January 2019 - ghc-tickets

[GHC] #15853: Unregisterised GHC panics when building test cgrun018
by GHC 04 Mar '19

04 Mar '19

#15853: Unregisterised GHC panics when building test cgrun018 -------------------------------------+------------------------------------- Reporter: iliastsi | Owner: (none) Type: bug | Status: new Priority: high | Milestone: Component: Compiler | Version: 8.4.3 (CodeGen) | Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Compile-time Unknown/Multiple | crash or panic Test Case: cgrun018 | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- An unregisterised ghc-8.4 on amd64 panics when building test cgrun018. {{{ Compile failed (exit code 1) errors were: [1 of 1] Compiling Main ( cgrun018.hs, cgrun018.o ) ghc-stage2: panic! (the 'impossible' happened) (GHC version 8.4.4 for x86_64-unknown-linux): pprStatics: float F32 F32 F32 F32 F32 F32 F32 F32 F32 F32 F32 Call stack: CallStack (from HasCallStack): callStackDoc, called at compiler/utils/Outputable.hs:1150:37 in ghc:Outputable pprPanic, called at compiler/cmm/PprC.hs:525:5 in ghc:PprC }}} We have seen this error while building packages for Debian on 64-bit architectures where GHC is unregisterised (e.g., mips64el, s390x): [https://buildd.debian.org/status/fetch.php?pkg=haskell- gloss&arch=mips64el&ver=1.13.0.1-1&stamp=1540810998&raw=0 gloss on mips64el], [https://buildd.debian.org/status/fetch.php?pkg=haskell- juicypixels&arch=s390x&ver=3.2.9.5-3&stamp=1540808360&raw=0 juicypixels on s390x] -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/15853> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 4

[GHC] #15960: Using -g causes differences in generated core.
by GHC 04 Mar '19

04 Mar '19

#15960: Using -g causes differences in generated core. -------------------------------------+------------------------------------- Reporter: AndreasK | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: 8.6.3 Component: Compiler | Version: 8.6.2 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- For nofib/shootout/fannkuch-redux: Normal results {{{ > /e/ghc-8.6.1/bin/ghc.exe Main.hs -O2 -XBangPatterns -fforce-recomp -ddump-simpl -dsuppress-all -dsuppress-uniques | less ... Result size of Tidy Core = {terms: 2,747, types: 2,736, coercions: 100, joins: 50/73} ... }}} {{{ > /e/ghc-8.6.1/bin/ghc.exe Main.hs -O2 -XBangPatterns -fforce-recomp -ddump-simpl -dsuppress-all -dsuppress-uniques -g | less ... Result size of Tidy Core = {terms: 2,654, types: 2,557, coercions: 100, joins: 38/71} ... }}} There are also structural differences in the generated code. I've hit the issue with a tree based on head as well so likely happening there as well. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/15960> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 8

[GHC] #9894: Presence of DWARF debug symbols doesn't result in working backtraces in GDB
by GHC 04 Mar '19

04 Mar '19

#9894: Presence of DWARF debug symbols doesn't result in working backtraces in GDB ----------------------------------+--------------------------------------- Reporter: tibbe | Owner: scpmw Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.9 Keywords: | Operating System: Linux Architecture: x86_64 (amd64) | Type of failure: None/Unknown Difficulty: Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Revisions: ----------------------------------+--------------------------------------- Now when GHC can output DWARF debug symbols we need to take the remaining steps needed to get downstream tool like GDB to do something useful with it. As of today GDB cannot produce a valid stack trace for this code: {{{#!hs module Main where import System.IO f name = print $ "Hello, " ++ name {-# NOINLINE f #-} g name = f name {-# NOINLINE g #-} main = do name <- getLine g name }}} when a breakpoint is set on the line that defines `f` before the program is run. The resulting backtrace looks like: {{{ (gdb) bt #0 Main_f_info () at /tmp/Test.hs:5 #1 0x0000000000481a00 in ?? () Backtrace stopped: frame did not save the PC }}} This bug tracks the remaining work that needs to be done before this will work. Peter, what is the remaining work here? -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9894> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 5

[GHC] #9910: ghc's dwarf (-g) output does not like Unicode (Char.intToDigit: not a digit 136)
by GHC 04 Mar '19

04 Mar '19

#9910: ghc's dwarf (-g) output does not like Unicode (Char.intToDigit: not a digit 136) -------------------------------------+------------------------------------- Reporter: slyfox | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.9 Keywords: | Operating System: Architecture: Unknown/Multiple | Unknown/Multiple Difficulty: Unknown | Type of failure: Blocked By: | None/Unknown Related Tickets: | Test Case: | Blocking: | Differential Revisions: -------------------------------------+------------------------------------- Easy to reproduce as: {{{ inplace/bin/ghc-stage2 -w -g -c testsuite/tests/parser/unicode/utf8_024.hs ghc-stage2: panic! (the 'impossible' happened) (GHC version 7.9.20141219 for x86_64-unknown-linux): Char.intToDigit: not a digit 136 Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug }}} {{{ petermw> Yes, Unicode doesn't work with DWARF yet }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9910> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 2

