On 06/26/14 12:37 PM, Johan Tibell wrote:

I think it's reasonable to say my change triggers the issue, but I don't know why and I need someone who understand the register allocator better (e.g. Simon M) to at least give me some pointers on how to debug this. If this is causing more trouble than build bot breakages (which is bad enough), let me know and I'll either revert temporarily and or try to work around the issue.

it does not hurt me as long as it does not go to 7.10 release so it's enough time to fix that. So feel free not to revert your patch especially if you do have some time to look into real culprit in regs alloc.

Also, how do I repro this if I don't have a x86 machine lying around? Have people been successful working on these kind of issues in a VM?

My simplest way is to install VirtualBox and inside it i386-ubuntu, give it 2 GB RAM, 15GB drive and this should get you going really quickly. Then you will need to: sudo apt-get install gcc ghc make git (in ubuntu) which should be enough and then just follow ghc's wiki to obtain source tree and bootstrap on linux... Roughly it should be that IMHO. If not, I may do that myself and send you exact stepts to follow... Thanks! Karel

On Thu, Jun 26, 2014 at 10:55 AM, Karel Gardas mailto:karel.gardas@centrum.cz> wrote:

Hello,

builders running on i386 building for this platform caught issue which shows as a build breakage:

ghc-stage1: panic! (the 'impossible' happened) (GHC version 7.9.20140624 for i386-unknown-linux): RegAllocLinear.__allocRegsAndSpill: no spill candidates allocating vreg: VirtualRegI n1Q6 assignment: [(c1PV,InMem 2),(n1Q5,InBoth (RealRegSingle 3) 0),(n1Q6,InMem 1),(n1Q7,InMem 3),(n1Q9,InReg (RealRegSingle 2))] freeRegs: FreeRegs 4282318848 initFreeRegs: FreeRegs 4282318861 Please report this as a GHC bug: http://www.haskell.org/ghc/__reportabug http://www.haskell.org/ghc/reportabug make[1]: *** [libraries/ghc-prim/dist-__install/build/GHC/__PrimopWrappers.o] Error 1 libraries/ghc-prim/ghc.mk:4 http://ghc.mk:4: recipe for target 'libraries/ghc-prim/dist-__install/build/GHC/__PrimopWrappers.o' failed

Have a look for example on linux-i386 buildot log here: http://haskell.inf.elte.hu/__builders/validator1-linux-x86-__head/18/7.html http://haskell.inf.elte.hu/builders/validator1-linux-x86-head/18/7.html

Anyway, this happens on Linux, FreeBSD and Solaris buildbots on i386 so it's OS independent and probably 32bit/i386 platform specific and it's two days old breakage now. The last two night builds fail on all mentioned buildbots. I'm not sure but perhaps:

commit d8abf85f8ca176854e9d5d0b12371c__4bc402aac3 Author: Johan Tibell mailto:johan.tibell@gmail.com> Date: Mon Jun 9 11:43:21 2014 +0200

triggers that issue? I'm not claiming that the commit is actual culprit, this may be just recently un-hidden issue in linear regs allocator on i386!

Thanks! Karel

On 06/26/2014 12:37 PM, Johan Tibell wrote:

I think it's reasonable to say my change triggers the issue, but I don't know why and I need someone who understand the register allocator better (e.g. Simon M) to at least give me some pointers on how to debug this. If this is causing more trouble than build bot breakages (which is bad enough), let me know and I'll either revert temporarily and or try to work around the issue.

Also, how do I repro this if I don't have a x86 machine lying around? Have people been successful working on these kind of issues in a VM?

I can give you shell access to the machine in the OP but it's not the fastest thing in the world (3GHz Core 2 Duo). Let me know if you want it.

On Thu, Jun 26, 2014 at 10:55 AM, Karel Gardas wrote:

...

Hello,

builders running on i386 building for this platform caught issue which shows as a build breakage:

ghc-stage1: panic! (the 'impossible' happened) (GHC version 7.9.20140624 for i386-unknown-linux): RegAllocLinear.allocRegsAndSpill: no spill candidates allocating vreg: VirtualRegI n1Q6 assignment: [(c1PV,InMem 2),(n1Q5,InBoth (RealRegSingle 3) 0),(n1Q6,InMem 1),(n1Q7,InMem 3),(n1Q9,InReg (RealRegSingle 2))] freeRegs: FreeRegs 4282318848 initFreeRegs: FreeRegs 4282318861 Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug make[1]: *** [libraries/ghc-prim/dist-install/build/GHC/PrimopWrappers.o] Error 1 libraries/ghc-prim/ghc.mk:4: recipe for target 'libraries/ghc-prim/dist- install/build/GHC/PrimopWrappers.o' failed

Have a look for example on linux-i386 buildot log here: http://haskell.inf.elte.hu/builders/validator1-linux-x86-head/18/7.html

Anyway, this happens on Linux, FreeBSD and Solaris buildbots on i386 so it's OS independent and probably 32bit/i386 platform specific and it's two days old breakage now. The last two night builds fail on all mentioned buildbots. I'm not sure but perhaps:

commit d8abf85f8ca176854e9d5d0b12371c4bc402aac3 Author: Johan Tibell Date: Mon Jun 9 11:43:21 2014 +0200

triggers that issue? I'm not claiming that the commit is actual culprit, this may be just recently un-hidden issue in linear regs allocator on i386!

Thanks! Karel

_______________________________________________ ghc-devs mailing list ghc-devs@haskell.org http://www.haskell.org/mailman/listinfo/ghc-devs

-- Mateusz K.

Thanks. It’s blocking me from building GHC at all on my laptop, so temporary reversion would be great Thank you! Simon From: ghc-devs [mailto:ghc-devs-bounces@haskell.org] On Behalf Of Johan Tibell Sent: 26 June 2014 11:38 To: Karel Gardas Cc: Simon Marlow; ghc-devs Subject: Re: Two days old build breakage on i386. I think it's reasonable to say my change triggers the issue, but I don't know why and I need someone who understand the register allocator better (e.g. Simon M) to at least give me some pointers on how to debug this. If this is causing more trouble than build bot breakages (which is bad enough), let me know and I'll either revert temporarily and or try to work around the issue. Also, how do I repro this if I don't have a x86 machine lying around? Have people been successful working on these kind of issues in a VM? On Thu, Jun 26, 2014 at 10:55 AM, Karel Gardas mailto:karel.gardas@centrum.cz> wrote: Hello, builders running on i386 building for this platform caught issue which shows as a build breakage: ghc-stage1: panic! (the 'impossible' happened) (GHC version 7.9.20140624 for i386-unknown-linux): RegAllocLinear.allocRegsAndSpill: no spill candidates allocating vreg: VirtualRegI n1Q6 assignment: [(c1PV,InMem 2),(n1Q5,InBoth (RealRegSingle 3) 0),(n1Q6,InMem 1),(n1Q7,InMem 3),(n1Q9,InReg (RealRegSingle 2))] freeRegs: FreeRegs 4282318848 initFreeRegs: FreeRegs 4282318861 Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug make[1]: *** [libraries/ghc-prim/dist-install/build/GHC/PrimopWrappers.o] Error 1 libraries/ghc-prim/ghc.mk:4http://ghc.mk:4: recipe for target 'libraries/ghc-prim/dist-install/build/GHC/PrimopWrappers.o' failed Have a look for example on linux-i386 buildot log here: http://haskell.inf.elte.hu/builders/validator1-linux-x86-head/18/7.html Anyway, this happens on Linux, FreeBSD and Solaris buildbots on i386 so it's OS independent and probably 32bit/i386 platform specific and it's two days old breakage now. The last two night builds fail on all mentioned buildbots. I'm not sure but perhaps: commit d8abf85f8ca176854e9d5d0b12371c4bc402aac3 Author: Johan Tibell mailto:johan.tibell@gmail.com> Date: Mon Jun 9 11:43:21 2014 +0200 triggers that issue? I'm not claiming that the commit is actual culprit, this may be just recently un-hidden issue in linear regs allocator on i386! Thanks! Karel

Just to make sure this is the same breakage, are you on an i386 Windows machine? If so git revert d8abf85f8ca176854e9d5d0b12371c4bc402aac3 and 04dd7cb3423f1940242fdfe2ea2e3b8abd68a177 to get unstuck. On Thu, Jun 26, 2014 at 2:13 PM, Simon Peyton Jones wrote:

Aaaargh! Once again the Windows build is broken. I am utterly stalled.

