* [RFA] Split Ipipe.exp test into seperate versions.
@ 2003-09-03 18:38 Michael Snyder
2003-09-03 19:03 ` Dave Brolley
0 siblings, 1 reply; 5+ messages in thread
From: Michael Snyder @ 2003-09-03 18:38 UTC (permalink / raw)
To: gdb-patches; +Cc: Dave Brolley
[-- Attachment #1: Type: text/plain, Size: 290 bytes --]
The fr400 wasn't really being tested, because of a typo.
When I fixed the typo, the fr400 failed the test, because
of an LSI difference (it places the address of the VLIW
bundle into EPCR0, rather than the address of the insn
causing the interrupt). So I've split the test case
into two.
[-- Attachment #2: Ipipe --]
[-- Type: text/plain, Size: 2154 bytes --]
2003-09-03 Michael Snyder <msnyder@redhat.com>
* sim/frv/interrupts/Ipipe-fr400.cgs: New file.
* sim/frv/interrupts/Ipipe-fr500.cgs: New file.
* sim/frv/interrupts/Ipipe.cgs: Remove (replaced by above).
Index: sim/frv/interrupts/Ipipe-fr400.cgs
===================================================================
RCS file: sim/frv/interrupts/Ipipe-fr400.cgs
diff -N sim/frv/interrupts/Ipipe-fr400.cgs
0a1,35
> # venus testcase
> # mach: fr400
>
> .include "testutils.inc"
>
> start
>
> .global Ipipe
> Ipipe:
> ; Clear the packing bit of the insn at 'pack:'. We can't
> ; simply use '.p' because the assembler will catch the error.
> set_gr_mem pack,gr10
> and_gr_immed 0x7fffffff,gr10
> set_mem_gr gr10,pack
> set_gr_addr pack,gr10
> flush_data_cache gr10
>
> and_spr_immed -4081,tbr ; clear tbr.tt
> set_gr_spr tbr,gr7
> inc_gr_immed 0x070,gr7 ; address of exception handler
> set_bctrlr_0_0 gr7
> set_spr_immed 128,lcr
> set_spr_addr ok0,lr
> set_psr_et 1
>
> bundle: add.p gr1,gr1,gr1
> pack: add gr2,gr2,gr2
> bad: add gr3,gr3,gr3
> fail
> ok0:
> test_spr_immed 1,esfr1
> test_spr_bits 0x3f,0,0xb,esr0
> test_spr_addr bundle,epcr0
>
> pass
Index: sim/frv/interrupts/Ipipe-fr500.cgs
===================================================================
RCS file: sim/frv/interrupts/Ipipe-fr500.cgs
diff -N sim/frv/interrupts/Ipipe-fr500.cgs
0a1,35
> # venus testcase
> # mach: fr500
>
> .include "testutils.inc"
>
> start
>
> .global Ipipe
> Ipipe:
> ; Clear the packing bit of the insn at 'pack:'. We can't
> ; simply use '.p' because the assembler will catch the error.
> set_gr_mem pack,gr10
> and_gr_immed 0x7fffffff,gr10
> set_mem_gr gr10,pack
> set_gr_addr pack,gr10
> flush_data_cache gr10
>
> and_spr_immed -4081,tbr ; clear tbr.tt
> set_gr_spr tbr,gr7
> inc_gr_immed 0x070,gr7 ; address of exception handler
> set_bctrlr_0_0 gr7
> set_spr_immed 128,lcr
> set_spr_addr ok0,lr
> set_psr_et 1
>
> add.p gr1,gr1,gr1
> pack: add gr2,gr2,gr2
> bad: add gr3,gr3,gr3
> fail
> ok0:
> test_spr_immed 1,esfr1
> test_spr_bits 0x3f,0,0xb,esr0
> test_spr_addr bad,epcr0
>
> pass
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [RFA] Split Ipipe.exp test into seperate versions.
2003-09-03 18:38 [RFA] Split Ipipe.exp test into seperate versions Michael Snyder
@ 2003-09-03 19:03 ` Dave Brolley
2003-09-03 21:52 ` Michael Snyder
0 siblings, 1 reply; 5+ messages in thread
From: Dave Brolley @ 2003-09-03 19:03 UTC (permalink / raw)
To: Michael Snyder; +Cc: gdb-patches
One minor change -- replace "venus" with "frv". Other than that, it's
approved to commit.
Dave
Michael Snyder wrote:
> The fr400 wasn't really being tested, because of a typo.
> When I fixed the typo, the fr400 failed the test, because
> of an LSI difference (it places the address of the VLIW
> bundle into EPCR0, rather than the address of the insn
> causing the interrupt). So I've split the test case
> into two.
>
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [RFA] Split Ipipe.exp test into seperate versions.
2003-09-03 19:03 ` Dave Brolley
@ 2003-09-03 21:52 ` Michael Snyder
2003-09-03 22:06 ` Andrew Cagney
0 siblings, 1 reply; 5+ messages in thread
From: Michael Snyder @ 2003-09-03 21:52 UTC (permalink / raw)
To: Dave Brolley; +Cc: gdb-patches
Dave Brolley wrote:
> One minor change -- replace "venus" with "frv". Other than that, it's
> approved to commit.
Changed and committed.
> Michael Snyder wrote:
>
>> The fr400 wasn't really being tested, because of a typo.
>> When I fixed the typo, the fr400 failed the test, because
>> of an LSI difference (it places the address of the VLIW
>> bundle into EPCR0, rather than the address of the insn
>> causing the interrupt). So I've split the test case
>> into two.
>>
>
>
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [RFA] Split Ipipe.exp test into seperate versions.
2003-09-03 21:52 ` Michael Snyder
@ 2003-09-03 22:06 ` Andrew Cagney
2003-09-03 22:36 ` Michael Snyder
0 siblings, 1 reply; 5+ messages in thread
From: Andrew Cagney @ 2003-09-03 22:06 UTC (permalink / raw)
To: Michael Snyder, Dave Brolley; +Cc: gdb-patches
> Dave Brolley wrote:
> One minor change -- replace "venus" with "frv". Other than that, it's approved to commit.
>
> Changed and committed.
>
> Michael Snyder wrote:
>
> The fr400 wasn't really being tested, because of a typo.
> When I fixed the typo, the fr400 failed the test, because
> of an LSI difference (it places the address of the VLIW
> bundle into EPCR0, rather than the address of the insn
> causing the interrupt). So I've split the test case
> into two.
FYI, the new sim/frv is having 8.3 problems. See:
http://sources.redhat.com/current/ari/
fnchange.lst needs a tweak.
Andrew
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [RFA] Split Ipipe.exp test into seperate versions.
2003-09-03 22:06 ` Andrew Cagney
@ 2003-09-03 22:36 ` Michael Snyder
0 siblings, 0 replies; 5+ messages in thread
From: Michael Snyder @ 2003-09-03 22:36 UTC (permalink / raw)
To: Andrew Cagney; +Cc: Dave Brolley, gdb-patches
Andrew Cagney wrote:
>> Dave Brolley wrote:
>> One minor change -- replace "venus" with "frv". Other than that, it's
>> approved to commit.
>>
>> Changed and committed.
>>
>> Michael Snyder wrote:
>>
>> The fr400 wasn't really being tested, because of a typo.
>> When I fixed the typo, the fr400 failed the test, because
>> of an LSI difference (it places the address of the VLIW
>> bundle into EPCR0, rather than the address of the insn
>> causing the interrupt). So I've split the test case
>> into two.
>
>
> FYI, the new sim/frv is having 8.3 problems. See:
> http://sources.redhat.com/current/ari/
> fnchange.lst needs a tweak.
Tweaked:
2003-09-03 Michael Snyder <msnyder@redhat.com>
* config/djgpp/fnchange.lst: Fix up sim/frv/profile-fr*.[ch].
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2003-09-03 22:36 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2003-09-03 18:38 [RFA] Split Ipipe.exp test into seperate versions Michael Snyder
2003-09-03 19:03 ` Dave Brolley
2003-09-03 21:52 ` Michael Snyder
2003-09-03 22:06 ` Andrew Cagney
2003-09-03 22:36 ` Michael Snyder
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox