x18, size=0x10 */
+ 0x00011640, 0x00000018, 0x00100000,
/* TSTORM_FCOE_RX_STATS_OFFSET(pf_id), */
- /* offset=0xc828, mult1=0x48, size=0x38 */
- 0x0000c828, 0x00000048, 0x00380000,
+ /* offset=0xc830, mult1=0x48, size=0x38 */
+ 0x0000c830, 0x00000048, 0x00380000,
/* PSTORM_FCOE_TX_STATS_OFFSET(pf_id), */
- /* offset=0x11090, mult1=0x20, size=0x20 */
- 0x00011090, 0x00000020, 0x00200000,
+ /* offset=0x11710, mult1=0x20, size=0x20 */
+ 0x00011710, 0x00000020, 0x00200000,
/* PSTORM_RDMA_QUEUE_STAT_OFFSET(rdma_stat_counter_id), */
- /* offset=0x3fd0, mult1=0x80, size=0x10 */
- 0x00003fd0, 0x00000080, 0x00100000,
+ /* offset=0x4650, mult1=0x80, size=0x10 */
+ 0x00004650, 0x00000080, 0x00100000,
/* TSTORM_RDMA_QUEUE_STAT_OFFSET(rdma_stat_counter_id), */
- /* offset=0x3c98, mult1=0x10, size=0x10 */
- 0x00003c98, 0x00000010, 0x00100000,
+ /* offset=0x3618, mult1=0x10, size=0x10 */
+ 0x00003618, 0x00000010, 0x00100000,
/* XSTORM_RDMA_ASSERT_LEVEL_OFFSET(pf_id), */
- /* offset=0xa868, mult1=0x8, size=0x1 */
- 0x0000a868, 0x00000008, 0x00010000,
+ /* offset=0xa968, mult1=0x8, size=0x1 */
+ 0x0000a968, 0x00000008, 0x00010000,
/* YSTORM_RDMA_ASSERT_LEVEL_OFFSET(pf_id), */
/* offset=0x97a0, mult1=0x8, size=0x1 */
0x000097a0, 0x00000008, 0x00010000,
/* PSTORM_RDMA_ASSERT_LEVEL_OFFSET(pf_id), */
- /* offset=0x11310, mult1=0x8, size=0x1 */
- 0x00011310, 0x00000008, 0x00010000,
+ /* offset=0x11990, mult1=0x8, size=0x1 */
+ 0x00011990, 0x00000008, 0x00010000,
/* TSTORM_RDMA_ASSERT_LEVEL_OFFSET(pf_id), */
- /* offset=0xf018, mult1=0x8, size=0x1 */
- 0x0000f018, 0x00000008, 0x00010000,
+ /* offset=0xf020, mult1=0x8, size=0x1 */
+ 0x0000f020, 0x00000008, 0x00010000,
/* MSTORM_RDMA_ASSERT_LEVEL_OFFSET(pf_id), */
/* offset=0x12628, mult1=0x8, size=0x1 */
0x00012628, 0x00000008, 0x00010000,
@@ -197,11 +197,11 @@ static const u32 iro_arr[] = {
/* offset=0x11da8, mult1=0x8, size=0x1 */
0x00011da8, 0x00000008, 0x00010000,
/* XSTORM_IWARP_RXMIT_STATS_OFFSET(pf_id), */
- /* offset=0xa978, mult1=0x30, size=0x10 */
- 0x0000a978, 0x00000030, 0x00100000,
+ /* offset=0xaa78, mult1=0x30, size=0x10 */
+ 0x0000aa78, 0x00000030, 0x00100000,
/* TSTORM_ROCE_EVENTS_STAT_OFFSET(roce_pf_id), */
- /* offset=0xd768, mult1=0x28, size=0x28 */
- 0x0000d768, 0x00000028, 0x00280000,
+ /* offset=0xd770, mult1=0x28, size=0x28 */
+ 0x0000d770, 0x00000028, 0x00280000,
/* YSTORM_ROCE_DCQCN_RECEIVED_STATS_OFFSET(roce_pf_id), */
/* offset=0x9a58, mult1=0x18, size=0x18 */
0x00009a58, 0x00000018, 0x00180000,
@@ -209,14 +209,14 @@ static const u32 iro_arr[] = {
