From: "Randhawa, Jagmeet" <jagmeet.randhawa@intel.com>
To: "Zeng, Oak" <oak.zeng@intel.com>
Cc: "igt-dev@lists.freedesktop.org" <igt-dev@lists.freedesktop.org>
Subject: Re: [PATCH i-g-t] tests/intel/xe_vm: Fix WB caching PAT index for coherency mode
Date: Wed, 27 Mar 2024 11:25:05 -0700 [thread overview]
Message-ID: <d9b33040-0372-4fd0-ba63-4f9e42ba31b0@intel.com> (raw)
In-Reply-To: <SA1PR11MB69916004963FDF5C1987A6B192352@SA1PR11MB6991.namprd11.prod.outlook.com>
On 3/26/2024 11:46 AM, Zeng, Oak wrote:
> Hi Jagmeet,
>
>> -----Original Message-----
>> From: Randhawa, Jagmeet <jagmeet.randhawa@intel.com>
>> Sent: Monday, March 25, 2024 5:27 PM
>> Cc: igt-dev@lists.freedesktop.org; Zeng, Oak <oak.zeng@intel.com>; Cavitt,
>> Jonathan <jonathan.cavitt@intel.com>; Auld, Matthew
>> <matthew.auld@intel.com>; kamil.konieczny@linux.intel.com; Randhawa,
>> Jagmeet <jagmeet.randhawa@intel.com>
>> Subject: [PATCH i-g-t] tests/intel/xe_vm: Fix WB caching PAT index for coherency
>> mode
>>
>> Resolve a coherency conflict by setting the
>> correct PAT index for WB caching with
>> intel_get_pat_idx_wb(fd), addressing the
>> mismatch between the default "XE_COH_NONE"
>> coherency mode and DRM_XE_GEM_CPU_CACHING_WB.
> I feel the description here is still not very clear. Isn't a confliction b/t the default cpu caching_mode and default pat_index?
>
> Again, XE_COH_NONE is a xekmd *internal* concept, not part of xekmd uAPI. User can't set the XE_COH_NONE mode. I think this is determined by pat_index....
>
> By looking at driver, I figured the only not allowed setting is: cpu_caching == WB && a pat_index which indicates a XE_COH_NONE coherence mode...
>
> So maybe something like: driver doesn't allow cpu_caching == WB && a pat_index which indicates a XE_COH_NONE coherence mode
>
> The code looks good to me. So it is reviewed-by: Oak Zeng <oak.zeng@intel.com>. Improve a little the description message if you want.
>
>
> Oak
I agree, the description has now been updated and better explains the
code change,
Thank you!
>
>
>> This conflict occurs for VLK:54352.
>>
>> Cc: Jonathan Cavitt <jonathan.cavitt@intel.com>
>> Cc: Matthew Auld <matthew.auld@intel.com>
>> Cc: Oak Zeng <oak.zeng@intel.com>
>> Signed-off-by: Jagmeet Randhawa <jagmeet.randhawa@intel.com>
>> ---
>> tests/intel/xe_vm.c | 1 +
>> 1 file changed, 1 insertion(+)
>>
>> diff --git a/tests/intel/xe_vm.c b/tests/intel/xe_vm.c
>> index 7c061c497..1f458174d 100644
>> --- a/tests/intel/xe_vm.c
>> +++ b/tests/intel/xe_vm.c
>> @@ -1789,6 +1789,7 @@ static void bind_flag_invalid(int fd)
>> bind.bind.range = bo_size;
>> bind.bind.addr = addr;
>> bind.bind.op = DRM_XE_VM_BIND_OP_MAP;
>> + bind.bind.pat_index = intel_get_pat_idx_wb(fd);
>> bind.num_syncs = 1;
>> bind.syncs = (uintptr_t)sync;
>>
>> --
>> 2.25.1
next prev parent reply other threads:[~2024-03-27 18:25 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-03-25 21:26 [PATCH i-g-t] tests/intel/xe_vm: Fix WB caching PAT index for coherency mode Jagmeet Randhawa
2024-03-25 21:27 ` Jagmeet Randhawa
2024-03-26 7:51 ` Matthew Auld
2024-03-26 10:00 ` Kamil Konieczny
2024-03-26 18:46 ` Zeng, Oak
2024-03-27 18:25 ` Randhawa, Jagmeet [this message]
2024-03-25 22:34 ` ✓ CI.xeBAT: success for " Patchwork
2024-03-25 22:47 ` ✓ Fi.CI.BAT: " Patchwork
2024-03-26 4:46 ` ✗ Fi.CI.IGT: failure " Patchwork
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=d9b33040-0372-4fd0-ba63-4f9e42ba31b0@intel.com \
--to=jagmeet.randhawa@intel.com \
--cc=igt-dev@lists.freedesktop.org \
--cc=oak.zeng@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox