Skip to contentSkip to conversation

re-enabling interrupts and interrupt collect

Linux-ia64 Chuck Fleckenstein 4 participants 4 messages Nov 8, 2000

lore
Share & tools
lore
#0 lore raw

Since the ssm of psr.i does not require serialization then I believe this would
be expected.

so as you said the guaranteed method of safely reenabling ic and i would be:

    ssm psr.ic
    ;;
    srlz.d
    ssm psr.i

Chuck

Quoted 46 lines · Jonathan Case Nicklin

All,

Recently, i was working on a section of code that disabled/enable interrupts
and interrupt collection like so.

    rsm psr.i | psr.ic
    ;;
    ... //body
    ;;
    ssm psr.i | psr.ic
    ;;
    srlz.d
    ;;

I found however that this implementation did not work under heavy loads.

It took a while to figure out that a pending interrupt that had occurred

in the body of code, executed while interrupts were turned off, was delivered
after the ssm call as expected. However, it was delivered before the ic bit
serialized. In the code this caused undesirable results. I found that the
proper way to implement the above is to re-enable the ic bit and serialize
before re-enabling the interrupt bit. Has anyone else come across this problem and can anybody shed some light on whether this is the expected operation.

Sincerely,
Jonathan Case Nicklin

Mission Critical Linux
www.missioncriticallinux.com

ps. The manual gives an example of the implementation that works, but does
not provide any explanation of implementation itself (as far as I can see ;-P )

_______________________________________________
Linux-IA64 mailing list
[email protected]
http://lists.linuxia64.org/lists/listinfo/linux-ia64

#1 lore raw

All,

Recently, i was working on a section of code that disabled/enable interrupts
and interrupt collection like so.

    rsm psr.i | psr.ic
    ;;
    ... //body
    ;;
    ssm psr.i | psr.ic
    ;;
    srlz.d
    ;;

I found however that this implementation did not work under heavy loads.

It took a while to figure out that a pending interrupt that had occurred

in the body of code, executed while interrupts were turned off, was delivered
after the ssm call as expected. However, it was delivered before the ic bit
serialized. In the code this caused undesirable results. I found that the
proper way to implement the above is to re-enable the ic bit and serialize
before re-enabling the interrupt bit. Has anyone else come across this problem and can anybody shed some light on whether this is the expected operation.

Sincerely,
Jonathan Case Nicklin

Mission Critical Linux
www.missioncriticallinux.com

ps. The manual gives an example of the implementation that works, but does
not provide any explanation of implementation itself (as far as I can see ;-P )

#2 lore raw
In reply to Chuck Fleckenstein

Yes, psr.ic needs to be on before psr.i can be turned on. That's what we are doing in ivt.S, for example. I thought the manual was quite clear on that topic.

	--david
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line · Jonathan Case Nicklin

Jonathan> All, Recently, i was working on a section of code that Jonathan> disabled/enable interrupts and interrupt collection like Jonathan> so.

Jonathan> rsm psr.i | psr.ic ;; ... //body ;; ssm psr.i | psr.ic
Jonathan> ;; srlz.d ;;

Jonathan> I found however that this implementation did not work Jonathan> under heavy loads.

Jonathan> It took a while to figure out that a pending interrupt Jonathan> that had occurred in the body of code, executed while Jonathan> interrupts were turned off, was delivered after the ssm Jonathan> call as expected. However, it was delivered before the ic Jonathan> bit serialized. In the code this caused undesirable Jonathan> results. I found that the proper way to implement the Jonathan> above is to re-enable the ic bit and serialize before Jonathan> re-enabling the interrupt bit. Has anyone else come across Jonathan> this problem and can anybody shed some light on whether Jonathan> this is the expected operation.

Jonathan> Sincerely, Jonathan Case Nicklin

Jonathan> Mission Critical Linux www.missioncriticallinux.com

Jonathan> ps. The manual gives an example of the implementation that Jonathan> works, but does not provide any explanation of Jonathan> implementation itself (as far as I can see ;-P )

#3 lore raw
In reply to Chuck Fleckenstein

Setting the PSR.ic or PSR.i is not implicitly serialized and setting both PSR.ic and PSR.i in a single instruction could create a timing window where PSR.ic=0 and PSR.i=1. So, software must set the PSR.ic explicitly before setting the PSR.i.

Asit

Quoted 89 lines

-----Original Message-----
From: Chuck Fleckenstein [mailto:[email protected]]
Sent: Wednesday, November 08, 2000 2:07 AM
To: Jonathan Case Nicklin
Cc: [email protected]
Subject: Re: [Linux-ia64] re-enabling interrupts and interrupt collect

Since the ssm of psr.i does not require serialization then I believe this
would
be expected.

so as you said the guaranteed method of safely reenabling ic and i would
be:

    ssm psr.ic
    ;;
    srlz.d
    ssm psr.i

Chuck

Quoted 15 lines · Jonathan Case Nicklin

All,

Recently, i was working on a section of code that disabled/enable interrupts
and interrupt collection like so.

    rsm psr.i | psr.ic
    ;;
    ... //body
    ;;
    ssm psr.i | psr.ic
    ;;
    srlz.d
    ;;
Quoted 1 line

I found however that this implementation did not work under

heavy loads.

Quoted 2 lines

It took a while to figure out that a pending interrupt that

had occurred

Quoted 4 lines

in the body of code, executed while interrupts were turned off, was delivered
after the ssm call as expected. However, it was delivered

before the ic

Quoted 2 lines

bit
serialized. In the code this caused undesirable results. I

found that

Quoted 4 lines

the
proper way to implement the above is to re-enable the ic bit and serialize
before re-enabling the interrupt bit. Has anyone else come

across this

Quoted 1 line

problem and can anybody shed some light on whether this is

the expected

Quoted 9 lines

operation.

Sincerely,
Jonathan Case Nicklin

Mission Critical Linux
www.missioncriticallinux.com

ps. The manual gives an example of the implementation that

works, but

Quoted 2 lines

does
not provide any explanation of implementation itself (as

far as I can

Quoted 6 lines

see ;-P )

_______________________________________________
Linux-IA64 mailing list
[email protected]
http://lists.linuxia64.org/lists/listinfo/linux-ia64

_______________________________________________
Linux-IA64 mailing list
[email protected]
http://lists.linuxia64.org/lists/listinfo/linux-ia64