CTB writes are now in the path of command submission and should be
optimized for performance. Rather than reading CTB descriptor values
(e.g. head, tail) which could result in accesses across the PCIe bus,
store shadow local copies and only read/write the descriptor values when
absolutely necessary
On Wed, Jul 07, 2021 at 01:21:35PM -0700, John Harrison wrote:
> On 7/7/2021 11:56, Matthew Brost wrote:
>
> > Ok, I sent it but I looks like patchworks didn't like it. Anyways we
> > should be able to review that patch.
> >
> > Matt
> Maybe because it came out as 6/56 instead of 6/7? Also, not s
On 7/7/2021 11:56, Matthew Brost wrote:
Ok, I sent it but I looks like patchworks didn't like it. Anyways we
should be able to review that patch.
Matt
Maybe because it came out as 6/56 instead of 6/7? Also, not sure if it
needs to be in reply to 0/7 or 6/7?
John.
On Wed, Jul 07, 2021 at 11:19:01AM -0700, John Harrison wrote:
> On 7/7/2021 10:50, Matthew Brost wrote:
> > On Tue, Jul 06, 2021 at 03:51:00PM -0700, John Harrison wrote:
> > > On 7/6/2021 15:20, Matthew Brost wrote:
> > > > CTB writes are now in the path of command submission and should be
> > >
On 7/7/2021 10:50, Matthew Brost wrote:
On Tue, Jul 06, 2021 at 03:51:00PM -0700, John Harrison wrote:
On 7/6/2021 15:20, Matthew Brost wrote:
CTB writes are now in the path of command submission and should be
optimized for performance. Rather than reading CTB descriptor values
(e.g. head, tail
On Tue, Jul 06, 2021 at 03:51:00PM -0700, John Harrison wrote:
> On 7/6/2021 15:20, Matthew Brost wrote:
> > CTB writes are now in the path of command submission and should be
> > optimized for performance. Rather than reading CTB descriptor values
> > (e.g. head, tail) which could result in access
On 7/6/2021 15:20, Matthew Brost wrote:
CTB writes are now in the path of command submission and should be
optimized for performance. Rather than reading CTB descriptor values
(e.g. head, tail) which could result in accesses across the PCIe bus,
store shadow local copies and only read/write the d
On 7/6/2021 12:33, Michal Wajdeczko wrote:
On 06.07.2021 21:19, John Harrison wrote:
On 7/6/2021 12:12, Michal Wajdeczko wrote:
On 06.07.2021 21:00, John Harrison wrote:
On 7/1/2021 10:15, Matthew Brost wrote:
CTB writes are now in the path of command submission and should be
optimized for pe
CTB writes are now in the path of command submission and should be
optimized for performance. Rather than reading CTB descriptor values
(e.g. head, tail) which could result in accesses across the PCIe bus,
store shadow local copies and only read/write the descriptor values when
absolutely necessary
On Tue, Jul 06, 2021 at 09:33:23PM +0200, Michal Wajdeczko wrote:
>
>
> On 06.07.2021 21:19, John Harrison wrote:
> > On 7/6/2021 12:12, Michal Wajdeczko wrote:
> >> On 06.07.2021 21:00, John Harrison wrote:
> >>> On 7/1/2021 10:15, Matthew Brost wrote:
> CTB writes are now in the path of co
On 06.07.2021 21:19, John Harrison wrote:
> On 7/6/2021 12:12, Michal Wajdeczko wrote:
>> On 06.07.2021 21:00, John Harrison wrote:
>>> On 7/1/2021 10:15, Matthew Brost wrote:
CTB writes are now in the path of command submission and should be
optimized for performance. Rather than read
On 7/6/2021 12:12, Michal Wajdeczko wrote:
On 06.07.2021 21:00, John Harrison wrote:
On 7/1/2021 10:15, Matthew Brost wrote:
CTB writes are now in the path of command submission and should be
optimized for performance. Rather than reading CTB descriptor values
(e.g. head, tail) which could resu
On 06.07.2021 21:00, John Harrison wrote:
> On 7/1/2021 10:15, Matthew Brost wrote:
>> CTB writes are now in the path of command submission and should be
>> optimized for performance. Rather than reading CTB descriptor values
>> (e.g. head, tail) which could result in accesses across the PCIe bu
On 7/1/2021 10:15, Matthew Brost wrote:
CTB writes are now in the path of command submission and should be
optimized for performance. Rather than reading CTB descriptor values
(e.g. head, tail) which could result in accesses across the PCIe bus,
store shadow local copies and only read/write the d
CTB writes are now in the path of command submission and should be
optimized for performance. Rather than reading CTB descriptor values
(e.g. head, tail) which could result in accesses across the PCIe bus,
store shadow local copies and only read/write the descriptor values when
absolutely necessary
CTB writes are now in the path of command submission and should be
optimized for performance. Rather than reading CTB descriptor values
(e.g. head, tail) which could result in accesses across the PCIe bus,
store shadow local copies and only read/write the descriptor values when
absolutely necessary
16 matches
Mail list logo
