Re: [PATCH][gdb/symtab] Don't create duplicate psymtab for forward-imported CU

[Tom de Vries <tdevries@suse.de>]

On 08-04-2020 12:22, Tom de Vries wrote:
> On 08-04-2020 12:19, Tom de Vries wrote:
>> Hi,
>>
>> Consider the executable generated for test-case gdb.dwarf2/imported-unit.exp.
>>
>> When loading the executable using various tracing:
>> ...
>> $ gdb \
>>   outputs/gdb.dwarf2/imported-unit/imported-unit \
>>   -batch \
>>   -iex "set verbose on" \
>>   -iex "set debug symtab-create 1"
>>   ...
>> Created psymtab 0x213f380 for module <artificial>@0xc7.
>> Created psymtab 0x20e7b00 for module imported_unit.c.
>> Created psymtab 0x215da20 for module imported_unit.c.
>> Created psymtab 0x2133630 for module elf-init.c.
>> Created psymtab 0x215b910 for module ../sysdeps/x86_64/crtn.S.
>> ...
>> we notice that there are two psymtabs generated for imported_unit.c.
>>
>> This is due to the following: in dwarf2_build_psymtabs_hard we loop over CUs
>> and generate partial symtabs for those, and if we encounter an import of
>> another CU, we also generate a partial symtab for that one, unless already
>> created.
>>
>> This works well with backward import references:
>> - the imported CU is read
>> - then the importing CU is read
>> - the import is encountered, but the imported CU is already read, so
>>   we're done.
>>
>> But with forward import references, we have instead:
>> - the importing CU is read
>> - the import is encountered, and the imported CU is read
>> - the imported CU is read once more
>>
>> Fix this by skipping already created psymtabs in the loop in
>> dwarf2_build_psymtabs_hard.
>>
>> Build in combination with the tentative fix for PR25718 - "[ada] symbol in
>> imported unit not found"
>> ( https://sourceware.org/pipermail/gdb-patches/2020-March/167152.html ) to
>> prevent this regression:
>> ...
>> FAIL: gdb.ada/char_enum.exp: ptype pck.Global_Enum_Type
>> ...
>>
>> Tested on x86_64-linux, with native and target board
>> unix/-flto/-O0/-flto-partition=none/-ffat-lto-objects.
>>
>> This causes this regression with the target board:
>> ...
>> FAIL: gdb.ada/dgopt.exp: list x.adb:16, 16
>> ...
>> which I consider a seperate PR, filed as PR25801 - "Filename of shared psymtab
>> is ignored".
>>
> 
> [ Forgot to mention ]
> 
> Using a 2.8GB libxul.so debug file from opensuse tumbleweed, build with
> lto, we have:
> ...
> $ time.sh gdb -batch -iex "maint set dwarf max-cache-age 1000" -iex "set
> language c" usr/lib/debug/usr/lib64/firefox/libxul.so-74.0-1.1.x86_64.debug
> maxmem: 10417196
> real: 116.62
> user: 114.85
> system: 2.37
> ...
> and with this patch:
> ...
> $ time.sh gdb -batch -iex "maint set dwarf max-cache-age 1000" -iex "set
> language c" usr/lib/debug/usr/lib64/firefox/libxul.so-74.0-1.1.x86_64.debug
> maxmem: 10436372
> real: 77.42
> user: 75.54
> system: 2.46
> 
> OK for trunk?
> 

Committed (
https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=3d5afab3393064e563305bc27264fde5a7598635
).

Thanks,
- Tom