Moreover -fregs-graph and -fregs-iterative now *silently* do nothing. At least they should elicit warnings saying that they are disabled pending the fix to X and Y.

Please can someone bisect to find out which patch is the culprit?

I wish we had a more systematic way to find this out. I hate being the main person who gets stuck because some unrelated change has broken the Windows build. (Thanks for Karel, who got to it a day before me.)

Thanks

Simon

| -----Original Message----- | From: ghc-devs [mailto:ghc-devs-bounces@haskell.org] On Behalf Of Karel | Gardas | Sent: 26 June 2014 09:56 | To: ghc-devs; Johan Tibell | Subject: Two days old build breakage on i386. | | | Hello, | | builders running on i386 building for this platform caught issue which | shows as a build breakage: | | ghc-stage1: panic! (the 'impossible' happened) | (GHC version 7.9.20140624 for i386-unknown-linux): | RegAllocLinear.allocRegsAndSpill: no spill candidates | allocating vreg: VirtualRegI n1Q6 | assignment: [(c1PV,InMem 2),(n1Q5,InBoth (RealRegSingle 3) | 0),(n1Q6,InMem 1),(n1Q7,InMem 3),(n1Q9,InReg (RealRegSingle 2))] | freeRegs: FreeRegs 4282318848 | initFreeRegs: FreeRegs 4282318861 | Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug | make[1]: *** | [libraries/ghc-prim/dist-install/build/GHC/PrimopWrappers.o] Error 1 | libraries/ghc-prim/ghc.mk:4: recipe for target | 'libraries/ghc-prim/dist-install/build/GHC/PrimopWrappers.o' failed | | Have a look for example on linux-i386 buildot log here: | http://haskell.inf.elte.hu/builders/validator1-linux-x86-head/18/7.html | | Anyway, this happens on Linux, FreeBSD and Solaris buildbots on i386 so | it's OS independent and probably 32bit/i386 platform specific and it's | two days old breakage now. The last two night builds fail on all | mentioned buildbots. I'm not sure but perhaps: | | commit d8abf85f8ca176854e9d5d0b12371c4bc402aac3 | Author: Johan Tibell | Date: Mon Jun 9 11:43:21 2014 +0200 | | triggers that issue? I'm not claiming that the commit is actual culprit, | this may be just recently un-hidden issue in linear regs allocator on | i386! | | Thanks! | Karel | _______________________________________________ | ghc-devs mailing list | ghc-devs@haskell.org | http://www.haskell.org/mailman/listinfo/ghc-devs

