public inbox for gcc-cvs@sourceware.org help / color / mirror / Atom feed
From: Pierre-Marie de Rodat <pmderodat@gcc.gnu.org> To: gcc-cvs@gcc.gnu.org Subject: [gcc r13-595] [Ada] Fix problematic underflow for Float_Type'Value Date: Wed, 18 May 2022 08:43:47 +0000 (GMT) [thread overview] Message-ID: <20220518084347.BB2E73857815@sourceware.org> (raw) https://gcc.gnu.org/g:7c77ec1199c3a3d1ac48c9d963b8389c10a2a5bf commit r13-595-g7c77ec1199c3a3d1ac48c9d963b8389c10a2a5bf Author: Eric Botcazou <ebotcazou@adacore.com> Date: Mon Apr 4 23:16:53 2022 +0200 [Ada] Fix problematic underflow for Float_Type'Value We need a couple of guards for boundary conditions in the support code. gcc/ada/ * libgnat/s-dourea.adb ("/"): Add guard for zero and infinite divisor. * libgnat/s-valuer.adb (Scan_Raw_Real): Add guard for very large exponent values. Diff: --- gcc/ada/libgnat/s-dourea.adb | 12 ++++++++++++ gcc/ada/libgnat/s-valuer.adb | 9 ++++++++- 2 files changed, 20 insertions(+), 1 deletion(-) diff --git a/gcc/ada/libgnat/s-dourea.adb b/gcc/ada/libgnat/s-dourea.adb index a6cf2a1ca0b..4f378d6cb6d 100644 --- a/gcc/ada/libgnat/s-dourea.adb +++ b/gcc/ada/libgnat/s-dourea.adb @@ -178,6 +178,12 @@ package body System.Double_Real is P, R : Double_T; begin + if Is_Infinity (B) or else Is_Zero (B) then + return (A.Hi / B, 0.0); + end if; + pragma Annotate (CodePeer, Intentional, "test always false", + "code deals with infinity"); + Q1 := A.Hi / B; -- Compute R = A - B * Q1 @@ -196,6 +202,12 @@ package body System.Double_Real is R, S : Double_T; begin + if Is_Infinity (B.Hi) or else Is_Zero (B.Hi) then + return (A.Hi / B.Hi, 0.0); + end if; + pragma Annotate (CodePeer, Intentional, "test always false", + "code deals with infinity"); + Q1 := A.Hi / B.Hi; R := A - B * Q1; diff --git a/gcc/ada/libgnat/s-valuer.adb b/gcc/ada/libgnat/s-valuer.adb index 4b4e8873c3e..b474f8429e5 100644 --- a/gcc/ada/libgnat/s-valuer.adb +++ b/gcc/ada/libgnat/s-valuer.adb @@ -645,7 +645,14 @@ package body System.Value_R is Ptr.all := Index; Scan_Exponent (Str, Ptr, Max, Expon, Real => True); - Scale := Scale + Expon; + + -- Handle very large exponents like Scan_Exponent + + if Expon < Integer'First / 10 or else Expon > Integer'Last / 10 then + Scale := Expon; + else + Scale := Scale + Expon; + end if; -- Here is where we check for a bad based number
reply other threads:[~2022-05-18 8:43 UTC|newest] Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
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=20220518084347.BB2E73857815@sourceware.org \ --to=pmderodat@gcc.gnu.org \ --cc=gcc-cvs@gcc.gnu.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: linkBe 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 read-only IMAP folder(s) and NNTP newsgroup(s).