From: Luc Van Oostenryck <luc.vanoostenryck@gmail.com>
To: linux-sparse@vger.kernel.org
Cc: Christopher Li <sparse@chrisli.org>,
Linus Torvalds <torvalds@linux-foundation.org>,
Dibyendu Majumdar <mobile@majumdar.org.uk>,
Luc Van Oostenryck <luc.vanoostenryck@gmail.com>
Subject: [PATCH v3 4/8] change the masking when loading bitfields
Date: Wed, 9 Aug 2017 01:06:30 +0200 [thread overview]
Message-ID: <20170808230634.16227-5-luc.vanoostenryck@gmail.com> (raw)
In-Reply-To: <20170808230634.16227-1-luc.vanoostenryck@gmail.com>
To make some simplification easier, change the way the masking
is done when loading bitfields.
Signed-off-by: Luc Van Oostenryck <luc.vanoostenryck@gmail.com>
---
linearize.c | 15 +++++++++------
validation/bitfield-size.c | 6 ++++--
2 files changed, 13 insertions(+), 8 deletions(-)
diff --git a/linearize.c b/linearize.c
index ba76397ea..f833fd8d9 100644
--- a/linearize.c
+++ b/linearize.c
@@ -1000,14 +1000,17 @@ static pseudo_t linearize_load_gen(struct entrypoint *ep, struct access_data *ad
{
struct symbol *ctype = ad->result_type;
pseudo_t new = add_load(ep, ad);
+ unsigned int off = ctype->bit_offset;
+ unsigned int siz = ctype->bit_size;
- if (ctype->bit_offset) {
- pseudo_t shift = value_pseudo(ctype->bit_offset);
- pseudo_t newval = add_binary_op(ep, ad->source_type, OP_LSR, new, shift);
- new = newval;
+ if (siz != type_size(ad->source_type)) {
+ pseudo_t mask = value_pseudo(((1ULL << siz) - 1) << off);
+ new = add_binary_op(ep, ad->source_type, OP_AND, new, mask);
+ }
+ if (off) {
+ pseudo_t shift = value_pseudo(off);
+ new = add_binary_op(ep, ad->source_type, OP_LSR, new, shift);
}
- if (ctype->bit_size != type_size(ad->source_type))
- new = cast_pseudo(ep, new, ad->source_type, ad->result_type);
return new;
}
diff --git a/validation/bitfield-size.c b/validation/bitfield-size.c
index ce78ecf21..c8c94bb15 100644
--- a/validation/bitfield-size.c
+++ b/validation/bitfield-size.c
@@ -35,7 +35,9 @@ unsigned int get_pbfi_b(struct bfi *bf) { return bf->b; }
* check-command: test-linearize -Wno-decl $file
* check-output-ignore
*
- * check-output-pattern-24-times: cast\\.
- * check-output-pattern-12-times: cast\\.4
+ * check-output-excludes: cast\\.4
+ * check-output-pattern-6-times: cast\\.
* check-output-pattern-6-times: lsr\\..*\\$6
+ * check-output-pattern-6-times: and\\..*\\$15
+ * check-output-pattern-6-times: and\\..*\\$960
*/
--
2.13.2
next prev parent reply other threads:[~2017-08-08 23:06 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-08-08 23:06 [PATCH v3 0/8] fix loading of partially defined bitfield Luc Van Oostenryck
2017-08-08 23:06 ` [PATCH v3 1/8] Remove single-store shortcut Luc Van Oostenryck
2017-08-08 23:06 ` [PATCH v3 2/8] new helper: def_opcode() Luc Van Oostenryck
2017-08-08 23:06 ` [PATCH v3 3/8] reuse nbr_pseudo_users() Luc Van Oostenryck
2017-08-08 23:06 ` Luc Van Oostenryck [this message]
2017-08-08 23:06 ` [PATCH v3 5/8] simplify ((A & M') | B ) & M when M' & M == 0 Luc Van Oostenryck
2017-08-08 23:06 ` [PATCH v3 6/8] transform (A & M) >> S to (A >> S) & (M >> S) Luc Van Oostenryck
2017-08-08 23:06 ` [PATCH v3 7/8] transform (A << S) >> S into A & (-1 " Luc Van Oostenryck
2017-08-08 23:06 ` [PATCH v3 8/8] fix: cast of OP_AND only valid if it's an OP_CAST Luc Van Oostenryck
2017-08-09 5:52 ` [PATCH v3 0/8] fix loading of partially defined bitfield Christopher Li
2017-08-09 19:37 ` [PATCH v4 0/9] " Luc Van Oostenryck
2017-08-09 19:37 ` [PATCH v4 1/9] testsuite: add support for commands with timeout Luc Van Oostenryck
2017-08-09 19:37 ` [PATCH v4 2/9] Remove single-store shortcut Luc Van Oostenryck
2017-08-09 19:47 ` Dibyendu Majumdar
2017-08-09 20:02 ` Luc Van Oostenryck
2017-08-09 22:23 ` Dibyendu Majumdar
2017-08-09 23:15 ` Luc Van Oostenryck
2017-08-09 23:32 ` Dibyendu Majumdar
2017-08-09 23:51 ` Luc Van Oostenryck
2017-08-12 18:51 ` Dibyendu Majumdar
2017-08-12 19:28 ` Luc Van Oostenryck
2017-08-12 19:55 ` Dibyendu Majumdar
2017-08-12 20:14 ` Luc Van Oostenryck
2017-08-09 19:38 ` [PATCH v4 3/9] new helper: def_opcode() Luc Van Oostenryck
2017-08-09 19:38 ` [PATCH v4 4/9] reuse nbr_pseudo_users() Luc Van Oostenryck
2017-08-09 19:38 ` [PATCH v4 5/9] change the masking when loading bitfields Luc Van Oostenryck
2017-08-09 19:38 ` [PATCH v4 6/9] simplify ((A & M') | B ) & M when M' & M == 0 Luc Van Oostenryck
2017-08-09 19:38 ` [PATCH v4 7/9] transform (A & M) >> S to (A >> S) & (M >> S) Luc Van Oostenryck
2017-08-10 1:16 ` Christopher Li
2017-08-10 23:41 ` Luc Van Oostenryck
2017-08-09 19:38 ` [PATCH v4 8/9] transform (A << S) >> S into A & (-1 " Luc Van Oostenryck
2017-08-09 19:38 ` [PATCH v4 9/9] fix: cast of OP_AND only valid if it's an OP_CAST Luc Van Oostenryck
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20170808230634.16227-5-luc.vanoostenryck@gmail.com \
--to=luc.vanoostenryck@gmail.com \
--cc=linux-sparse@vger.kernel.org \
--cc=mobile@majumdar.org.uk \
--cc=sparse@chrisli.org \
--cc=torvalds@linux-foundation.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).