git revert d8abf85f8ca176854e9d5d0b12371c4bc402aac3 [master f958079] Revert "Add more primops for atomic ops on byte arrays" 23 files changed, 86 insertions(+), 1016 deletions(-) rewrite compiler/nativeGen/CPrim.hs (62%) delete mode 100644 libraries/ghc-prim/cbits/atomic.c delete mode 100644 testsuite/tests/concurrent/should_run/AtomicPrimops.hs delete mode 100644 testsuite/tests/concurrent/should_run/AtomicPrimops.stdout HEAD $ git revert 04dd7cb3423f1940242fdfe2ea2e3b8abd68a177 fatal: bad object 04dd7cb3423f1940242fdfe2ea2e3b8abd68a177 HEAD $ What now? Simon From: Johan Tibell [mailto:johan.tibell@gmail.com] Sent: 26 June 2014 13:25 To: Simon Peyton Jones Cc: Karel Gardas; ghc-devs Subject: Re: Two days old build breakage on i386. Just to make sure this is the same breakage, are you on an i386 Windows machine? If so git revert d8abf85f8ca176854e9d5d0b12371c4bc402aac3 and 04dd7cb3423f1940242fdfe2ea2e3b8abd68a177 to get unstuck. On Thu, Jun 26, 2014 at 2:13 PM, Simon Peyton Jones mailto:simonpj@microsoft.com> wrote: Aaaargh! Once again the Windows build is broken. I am utterly stalled. Moreover -fregs-graph and -fregs-iterative now *silently* do nothing. At least they should elicit warnings saying that they are disabled pending the fix to X and Y. Please can someone bisect to find out which patch is the culprit? I wish we had a more systematic way to find this out. I hate being the main person who gets stuck because some unrelated change has broken the Windows build. (Thanks for Karel, who got to it a day before me.) Thanks Simon | -----Original Message----- | From: ghc-devs [mailto:ghc-devs-bounces@haskell.orgmailto:ghc-devs-bounces@haskell.org] On Behalf Of Karel | Gardas | Sent: 26 June 2014 09:56 | To: ghc-devs; Johan Tibell | Subject: Two days old build breakage on i386. | | | Hello, | | builders running on i386 building for this platform caught issue which | shows as a build breakage: | | ghc-stage1: panic! (the 'impossible' happened) | (GHC version 7.9.20140624 for i386-unknown-linux): | RegAllocLinear.allocRegsAndSpill: no spill candidates | allocating vreg: VirtualRegI n1Q6 | assignment: [(c1PV,InMem 2),(n1Q5,InBoth (RealRegSingle 3) | 0),(n1Q6,InMem 1),(n1Q7,InMem 3),(n1Q9,InReg (RealRegSingle 2))] | freeRegs: FreeRegs 4282318848 | initFreeRegs: FreeRegs 4282318861 | Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug | make[1]: *** | [libraries/ghc-prim/dist-install/build/GHC/PrimopWrappers.o] Error 1 | libraries/ghc-prim/ghc.mk:4http://ghc.mk:4: recipe for target | 'libraries/ghc-prim/dist-install/build/GHC/PrimopWrappers.o' failed | | Have a look for example on linux-i386 buildot log here: | http://haskell.inf.elte.hu/builders/validator1-linux-x86-head/18/7.html | | Anyway, this happens on Linux, FreeBSD and Solaris buildbots on i386 so | it's OS independent and probably 32bit/i386 platform specific and it's | two days old breakage now. The last two night builds fail on all | mentioned buildbots. I'm not sure but perhaps: | | commit d8abf85f8ca176854e9d5d0b12371c4bc402aac3 | Author: Johan Tibell mailto:johan.tibell@gmail.com> | Date: Mon Jun 9 11:43:21 2014 +0200 | | triggers that issue? I'm not claiming that the commit is actual culprit, | this may be just recently un-hidden issue in linear regs allocator on | i386! | | Thanks! | Karel | _______________________________________________ | ghc-devs mailing list | ghc-devs@haskell.orgmailto:ghc-devs@haskell.org | http://www.haskell.org/mailman/listinfo/ghc-devs

Herbert pushed my revert for me a minute ago. Everyone should be good once they sync. On Thu, Jun 26, 2014 at 2:51 PM, Johan Tibell wrote:

I guess you don't have 04dd7cb3423f1940242fdfe2ea2e3b8abd68a177 (which I pushed this morning) which is fine. You should be in a good state now when d8abf85f8ca176854e9d5d0b12371c4bc402aac3 is reverted.

On Thu, Jun 26, 2014 at 2:49 PM, Simon Peyton Jones

...

wrote:

...

git revert d8abf85f8ca176854e9d5d0b12371c4bc402aac3

[master f958079] Revert "Add more primops for atomic ops on byte arrays"

23 files changed, 86 insertions(+), 1016 deletions(-)

rewrite compiler/nativeGen/CPrim.hs (62%)

delete mode 100644 libraries/ghc-prim/cbits/atomic.c

delete mode 100644 testsuite/tests/concurrent/should_run/AtomicPrimops.hs

delete mode 100644 testsuite/tests/concurrent/should_run/AtomicPrimops.stdout

HEAD $ git revert 04dd7cb3423f1940242fdfe2ea2e3b8abd68a177

fatal: bad object 04dd7cb3423f1940242fdfe2ea2e3b8abd68a177

HEAD $

What now?

