Re: top enhancements + 2 bugs swatted

Jim Warner <[email protected]> Mon, 12 Sep 2022 14:43:45 -0500
Newsgroups gmane.linux.procps.devel
Message-ID <[email protected]>
On 9/12/22 7:35 AM, Craig Small (csmall) wrote:
> My desktop has 12 cores so that I can get 8 across, nice!
> 
> My question is how the memory is displayed. It's more of a style thing 
> than anything significant.
> I have a monitor with strange dimensions, 282 columns by 55 lines. So 
> the CPUs nicely stretch across the 282 columns 8-across.
> The memory when it's just numbers seems to stay firmly in 80-column land.
> The graphs get squished up like the CPUs, even though there is real 
> estate to the right of them.

Hi Craig,

Thanks for the prompt push.

I may have you bested.  I recently invested in a couple of new AMD 
platforms with 8 cores and 16 threads.  That's what provided the impetus 
for this top extension.

[ and when I maximize an rxvt window I get 317 columns and 77 rows ]

Anyway, that memory approach was a conscious design decision to keep the 
memory graphs aligned with the first two cpu columns.  I've now 
redesigned things to maximize the graph widths and minimize data loss. 
They'll now behave as they did prior to the '4' toggle extension.

Here is what I've attached:

1) The most recent top newlib patches ported to the master branch.  The 
library patch wasn't applicable and the old library is frozen anyway.

2) Two patches for newlib.  The first is a collection of miscellany that 
definitely should be pushed.  The second is the new (optional?) memory 
implementation.  I can live with whatever you decide to do with that one.

Regards,

Jim

p.s. I'll send any additional master catch-up patches after the newlib 
hash is known.
master-catchup-for-8.tgz (application/x-compressed-tar, 11.2 KB) - not displayed
newlib-2-more.tgz (application/x-compressed-tar, 4.1 KB) - not displayed