[gcc r10-10146] ipa-fnsummary: Remove inconsistent bp_pack_value
Kewen Lin
linkw@gcc.gnu.org
Wed Sep 22 03:18:13 GMT 2021
https://gcc.gnu.org/g:406cca66bc867d60351a5a3976bf1ed81781c582
commit r10-10146-g406cca66bc867d60351a5a3976bf1ed81781c582
Author: Kewen Lin <linkw@linux.ibm.com>
Date: Tue Sep 21 19:13:57 2021 -0500
ipa-fnsummary: Remove inconsistent bp_pack_value
There is one inconsistent bit-field streaming out and in.
On the side of streaming in:
bp_pack_value (&bp, info->inlinable, 1);
bp_pack_value (&bp, false, 1);
bp_pack_value (&bp, info->fp_expressions, 1);
while on the side of the streaming out:
info->inlinable = bp_unpack_value (&bp, 1);
info->fp_expressions = bp_unpack_value (&bp, 1)
The removal of Cilk Plus support r8-4956 missed to remove
the streaming out of the bit, instead just change the value
for streaming out to be always false.
By hacking fp_expression_p to always return true, I can see
it reads the wrong fp_expressions value (false) out in wpa.
GCC12 adopts commit 63c6446f77b9001d26f973114450d790749f282b
which removes the inconsistent streaming out instead.
gcc/ChangeLog:
* ipa-fnsummary.c (inline_read_section): Unpack a dummy bit
to keep consistent with the side of streaming out.
(cherry picked from commit cc11a171eee059b645870f2e208c530f301239c9)
Diff:
---
gcc/ipa-fnsummary.c | 15 +++++++++++----
1 file changed, 11 insertions(+), 4 deletions(-)
diff --git a/gcc/ipa-fnsummary.c b/gcc/ipa-fnsummary.c
index 1105f6ae5c2..825db3c973d 100644
--- a/gcc/ipa-fnsummary.c
+++ b/gcc/ipa-fnsummary.c
@@ -4239,13 +4239,20 @@ inline_read_section (struct lto_file_decl_data *file_data, const char *data,
bp = streamer_read_bitpack (&ib);
if (info)
{
- info->inlinable = bp_unpack_value (&bp, 1);
- info->fp_expressions = bp_unpack_value (&bp, 1);
+ info->inlinable = bp_unpack_value (&bp, 1);
+ /* On the side of streaming out, there is still one bit
+ streamed out between inlinable and fp_expressions bits,
+ which was used for cilk+ before but now always false.
+ To remove the bit packing need to bump LTO minor version,
+ so unpack a dummy bit here to keep consistent instead. */
+ bp_unpack_value (&bp, 1);
+ info->fp_expressions = bp_unpack_value (&bp, 1);
}
else
{
- bp_unpack_value (&bp, 1);
- bp_unpack_value (&bp, 1);
+ bp_unpack_value (&bp, 1);
+ bp_unpack_value (&bp, 1);
+ bp_unpack_value (&bp, 1);
}
count2 = streamer_read_uhwi (&ib);
More information about the Gcc-cvs
mailing list