Simon

*From:* Johan Tibell [mailto:johan.tibell@gmail.com] *Sent:* 26 June 2014 13:25 *To:* Simon Peyton Jones *Cc:* Karel Gardas; ghc-devs *Subject:* Re: Two days old build breakage on i386.

Just to make sure this is the same breakage, are you on an i386 Windows machine? If so git revert d8abf85f8ca176854e9d5d0b12371c4bc402aac3 and 04dd7cb3423f1940242fdfe2ea2e3b8abd68a177 to get unstuck.

On Thu, Jun 26, 2014 at 2:13 PM, Simon Peyton Jones < simonpj@microsoft.com> wrote:

Aaaargh! Once again the Windows build is broken. I am utterly stalled.

Moreover -fregs-graph and -fregs-iterative now *silently* do nothing. At least they should elicit warnings saying that they are disabled pending the fix to X and Y.

Please can someone bisect to find out which patch is the culprit?

I wish we had a more systematic way to find this out. I hate being the main person who gets stuck because some unrelated change has broken the Windows build. (Thanks for Karel, who got to it a day before me.)

Thanks

Simon

| -----Original Message----- | From: ghc-devs [mailto:ghc-devs-bounces@haskell.org] On Behalf Of Karel | Gardas | Sent: 26 June 2014 09:56 | To: ghc-devs; Johan Tibell | Subject: Two days old build breakage on i386. | | | Hello, | | builders running on i386 building for this platform caught issue which | shows as a build breakage: | | ghc-stage1: panic! (the 'impossible' happened) | (GHC version 7.9.20140624 for i386-unknown-linux): | RegAllocLinear.allocRegsAndSpill: no spill candidates | allocating vreg: VirtualRegI n1Q6 | assignment: [(c1PV,InMem 2),(n1Q5,InBoth (RealRegSingle 3) | 0),(n1Q6,InMem 1),(n1Q7,InMem 3),(n1Q9,InReg (RealRegSingle 2))] | freeRegs: FreeRegs 4282318848 | initFreeRegs: FreeRegs 4282318861 | Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug | make[1]: *** | [libraries/ghc-prim/dist-install/build/GHC/PrimopWrappers.o] Error 1 | libraries/ghc-prim/ghc.mk:4: recipe for target | 'libraries/ghc-prim/dist-install/build/GHC/PrimopWrappers.o' failed | | Have a look for example on linux-i386 buildot log here: | http://haskell.inf.elte.hu/builders/validator1-linux-x86-head/18/7.html | | Anyway, this happens on Linux, FreeBSD and Solaris buildbots on i386 so | it's OS independent and probably 32bit/i386 platform specific and it's | two days old breakage now. The last two night builds fail on all | mentioned buildbots. I'm not sure but perhaps: | | commit d8abf85f8ca176854e9d5d0b12371c4bc402aac3 | Author: Johan Tibell | Date: Mon Jun 9 11:43:21 2014 +0200 | | triggers that issue? I'm not claiming that the commit is actual culprit, | this may be just recently un-hidden issue in linear regs allocator on | i386! | | Thanks! | Karel

| _______________________________________________ | ghc-devs mailing list | ghc-devs@haskell.org | http://www.haskell.org/mailman/listinfo/ghc-devs

Here's some more debug output. Can someone interpret it: genRaInsn cmpxchgl %vI_n1nF,8(%vI_n1nD,%vI_n1nE,4) r_dying = [%vI_n1nD, %vI_n1nE, %vI_n1nF] w_dying = [] virt_read = [%vI_n1nD, %vI_n1nE, %vI_n1nF] virt_written = [] freeregs = FreeRegs 4282318848 assig = [n1nD :-> InMem 0, n1nE :-> InReg (RealRegSingle 2), n1nF :-> InReg (RealRegSingle 3)] ghc-stage1: panic! (the 'impossible' happened) (GHC version 7.9.20140626 for i386-unknown-linux): RegAllocLinear.allocRegsAndSpill: no spill candidates allocating vreg: VirtualRegI n1nD assignment: [(n1nD,InMem 0),(n1nE,InReg (RealRegSingle 2)),(n1nF,InReg (RealRegSingle 3))] freeRegs: FreeRegs 4282318848 initFreeRegs: FreeRegs 4282318861 (i.e. it's the cmpxchg instruction which is causing the failure.) On Thu, Jun 26, 2014 at 8:17 PM, Johan Tibell wrote:

I'm trying to understand the output from the register allocator:

(GHC version 7.9.20140626 for i386-unknown-linux): RegAllocLinear.allocRegsAndSpill: no spill candidates

allocating vreg: VirtualRegI n1nD assignment: [(n1nD,InMem 0),(n1nE,InReg (RealRegSingle 2)),(n1nF,InReg (RealRegSingle 3))] freeRegs: FreeRegs 4282318848 initFreeRegs: FreeRegs 4282318861

Without understanding exactly what's wrong, the message above suggests that we're trying to allocate a reg for n1nD, but there's already an assignment for that virtual reg, is that right?

On Thu, Jun 26, 2014 at 3:05 PM, Johan Tibell wrote:

...

Herbert pushed my revert for me a minute ago. Everyone should be good once they sync.

On Thu, Jun 26, 2014 at 2:51 PM, Johan Tibell wrote:

...

I guess you don't have 04dd7cb3423f1940242fdfe2ea2e3b8abd68a177 (which I pushed this morning) which is fine. You should be in a good state now when d8abf85f8ca176854e9d5d0b12371c4bc402aac3 is reverted.

On Thu, Jun 26, 2014 at 2:49 PM, Simon Peyton Jones < simonpj@microsoft.com> wrote:

...

git revert d8abf85f8ca176854e9d5d0b12371c4bc402aac3

[master f958079] Revert "Add more primops for atomic ops on byte arrays"

23 files changed, 86 insertions(+), 1016 deletions(-)

rewrite compiler/nativeGen/CPrim.hs (62%)

delete mode 100644 libraries/ghc-prim/cbits/atomic.c

delete mode 100644 testsuite/tests/concurrent/should_run/AtomicPrimops.hs

delete mode 100644 testsuite/tests/concurrent/should_run/AtomicPrimops.stdout

HEAD $ git revert 04dd7cb3423f1940242fdfe2ea2e3b8abd68a177

fatal: bad object 04dd7cb3423f1940242fdfe2ea2e3b8abd68a177

HEAD $

What now?

Simon

*From:* Johan Tibell [mailto:johan.tibell@gmail.com] *Sent:* 26 June 2014 13:25 *To:* Simon Peyton Jones *Cc:* Karel Gardas; ghc-devs *Subject:* Re: Two days old build breakage on i386.

Just to make sure this is the same breakage, are you on an i386 Windows machine? If so git revert d8abf85f8ca176854e9d5d0b12371c4bc402aac3 and 04dd7cb3423f1940242fdfe2ea2e3b8abd68a177 to get unstuck.

On Thu, Jun 26, 2014 at 2:13 PM, Simon Peyton Jones < simonpj@microsoft.com> wrote:

Aaaargh! Once again the Windows build is broken. I am utterly stalled.

Moreover -fregs-graph and -fregs-iterative now *silently* do nothing. At least they should elicit warnings saying that they are disabled pending the fix to X and Y.

Please can someone bisect to find out which patch is the culprit?

I wish we had a more systematic way to find this out. I hate being the main person who gets stuck because some unrelated change has broken the Windows build. (Thanks for Karel, who got to it a day before me.)

Thanks

Simon

| -----Original Message----- | From: ghc-devs [mailto:ghc-devs-bounces@haskell.org] On Behalf Of Karel | Gardas | Sent: 26 June 2014 09:56 | To: ghc-devs; Johan Tibell | Subject: Two days old build breakage on i386. | | | Hello, | | builders running on i386 building for this platform caught issue which | shows as a build breakage: | | ghc-stage1: panic! (the 'impossible' happened) | (GHC version 7.9.20140624 for i386-unknown-linux): | RegAllocLinear.allocRegsAndSpill: no spill candidates | allocating vreg: VirtualRegI n1Q6 | assignment: [(c1PV,InMem 2),(n1Q5,InBoth (RealRegSingle 3) | 0),(n1Q6,InMem 1),(n1Q7,InMem 3),(n1Q9,InReg (RealRegSingle 2))] | freeRegs: FreeRegs 4282318848 | initFreeRegs: FreeRegs 4282318861 | Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug | make[1]: *** | [libraries/ghc-prim/dist-install/build/GHC/PrimopWrappers.o] Error 1 | libraries/ghc-prim/ghc.mk:4: recipe for target | 'libraries/ghc-prim/dist-install/build/GHC/PrimopWrappers.o' failed | | Have a look for example on linux-i386 buildot log here: | http://haskell.inf.elte.hu/builders/validator1-linux-x86-head/18/7.html | | Anyway, this happens on Linux, FreeBSD and Solaris buildbots on i386 so | it's OS independent and probably 32bit/i386 platform specific and it's | two days old breakage now. The last two night builds fail on all | mentioned buildbots. I'm not sure but perhaps: | | commit d8abf85f8ca176854e9d5d0b12371c4bc402aac3 | Author: Johan Tibell | Date: Mon Jun 9 11:43:21 2014 +0200 | | triggers that issue? I'm not claiming that the commit is actual culprit, | this may be just recently un-hidden issue in linear regs allocator on | i386! | | Thanks! | Karel

| _______________________________________________ | ghc-devs mailing list | ghc-devs@haskell.org | http://www.haskell.org/mailman/listinfo/ghc-devs

On Fri, Jun 27, 2014 at 1:14 PM, Simon Marlow wrote:

The problem is that this instruction requires three separate registers, but cmpxchgl already reads and writes %eax leaving only two free registers (%ecx and %edx).

You'll need to arrange to not use the complicated addressing modes with cmpxchg on i386, and keep the number of free regs required <= 2.

Where's the best place to arrange for that? If I switch from using getAmode to computing the address into a temp register in StgCmmPrim will that ensure that the address is simple, or could some optimization replace my temp register computation with a complex address again?

I fixed the x86 issue and re-commited my work as 4ee4ab01c1d97845aecb7707ad2f9a80933e7a49. On Fri, Jun 27, 2014 at 1:27 PM, Simon Marlow wrote:

On 27/06/2014 12:23, Johan Tibell wrote:

...

On Fri, Jun 27, 2014 at 1:14 PM, Simon Marlow wrote:

...

The problem is that this instruction requires three separate registers, but cmpxchgl already reads and writes %eax leaving only two free registers (%ecx and %edx).

You'll need to arrange to not use the complicated addressing modes with cmpxchg on i386, and keep the number of free regs required <= 2.

Where's the best place to arrange for that? If I switch from using getAmode to computing the address into a temp register in StgCmmPrim will that ensure that the address is simple, or could some optimization replace my temp register computation with a complex address again?

There aren't any optimisations that happen on the instructions after codegen, and if there were, they would have to respect the same rule. So using a register is the right thing, yes.