[GHC] #10236: DWARF unwind info is broken
by GHC 04 Mar '19

04 Mar '19

#10236: DWARF unwind info is broken -------------------------------------+------------------------------------- Reporter: | Owner: thoughtpolice | Status: new Type: bug | Milestone: 7.10.2 Priority: high | Version: 7.10.1 Component: Compiler | Operating System: Unknown/Multiple (Debugging) | Type of failure: Incorrect result Keywords: dwarf | at runtime Architecture: | Blocked By: Unknown/Multiple | Related Tickets: Test Case: | Blocking: | Differential Revisions: Phab:D792 | -------------------------------------+------------------------------------- As reported by `bitonic` and `petermw` on `#ghc` (April 2nd): {{{ 07:02 < bitonic> I'm trying to get a meaningful backtrace with DWARF, using <https://ghc.haskell.org/trac/ghc/wiki/DWARF> as a guide. however, all I get is `Backtrace stopped: previous frame identical to this frame (corrupt stack?)` 07:03 < bitonic> I've re-built GHC 7.10.1 using `GhcRtsHcOpts += -g` and `GhcLibHcOpts += -g`, even if I'm not sure it's even necessary 07:04 < bitonic> are there any additional steps I should take? or any way to make sure that the binary I'm generating is sane? }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10236> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 6

[GHC] #10915: Statistical profiling support in the RTS
by GHC 04 Mar '19

04 Mar '19

#10915: Statistical profiling support in the RTS -------------------------------------+------------------------------------- Reporter: bgamari | Owner: bgamari Type: feature | Status: new request | Priority: normal | Milestone: 8.0.1 Component: Compiler | Version: 7.10.2 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): Phab:D1215, | Phab:D1214 | -------------------------------------+------------------------------------- Now since GHC can produce useful debugging information (e.g. DWARF annotations and source notes) in compiled objects, it would great if we could leverage this for more efficient profiling. While ideally we would be able to rely on external tools like `perf` for this task, this would require that the STG stack be walkable like a standard C stack. A more feasible alternative is to incorporate a simple statistical profiler into the RTS. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10915> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 12

[GHC] #11261: Implement DWARF debugging on powerpc64
by GHC 04 Mar '19

04 Mar '19

#11261: Implement DWARF debugging on powerpc64 -------------------------------------+------------------------------------- Reporter: trommler | Owner: trommler Type: bug | Status: new Priority: normal | Milestone: 8.2.1 Component: Compiler | Version: 7.11 (CodeGen) | Keywords: | Operating System: Linux Architecture: powerpc64 | Type of failure: Incorrect result | at runtime Test Case: debug, T10667 | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- debug: {{{ ghc-stage2: panic! (the 'impossible' happened) (GHC version 7.11.20151219 for powerpc64-unknown-linux): dwarfReturnRegNo: Unsupported platform! CallStack (from ImplicitParams): error, called at compiler/nativeGen/Dwarf/Constants.hs:224:19 in ghc:Dwarf.Constants Please report this as a GHC bug: http://www.haskell.org/ghc/reportabug }}} Provide DWARF constants for registers. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11261> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 12

[GHC] #11337: Unwind information incorrect between Sp adjustment and end of block
by GHC 04 Mar '19

04 Mar '19

#11337: Unwind information incorrect between Sp adjustment and end of block -------------------------------------+------------------------------------- Reporter: bgamari | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.10.3 (CodeGen) | Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- As pointed out in Phab:D1532, the DWARF unwinding information that we produce is currently a bit oversimplified. Namely, we produce exactly one unwind table per Cmm block. This works reasonably well in most cases since we most Cmm blocks have the form, {{{ aProcedure() { casl: -- we just entered the procedure, so the unwinding is trivial. unwind Sp = Sp -- we push some values onto the stack... I64[Sp - 16] = ... I64[Sp - 8] = ... -- and before leaving the block we update Sp. Sp = Sp - 16; -- technically our unwind information is now a lie call aFunction() returns to casd; casd: -- we inherit the unwind information from the state of the stack when we -- left the preceding block (casl) unwind Sp = Sp + 16 R2 = I64[Sp + 8]; -- pop off that which we pushed Sp = Sp + 16; call GHC.List.$wunsafeTake_info(R3, R2) args: 8, res: 0, upd: 8; }}} Here there is a narrow window where our unwind information is technically wrong: between updating `Sp` in `casl` and calling into `aFunction`. Note that after we arrive in `aFunction` we are safe, since our return address is `casd`, which has the correct unwinding information. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11337> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 10

[GHC] #11338: Unwind information is incorrect in region surrounding a safe foreign call
by GHC 04 Mar '19

04 Mar '19

#11338: Unwind information is incorrect in region surrounding a safe foreign call -------------------------------------+------------------------------------- Reporter: bgamari | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.10.3 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: #11337 Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- Safe foreign calls can occur at the end of a block. Like #11337, these produce incorrect debug information. However, the situation is worse in the case of foreign calls since these return to the calling block. This means that we are currently unable to unwind out of a safe foreign call. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11338> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 7

[GHC] #11353: DWARF call frame information incorrect in the presence of unsafe foreign calls
by GHC 04 Mar '19

04 Mar '19

#11353: DWARF call frame information incorrect in the presence of unsafe foreign calls -------------------------------------+------------------------------------- Reporter: bgamari | Owner: Type: bug | Status: new Priority: normal | Milestone: 8.2.1 Component: Compiler | Version: 7.10.3 (CodeGen) | Keywords: | Operating System: Unknown/Multiple Architecture: x86_64 | Type of failure: Debugging (amd64) | information is incorrect Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- Unsafe foreign calls result in adjustments to `$rsp` made in the NCG to comply with calling convention alignment requirements (see `X86.CodeGen.genCCall64'`). Unfortunately, this happens after we've generated unwind information. This results in incorrect unwinding information for `$rsp` when inside of a foreign call. The results can be quite catastrophic (e.g. segmentation faults while unwinding). Unfortunately it's really not clear what can be done about this given that these adjustments aren't present in the Cmm representation that we use to produce frame information. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11353> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 13