/* offset=0x9bd8, mult1=0x8, size=0x8 */
0x00009bd8, 0x00000008, 0x00080000,
/* PSTORM_ROCE_DCQCN_SENT_STATS_OFFSET(roce_pf_id), */
- /* offset=0x13398, mult1=0x8, size=0x8 */
- 0x00013398, 0x00000008, 0x00080000,
+ /* offset=0x13a18, mult1=0x8, size=0x8 */
+ 0x00013a18, 0x00000008, 0x00080000,
/* USTORM_ROCE_CQE_STATS_OFFSET(roce_pf_id), */
/* offset=0x126e8, mult1=0x18, size=0x18 */
0x000126e8, 0x00000018, 0x00180000,
/* TSTORM_NVMF_PORT_TASKPOOL_PRODUCER_CONSUMER_OFFSET(port_num_id,taskpool_index), */
- /* offset=0xe608, mult1=0x288, mult2=0x50, size=0x10 */
- 0x0000e608, 0x00500288, 0x00100000,
+ /* offset=0xe610, mult1=0x288, mult2=0x50, size=0x10 */
+ 0x0000e610, 0x00500288, 0x00100000,
/* USTORM_NVMF_PORT_COUNTERS_OFFSET(port_num_id), */
/* offset=0x12970, mult1=0x138, size=0x28 */
0x00012970, 0x00000138, 0x00280000,
diff --git a/drivers/net/qede/qede_ethdev.h b/drivers/net/qede/qede_ethdev.h
index 1ac2d086a..21115a077 100644
--- a/drivers/net/qede/qede_ethdev.h
+++ b/drivers/net/qede/qede_ethdev.h
@@ -46,7 +46,7 @@
#define QEDE_PMD_VER_PREFIX "QEDE PMD"
#define QEDE_PMD_VERSION_MAJOR 2
#define QEDE_PMD_VERSION_MINOR 11
-#define QEDE_PMD_VERSION_REVISION 0
+#define QEDE_PMD_VERSION_REVISION 3
#define QEDE_PMD_VERSION_PATCH 1
#define QEDE_PMD_DRV_VERSION qede_stringify(QEDE_PMD_VERSION_MAJOR) "." \
diff --git a/drivers/net/qede/qede_main.c b/drivers/net/qede/qede_main.c
index c9caec645..4eb79d0fb 100644
--- a/drivers/net/qede/qede_main.c
+++ b/drivers/net/qede/qede_main.c
@@ -18,7 +18,7 @@
char qede_fw_file[PATH_MAX];
static const char * const QEDE_DEFAULT_FIRMWARE =
- "/lib/firmware/qed/qed_init_values-8.40.25.0.bin";
+ "/lib/firmware/qed/qed_init_values-8.40.33.0.bin";
static void
qed_update_pf_params(struct ecore_dev *edev, struct ecore_pf_params *params)
--
2.18.0
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [dpdk-dev] [PATCH 1/3] net/qede/base: fix for FW bug
2019-10-20 5:20 Rasesh Mody
@ 2019-10-21 10:58 ` Kevin Traynor
2019-10-30 11:25 ` Jerin Jacob
0 siblings, 1 reply; 10+ messages in thread
From: Kevin Traynor @ 2019-10-21 10:58 UTC (permalink / raw)
To: Rasesh Mody, dev, jerinj, ferruh.yigit; +Cc: GR-Everest-DPDK-Dev
On 20/10/2019 06:20, Rasesh Mody wrote:
> In our testing we have identified a critical FW bug. Performance
> is degraded significantly for certain packet sizes with 8.40.25.0 FW.
>
> This patch updates the FW to version 8.40.33.0. The updated FW has a
> fix to performance issue.
>
> The patch also adds initialization for FW overlay RAM as part of
> hardware initialization which is required by the new FW.
>
> Fixes: 066f0d380a2f ("net/qede/base: update FW to 8.40.25.0")
>
Hi Rasesh, it should be:
Fixes: 21c959e2cb10 ("net/qede/base: update FW to 8.40.25.0")
> Signed-off-by: Rasesh Mody <rmody@marvell.com>
> ---
> doc/guides/nics/qede.rst | 4 +-
> drivers/net/qede/base/common_hsi.h | 6 +-
> drivers/net/qede/base/ecore.h | 4 +-
> drivers/net/qede/base/ecore_dev.c | 19 ++-
> drivers/net/qede/base/ecore_hsi_debug_tools.h | 3 +
> drivers/net/qede/base/ecore_hw.c | 4 +-
> drivers/net/qede/base/ecore_init_fw_funcs.c | 138 ++++++++++++++++++
> drivers/net/qede/base/ecore_init_fw_funcs.h | 35 +++++
> drivers/net/qede/base/ecore_int.c | 14 +-
> drivers/net/qede/base/ecore_iro_values.h | 122 ++++++++--------
> drivers/net/qede/qede_ethdev.h | 2 +-
> drivers/net/qede/qede_main.c | 2 +-
> 12 files changed, 274 insertions(+), 79 deletions(-)
>
> diff --git a/doc/guides/nics/qede.rst b/doc/guides/nics/qede.rst
> index 471d98014..2f4045795 100644
> --- a/doc/guides/nics/qede.rst
> +++ b/doc/guides/nics/qede.rst
> @@ -78,9 +78,9 @@ Supported QLogic Adapters
> Prerequisites
> -------------
>
> -- Requires storm firmware version **8.37.7.0**. Firmware may be available
> +- Requires storm firmware version **8.40.33.0**. Firmware may be available
> inbox in certain newer Linux distros under the standard directory
> - ``E.g. /lib/firmware/qed/qed_init_values-8.37.7.0.bin``.
> + ``E.g. /lib/firmware/qed/qed_init_values-8.40.33.0.bin``.
Not available in latest Fedora and Ubuntu releases
> If the required firmware files are not available then download it from
> `linux-firmware git repository <http://git.kernel.org/pub/scm/linux/kernel/git/firmware/linux-firmware.git/tree/qed>`_
Not available here either
> or `QLogic Driver Download Center <http://driverdownloads.qlogic.com/QLogicDriverDownloads_UI/DefaultNewSearch.aspx>`_.
Maybe here? 8.37 is the latest elsewhere.
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [dpdk-dev] [PATCH 1/3] net/qede/base: fix for FW bug
2019-10-21 10:58 ` Kevin Traynor
@ 2019-10-30 11:25 ` Jerin Jacob
0 siblings, 0 replies; 10+ messages in thread
From: Jerin Jacob @ 2019-10-30 11:25 UTC (permalink / raw)
To: Kevin Traynor
Cc: Rasesh Mody, dpdk-dev, Jerin Jacob, Ferruh Yigit, GR-Everest-DPDK-Dev
On Mon, Oct 21, 2019 at 4:29 PM Kevin Traynor <ktraynor@redhat.com> wrote:
>
> On 20/10/2019 06:20, Rasesh Mody wrote:
> > In our testing we have identified a critical FW bug. Performance
> > is degraded significantly for certain packet sizes with 8.40.25.0 FW.
> >
> > This patch updates the FW to version 8.40.33.0. The updated FW has a
> > fix to performance issue.
> >
> > The patch also adds initialization for FW overlay RAM as part of
> > hardware initialization which is required by the new FW.
> >
> > Fixes: 066f0d380a2f ("net/qede/base: update FW to 8.40.25.0")
> >
>
> Hi Rasesh, it should be:
>
> Fixes: 21c959e2cb10 ("net/qede/base: update FW to 8.40.25.0")
# Updated the Fixes git id
# Added cc: stable@dpdk.org
# Changed the subject to "net/qede/base: update FW to 8.40.33.0"
Series applied to dpdk-next-net-mrvl/master. Thanks.
> > Signed-off-by: Rasesh Mody <rmody@marvell.com>
^ permalink raw reply [flat|nested] 10+ messages in thread
end of thread, other threads:[~2019-10-30 19:16 UTC | newest]
Thread overview: 10+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-10-21 18:52 [dpdk-dev] [PATCH 1/3] net/qede/base: fix for FW bug Rasesh Mody
2019-10-21 20:10 ` Kevin Traynor
2019-10-22 8:45 ` Ferruh Yigit
2019-10-22 9:13 ` Kevin Traynor
2019-10-30 19:15 ` Rasesh Mody
-- strict thread matches above, loose matches on Subject: below --
2019-10-21 20:46 Rasesh Mody
2019-10-21 21:41 ` Kevin Traynor
2019-10-20 5:20 Rasesh Mody
2019-10-21 10:58 ` Kevin Traynor
2019-10-30 11:25 ` Jerin Jacob
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).