Cheers, Simon

After your reversion, I can build again. Thanks! From: Johan Tibell [mailto:johan.tibell@gmail.com] Sent: 26 June 2014 13:51 To: Simon Peyton Jones Cc: Karel Gardas; ghc-devs Subject: Re: Two days old build breakage on i386. I guess you don't have 04dd7cb3423f1940242fdfe2ea2e3b8abd68a177 (which I pushed this morning) which is fine. You should be in a good state now when d8abf85f8ca176854e9d5d0b12371c4bc402aac3 is reverted. On Thu, Jun 26, 2014 at 2:49 PM, Simon Peyton Jones mailto:simonpj@microsoft.com> wrote: git revert d8abf85f8ca176854e9d5d0b12371c4bc402aac3 [master f958079] Revert "Add more primops for atomic ops on byte arrays" 23 files changed, 86 insertions(+), 1016 deletions(-) rewrite compiler/nativeGen/CPrim.hs (62%) delete mode 100644 libraries/ghc-prim/cbits/atomic.c delete mode 100644 testsuite/tests/concurrent/should_run/AtomicPrimops.hs delete mode 100644 testsuite/tests/concurrent/should_run/AtomicPrimops.stdout HEAD $ git revert 04dd7cb3423f1940242fdfe2ea2e3b8abd68a177 fatal: bad object 04dd7cb3423f1940242fdfe2ea2e3b8abd68a177 HEAD $ What now? Simon From: Johan Tibell [mailto:johan.tibell@gmail.commailto:johan.tibell@gmail.com] Sent: 26 June 2014 13:25 To: Simon Peyton Jones Cc: Karel Gardas; ghc-devs Subject: Re: Two days old build breakage on i386. Just to make sure this is the same breakage, are you on an i386 Windows machine? If so git revert d8abf85f8ca176854e9d5d0b12371c4bc402aac3 and 04dd7cb3423f1940242fdfe2ea2e3b8abd68a177 to get unstuck. On Thu, Jun 26, 2014 at 2:13 PM, Simon Peyton Jones mailto:simonpj@microsoft.com> wrote: Aaaargh! Once again the Windows build is broken. I am utterly stalled. Moreover -fregs-graph and -fregs-iterative now *silently* do nothing. At least they should elicit warnings saying that they are disabled pending the fix to X and Y. Please can someone bisect to find out which patch is the culprit? I wish we had a more systematic way to find this out. I hate being the main person who gets stuck because some unrelated change has broken the Windows build. (Thanks for Karel, who got to it a day before me.) Thanks Simon | -----Original Message----- | From: ghc-devs [mailto:ghc-devs-bounces@haskell.orgmailto:ghc-devs-bounces@haskell.org] On Behalf Of Karel | Gardas | Sent: 26 June 2014 09:56 | To: ghc-devs; Johan Tibell | Subject: Two days old build breakage on i386. | | | Hello, | | builders running on i386 building for this platform caught issue which | shows as a build breakage: | | ghc-stage1: panic! (the 'impossible' happened) | (GHC version 7.9.20140624 for i386-unknown-linux): | RegAllocLinear.allocRegsAndSpill: no spill candidates | allocating vreg: VirtualRegI n1Q6 | assignment: [(c1PV,InMem 2),(n1Q5,InBoth (RealRegSingle 3) | 0),(n1Q6,InMem 1),(n1Q7,InMem 3),(n1Q9,InReg (RealRegSingle 2))] | freeRegs: FreeRegs 4282318848 | initFreeRegs: FreeRegs 4282318861 | Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug | make[1]: *** | [libraries/ghc-prim/dist-install/build/GHC/PrimopWrappers.o] Error 1 | libraries/ghc-prim/ghc.mk:4http://ghc.mk:4: recipe for target | 'libraries/ghc-prim/dist-install/build/GHC/PrimopWrappers.o' failed | | Have a look for example on linux-i386 buildot log here: | http://haskell.inf.elte.hu/builders/validator1-linux-x86-head/18/7.html | | Anyway, this happens on Linux, FreeBSD and Solaris buildbots on i386 so | it's OS independent and probably 32bit/i386 platform specific and it's | two days old breakage now. The last two night builds fail on all | mentioned buildbots. I'm not sure but perhaps: | | commit d8abf85f8ca176854e9d5d0b12371c4bc402aac3 | Author: Johan Tibell mailto:johan.tibell@gmail.com> | Date: Mon Jun 9 11:43:21 2014 +0200 | | triggers that issue? I'm not claiming that the commit is actual culprit, | this may be just recently un-hidden issue in linear regs allocator on | i386! | | Thanks! | Karel | _______________________________________________ | ghc-devs mailing list | ghc-devs@haskell.orgmailto:ghc-devs@haskell.org | http://www.haskell.org/mailman/listinfo/ghc-devs